UBS ISM (HMVVnilVlH flffiH TREATISE ON THE STEAM ENGINE BY JAMES V RENWICK, LL.D. PKOFESSOR OF NATURAL AND EXPER1SIENTAL PHILOSOPHY AND CHEMISTRY iN COLUMBIA COLLEGE, NEW-YORK. REVISED AND ENLARGED. TO WHICH IS ADDED AN APPENDIX, BEING AN ANALYSIS OF A NEW THEORY OF THE STEAM ENGINE, BY The CH. G. DE PAMBOUR. GEORGE CARVILL, C. S. FRANCIS & CO., 252 Broadway, and CAREY & HART, Philadelphia. 1848. TA Entered according to the Act of Congress, in the year 1839, by CARVILL & CO. In the Clerk's Office of the District Court of the Southern District of New- York. A^ ■V u t V, ^i* K Mtw-YoKK: Van Norden Fig. 5 and 6. The best apparatus for ascertaining the level of water in a boiler, is a straight glass tube, open at both ends, which are plac- ad in two cups that communicate by means of tubes with the 9 66 BOILERS. interior of the boiler. To these cups the tube is cemented, and the water will stand in it at the same height that it does in the boiler. It was at one time supposed that this simple method could not be adapted either to high pressure boilers, or to those made of materials other than copper, but it has of late been suc- cessfully introduced in both cases. Another mode, which, however, has rarely been used with steam engines, is to adapt a tube, made like the pipe of an or- gan, to the boiler, the lower end of which descends as far as the level, below which the water ought not to be permitted to fall. So soon as the water decends the least space below this point, steam will escape through the pipe, and give warning of the necessity for a supply by the sound it causes. In high pressure boilers, this tube cannot be applied in its simple form, as it would become of inconvenient length. We have, however, seen more than one ingenious modifica- tion of this apparatus, in which the communication with the steam is effected by a valve that is intended to open when the water falls below its proper level, and to close after the supply has been introduced. If a substance of convenient form, denser than water be taken, and be made to communicate with a lever on the outside of the boiler, by means of a wire passing through a stuffing box, it may be made just to float at the surface of the water by a counterpoising weight suspended from the opposite end of the lever. When the water falls, the floating substance will pre- ponderate, and the lever will incline towards it ; when the water is too high, the inclination will be in the opposite direc- tion ; and these deviations of the lever from the horizontal position, may be marked by an index at right angles to it, and affixed to its centre of motion. In stationary engines, particularly those employed in manu- facturing establishments, this method of indicating the position of water in the boiler has been successfully employed ; but in steam-boats its use may not only lead to uncertain ty, but be actually productive of danger. In the motion to which boilers are often subjected in the latter case, the water is con- tinually shifting, and the whole boiler is sometimes filled with BOILERS. 67 a foam composed of an intimate mixture of water and steam. The same remark applies to the case of locomotive engines, and it is by no means certain that any floating apparatus will act when the water foams. 45. The boiler may be supplied as often as appears necessary from the indications of either of these apparatus, by different means, that must vary according to the elastic force of the ge- nerated steam. If the steam have a force but little greater than an atmosphere, a simple tube, having a top shaped like a fun- nel, will answer the purpose. This must be inserted into the boiler through a steam-tight joint, until it nearly reaches the bottom, and it must be high enough to contain a column of wa- ter equivalent to the excess" of the power of steam over the atmospheric pressure. The steam being of a constant elasticity, the water will stand in this tube at a constant height above the level of the water in the boiler, and if water be poured into it, an equal quantity must pass into the boiler. To adapt this to high pressure engines, a tube of inconvenient length would be necessary ; recourse must therefore be had to other means. The most simple of these is a spherical vessel, connected with the boiler from beneath by a tube and stop-cock, and closed at top by another stop-cock, surmounted by a funnel. The stop-cock is at first closed, and the sphere filled through the funnel and upper stop-cock ; the upper stop-cock is then shut : and, when it is known that the boiler needs a supply, the lower stop-coclc is opened, through which the water will now pass to the boiler. The alternate action of these stop-cocks, in the first instance, prevents the steam from escaping, and from interfering with the entrance of the water ; and, in the second, permits the water to enter the boiler, while it is replaced by the entrance of steam into the spherical vessel. See PL II, Fig. 2. This apparatus is no longer applied to its original pur- pose, but under the name of a globe-cock is employed to intro- duce grease into the parts of high pressure engines. The operation of such an apparatus may be facilitated by making two parallel passages through each of the cocks. One of these is in each made, when the stop-cock is open, to 68 BOILERS. adapt itself to a tube, which reaches from each stop- cock nearly to the opposite side of the spherical vessel. "When the upper stop-cock is opened, the water will enter through the tube, and the enclosed air or steam will escape at the other passage of the cock. When the lower stop-cock is open- ed, the water will rush through the naked opening, and steam rise to replace it through the tube. A forcing pump may also be used to supply high pressure boilers, the pressure on the piston of which must be greater than the elastic force of the steam. A valve, by which the water the pump propels, may be made either to enter the boiler, or run to waste, at pleasure, will then supply the water that may be needed. This is the mode which is now universally used in all boilers which generate steam of a tension of more than one and a half atmospheres. See PL IV.. Fig. 6. In all cases, however, it would be better, if possible, that the feediug apparatus should be self-acting : or, to speak more pro- perly, that it should be governed in its operation by the level at which the water stands in the boiler. For boilers which generate steam not exceeding H atmos- pheres, the construction of such an apparatus is attended with but little difficulty ; but as methods of applying it safely at higher tensions are now coming into almost general use. what we are about to say on this subject is fast becoming a matter of history, rather than one of practical utility. We shall therefore describe them for the purpose of exhibiting the inge- nuity by which this part of the low pressure boiler was made to conform to the action of the engine, and not as applicable in the present state of our knowledge of the best mode of using steam. The most obvious and simple of all, and it is equally applica- ble to high and low pressure boilers, is a ball floating on the surface of the water, and attached by an arm to a stop-cock upon the supply pipe. This stop-cock is opened when the ball falls, and shuts when it rises. It may be attached in a low pressure boiler to a pipe, of a length sufficient to bear a column of water equivalent to the excess of the force of the BOILERS. 69 steam above an atmosphere ; but in a high pressure engine it must be adapted to the pipe proceeding from a forcing pump. This method is liable to the objection of being subject to get out of order, and of being out of sight and reach ; it may, therefore, fail at the moment it is least expected. The floating ball may be made to act, through the interven- tion of a lever and a rod, upon a conical valve in the feed pipe. A floating apparatus, similar to that which has already been mentioned for indicating the level of water in the boiler, may be made to regulate the supply of a low-pressure boiler. In the first place, the lever there described may have its centre of mo- tion in the axis of a stop-cock to which it is attached. In the second place, a conical valve may be attached, by a rod, to the arm of the lever that carries the weight. Around this rod is a small reservoir of water, communicating with the boiler be- neath by a pipe reaching nearly to the bottom. The conical valve is adapted to the junction of this pipe with the reservoir in such a manner as to open the communication when the float falls, and close it when the float rises. See PI. II., Fig. 8. For reasons which have already been mentioned, this method is not applicable to the boilers of steam-boat and locomotive en- gines. In most cases engines do not require to be kept in constant action. Those employed in manufactories are always stopped at night, and generally wmile the workmen are at their meals. As the action must be kept up until the end of the working hours, a great loss of heat is caused by the sudden cessation of the motion of the engine at a moment when the steam has its full tension. This loss has been in a great degree prevented by a modification of this feeding apparatus invented by Hall of Glasgow. In this modification the float is double, one part lying at the level at which the steam is to be maintained while the engine is at work ; the other within a few inches of the top of the boiler. The counterpoise is made up of two weights. These, when united, counterpoise the floats when the lower one is at the surface of the water in the boiler. On removing one of the weights, that which remains becomes a counterpoise to the double float when its upper part reaches the surface of the 70 BOILERS. water. The water will, therefore, flow from the cistern of the feeding apparatus until the lower re-assumes its horizontal po- sition. On the replacing this weight, the engine will work off the water which has been thus introduced before the feeding apparatus will begin to admit water. By this arrangement, water is introduced at tie temperature of condensation as soon as the engine is stopped, and the fire will be employed in bringing; up to its working temperature during the interval of work. In the application of it to the en- gine of the Glasgow Water Works, a saving of nearly twenty- live per cent in the fuel consumed was obtained. The same me- thod is. of course, applicable to ferries, and to passage boars which have occasion to stop at landings ; and there would be no difficulty in graduating the water thus admitted to the in- tervals in the working of the engine. In the great change which is taking place in the manner of working condensing engines by which high steam is advan- tageously employed, all these modes so ingeniously planned to regulate the supply of boilers, may be said to have become in a great degree obsolete. In particular, it may be questioned whether any apparatus governed by a float will be sure to work in a steam- boat, a locomotive, or even a fixed high pressure en- gine. The supply of high pressure boilers, as has already been sta- ted, is always effected by means of a forcing pump. Set PI. IT., Fig. 6.' This propels, by the action of the piston a b, a stream of wa- ter into a pipe furnished with two stop-cocks or valves, c and c/, that act alternately : by one of these, c, water is admitted into the boiler, by the other, d. it is allowed to run to waste. These valves are usually left to the care of the engineer, as an appa- ratus to render them self-acting is necessarily complex. We, however, give a drawing of one. the invention of a Air. Frank- lin, that has received the medal of the British Society of Arts. Set Pi DL, Fig. 1. All feeding apparatus should be sufficient to supply the boil- er with considerably more water than it usually evaporates. Generally speaking, it is made to furnish five or six times as BOILERS. / 1 much, as it is far better that water should run to waste, than that there should be at any time a want of a full supply. It will be obvious, that a self-acting feeding apparatus that will perform its duty when the engine is at rest, is still a de- sideratum for high pressure boilers. It is in the case of steam boats and locomotive engines that such an apparatus is almost indispensable, in order to place them wholly beyond the reach of danger, and to the want of it many fatal explosions are to be attributed. Many propositions have been recently made to sup- ply this desideratum, but none have come into general use. In the absence of a self-acting apparatus, force pumps, to be work- ed by hand, are applied to the boilers of steam boats and to locomo- tive engines. They also serve to fill the boiler in the first in- stance. 46. A regular supply of water is not only necessary to keep up the flow of steam, but is of great importance to the safety of a boiler ; we have therefore treated ot it next in order to the material. Whatever precautions may be taken in the choice and adjust- ment of the strength of the material, and in giving a regular supply of water, it is indispensable, before a boiler is made use of, that it should be proved. This is necessary, because the proof shows defects that would otherwise escape notice, particularly at the joints of the wrought metals, while in cast iron there are frequently cavities that are not seen upon the surface. These different defects might cause a boiler to burst with violence if it were to be subjected to the action of steam before proof had been performed in some other manner. This pre- liminary proof is best effected by means of the hydraulic press, or water pressure pump of Bramah, whose principle has been explained on page 9. This method is, however, still defec- tive, inasmuch as it must be performed, if not with cold water, with that which is far below the heat to which parts of the boil- er must be afterwards subjected. It has been proposed to apply a pressure five or six times as great as the boiler is intended to bear. Nor is this too great a precaution, for the water proof is performed when cold, and we 72 BOILERS. have seen that some metals are more tenacious when cold than when heated, and the proportion of six to one, at least, is neces- sary before this difference is obviated. If a boiler be not sub- jected to such a proof, it may be possible that when heated its limit of rupture may be reached before the safety valve opens. The water proof having been performed, the boiler should next be subjected to a similar trial by steam, say of four or five atmospheres more than is usually to be generated in the boiler without causing its safety valves to act. In France, it is re- quired by law, that all high pressure boilers be subjected to a proof five times as great as the boiler is intended to bear when in service. It is, however, to be considered that, when the boiler is to be used to generate high steam, such excessive strain would rather tend to increase the danger than to ensure safety; for if the strain to which the metal is exposed exceed the limit of its elasticity, is will be materially weakened, although it may not explode under the proof. 47. The next precaution to be taken is, that the boiler be furnished with safety valves. A safety valve is a conical or cylindrical stopper inserted into, or resting upon a seat of the same shape, and kept in its place by a weight equal to the most intense pressure that is intended to be exerted upon it by the vapour from within the boiler. When the steam acquires a force greater than this, the safety valve will open and permit the steam to escape ; at all inferior temperatures it will remain shut. Three things must therefore be investigated in order to their preparation, viz : the size of the opening to which they are to be adapted, the load they are to bear, and the proper mode of placing them. The openings must be large enough to permit all the vapour that can be formed to escape. This may be estimated at the conversion into steam of a cubic foot per hour from every eight or ten feet of fire and flue surface. But as the safety-valve will probably be most needed when the fire lias been augmented beyond its proper quantity, it will be well to prepare for the escape of, at least, four times that quantity, say a cubic foot of water evaporated for every two feet of fire surface ; this is the BOILERS. 73 maximum of steam that can be formed under any ordinary cir- cumstance. The vapour will escape with a velocity that will depend upon its elastic force, but which increases much less rapidly than that does. We subjoin the velocities under different ex- pansive forces. Table of the Velocity of steam at different temperatures. r Expansive force. i Velocity per second. n Atrr lospheres, - - - - 873 H do. . - - - 1145 n do. - - - - 1296 2 do. - - i. - 1405 3 do. - _ - - 1548 4 do. . . - - 1663 5 do. - . - . 1725 6 do. . . . . 1785 8 do. . . - . 1852 10 do. - . - . 1993 12 do. - - . . 2029 14 do. . . . . 2052 16 do. . . . . 2072 18 do. . • * . 2084 20 do. . . . . 2098 IL- ■ ■mi ■!■■■■ ml' _ j The quantity obtained, by multiplying these velocities by the area of the opening to which the valve is adapted, must be diminished by the constant fraction that represents the section of the vena contracta in fluids ; this, in such an orifice, is about fths or .75. To determine the quantity then, that will issue by a given safety valve, three-fourths of its area must be multiplied by the velocity under the anticipated expansive force, When the quantity to be discharged per second is given, the reverse of the operation will give the proper area of the safety valve. The bulk of steam generated by the evaporation of any given quantity of water, may be found by multiplying the bulk of the water by the number representing the volume of steam of that temperature, on page 24. It will not be attended with any inconvenience to make the 10 74 BOILERS. safety valves of high pressure boilers as large as those used for steam of less elasticity, and this is the method which is adopted in practice. The weight, with which the upper surface of a safety valve is to be loaded, should be equal to the pressure which the va- pour, at the maximum temperature for which the boiler is cal- culated, is capable of exerting upon the lower side of the safety valve. "When this expansive force of the steam, at the given temperature, is estimated in atmospheres, one atmosphere, or 151bs. per square inch, is to be deducted from the estimate, in- asmuch as the escape of the steam is opposed by the pressure of the atmosphere itself, which therefore acts as a part of the weight with which the safety valve is loaded. Therefore, to find the weight : The area of the safety valve in square inches must be ?nul- tiplied by 15 times the number of atmospheres to which the expansive force of steam at the given, temperature is equiva- lent, less one : the product is the weight in pounds. The weight in most cases acts upon a lever of the second kind, by the intervention of which the pressure is increased. As the foregoing rule gives the pressure that ought to act upon the safety valve, the weight that is suspended from the lever must be diminished, in the ratio by which the whole length of the lever exceeds the distance between the fulcrum of the lever and the safety valve. The number of atmospheres, to which the expansive force of steam, at different temperatures, is equal, may be found by the table upon page 23. There is a curious phenomenon which occurs when steam issues from a safety valve, or other orifice ; the temperature of the vapour just without the opening, is lower, the higher the tension of the steam is within the boiler. Thus steam issuing from a boiler, the water within which is at 212°, scalds the hand ; while if it had a tension of several atmospheres, the heat would be easily borne without injury. This phenomenon, which at first sight appears almost paradoxical, grows out of the rapid dilatation of dense steam, and the consequent increase of its capacity for specific heat. The greater the tension, and BOILERS. 75 consequent density of the steam, the greater will be the diminu- tion of temperature. This explanation would teach us that the size of safety valves, as calculated from the table on p. 24, is less than they ou^ht to be in practice, and no inconvenience can arise from making them of such size as will allow the escape of low steam ; for the weight will close the valve as soon as the tension of the steam is sufficiently diminished. Safety valves, generally speaking, are of the figure of a frus- tum of a solid cone, ground to fit a hollow frustum of an equal hollow cone. They may, as has been stated, be pressed down by a weight, acting directly, or through the intervention of a lever. In the former case the weight may either hang from the valve within the cylinder, or be fastened to its exterior surface. A valve of this description furnishes a constant pres- sure, and ought to be adapted to the highest temperature the boiler is intended to bear. If it act by means of a lever, the pressure of the weight, when at the extremity of the lever, ought to be equal in its action to the same maximum tension ; but, by making it act nearer to the fulcrum, its action may be made equivalent to the expansive force of steam of lower tempe- ratures. The latter is, therefore, best adapted to the case where the action of the boiler is left to the discretion of the fire- man ; while those where the weight acts directly, may be en- closed and kept beyond his reach. On Plate I. are to be seen several varieties of the safety valve. Pig. 1 0, is a conical valve, whose weight is suspended beneath it, and hangs within the boiler. Fig. 11, is one, also conical, whose weight lies above it, and without the boiler. Fig. 12, is another of the same shape, and bearing a weight upon it, which is enclosed in a cylinder in such a way that it may be shut up beyond the reach of the workmen. Fig. 13, is a cylindrical safety valve, working in a pipe and pressed down by a spring ; when the pressure of the steam overcomes the elasticity of the spring, the lateral openings in the pipe are un- covered in succession, and the space for the escape of steam in- creases with its tension. A safety valve, pressed down by a lever bearing a weight, is represented upon PL II. at Fig. 7, 76 BOILERS. The best mode of regulating the length of the arms of the lever to each other, is to make them in the ratio which the sur- face of the valve bears to the unit of superficial measure. Thus, if the surface of the valve be five square inches, the arms of the lever should be in the ratio of 5 : 1. The advantage of this method is, that the weight which is applied to the lever is the exact measure of the pressure on each square inch of the valve. In locomotive engines, a suspended weight is liable to an ob- cillating motion, which varies its pressure upon the valve, and may cause it to be continually opening and shutting. For this reason, instead of a weight, a spring has been substituted. This, being made upon the principle of the common spring weighing machine, may give any required pressure on the lever of the safety valve beneath a given limit. The tension of the spring is adjusted by means of a screw, which draws out the spring until its index marks upon a scale the weight for which the tension is intended to be a substitute. In all boilers there ought always to be two valves, one of which is left to the care of the fireman or engineer, the other fitted to the intended maximum pressure of the contained steam, and closed up. Very serious accidents have frequently occur- red from leaving safety valves wholly to the control of a work- man, or even of the captains of steam vessels, who may feel a temptation to increase the pressure of steam beyond what the boiler is capable of bearing. The proper situation of the safe- ty valve is upon the top of the boiler: and when there are two, one should be at each end : that left to the discretion of the workmen at the end next the fire, so as to be within their reach ; that which is beyond their control, at the farther extremitv. When the aperture by which the boiler is entered for the pur- pose of cleansing it, is situated on the top, the safety valve is often placed in its cover. In consequence oi a remarkable fact that has recently been observed, much doubt has existed as to the certainty of the action of a safety valve. When air is strongly compressed in a vessel or pipe, and issues thence by an orifice in a plane sur- face, if a plate or disk be presented to the orifice by one of its plane surfaces, so far from being driven away, it will be retain- BOILERS. 77 ed at a very small distance from the orifice. In this case the air escapes in the form of the surface of a very obtuse angled cone around the edges of the disk, leaving a conical vacuum beneath ; and in consequence of the well-known fact that there is a lateral communication of motion from a current of fluid to neighbouring portions of the same or other fluids, the air above the plate moves towards it in order to join the stream ; the vacuum beneath, and current from above, united, retain the disk at a constant, but small distance, from the plane in which the orifice is pierced. It has been found that the escape of steam is attended with similar phenomena. Still, however, in conical safety valves, if thin, there is no reason to apprehend any danger from this source. In those of the usual form, the increased resistance growing out of this cause, will not exceed one-twentieth of an atmosphere, or three-fourths of a pound upon each square inch. But were the valve to have the form of a frustum of a cone, whose height is great in proportion to the aperture, the resistance might become enormous ; and, in the case of some experiments that were made in reference to this subject, it amounted to upwards of thirty atmospheres. Here, then, we have a strong reason in confirmation of the pro- priety of the usual practice of making the safety valves of high pressure boilers of the size of those of condensing engines. However carefully a safety valve may have been construct- ed, it may, notwithstanding, cease to act, in consequence of rust, which will fix it to its seat. This is much more likely to hap- pen when it has been long shut, but may occur even to valves in frequent action. Still, however, to open the valve from time to time is the best preventative ; for a safety valve that has remained closed for a week, no longer deserves the name. Large boilers of but little strength, as employed for genera- ting low steam, are sometimes exposed to a danger of an oppo- site nature. When the fire is extinguished, the steam within will be condensed and a partial vacuum formed, the external at- mosphere will now act, and may cause the boiler to collapse. It has been proposed to remedy this defect by an air valve. A conical valve opening inwards, and kept in its place either by a counterpoise or a weak spring ; if either of these be little more 78 BOILERS. powerful than the weight of the valve itself, they will keep it in its seat so long as the tension of the steam within exceeds that of the atmosphere ; but when the latter becomes the most powerful, the valve will open and admit air. We have stated, that when the space intended to be occupied by water becomes narrow, it may be wholly displaced by steam. In such case danger will arise from the burning of the metal, and from the heating of the steam after it is generated. To meet this danger, an ingenious person of the name of Doug- las has proposed to place in the bottom of boilers a valve opening inwards ; and he asserts that he has seen it in action even when the boiler contained steam of 4 or 5 atmospheres. We do not pretend to vouch for the fact, which appears contrary to mechanical laws ; but that it should be true, is not more extra- ordinary than many of the cases of violent explosion. 49. Lest the safety valve or valves should by some accident become fixed to their seat, it is important that there should be some means of determining, at any moment, the elastic force of the steam within the boiler. Apparatus for this purpose are called Steam Guages. The simplest form of these is a bent tube, the two branches of which are parallel ; one of these branches is open to the air, the other is bent in such a manner as to be adapted to an opening in a part of the boiler above the level of the water it contains, or to the steam pipe, and is soldered or sealed to it in such a manner that steam cannot escape by the joint ; this, tube, if empty, would permit steam to escape through it, but mercury is poured in to fill the bend, and rise some inches in each branch of the tube. When the expansive force of the contained steam is just equal to an atmosphere, the mercury will stand at equal heights in each branch of the tube ; when the pressure increases, the level of the mercury, in the branch nearest the boiler, will be depressed, and that in the open branch raised ; the sum of the two equal changes of level will be the measure, in inches of mercury, of the expansive force. As the changes of level in the two branches are equal, it is sufficient to measure that of the outer tube alone ; this is done by a scale on the side, if the tube be of glass : but if of iron, BOILERS. 79 by placing in it an iron rod, which floats upon the mercury ; and which just reaches the top of the tube when the two columns of that fluid are of equal height. The tube or rod might be graduated by division into half inches, each of which would correspond to a difference in the two levels of a whole inch, the thirtieth part of an atmosphere, or half a pound upon each square inch of surface, over and above one atmosphere. Were the guage to be graduated to inches, each inch would correspond to one pound of internal pressure against the weight with which the safety valve is loaded ; and this is the mode of graduation most usually employed in this country. A gua^eof this form is represented on PL I., Fig. 7. By add- ing to the length of both branches of the tube, making it of a strong material, and increasing the quantity of mercury, this guage may be fitted for steam of any elastic force, but it might in that case become inconvenient to observe its indication by means of a graduated rod. In such cases a float of iron may rest on the surface of the mercury in ihe open branch, and be attached, by a cord passing over a pulley, to a counterpoise ; the ascent and descent of which, along a graduated scale, would mark the difference of level upon the same principle as the rod. A straight tube, inserted nearly to the bottom of a close cistern, which communicates at top with the steam of the boiler, and which contains a mass of mercury, will also answer this pur- pose. If the surface of the cistern be large in proportion to to the area of the tube, the change of level within it may be neglected, and the height of the mercury in the tube measured in inches. See Plate I., Fig. 8. The steam guage may be made to act as an additional safety valve. In this case its height must not exceed that which will measure, in a column of mercury, the maximum pressure the boiler is intended to bear ; and the top must be widened into the form of a funnel, sufficiently large to contain all the mercury with which the tube is supplied. Such an apparatus, with the arrangement of float and counterpoise sliding in a scale, is re- presented PI. I.j Fig. 8. A guage for a high pressure boiler may be made by immers- 80 BOILERS. ing the lower end of a glass tube in a basin of mercury ; the upper end of the tube is closed, and it contains atmospheric air. The basin of mercury is enclosed in a case communicating with the boiler. The steam, acting on the surface of the mercury, will force it up the tube and compress the air, and the space that it occupies, being according to the law stated on p. 14, inverse- ly as the pressure, will show the tension of the steam. Such a guage is represented at Fig. 8, on PL I. 50. Should the fire be more intense than is consistent with a regular supply of steam of the required temperature and pres- sure, an apparatus has been contrived to moderate its action, by the very increase in elastic force which it communicates to the steam. This is called the self-acting or self- regulating damper. Hitherto they have only been applied to boilers con- taining steam of so small an elasticity as to be capable of being supplied with water by an open feed-pipe, as described § 45. The water stands in this pipe at a height above that in the boiler, which depends on the difference between the elastic force of the steam and the pressure of the atmosphere. A plate of iron, sliding in a vertical groove at the throat of the chim- ney, is attached to a float resting on the surface of the water in the feed-pipe by a cord passing over puliies ; when the float rises by an increased action of the steam, raising the water in the feed-pipe, the damper will descend, and when the float descends, the damper will be raised. It will, in the former case, lessen, and in the latter, increase the aperture of the chimney, and the draught will vary with the size of the aperture, as has already been stated. Such an apparatus is represented as attached to the boiler, Fig. 1, PL I. n is the float, o the pully. The use of a self-acting damper is liable to the same difficulties as that of a self-acting feeding apparatus. It is hence scarcely or ne- ver used, except the steam is of very moderate tension. 51. In addition to a self-regulating damper, there should be another, to be worked by hand as occasion may require ; and in order to place the fuel completely under the control of the fireman, the passage by which air is admitted to the ashpit ought also to be capable of being opened and shut at pleasure. BOILERS. 81 Doors and valves for this purpose should therefore be provided, and the apparatus is called a Register. Such dampers are placed in the chimneys of almost all our steam-boats, and tem- porary doors to the ashpit are made of plates of sheet iron. 52. There are dangers, however, to which boilers are expos- ed, against which safety valves and self-acting dampers present no security, and of which steam guages give no notice. As a general rule, it is no doubt true that the temperature and ten- sion of vapour bear a constant relation to each other ; but it may so happen that steam, after being generated, is raised to a high temperature without exerting a proportionate expansive force. Thus, it a portion of a boiler should acquire a heat great- er than the water contained in the other parts, as it may do when not covered with water, the steam will receive an excess of heat without acquiring a proportionate elasticity. In ex- periments made by Mr. Perkins, steam was heated to a tem- perature at which, if of a corresponding density, it ought to have exerted a force of 56000 lbs. per square inch, but which did not exert a pressure of more than 150 lbs. The reason is ob- vious, for it was enclosed in a separate vessel, and its quantity remaining constant, it did not increase* in density. Had, how- ever, a small additional quantity of water, heated under pres- sure to a high temperature, been injected, it might be inferred, that the steam would have acquired the density necessary to enable it to exert the force corresponding to its temperature. Perkins also established the truth of this inference by actual experiment. Water was heated in one of his generators, the safety valve of which was loaded with a weight of 60 atmos- pheres, to a temperature of upwards of 900° ; a receiver was prepared, void of both air and steam, and heated to upwards of 1800° ; a small quantity of water was then made to pass from the generator to the receiver ; this was instantly converted into steam, whose heat was sufficient to inflame the hemp that coated the tube, at a distance of 10 feet from the generator ; its temperature was therefore estimated at not less than 1400°. In spite of this high temperature at which the steam was formed, its pressure did not exceed five atmospheres. But by injecting 11 82 BOILERS. more water, although the temperature was lessened, the elastic force was gradually increased to 100 atmospheres. In phe- nomena of this description we may find the cause of many explosions that cannot be explained on any other principle. If we suppose that the supply of water is impeded, neglect- ed, or checked altogether, the level of that in the boiler must descend, and parts exposed to the action of the fire may become dry ; such parts may then be heated far beyond the tempera- ture of the water beneath : and the vapour may be rendered by them sufficiently hot to make other parts of the boiler lumi- nous. If, by any cause, the water from beneath be brought in- to contact with the vapour and heated surfaces of the boiler, it will be instantly converted into steam of great expansive force, and in quantities for which the usual safety valves are not suf- ficient to provide an escape. An explosion must therefore ensue. The water may be brought into contact with these heated parts of the boiler, or with the hot vapour, by the very means that in other cases would be applied to diminish the danger. Thus, if the safety or throttle valve should be opened, the wa- ter, which was before boiling quietly, will suddenly rise with violent ebullition, or if the feeding apparatus begin again to act, the level of the water will be raised. In both cases, a con- tact will take place with the red hot surfaces, and with the in- tensely heated steam. There are also other cases in which the space usually occupied by water, and even the whole boiler, becomes filled with a foam- ing mixture of steam and water. The circumstances are similar to those of a pot boiling over. In such cases the metal of the boil- er and flues may be heated to incandescence, for such a mixture is a bad conductor of heat. Here again the injection of water into the boiler, or the opening of the valves, may be attended with danger. Water also, as has been stated (§16,) if heated to 680°, tends to assume the gaseous form, and then exerts a pressure which no vessel, constructed as boilers usually are, is capable of resisting. It is also within the limit of possibility that in iron boilers an explosive mixture may be generated. The metal, when red hot, BOILERS. 83 will decompose the steam, and hydrogen will be liberated. If by any means oxygen can be introduced, the same heat will cause it to explode ; but oxygen cannot enter until the tension within the boilers become less than an atmosphere. These have been admitted to be the only causes of the explo- sion of boilers, whether of low or high pressure. When boilers give way under the force of steam alone, dangerous consequen- . ces appear to have rarely happened. We have ourselves been twice in steam-boats, working with steam of not less than an at- mosphere and a half, when the boiler has given way ; and in neither case was the accident known to the passengers, except by stopping of the machinery. The wrought iron boiler of a high pressure engine, working with steam of the tension of six atmospheres, gave way some years since in a manufacturing establishment in the city of New- York, and the only bad effect was the extinction of the fire by the efflux of water. At Paris, the lower part of a boiler of cast iron, working with steam of the same force, gave way, and no other bad consequen- ces followed. In nearly all the cases where fatal accidents have occurred, the explosion appears to have been due to other causes than the mere expansive force of the steam that would be formed when the boiler is in proper order and supplied with water. There are, however, a few instances where the escape of large quantities of steam into close cabins, or its partial decomposition ill pass- ing through the fire, have produced, suffocation. Neither can it be doubted that a boiler of equal or nearly equal strength throughout may give way with explosion under the action of steam gradually and steadily increasing in temperature. In the fatal accidents of the Chief Justice Marshall and Helen McGregor, the explosions took place after delay at stopping places, and followed almost instantly the opening of the throttle valve to set the engine again in motion. In the former, where the main internal flue gave way, the safety valve was either open, or had just been closed ; one of the persons on board re- marked a peculiar shrillness in the sound of the escaping steam, that can only be ascribed to its being intensely heated, without S4 BOILERS. having a corresponding density ; another observed that it had a violet hue. which may perhaps be explained by supposing it to have been heated until it would have been luminous bv night. In opposition to the opinion that the water had fallen too low and left the flues bare, it was stated by the captain that the guage cocks had been tried, but on examining the boiler it was found that they were situated on the side of the boiler nearest the landing : and it is well-known that on such occasions the influx of passengers to that side is often so great as to change the level of the boat so much as to render the guage cocks, when so situated useless. It is also possible that the fireman, who was by no means skil- ful, may have mistaken water of condensation in the tube for that coming from the boiler. This last mistake is one that ought to be carefully guarded against, by leaving the cock open several seconds. Of the intense heat that steam sometimes attains, even with- out causing explosion, the following instance may be cited : the packing of the piston of a steam-boat, working with steam of a tension no greater than an atmosphere and a half, burst into flame on opening the cylinder at least half an hour after the fire had been extinguished. Here it is evident that any mixture of heated water with this steam might have caused ex- v : :ion. Even the injection of water into an empty space, whose tem- perature is not below 650 : , may cause explosion, by its being wholly and suddenly converted into steam of great density and expansive force. The committee of the Franklin Institute have, in their report of a series of interesting and valuable experiments made by them, thrown some doubt upon the explanation given by Per- kins of the cause of the explosion of boilers. We do not, how- ever, consider their experiments as absolute proofs of the in- accuracy of his positions. TVe may remark, that in their expe- riments the density of the steam was never increased even to an approximation to what would have been consistent with the saturation of the space at its final temperature. Thus, in one of their experiments, the tension of the steam obtained by the BOILERS. 85 injection of water upon the heated sides of a vessel was 12 at- mospheres, while the temperature was that of steam which, if saturated, would have exerted a force of 27^ atmospheres. Cold water was also used and injected, while in practice the water, which would be most likely to be mixed with steam heated af- ter it had been generated, would be that rising in foam from the lower part of the same boiler. If the committee of the Franklin Institute have left us in doubt as to the accuracy of Perkins' explanation, they have raised none as to the certainty of danger when portions of the metal of boilers become intensely heated. In addition to the causes assigned by him, they have shown that the tenacity of copper decreases as it is heated, even from low temperatures ; and that, although that of iron increases with the temperature up to a limit which is above that at which steam is usually employed, it decreases rapidly at temperatures above that limit, and at a red heat is no more than one-sixth of what it is when cold. Boilers, when the fire is made within, or when the return flues pass through them, are obviously far more subject to acci- dents arising from this cause than those heated from without ; low pressure boilers are as liable to them as high ; and it is con- fidently believed that very many explosions are to be attributed to this cause, against which the usual safety apparatus furnish- es no protection. To pay the greatest attention to keeping the feeding apparatus in order, and to have the means of ascertain- ing at every moment the height of the water in the boiler, are the surest means of defence; but as the first of these may fail, and does not act in many boilers after the engine is stopped ; and as the second depends upon the faithfulness of the engineer, or may also be clogged, and cease to give true indications ; other means have been proposed. 53. The first of these is a thermometer, inserted through a collar into the part of the boiler occupied by the steam, and which will therefore indicate its temperature. It must be made to mark the higher temperatures only, and may be graduated by a standard instrument, in a bath of hot oil. This is, how- 86 BOILERS. ever, but a fragile instrument, and may also be neglected by the workmen. 54. Another method, which promises to be effectual in many cases, is to form a part of the boiler of a plate of metal fusible at a comparatively low temperature. Such is an alloy of bis- muth, lead, and tin, by varying the proportions of which a con- siderable difference in fusibility may be attained. They ought to be of such a mixture as not to melt until heated beyond the temperature assumed as the limit of the heat to which it is ever desired to raise the steam, but fusible at one considerably be- low that at which the boiler becomes red hot. From 20° to 40° above the maximum heat the steam is meant to attain, will be well suited to the purpose ; for they will then melt before any part of the boiler can become red hot. These plates must be adapted to the upper part of the boiler, and be of course in contact with the steam ; they are inserted at the end of tubes fitted steam-tight to the boiler. As they are apt to soften long before they melt, they ought to be covered ,by a diaphragm of wire gauze. When thus protected, they have been found not to give way until they actually melt. As different parts of the boiler may acquire different temperatures, two such plates will be needed upon its outer surface, at the two ends ; they ought to be as near to the body of the boiler as possible. When flues pass through the boiler, we conceive that it would be a proper precaution to furnish them also with plates of this description, but in this case the metal might be less fusible, and lead unal- loyed would suffice. It has been objected to plates of fusible metal, that when they give way, the engine becomes useless ; and that in steam-boats in particular, danger may arise from the proposed means of safety. But this objection has been fully obviated by an invention of Professor Bache of the University of Pennsylvania. In this, the tube in which the fusible plate is inserted, is prolonged above it far enough to allow the ap- plication of a safety valve, which may therefore be adapted, and close the opening, until a new fusible plate can be pro- cured. It has, however, been found easy to protect the fusible BOILERS. 87 plates from the heat within the boiler, and they are thus ren- dered nugatory. When fusible plates are not used, and when from a ther- mometer, or from other appearances, there is reason to appre- hend that the water has fallen too low in the boiler, and that the temperature of parts of it have been raised to a dangerous degree of heat, the only means of safety are, to check the draught of the chimney by the damper, to lessen or extinguish the fire as soon as possible, or even to procure rapid cooling by pouring water on the outer surface of the boiler. The safety valve may then be opened ; and, after the cooling is com- plete, the boiler filled up by a hand force pump. A damper kept open by the action of the engine, and closing the instant it stops, would have a good effect, and might be easily adapted to a centrifugal apparatus. 55. It has been proposed, as a mode of securing safety in cases of great increases of temperature in the upper part of the boiler, to provide safety valves that would open at the limit of temperature beyond which danger might ensue. A safety valve of the usual form, but loaded with a great weight, and placed upon a tube containing a cylinder of metal, will answer this object : let the metallic cylinder be supported from be- neath, and of such a length that the dilatation by heat shall bring it in contact with the safety valve at the required tem- perature ; any further increase of temperature will open the safety valve, and permit the escape of steam ; its action is cer- tain, for the expansive force of the metals when heated, is capa- ble of overcoming the most powerful resistances : but it is rather to be used as an indicator of the necessity of moderating the fire, and putting the feeding apparatus in order, than as af- fording perfect security from explosion. It is difficult to point out methods that are of themselves en- tirely to be relied upon to prevent explosions. However per- fectly a boiler may be constructed or furnished with safety ap- paratus, it will still depend much upon the carefulness and in- telligence of the persons entrusted with its management. One thing, however, appears certain, although contrary to general 88 BOILERS. belief, that as the most usual causes of explosion affect low pres- sure boilers equally with those which generate high steam, the latter are not more subject to accident than the former. There are precautions, however, which, if resorted to, may diminish the risk of such accidents in a very great degree ; so far, indeed, that without the greatest carelessness, they cannot occur. These may, perhaps, be recapitulated to advantage. 1. Cylindrical boilers, without any return flue, either without or within, are safer than any others. 2. Internal flues should be avoided wherever it is possible, and especially the chimney, or vertical flue, should never be permit- ted to pass through the boiler. But if internal flues must be used, and they cannot be avoided in steam boats and locomotive engines, the plan of diminishing them to mere tubes is the best, and care must be taken that the spaces between them are not too small. 3. Every boiler should be furnished, in addition to the usual safety valve, with one not under the control of the fireman. 4. All boilers should be furnished with griaofe cocks, or other apparatus, to show the level of the water, and these should be so placed in steam boats, that no error in their indication can take place when the vessel heels or rolls. 5. Plates of fusible metal should be provided, of a composi- tion melting so far above the usual temperature of the water and vapour, that they will not open on any ordinary occasion, but will give way before they attain a temperature that can be dangerous ; and these should have the addition proposed by Pro- fessor Bache. 6. A thermometer may be introduced into the boiler, whose in- dications may be seen from without. 7. Self-acting feeding apparatus should be adapted to the boiler, by which water will enter, and keep the fluid within at a constant level ; and this should depend upon the waste of wa- ter, and not on the action of the engine. It unluckily happens that no such apparatus has yet been introduced into use which is adapted to high pressure engines, nor indeed for any where the tension of the steam exceeds 1 f atmospheres. Neither are they always applied even to low pressure engines ; and in those BOILERS. 89 intended for steam boats, they would be worse than useless, from the uncertainty of their action.' 8. The chimney should be provided with a damper by which the draught of the flues may be suddenly checked, and doors should, if possible, be placed upon the ashpit. A damper that would close as soon as the engine ceased to move, would be of great service in lessening the liability to explosion, and this does not appear to be difficult of attainment. 9. The proof of the boiler should be conducted with the greatest care, first with water, at a pressure five or six times as great as the boiler is intended to carry, and afterwards with steam of more than the highest possible tension. The water proof should be repeated from time to time, and every part care- fully examined to ascertain that all the safety apparatus is in working order. In high pressure boilers, the force pump with which they are fitted is well adapted for giving the water- proof. Few or none of these precautions are usual in our American steam boats : the boilers, even if cylinders, have both internal flues and furnaces, and the vertical chimney frequently rises in the boiler ; there is never more than one safety valve ; plates of fusible metal are unknown ; the feeding apparatus is merely a forcing pump, which is turned on or thrown off at the pleasure of the engineer, and which does not act at all at the time the engine is not in motion ; but a very few steam boats have dampers upon their flues ; in fine, the proof is wholly a matter between the maker and proprietor, and for its proper performance the public have no guarantee. Thus, of all the precautions that have been proposed in order to insure indemnity from ex- plosion, but two are in use among our steam boats ; namely, the safety valve and the guage cocks ; the former being still sub- ject to the caprice of the persons employed, and the latter having an uncertainty in their indications, both when the boat inclines to either side, and when they contain, as they most frequently will do, water of condensation. They are also of no value when the water in the boiler foams. The means which are used are noi certain to insure safety, even where the care of the officers of the vessel, and of the per- 12 90 BOILERS. sons employed about the engine, is unremitting": and directed by the utmost intelligence ; hence dangerous accidents occur with- out giving rise to blame, and thus diminish a proper feeling of res- ponsibility. On the other hand, were the list of precautions that we have given, to be completed by a self-acting feeding appara- tus, independent of the action of the engine, for a high pressure boiler, we conceive that no accident could possibly happen where they are employed, except through carelessness, inatten- tion, or fool-hardiness. Should it appear that the feeding apparatus does not act to supply as much water as is evaporated, the damper should be closed, and the boiler might even be cooled by the gentle applica- tion of water from without ; but it will always be a sure source of danger to inject water in abundance, or even to open the safety valve suddenly, after the water has once fallen below its proper level, and before it is ascertained that neither the tem- perature of the steam within, or of the sides of the boiler, are such as to cause a sudden conversion of the water that comes into contact with them into steam. 55. In connexion with this subject, we have to mention and condemn an addition which is often made to the boilers em- ployed in our American steam boats. In consequence of the space which can be allowed for the boiler being: limited, it has been found that the flame often passes into the chimney, and even issues from its upper opening. As much heat would be thus lost, it has been attempted to apply it to the steam, in its passage from the boiler to the engine, by enclosing the chimney in a cylindrical case called a steam-chimney, through which the steam must pass. This method is. however, the least advantageous mode of applying heat, for the steam, heated out of contact with water, is not rendered more elastic than air would be under similar circumstances, and has its energy far less increased than if the same heat was applied to the water in the boiler. On the other hand, dangers similar to those of which we have just spoken, occur ; and if the risk of the heated steam being mixed with water is less than if it were in the boiler, another source of accident is to be found in the rapid BOILERS. 91 oxidation of the iron of the chimney, thus heated in contact with steam. This part of the chimney will, therefore, require fre- quent repairs, and if they be omitted, may give way with vio- lent explosion. To this cause the explosion of the steam boat William Gibbons, in the spring of 1836, is to be ascribed. It may therefore be stated, that the necessity for the use of a steam chimney is a proof of bad calculation in the plan of the boiler, and that the heat which it is intended to save by this means, may be much better applied. 56. There is another species of danger which arises from the deposit of solid substances. Almost every kind of water that is used for boilers, contains more or less earthy and saline mat- ter. The constant evaporation is replaced by new supplies of the same impure water, and the soluble portion or mechanical im- purity is consequently accumulating. The soluble parts be- come greater in quantity than the contained water can hold in solution, and these are deposited, along with those that are merely suspended. Crusts thus form on the lower part of the boiler, and the surface covered by them, being no longer in con- tact with the water, may be heated red hot, and may be corrod- ed in consequence of the property that some of these salts have, of being decomposed by the metals at a red heat. The boiler will become weak in these places, and be liable to burst. It hence becomes necessary to cleanse the boilers frequently ; for this purpose, as well as for examining the interior, an opening is made in the boiler, large enough to admit a man. This has a cover, which, when the boiler is in use, is fastened down by screw bolts and nuts, and is packed in such a manner as to be steam-ti^ht. This opening is called the Man-hole. These deposits become frequent and copious when sea-water is used, and it has be found necessary, in consequence, to cleanse the boilers of steam boats that navigate salt water at least once a week. So soon as the boiler has received succes- sive supplies of salt water, amounting together to nine times its own capacity, a crust of a double sulphate of line and soda will begin to collect. A farther evaporation will cause the de- posit of salt, and finally of the chloride of magnesia. 92 BOILERS. When the. water is fresh, and the deposit principally consists of sulphate of lime, as is the case with hard pump waters, vege- table feculae will suspend the impurities. To furnish this, potatoes may be thrown into the boiler, in the proportion of half the weight of bituminous coal that is consumed per hour. This quantity of that root once added, iurnishes starch enough to keep the earthy matter suspended by the water for a long space of time, and it has not been found necessary to cleanse boilers, when this addition is used, oftener than once a month. It is not known whether the same method will be ef- fectual in preventing the saline deposits of sea water. The necessity of cleansing and scraping boilers in which sea water is used, may be in a great degree prevented by blowing off a part of the water from time to time. This method is, however, attended with a great waste of heat and consequent consumption of fuel. We shall have occasion to cite an improvement in the engine, by which it is ensured that the boiler shall be fed with distilled water; and by the adoption of this improvement, all the dangers and inconveniences of which we have spoken may be obviated. 57. It merely remains that we should speak of the pipes by which the steam is conveyed from the boiler to the engine. The size of these will depend on the quantity of steam the boiler is intended to furnish, the resistance the pipe itself oppo- ses to the passage of the steam, and the loss of heat. Steam issues into a space containing atmospheric air with the velocities given in the table on p. 73. The velocities with which it rushes into a vacuum are as follows, viz. BOILERS. 93 Table of the Velocities with which Steam flows into a Vacuum. — —— ———— — — Force of Steam. Velocity per second. 1 Atmosphere, - 1908 2 do. - 1977 3 do. .... 2006 4 do. .... 2022 5 do 2038 10 do. 2098 15 do. 2121 20 do. 2141 It will be seen from this table, that the velocity of effluent steam increases very slowly with increase of temperature. This grows out of the fact that the density of vapour increases under ordinary circumstances nearly as fast as its elastic force. Did both follow the same law, the velocity would not increase at all ; but the weight of steam expended by a given orifice, in- creases rapidly, for the density of hot steam is much greater, and the weight that passes out is in the compound ratio of the density and the velocity. The table on page 73 gives the velocity for high pressure engines, for they, as we shall see hereafter, are resisted by the pressure of the atmosphere. The table just given contains the velocities for condensing engines. Both of these, however, require a correction for the friction, and the loss of motion by cooling ; but for these it is impossible to give any general rule. A me- thod that has been found to succeed in practice is, to make the orifice, or nozzle, by which the pipe communicates with the en- gine, such as would be calculated from the velocities of the ta- bles, and to make the rest of the pipe larger. The greater the distance the steam has to pass, the larger should be the pipe. To prevent the loss of heat growing out of the increased sur- face, the metal might be kept bright, in which state it will be a bad radiator of heat ; but this method is not applicable in prac- tice. The steam pipes are, therefore, often covered with a bad conductor of heat. The principles for calculating the area of the orifice by which such pipes communicate with the engine, and the sur- 94 BOILERS. faces of safety valves, are therefore identical ; we shall, in order to render them more intelligible, reduce them to the form of a Rule. To find the area of the passage by which steam shall reach the engine from the boiler in which it is generated : Divide the quantity of water in cubic inches evaporated per hour, by 3600, {the number of seconds in an hour,) multi- ply the quotient by the volume of steam of the given tempera- ture, from the table on page 24 : This will give the num- ber of cubic inches of steam that must pass per second. Di- vide this by the velocity per second, taken in the cases of high steam pipes, from the table on page 73, and in the case of low steam pipes from the table on page 93. The quotient is three-fourths of the required area, v;hence the diameter of the circular section can be obtained in the usual manner. Cylindrical boilers, placed in a vertical position, and having internal fireplaces and flues, have been much used in the United States. In our first edition we noticed one planned by Col. Miller of Charleston, S. C. ; but of this we have seen but a sin- gle instance, and it has not come into use. The most familiar case of this sort is in the locomotive engines of the Baltimore and Ohio Rail -road. We deem it due to justice to state, that Mr. John Stevens, of Hoboken, communicated to us, some years since, a plan of a boiler on similar principles, that we do not doubt would have been equally efficacious. In the use of anthracite coal, blowing engines to excite the combustion have been found advantageous. Such engines have also been applied to other kinds of fuel. In locomotive engines, the steam, after it has completed its action, is permitted to escape into the chimney, and by forcing out the air in its ra- pid expansion, acts to cause a more intense draught, and is thus superior to the best blowing engines. 58. Besides boilers of the various kinds we have mentioned, Mr. Perkins proposed one upon very different principles, which, for the sake of distinction, is called a Generator. It is a strong vessel, completely filled with water, which is heated to a very high temperature, and prevented from being converted BOILERS. 95 into steam by the strength of the vessel and the pressure upon its safety valve. A small quantity of water is flashed in from a forcing pump, which causes the escape of an equal quantity, that is instantly converted into steam of high temperature and consequent elasticity. As it is our intention to confine our- selves to forms that have actually come into general use, and as the generator of Perkins is still a subject of experiment, it does not enter into our views to describe it more particularly. For the same reason that we do not dwell upon the genera- tor of Perkins, we shall pass, with a slight notice, the boiler in- vented by Bennett of Ithaca, N. Y. In this the fuel is placed in a tight chamber furnished with two valves. Through one of these air is forced in by a blowing engine. The other opens into the boiler. Air will, therefore, accumulate in the fire-place until, by the joint effect of heat and compression, it assumes a tension even greater than that of the steam. It will then make its way into the boiler, and join the steam in its action on the en- gine. In experiments made with this apparatus, it appeared to produce a greater effect with a given quantity of fuel than any other boiler. Difficulties of a practical character have hitherto prevented its being brought into use. On Plate I. will be seen various forms of boilers. Fig. 1 and 2, are a longitudinal section and front elevation of the low pressure boiler of Watt, together with its furnace. a a a a, body of the boiler. 6, furnace, with its grate, c, ashpit. d d d ) flues. e, man hole, for entering the boiler to cleanse it. /, steam pipe, g, steam guage of the form shewn on a larger scale at Fig. 7. h, safety valve of the form shewn at Fig. 12. i, float of the feeding apparatus. k k : lever of the feeding apparatus, Z, valve of the feeding ap- paratus. m, supply cistern fed by the hot water pump of the engine, below this is the tube that conveys the water to the boiler, and which contains the float n of the self-acting damper. o o, pullies of the self-acting damper p. 96 BOILERS. V q, feed pipe. r r* guage cocks. It has been already mentioned that this form is now but rarely used, and, with its apparatus, is rather a mat- ter of history than a model to be copied in the existing state of the steam engine. Fig. 3, is a transverse section of a cylindrical boiler, the same letters are employed to designate such of the parts as are repre- sented. When used for low steam, all the parts represented in the preceding figures may be applied with equal facility to it. The faint circle within shews how the return flues might be made to pass through this boiler. Fig. 4, is a cylindrical boiler, with internal furnace and flues. On PI. VII. may be seen an outside view of a low pressure boiler for a steam boat ; this has an interior furnace and flues. On PL VILI. is a plan and section of an English steam boat boiler. On PL VI. are end views of a pair of cylindrical boilers for a high pressure engine. Perhaps the most perfect form which has yet been given to boilers, is that now adopted almost universally in locomotive engines, and which we have more than once referred to. The body of the boiler is cylindric. One of its ends rises no higher than the level of the water in the boiler, and forms the back of a furnace, in which by the prolongation of the plates of the boiler, the fuel is surrounded by water, except towards the ashpit. The flame and heated air are conveyed through the boiler in a number of pipes. In fixed engines, they are so much more costly than those of the form of a simple cylinder, as to be inapplicable ; but wherever it is important to save both room and weight, they are to be preferred to all others. They are, therefore, not only in general use upon rail-roads, but have been much employed in American steam boats. An outside view of a locomotive boiler may be seen on Plate IX. Having thus explained the structure of the boiler, and of the various accessories with which it may be furnished in order to render its action more regular and safe, we shall next proceed to treat of the action of steam as a mover of machinery, and of the different forms of engine that are at present in use. CHAPTER IT. general view of the double-acting condensing* ENGINE. Of Prime Movers in general. — Principles of the action ofMd* chines. — Modes of applying Steam as a prime mover.— Application of Steam to the Double- Acting Condensing Engine. — Modes of removing Water of Condensation and Vapour. — Modes of changing the reciprocating Rectilineal ► Motion of the Piston Rod into a reciprocating circular mo- tion. — Method of changing the reciprocating circular mo- tion into a continuous one. — Mode of regulating the vary- k ing motion of the Engine^ and making it produce one with uniform velocity. — Other methods of obtaining a rotary mo- tion. — Effect of the joint action of two Engines. — Water used to produce condensation. — Water that has been em* ployed in condensation applied to feed the boiler .—Manner of ascertaining the state of the Vacuum formed by conden- sation. — Mode of regulating the supply of Steam. — Accu* mutation of Steam in the boiler ', and mode of preventing it.— Double-acting condensing Engine considered as self acting. —^■Packing and Cements. — -Estimate of the power of the double-acting Condensing Engine. — Estimate of the quan- tity of w at er evaporated for each unit of force. — Estimate of the supply of water for the boiler. 59. The agents which we employ for the production of mechanical effects through the intervention of machines, may be divided into three classes. 13 95 DOUBLE-ACTIXG 1. The muscular force of man and living: animals : 2. The force of gravity producing the descent of heavy bo- dies, whether solid or fluid ; 3. Heat, applied either to change the volume of bodies that do not change their mechanical state during: its action, or to convert bodies into elastic fluids, acting with a powenul ex- pansive force. To the second of these classes we reter the force of running water, descending in channels to find the lowest accessible le- vel ; to the third, the currents of the atmosphere or wind, the more powerful agency of inflamed gunpowder, and of liquids converted into steam. 60. Machines are instruments by which we change the di- rection or intensity of the moving force. They can all be re- duced to six simple iorms, called Mechanic powers, and f; again to two still more simple modifications. In their action there is but one principle involved, which is as follows : The product of the moving force, estimated in some nal unit, into the space through which the point to which this force is applied, is, in all cases, equal to the sum of the pro- ducts of all the resistances hito the spaces described by their respective points of application. This principle has two distinct cases ; in the first, the ma- chine is at rest, or in equilibrio, under the action of the power and the resistances. In this case the points of application must be supposed to move, and the space employed in the calcula- tion is that through which they would move without altering the conditions of equilibrium. The principle is, in this case,, called that of Virtual Velocities. In the second case the ma- chine moves with uniform velocity under the action of the op- posing forces, and is said to have attained a state of permanent working, or to be in dynamical equilibrium. A machine passes from the state of rest, in consequence of the conditions of equilibrium being violated, and of the moving power acquiring, in consequence, a preponderance over the re- sistances. It leaves the state of rest gradually, and therefore moves at first with accelerated velocity, the conditions of equi- CONDENSING ENGINE. 99 librium, so long as this acceleration is going on, no longer hold good ; and there is one case in which the acceleration might continue as long as the motion. This is when the moving force is capable of acting with equal intensity upon a body at rest and upon a body in motion. Of the three classes of forces we have mentioned, gravity is the only one that thus acts, and it is limited by the check which the motion meets, in conse- quence of the body acted upon reaching the solid mass of the earth, the resistance of which speedily brings it to rest. But, even in the case of this force, the bodies which are propelled by it meet with resistances, that may finally render their motion uniform. Thus a stream of water, although propelled by the force of gravitation, moves in a pipe or channel of constant section with uniform velocity. In all other cases the action of the moving force does not depend upon the velocity with which the body in which it resides moves, or has a tendency to move, but upon the difference between this and the velocity of the machine to which it is applied. Hence, when the point of application is at rest, the force acts upon it with the whole in- tensity it is capable of exerting ; but when this point has a ve- locity equal to that of the body through which the force acts, the former no longer receives any impulse from the latter. As the motion grows out of the superiority of the moving force, and as the action of this force diminishes with every increase of the velocity of the point to which it is applied, equilibrium between its action and that of the resistances must again take place, and if they both act upon a machine, it will assume a state of perma- nent working. We have used the term resistance ; for the machine must not only do the work for which it is constructed, but must also overcome retarding forces that exist in the very nature of ma- terials and workmanship, or which grow out of extrinsic caus- es. Friction is the retarding force from which no material is free, and which no perfection of workmanship can wholly re- move ; the more important of extrinsic forces is the resistance of the fluids, in which machines may be placed, and which in most cases is that of the air of the atmosphere. We measure the mechanical action of a force not merelv by 100 DOUBLE-ACTING the weight it is capable of raising, but by the space through which it raises that weight in given time. Hence, as the pro- ducts of the moving and resisting forces, into the respective spaces through which their points of application pass, are equal either in the states of ordinary or dynamical equilibrium, the measure of these forces is also equal ; and even were there nei- ther friction nor resistance from the air, the utmost work a moving force is capable of performing, is no more than its own measure. Thus nothing is gained by any machine, if consi- dered abstractly, while the whole amount of friction and the resistance of the air is absolute loss. In practice, however, machines are of very great value, in spite of this actual waste of moving power ; we are enabled by them to accommodate the direction of the motion of the agent employed to that of the work to be performed ; we can render a power that has a fixed and determinate velocity, capable of doing work with any other given velocity ; we can apply a natural agent, whose intensity is determinate and invariable, to overcome a resistance of far greater intensity, although at the expense of a loss of velocity ; and we can in either, or all of these cases, bring to the aid of the power of man the action of the great natural agents, water, wind, and steam. Thus, then, the exertion that man must apply, when furnished with proper machines to enable him to make use of these great agents, may frequently become wholly intellectual, and he will no longer have need of his mere physical energies ; or at any rate, a sin- gle man will be able to direct the action of a power to perform a work, for which the united strength of thousands would be in- sufficient. It is in the application of steam to machinery, that this tri- umph of human mind over matter and the elements is most remarkable. 61. Steam may be applied as a moving power in four dif- ferent modes : 1. It may act against a space, wholly or partially void ; in this case, if proceeding from a water of the temperature of 212°, it exerts a force equivalent to the difference between the pres- CONDENSING ENGINE. 101 sure of the atmosphere and the tension of the matter contained in the space against which it acts ; or, if heated to a higher de- gree in a close vessel, with a force corresponding to the in- creased temperature, according to the law stated on page 24. 2. It may be admitted, at a high temperature, into a space greater than it is capable of filling at the density that corres- ponds to its heat, and act against a space void of air by its ex- pansive force. 3. It may, if proceeding from water heated to a high degree, in a close vessel, be able not merely to overcome the resistance of the atmosphere, but to exert, in addition, a great mechanical force. Or, 4. It may cause motion by its re-action. i In the two first cases it is necessary to have the means to form and keep up a vacuum. The mode universally employ- ed for this purpose consists in taking advantage of the conden- sation of steam itself into a liquid form. By the table upon page 24 it appears that the volume of steam at the temperature of 212° is 1696 times as great as that of the water whence it is generated ; hence, its complete condensation would leave but TbTb °f tne space it previously occupied, filled with any ma- terial substance. Such complete condensation is indeed im- possible, for reasons we shall hereafter refer to ; but it is yet ob- vious that a vacuum of a considerable degree of perfection may be thus attained. The condensation of steam is effected by withdrawing its la- tent heat. This is done in the steam engine by the application of cold water, that may either be applied to the surface of the vessels which contain it, or actually brought into contact with the steam itself. In the method now universally adopted into practice, the vessel is not only kept cool by immersion in a cis- tern of water, but a jet of cold water is constantly flowing into it. 62. Let a piston be fitted steam tight to a cylinder, closed at each end, and let the space both above and below it be filled with steam ; if the steam beneath the piston be suddenly con- densed, and fresh steam be permitted to flow into the upper part, the piston will be depressed to the bottom of the cylinder 102 DOUBLE-ACTING. by the whole energy of the steam, as given in the table on page 24 ; if, so soon as the piston has reached this lowest position, steam be admitted beneath it, and the steam resting upon the up- per side be suddenly condensed, the piston will now be forced upwards with a force equal to that by which it was caused to descend ; after reaching the top, the piston may again be forced down, and this alternating action may be kept up as long as steam can be supplied on the one hand, and the means of con- densing it found upon the other. If now a rod be passed through a collar in one of the lids of the cylinder, and fastened at one of the ends to the piston, this rod may be made the means of conveying the force which the steam exerts upon the piston, both in its ascent and descent, ei- ther directly, or through the intervention of other bodies, to some point at which it may be made to perform some regular work, or overcome some resistance. 63. If the steam be condensed within the cylinder, there will be a great loss of heat, and consequent increase in the expense •of supplying the moving power. Whether this condensation be effected by affusion of water upon the outer surface of the cylinder, or by the injection of a stream into its interior, the temperature of the enclosed space and of the sides of the vessel will be lowered, and the heat the steam has communicated to them, wholly or partially withdrawn. When the motion of the piston is to be reversed, and steam begins to enter on the side on which it was before condensed, it must again heat the piston and the adjacent parts of the cylinder up to its own tempera- ture ; this it does by parting with its latent heat, and it is con- sequently condensed : the steam flowing from the boiler, there- fore, exerts no mechanical action until the heat, before abstract- ed, is again replaced ; as the piston moves, fresh portions of cooled surface are exposed, and fresh quantities of steam must be expended in heating them. Such is the effect produced by the alternate heating and cooling of the parts, that it has been found, by actual experiment, that at least five times as much steam is expended upon them as is necessary simply to fill the cylinder. CONDENSING ENGINE. 103 Hence, it is obvious that the steam ought to be condensed in a separate vessel, having a communication alternately with the upper and lower sides of the piston. 64. Water is capable of forming vapour at all temperatures whatsoever. Its tendency to rise is, however,' impeded by pres- sure, and thus it does not boil in an open vessel where the ris- ing of steam is impeded by the resistance of the atmosphere, until it reaches the temperature of 212°. But with each di- minution of pressure, the boiling temperature becomes lower, until, in the vacuum of an air pump, it boils at 90° : Hence, so soon as a portion of the steam is condensed, fresh vapour will be rapidly formed at a lower temperature, and although the ex- pansive force of this diminishes in a geometric ratio, yet it is still capable of opposing a resistance to. the motion of the piston. This resistance is such that it has been found by experience that the vapour of water of 212°, whose expansive force is equi- valent to a pressure of 151bs. on every square inch, had never acted upon the piston with a mean force of more than lUlbs. until means were applied to remove or obviate this resistance. It may be removed, or at least very much lessened, by taking care to keep up a vacuum in the separate condenser. Two modes present themselves for doing this : the engine may be plac- ed at least 34 feet above the level of a cistern of water, and the condenser may be made to communicate with it by a pipe. As that height is the maximum distance to which the pressure of the atmosphere can raise a column of water, the water of con- densation and the condensed steam will flow into the pipe, and as much will pass out at its lower end ; the water being sup- ported at a constant level by this pressure. It, however, hap- pens so rarely that a proper situation can be found to carry this plan into effect, that it has never been applied to practice, and has ceased even to be thought of by those concerned in the construction of steam engines. A pump has therefore been resorted to, in order to keep up a vacuum in the condenser, by carrying off the water of conden- sation, and the vapour that may remain, or be again generated. This pump is called the Air Pump. It may be, with the conden- 104 DOUBLE-ACTING ser, immersed in a cistern of cold water, and a jet of that fluid may play through an aperture into the condenser. In this manner a greater cooling surface is brought into contact with the steam, and the condensation is effected more rapidly than could be done by simply cooling the surface of the condenser. The water that thus enters is regulated to the working of the engine by a valve called the Injection Cock. In some cases, however, as in steam boats, a cold water cistern cannot be em- ployed. The modifications which the structure of boat engines has undergone, will be described hereafter. 65. The alternating rectilineal motion of the piston in the Cylinder of the engine, can of course be only directly applied, to perform a work with the same species of motion and with equal velocity. Thus, by passing the piston-rod through the bottom of the Cylinder, it might be made to work a pump, or by laying it in a horizontal position, to drive a horizontal blow- ing machine. But the cases where this direct application is possible are very few and unimportant, and they have never been introduced into practice. Even where a motion of the same kind, and with equal velocity, is required, it is the nfore usual custom to carry the motion of the piston-rod to the place where the work is to be performed, through the intervention of a Balance or Lever beam, resting upon pivots. This beam, having the axis that passes through these pivots fixed, its ends move in circular ares with reciprocating motion. Now, as the motion of the piston-rod, although reciprocating, is rectilinear, it becomes necessary to make the connexion be- tween the piston-rod and the beam of such a nature as will permit the one of these motions to be accommodated to the other. The simplest, and it might almost be said, the most obvious plan, is to affix a bar to the end of the piston-rod, at right an- gles to its direction, and make the ends of this bar describe straight lines, by adapting them to straight guides of iron ; the end of the piston-rod being thus kept to its rectilineal course, the end of the beam is attached to it by a bar, which has a mo- tion upon cylindrical gudgeons, affixed both to the piston-rod and the beam. Through this bar the force that impels the rod CONDENSING ENGINE. 105 in its ascent and descent, is conveyed to the beam, and the gud- geons allow the bar to change its position in such a way that one end may always move in a straight line, and the opposite one in an arc of a circle. We have said that this method would seem the most ob- vious ; it is not, however, the earliest, and has only been used in the United States, where it has entirely surperseded the earlier method we are about to describe. This method is called the Parallel Motion. A bar similar to that we have des- cribed connects the piston-rod to the end of the beam, but the former has no guides ; a parallelogram is then formed of this rod, of a part of the lever-beam, and of two bars equal and pa- rallel to them ; the two gudgeons we have mentioned are si- tuated at two of the angles of this parallelogram, and at the other angles the connexion between the pieces that form the sides is also effected by gudgeons or pivots, which are called Centres. This parallelogram has therefore sides of a constant magni- tude, but the angles are capable of variation in size by the mo- tion of the sides upon the centres which connect them. The cen- tre at the angle diagonally opposite to that where the end of the beam is joined to the bar which connects it to the piston-rod, is attached by a bar to an immovable pivot in the frame of the instrument, or in an adjoining wall. By this last connexion, the point at this last-named angle, will, when the beam oscil- lates, describe a circle around the centre of the fixed pivot. The points at the two angles of the parallelogram which are situated at the end of and upon the beam, will also describe circular arcs, whose convexity is opposed to that of the arc described by the point attached to the fixed pivot. When the radii of these three different arcs bear a proper relation to each other, the remaining angular point of the parallelogram will des- cribe a straight line. Its path is, in truth, a portion of a curve of contrary flexure, but within the limits of the oscillations of the beam, it does not differ sensibly from a straight line. But as it is not really and truly a straight line, this method, how- ever ingenious, is both less perfect in theory and more com- plex in practice than the other. The side of the parallelogram opposite to that which is a part of the working beam, is called 14 106 DOUBLE-ACTING the Parallel Bar ; the remaining two sides are called Straps ; the bar which connects the lower angle of the parallelogram, to which the piston-rod is not fastened, with a fixed pivot, is called the Radius Bar. We have spoken of the bars, that, with a part of the beam, make up the parallel motion, as single. So far as their theory is concerned, this is sufficient, but for the sake of a proper ad- justment of the pivots, the straps are made double. In the pair of straps nearest the fulcrum of the lever beam, there is another parallel motion, which is applied to work the air pump. This consists in adapting a pivot, to the two straps to which the pump-rod is attached, by a circular socket, in such a way that the direction of the rod is not changed by the motion of the pivot; this pivot, thus placed between two points which describe circular arcs convex towards opposite directions, may be so adjusted in its distance from each respectively, as con- stantly to describe a straight line. The principle of these pa- rallel motions will be understood by reference to the following description and figure. m b is a part of the lever beam in its lowest position, on being the centre on which it vibrates ; to the points a and b are at- tached the straps afc and b d, and to these the parallel bar c d ; the axis of the radius bar is in a line passing through b, and its other end is attached to the point c. The four angles a, b, c, d are formed by pivots so as to have a free motion, and the ra- dius bar has pivots both at b and c. Thus the points a and b will, when the beam moves, describe the circular arcs a g i, and b I k, while the point c will describe the circular arc c e a whose convexity is opposite to the two former arcs ; and they will compel the point d to describe the straight line d b h. In this figure the line a b is half the length of one arm of the le- ver beam, and the radius bar is equal to the same line, but there may be other proportions ; all that is necessary, is that the ra- dius bar shall be equal in length, between its centres, to the dis- tance between the points m and a. The second parallel motion is formed by placing a pivot/, at the point where the line m d cuts the side a c of the parallel- CONDENSING ENGINE. 107 PARALLEL MOTION. ^s Hi 4 i 108 DOUBLE-ACTING ogram, this point will then be compelled to describe the straight line/ a n. For the parallel motion has recently been substituted the fol- lowing arrangement, to which we have already referred as still simpler. To the end of the piston-rod is attached a bar, or cross head, at right angles to it. the ends of this are placed between parallel vertical guides, situated in the plane passing through the line d b h. The cross beam is turned, at two places, into the form of pivots, to which the straps that unite the piston-rod to the working beam are applied. This method has several advantages over the parallel motion. It is much more easy of construction, and requires no geometric skill in the workmen. It is less costly. It, in addition, will permit the beam to describe an arc of greater amplitude, and thus the space occupied by the engrine may be diminished. 66, The end of the beam, opposite to that which is attached to the piston-rod, has also a reciprocating circular motion, ris- ing as the other end falls, and falling as it rises. This species of motion is hardly adapted to be applied directly to any usual species of work. In most of the important applications of the steam engine, the required motion is circular and continuous. It hence becomes necessary to convert the reciprocating motion of the working end of the beam into the last-named variety of motion. This change is effected by the intervention of the Connecting Rod or shackle bar, and the Crank. The connecting rod is a bar of iron attached to the working end of the lever beam by a cylindrical pivot, and a circular socket, in which it has a free motion. The crank is an arm or radius of iron, having a pivot at each end, one of these is fixed in a horizontal position to a socket in a solid support, and the arm has a free motion around the axis of this pivot ; the pivot at the other end projects from the arm, and is inserted in a socket on the lower end of the connecting rod. The length of the crank between the axes of the two pivots is equal to half the space passed through by the piston in the Cylinder, or what is called the length of the stroke. It is at least so when the arms of the CONDENSING ENGINE. 109 beam are of equal lengths, as is most usually the case ; and when they are not, this distance has the same ratio to half the length of stroke as the arms of the beam, to which the connect- ing rod and piston are respectively attached, have to each other. The working end of the beam, rising and falling in a circular arc, under the impulse conveyed from the Cylinder through the parallel motion, will act upon the crank through the inter- vention of the connecting rod ; the moveable end of the crank will describe, under this influence, a semicircle during the time that the beam either rises or descends : this semicircle may be directed to either side of the vertical line passing through the axis of the crank ; and a slight force applied to it in a proper direction, at its highest or lowest positions, will cause it to de- scribe a complete circle. This apparatus may be better understood by reference to Fig. 5, on PL IV, where A represents the end of the lever beam, b the part of the connecting rod, which is forked at the end, and embraces the beam ; c, the connecting rod represented in its highest position ; d, the pivot on the crank to which the connecting rod is attached ; E, the arm of the crank of which f is the centre ; g, h % i } k, represent four other positions of the crank. 67. The force that renders the rotary motion of the crank continuous, is derived from the fly-wheel, which also fulfils another most important purpose. No motion can well be imagined more irregular than that of the piston of a steam Cylinder. When it is in contact with ei- ther end of the Cylinder, the entrance of the steam gradually impels it from a state of rest, until it acquires a maximum of velocity, whose magnitude depends upon the relation between the supply of steam and the work to be performed. When it reaches the opposite end of the cylinder it again comes to rest, more or less suddenly, according to the manner in which the steam is supplied and cut off. A motion in the opposite direc- tion next succeeds, gradually increasing at first, and again ceas- ing when the piston reaches the opposite limit of its motion. It will be thus seen that not only is the direction of the mo- 110 DOUBLE-ACTING tion alternating, but that its velocity is continually varying, and that at two instants there is no motion whatever. Now, in very many applications of steam, it is not only necessary that the action be continuous and circular, but that its rate should be uniform. To effect these two objects, advantage is taken of the nature of matter, which has not the power either of setting itself in motion, or bringing itself to rest ; hence, when a mass is once set in motion, it will have a tendency to move forward continually, and with uniform velocity ; this it will tend to do, although the force that set it in motion cease to act ; and if its motion be resisted, the moving mass will communicate motion to the bodies which oppose it. The part of a machine in which this principle is called into action, is called a Fly-Wheel. It is usually a heavy circular ring, attached, by radiating arms, to the axis of a part of the machine that has a rapid motion. In steam engines it is fixed to the axis of the crank. The fly-wheel, like every other part of a machine, opposes a resistance to the moving power, and requires a certain expenditure of force to set it in motion ; but when once it is set in motion, it requires but small accessions of force, and these may be exerted at in- tervals, to keep it moving with the greatest mean velocity which the moving power, acting through the intervention of the ma- chine, is capable of communicating. If the power be variable, and therefore have a tendency to cause irregularity in the motion of the machine, the fly-wheel resists acceleration, on the one hand, because it cannot suddenly acquire a new velocity ; but will oppose any increase with a force equivalent to the product of its mass into the difference between the velocity it has when the acceleration begins to act, and that which the accelerating force is capable of giving ; on the other hand, as its motion cannot be suddenly checked when the force is either lessened or ceases to act, it therefore goes on, with a velocity decreas- ing only in consequence of the resistances it meets. In parting with its motion, it will communicate as much to the bodies which resist it, and will thus keep up the velocity of the ma- chinery driven by the engine, and render that of the engine itself regular until the acceleration again commences. Hence, in the varying action of the piston of a steam engine, the fly-wheel CONDENSING ENGINE. Ill moderates the speed when it has a tendency to become greatest, receiving" then an accession of force ; this it distributes again among the parts of the machine that are in motion, when the speed of the piston lessens, or actually becomes nought, which happens when it reaches its highest and lowest points. If the mass and velocity of the fly-wheel be made great, this tenden- cy to uniformity will become absolute, and it will go on with uniform velocity, under the constant variation of the motion originally received from the prime mover, giving to the machi- nery driven by the steam engine a regular and constant velo- city. This tendency of the fly-wheel to go forward with con- tinuous rotary motion, accelerated at first until it reach a mean between the maximum and minimum velocity which the piston is capable of communicating to it, through the intervention of the parallel motion of the working beam and the crank, is attain- ed by its passing through a single semicircle, or by performing no more than half a revolution ; hence, when the piston reaches its upper or lower position, and the steam ceases for an instant to act, the fly-wheel carries the crank forward beyond the vertical line ; the new impulse derived from the steam when it acts on the opposite side of the piston, is exerted to compel the crank to move forward in the opposite half of the circle it before described, and therefore with continuous rotary motion. The form and mode of action of the crank has a very benefi- cial influence, in permitting the uniform motion of the fly to be attained without exerting any injurious action upon the engine itself. The force of the crank is always applied to the fly- wheel, in the direction of a tangent to the circle the crank it- self describes ; the force of the steam acts upon the crank in the direction of the connecting rod. When the force of the steam is nothing, in consequence of the piston being in the act of changing the direction of its motion, these two lines are at right angles to each other ; the crank may therefore be carried forward by the fly-wheel, without being interrupted by the ab- solute cessation and subsequent change in the direction of the motion of the piston. But when the steam is exerting its maximum force upon the piston, these two lines nearly coincide, and the crank receives the whole force of the steam. Among 112 DOUBLE-ACTING all the modes, therefore, by which a variable and alterna: motion is converted into one that is continue as, none is more advantage ::s than the crank, and few as much so. ODe, which will be mentioned in the history of the Steam Engine, has equally good properties in this respect, and we know of no other. Persons ignorant of the principles of mechanics are in the habit of considering and declaring that much power is lost when motion is conveyed through the intervention of a crank. This idea appears to have been originally founded upon what occurs when a man works by means of a winch, an apparatus similar to a crank, and acting upon the same principles. Here a power which, when constantly and directly exerted, is capa- ble of balancing a pressure of 70lbs., is not capable of overcom- ing a resistance of more than 251b& This, however, ar:^ from the force itself actually falling, during one part of the re- volution of a winch, as low as the last-named limit ; and hence the revolution cannot be completed if the constant resistance exceed that amount. The power of a man depends not only upon his muscular force, but upon the manner and direction in which that muscuiar force is exerted ; and in some parts of the motion of a winch, this manner is extremely unfavourable. The crank or winch still acts upon the resistance, with the lie force the man applies to it, but this is less at some parts of the revolution that it is in others. Ln the steam entice, a similar variation in the intensity of the prime mover occurs, and it is greater in amount ; but while a man is as much, and even more fatigued in applying his force in the unfavourable positions of the winch, the varying motion of the piston of the steam cylinder corresponds almost exactly with a variation in the expenditure of steam. As a general principle in mechanics, no force can be lost ; it may be applied to resistances which do not enter into the esti- mate of the work performed ; for instance, to overcome the fric- tion of the machine ; or it may, by improper or disadvanta- geous direction, be wasted upon the machine : arts it thus tends to tear asunder or wear away. This last circum- stance does occur in the action of a steam engine, such as we CONDENSING ENGINE. 113 have described it, but the crank is not the only part which is liable to this objection. The rod or strap, which forms a part of the parallel motion, does not always act in the direction of a tangent to the arc described by the end of the working beam, with which it is connected. Hence, it at times expends a part of the force of the engine upon the beam, tending to draw it from its place. A similar obliquity occurs where the connect- ing rod is attached to the opposite end of the beam, and a simi- lar waste of power. In the crank, the connecting rod acts upon it at all angles with its tangent, from 0° to 90° : and hence a part of the force is wasted to draw the axle of the crank from its seat. Were the force of the steam constantly exerted upon a connecting rod three times as long as the stroke of the engine, the power thus wasted would bear to the whole power of the steam the proportion of 0.225 to 1, but as the steam actually ceases to exert any force at the upper and lower points of the crank's revolution, here no loss can occur, and the waste can- not exceed the ratio of 0.139 to one, or about one-seventh part ; while, if, as usually happens, the pressure of the steam first gradually increases, and then again diminishes, the real waste need not exceed one-tenth part of the force of the engine. A longer connecting rod causes the power to act more directly, and its waste to be consequently less. This waste is less than the friction of the engine, and still less than the increase the friction would acquire in any of the methods that have yet been proposed, of making the steam act directly upon a body so disposed as to be capable of acquiring a rotary motion, instead of applying it to a piston working with alternate strokes in a cylinder. We are therefore disposed to think that most of the plans which have been hitherto proposed of constructing rotary engines, have been a sheer waste of inge- nuity. 68. The method we have described, of converting the alter- nating motion of the piston-rod into a continuous rotary one, through the intervention of a parallel motion, a working beam, a connecting rod, and a crank, is not universal. The change is sometimes effected more immediately by affixing the connecting 15 114 DOUBLE-ACTING rod to a cross-head on the end of the piston-rod, which is then made to work between guides. When the Cylinder is ver- tical, the connecting rod and crank are usually double, the for- mer descending on each side of the Cylinder. We have seen more than one plan, in which the Cylinder itself was suspended upon trunnions, permitting it to have a vibratory motion. In this last form the connecting rod may be dispensed with, and the piston -rod acts immediate n the crank. The steam is admitted to the Cvlinder through the trunnions. Such was the condensing engine of French, placed in a boat on the Hudson river in 1808, and such is the high pressure engine construct- ed recently by an ingenious workman in the employ of the West Point Foundry : and which,, since the first edition was publish- ed, has been used both in stationary and locomotive engines constructed at the Novelty Works. New- York. This mode of suspension is. however, only suited to small engines, where the Cylinder has but little weight. When the beam is suppressed, there results a very considerable saving of room, and there are occasions where this is very important. An engine which has no beam, will occupy a space whose length is less than half that taken up by one that has. Tn many of the American steam boats, and particularly in most oi those constructed under the direction of Fulton, the engine has this form. In the Western States, it is usual not only to suppress the beam, but to lay the cvlinder in a horizontal position. This last method has many advantages, among: which may be mentioned as the principal, that a vessel is far less injured by a force acting in the direction of its length, than by one exerted vertically; and that the en- gine may be laid entirely under deck without interfering with any of its more valuable properties. On the other hand, the suppression of the working beam has this disadvantage, that the obliquity of the action of the piston upon the connecting rod is greater than occurs when the pa- rallel motion and beam are used ; and that the loss growing out of this obliquity is greatest in proportion to the power when the latter is a maximum : hence, the waste, compared with the mean power, is greater than in the other case. This, however, does not apply to Cylinders hanging upon trunnions, for, in CONDENSING ENGINE. 115 them, the power is applied directly when at its maximum of intensity. 69. In some few cases the motion communicated to the fly- wheel is rendered more uniform by using two complete en- gines, whose cranks are adapted to the same axle, but are situ- ated in planes at right angles to each other. When the piston of one of these Cylinders has reached the top or bottom, that of the other will be in the middle of its stroke. One of them will, therefore, be acting at its maximum of force when the other ceases to act altogether. This plan is far preferable in effect to that of a single engine of the same nominal power, but it is more expensive,* as a single engine of twice the force of each of them costs considerably less than the two. In many of the best loco- motive engines this method has been successfully used ; but when two engines are applied to a boat, it has been found that it was difficult to keep them at the same rate of working ; hence each is now usually applied to a separate shaft, and moves only one of the wheels. In the British steamers, however, the two engines act at right angles to each other upon the same axle. A fly-wheel is not always an indispensable part of an engine, for there may be some of the machinery which is driven, that will act as a regulator in its stead. Thus, in steam-boats where the wheels have a rapid motion, and in rail-way car- riages, no fly-wheel need be employed. 70. The condensation of the steam is effected in the Condens- er, both by keeping it constantly cool, and by admitting a jet of cold water into that vessel. To accomplish these objects, it is wholly immersed in a cistern supplied with cold water ; and a stream constantly spouts through an aperture in the side of the condenser, to which a stop-cock is adapted ; the quantity of this stream is regulated by the greater or less aperture which the stop-cock affords for the passage of the water. Steam at 212°, is capable, as may be inferred from what has been stated on page 21, of heating six times its weight of water to the same temperature, and the united bulk is seven. The temperature of condensation is usually 100°, and to cool seven measures of 116 DOUBLE-ACTING water of 212° to 100°, will require about sixteen measures of wa- ter, which, added to the six employedin condensation, is twenty- two. That is to say, twenty-two times the bulk of water eva- porated by the boiler, is the least quantity that will suffice for the proper condensation of steam, and cooling the condensed water. There must, besides, be a supply to prevent the water of the cistern from growing warm ; and it has hence been usual to make the cold water pump capable of supplying a pint of wa- ter for every cubic inch evaporated from the boiler. 71. In order to save a part of the heat, the condensed steam and water of condensation are delivered by the air pump into a vessel called the Hot Water Cistern, whence the water is rais- ed, by the Hot Water Pump, to the feeding apparatus of the boiler. These two pumps are worked by rods, attached to the working beam, when the engine has one ; in other cases these rods, with the rod of the air pump, are attached to a bar or beam, one end of which is adapted to the piston-rod, and rises and falls with it ; the other is fastened to a fixed centre upon which it oscillates. 72. The power of a condensing engine depends upon the state of the vacuum that is kept up in the condenser, as well as upon the pressure of the steam flowing from the boiler ; hence it is important to be able to know whether the rarefac- tion produced by the condensation of the steam, and the action of the air pump, be more or less perfect. This knowledge is at- tained by the Vacuum Guage. A glass tube, open at both ends, has its lower extremity immersed in a basin of mercury, the other end communicates by a pipe with the interior of the con- denser. When the steam is condensed in that vessel, the pressure of the atmosphere forces the mercury to rise in the tube to a height which is the measure of the exhaustion ; the difference between the height of this column and the height at which the mercury stands in a barometer, is the measure of the force which acts in opposition to the pressure of the steam upon the piston of the engine. This must, therefore, be deduct- ed, in estimating the actual performance of the engine, from the CONDENSING ENGINE. 117 indications of the steam-gunge after an atmosphere has been added to the latter. An apparatus called the Indicator, in which a spiral spring is alternately opposed to the steam and the va- cuum, has been proposed as a substitute for both the Steam and Vacuum Guages ; but it has not yet come into general use. It is, however, the only apparatus by which a true estimate can be obtained of the force actually employed in a steam engine, and, when compared with the steam and vacuum guages, would illustrate that part of the theory which is yet deficient, namely, the determination of the pressure which is exerted by steam of a given tension upon a piston moving with a given velocity. 73. The action of the fly, in producing regularity of motion, reaches only to the inequalities that take place in the motion of the piston during a single stroke. Should the flow of steam increase, the mean motion of the fly-wheel will be accelerated, and, should the flow be diminished, the fly-wheel will be uni- formly retarded. Neither does it control any change in the mo- tion of the machinery, driven by the steam, unless that change be periodic. But it frequently happens that the quantity of. steam supplied by the boiler, fluctuates. Some regulator is therefore necessary, whenever work is to be done with regulari- ty, which shall control the prime mover itself. For this pur- pose a Governor is adapted to the steam engine. This is also required in cases where the quantity of work to be performed is fluctuating, as is the case in many branches of manufactures, where a part of the machinery may be suddenly stopped, or may be as suddenly connected with the engine. The govern- or is an apparatus that is sometimes called a Conical Pendu- lum. Two heavy balls are suspended by bars to the opposite sides of a vertical axis. This axis is set in motion by the en- gine ; as it turns, the balls of the governor acquire a centrifugal force, which may be sufficient to overcome their weight, and cause them to diverge and fly off, performing in their course a larger circle than before. As the balls fly off, they act, through the intervention of a system of levers, upon a valve that is si- tuated in the steam pipe. This, which is called the Throttle- valve, has the form of a circular disk of metal, exactly filling 118 DOUBLE-ACTING up the pipe when placed across it. It turns upon pivots placed at the opposite ends of one of its diameters, and may thus either present its edge to the steam that passes along the pipe, in which case it hardly resists its course ; or may assume any intermediate position until it close the pipe altogether. When the balls of the governor revolve with so little velocity that the centrifugal force cannot overcome their weight, the levers place the throttle valve in the position that presents its edge to the steam ; when the velocity becomes great enough to throw out the balls to their utmost limit, this valve is thrown across the pipe, and shuts the passage completely ; with intermediate po- sitions of the valves, the passage is more or less open, accord- ing to the rotary velocity of the governor. The governor is driven, by a strap that passes over a drum on the axis of the crank, or by wheels and pinions, deriving their motion from the same part of the engine. This apparatus is of no use in navigation or locomotion, but is indispensable in engines used for manufacturing purposes. 74. When the throttle valve acts under the influence of the governor to lessen the efflux of steam from the boiler, the elastic fluid will accumulate in that vessel, and its density and elasticity will increase along with its temperature. In this event it will act upon the float which counterpoises the self-regulating damp- er, and the latter will descend and lessen the draught of the chimney. A diminution in the expenditure of steam thus acts to diminish the intensity of the fire by which it is generated, while, if it accumulate too suddenly, the safety valve affords it a vent. It is, however, to be remarked, that such floats are inadmissible, except when the tension of steam does not much exceed a sin- gle atmosphere ; and that self-regulating dampers are never used in steam-boats or locomotive engines. The valves, by which steam is admitted into the upper and lower parts of the cylinder alternately, and by which the com- munication with the boiler is opened and closed, are worked by machinery attached to the engine. Rack work upon the rod of the air-pump was originally used for this purpose, but it is now CONDENSING ENGINE. 119 more usual to adapt an Eccentric to the axle of the crank. The eccentric is a circular plate of metal, which has an opening within it that just fits a part of the axle of the crank. This open- ing is placed in a position eccentric to the plate itself, and hence the apparatus derives its name. The eccentric plate is attach- ed to ihe axle of the crank, and revolves with it. A circular ring fits upon the eccentric, but leaves the latter a free motion within it ; any given point in this ring will, therefore, have its distance from the axis of the crank changed within certain lim- its ; this change is conveyed to a bent lever which works the valves, through the intervention of an open frame-work of the figure of an isosceles triangle, whose two equal sides are tan- gents to the circular ring that encloses the eccentric plate. 75. The Double- Acting Condensing Steam Engine, then, is in a great measure self-acting. In truth, when applied to per- form work with an uniform velocity, little is left to be done, ex- cept to supply the fire with fuel, and to observe the indications of the guagesfrom time to time. Even the supply of fuel has been regulated by machinery driven by the engine, in such a way that it need not be fed for several hours. It is therefore not to be wondered that the condensing steam engine, worked by steam of a tension little exceeding an atmosphere, was considered for a time, and is still considered by many, as the most perfect of all human inventions. We shall, however, have occasion to des- cribe another method of working the condensing engine, by which its efficient power has been more than quadrupled. It unluckily happens that much of the beautiful and ingenious ap- paratus which, in their application to the engine or the boiler, tend to render the former self-acting, are rendered useless in the new mode of working. 76. The pistons of the Cylinder and air-pump, and the open- ings in the covers of those parts of the engine through which they move, are rendered steam tight by packing. The substance for- merly solely employed for this purpose was hemp, in the form of plaited bands, and it is coated with grease. The joints of the se- veral parts are closed by plaited hemp, or felt, coated with white 120 DOUBLE-ACTING lead ground in oil, or where one part is made to fit into another, by an iron cement, composed of iron filings, or gun borings, mu- riate of ammonia, and flour of sulphur ; the proportions are sixteen parts by weight of the first, two parts of the second, and one of the third substance. The joints are, generally speaking, formed by flaunches cast upon the pieces, in which holes are drilled \ through the latter are passed screw bolts, that are fasten- ed by nuts. The power of machines is estimated in terms of some con- ventional force, taken as the unit. Steam engines having been originally introduced as a substitute for the action of horses, it became the practice to compare the force of an engine with the strength of a number of horses. The unit, which is employed in the estimate, is, therefore, a horse-power ; and we speak of engines as being of the power of a certain number of horses. As the strength of horses is very various, this is still a vague method, and it becomes necessary that the estimate of the work a horse is capable of performing, should also be agreed upon. Different engineers have at different times made use of different values ; but the modes of estimating the horse-power resolve themselves into the expression of the number of avoirdupois pounds raised one foot high in a minute. Desagnliers estimates this number at 27,5001bs., and Smea- ton 22,9161bs. Watt supposes that a horse is able to raise 32,000lbs. ; but in calculating the power of the engines of Watt and Bolton, the estimate has been taken as high as 44,0001bs. The force which acts is the pressure of the steam, and as much pressure as is indicated by the steam guage is supposed to act upon the piston ; this, multiplied by the velocity of the piston, gives the whole power of the steam ; but before the steam that issues from the boiler can reach the piston, it is retarded by the friction of the pipes, and loses by cooling a part of the ex- pansive force indicated by the steam-guage ; its action is next diminished by the cooling it undergoes in the Cylinder itself j and before the power is transmitted to the working point of the engine, it must overcome the friction of the piston, open and shut the valves, force the steam into the condenser, and work the air-pump and the hot and cold water pumps. It has, next, CONDENSING ENGINE. 121 to overcome the friction of the axles of the lever-beam, parallel motion, and crank ; and the piston is besides resisted by the uncondensed steam remaining in the condenser. Of the last, the vacuum guage furnishes a measure ; but of all the rest no- thing is known perfectly, except by comparing the work ac- tually performed with the original force of the steam. We have further to remark, what appears to have been neg- lected by all former writers, that the actual tension of the steam is not the measure of its pressure upon the piston when in mo- tion. It will be obvious that the whole of such a force can only be exerted upon a body at rest, and that when the velocity of a body is as great as that with which the steam can fol- low it, all pressure ceases. It might be a mathematical inves- tigation of no little theoretic interest to determine at what ve- locity within these limits a maximum effect is produced, and what will be the pressures of steam of a given tension upon a piston moving with given velocities. It does not, however, seem probable that at the present moment such an investigation would be attended with any valuable practical result. It has been deduced from observation upon the working of engines, that more than 40 per cent, of the original power of the steam is lost from these several causes ; hence the indication of the steam-guage must be diminished in that ratio at least, before it is employed in the calculation of the force of the en- gine. In this country it has been usual to estimate the horse power at 33,0001bs., raised one foot per minute, and the mean pres- sure of the steam, in a condensing engine, at lOlbs. per square inch. We hence have the following rule : Multiply the area of the piston in square inches by 10, and by the velocity of the piston in feet per minute ; divide the continued product of these three quantities by 33,000, the quo- tient of the estimated force of the engine in horse power. The rule of Brunton gives 44,0001bs. for the divisor, and that of Tredgold reduces the mean pressure to 9,10 pounds per square inch, which would correspond to a divisor of near- ly 36,000 when the pressure is assumed at lOlbs. The tension of the steam is supposed to be five inches of mercury, marked 16 122 DOUBLE-ACTING by 2£ inches in the mercurial guage ; and equivalent to a pres- sure of 2^1 bs. more than an atmosphere, or 17|lbs. per square inch. The safety valve is loaded with a weight of three pounds per square inch, which, with the aid of the atmosphere, will retain steam whose expansive force is not greater than 1 81bs. per inch. 77. The quantity of water to be evaporated in order to do the work of a horse in a double-acting condensing engine regu- lated as we have just stated, may be estimated as follows, viz : A cubic foot of water, evaporated under the ordinary pres- sure of the atmosphere, occupies a space 1696 times as great as it did before ; but the space it occupies under a pressure of \7\ pounds is, if we abstract from the expansion by tempera- ture, less in the ratio of 15 to 17^ ; for elastic fluids occupy spaces inversely proportioned to the pressures by which they ere confined, (see p. 14); hence the space occupied by steam having an expansive force of 17£lbs. is 1454 times the original bulk of the steam. A cubic foot of water, therefore, occupies a space, in the form of such steam, of 1454 cubic feet ; and the effective pressure, as we have before stated, is lOlbs. per square inch, or 14401bs. per square foot ; the power of a cubic foot of water is, therefore, to lift 1440x1454 or 2,0937601bs. through the space of a foot. If this be divided by 33,000, which is the conventional weight to be lifted by a horse power per minute, it will give the num- ber of minutes in which, if a cubic foot of water be evaporated, it will keep up this conventional unit of force. The quotient is 63, or three minutes more than an hour. It is therefore usual to allow the evaporation of a cubic foot of water per hour to be equal, in the engine under consideration, to a horse power : and as it is well to be always certain of a supply of steam, boilers are made to furnish more units of steam than the en- gine is estimated at ; the waste of heat in small boilers being greater in proportion than in large ones, this excess is a con- stant quantity, the boiler being calculated to produce steam equivalent to two horse power more than the estimated force of the engine. We have seen that a surface of boiler in con- CONDENSING ENGINE. 123 tact with flame and hot air of 8 square feet (see p. 57) is equal to the conversion of this quantity of water into steam. 78. The feeding apparatus of the boiler, which is, in this form of engine, composed of a pump that raises the water of conden- sation from the hot water cistern to a cistern at the top of the feed pipe, must therefore supply at least one cubic foot per hour for each horse power at which the force of the engine is , estimated : or 14 1 64 part by bulk of the capacity of the cylinder, ' at each stroke of its piston. As, however, it is better to have an excess than a defect of water, the hot water pump usually raises at each stroke g-fo-th part of capacity of the cylinder. Such are the general principles of action of one form of the condensing engine, which, to distinguish it from others in which the same operation is employed to form a vacuum, is called the Double- Acting Engine, to which epithet is also add- ed the name of the inventor, Watt. We are now prepared to enter into a more particular view of its several parts, the use and operation of which would have been unintelligible, had we not previously investigated their uses, and the relation in which they stand to each other. CHAPTER Y. DESCRIPTION OF THE DOUBLE-ACTING CONDENSING ENGINE. Usual form of Double- Acting Condensing Engine. — Steam- pipe. — Jacket. — Side Pipes. — Side Valve. — Puppet Valve. — Balance Valve. — Cylinder. — Cylinder Lid. — Cylinder Bottom. — Piston. — Woolfs Piston. — Metallic Packing. — Condenser . — Air Pump. — Delivering Door. — Air Pump Bucket. — Hot Water Cistern and Pump. — Cold Water Cistern. — -Injection Cock. — Water of Condensation. — Cold Water Pump. — Parallel Motion. — Lever Beam. — Pump Rods. — Connecting Rod. — Crank. — Fly Wheel. — Tumbling Shaft. — Eccentric. — Double Eccentric. — Ad- justment of Eccentric. — Governor. — -Throttle Valve. — Other forms of Double-Acting Condensing Engine, — Mode of setting these Engines in motion. 79. Having in the last chapter explained the general princi- ples of action of the Double Condensing Engine, we shall now proceed to describe the several parts more particularly, and in reference to a plate, on which they are figured in connexion with each other. See. PI. III. As the condensing engine, in its most complete form, and adapted for general purposes, is in more general and frequent use in Great Britain than in this country, an engine constructed by Messrs. Murray, Fen ton & Wood, of Leeds, has been chosen for the illustration of this part of our subject Fig. I. is an external elevation of this engine. Fig. II. a section. Fig. III. a horizontal plan. Fig. IY. a view of the DOUBLE-ACTING CONDENSING ENGINE. 125 lower part of the apparatus from the opposite side. The same letters apply to the same parts in these four several figures. In the engine before us, the steam reaches a part of the steam- pipe marked s, whence it flows into a space formed around the Cylinder by a cylindrical case called the Jacket. The use of this is to keep the Cylinder itself at an uniform temperature. All engines have not this additional part, and in this country in particular we never recollect to have seen it used. For it, a simple casing of wood is frequently substituted, which, being a bad conducter, has been supposed to be well adapted to pre- serve the heat of the cylinder. From what has been said in respect to the mode in which heat is carried off under certain circumstances, it will appear that both Jacket and wooden casing are liable to objections. In the air, but little heat is carried off in consequence of the conduct- ing power of the surface, and by far the greatest part of the loss is due to radiation. Now, of the metals, the rough blackened surface of cast iron is among the best radiators, and wood stands high in the general order of radiating power ; and hence, in the first case, the steam will be cooled before it reaches its place of action ; and in the second, the temperature of the Cylinder will be more affected than if it had not been cased. The principles we have discussed would point out as a sure method of retain- ing the heat, to enclose the Cylinder in an air-tight cylindrical case of some bright metal, with a thin body of air between them. The confined air will convey but little heat to the casing, and that which is conveyed will radiate very slowly. 80. In the engine before us, the steam passes from the jacket to the side-pipes, marked a a, through the opening marked b. The form and arrangement of these pipes depends upon the structure of the valves ; in this engine the valves are of that description called the Slide- Valve. This was originally in- vented by Murray, of Leeds, but was compressed by him with- in a shorter space ; the valve before us, which occupies the whole side-pipe, is an improvement of Watt's. 81. The side-pipe has the general figure of a half cylinder, 126 DOUBLE-ACTING the plane face of which is turned towards the Cylinder of the engine, and is terminated at top by a square box. The steam en- ters this pipe by a channel b. that communicates with the jacket. In engines that have no jacket, the steam-pipe usually enters the side-pipe from behind it, about the middle of its height. Within this pipe is placed another, which exactly fills it at the upper and lower extremity, but which is made less in the middle, so that the steam, on entering the side-pipe, fills up the space between the two pipes. The inner pipe is moveable, and attached to a rod that passes through an air-tight collar in the square box, of which we have spoken, and by which it is drawn up and pushed down alternately, under the action of a mechanism that will be hereafter described. Between the Cylinder and the outer pipe are two channels, whose section is rectangular. One of these forms a communi- cation with the upper, the other with the lower part of the Cylinder. The length of the inner pipe is so adjusted that when that part, at one of its extremities, which just fills up the outer pipe, is opposite to the corresponding rectangular passage, the other rectangular passage shall be opposite to the space that we have described as left between the middle part of the inner and the outer pipe. Hence, steam will flow into the Cylinder from this space. In the plane surface of the part of the inner pipe that is applied to the first-named rectangular passage, there is a corresponding rectangular opening, by which the steam, from the adjacent side of the piston, will pass into the inner pipe, and thence by a passage marked o into the condenser n. In the position in which the engine is represented in the figure, the steam is flowing into the lower part of the Cylinder, and beneath the piston, while it is passing out at the opposite end, and through the inner pipe to the condenser. The inner pipe has a similar rectangular opening at its op- posite extremity ; when, by the action of the engine, the inner pipe changes its position, this opening adapts itself to the ad- jacent rectangular passage ; while the other communicates with the space between the two pipes, and thus the direction of the steam and the motion of the piston are reversed. CONDENSING ENGINE. 127 It will therefore be seen that there is a constant communication between the space contained between the two pipes and the boil- er, while the inner pipe has a constant communication with the condenser. A change in the position of the inner pipe brings the openings of the Cylinder alternately into communi- cation with the boiler, and condenser. It is obvious that this species of valve requires very perfect workmanship ; the plane surfaces of the outer and inuer pipe must be ground in the most careful and exact manner, and the ""»"" m 128 DOUBLE-ACTING circular surfaces, where they come in contact, at the upper and lower extremities, must also be accurately fitted. The structure and use of this species of valve will be better understood by reference to the figures on the preceding page, in which it is represented in two different positions. In order to give more variety, we have taken a form different from that of the engine in PI. III. in which the spindle enters the side-pipe from above, while in the figure, the spindle is applied beneath. This valve, being as long as the Cvlinder. has been called the Long Slide-valve, in order to distinguish it from one acting upon the same principle, but which does not occupy so great a space, and which is called the Short Slide-valve. PI. I Fig. 6, re- presents a section of a cylinder, and side-pipes adapted for the occupation of a valve of the latter description ; and we shall describe it more fully hereafter, in treating of the kind of en- gine to which it is most frequently adapted. The valve which is most frequently used in modern English engines is also of the sliding form, but is divided into two parts, the one corresponding to the upper, the other to the lower end of the cylinder. The slides are connected by a rod. This form is called the double D valve, and is placed, like the puppet valve we are about to describe, between two side-pipes, one of which communicates with the boiler, the other with the condenser. Sliding valves have the advantage, which is in many cases important, of opening gradually, and thus causing no sudden shock when the motion of the engine is changed ; and by a proper adjustment of the distances between the openings of the inner pipe, and of the apparatus by which it is driven, it may be made to cut off the steam before the motion of the piston is completed, and thus again render the change less sudden. On the other hand, the accuracy of workmanship it requires may not be always attainable, and its repair cannot be effected in situations remote from well-organized workshops. It is also to be stated, that when impure water is used, it has been found to wear rapidly and unequally ; and thus, after having been intro- duced in many of our steam-boats, it was laid aside, and an- other and more ancient species of valve restored. Of this we shall proceed to give a description. CONDENSING ENGINE. 129 82. This species of valve, usually called the puppet valve, is represented on PL II, Fig. 3. The side pipes are two in number, of which that marked A, is continually receiving steam from the boiler, through the steam pipe E, while that marked B is constantly conveying it to the condenser. These pipes are united by being both inserted at each end into the same cylindrical case, or box, of which there are consequently two, at C and D. These are called Steam Chests, and each of them is divided into three spaces by two diaphragms, having each an opening of the form of a trun- cated cone, whose least base is lowermost. These appertures are called nozzles, and are the seats of the valves ; to these nozzles, four solid frusta of cones, a, 6, c, d, are accurately ground, and form the valves ; from the space between the two diaphragms in each box, is an opening that allows the steam to pass to and from the cylinder. It will be obvious, that when the two upper valves, a and c, in each box, are raised, steam must flow from the boiler into the Cylinder, and when the two lower of each set, b and d, are raised, it must flow from the cylinder to the condenser. Hence, it is necessary that they should reciprocate, the valves a and d opening when the valves b and c close, and vice versa. Hence, the two valves, a and d, are united, and made to open and shut at the same time ; as are the two valves b and c. It will be perceived by the drawing, that each valve has a cylindrical spindle or stem attached to it, and the four nozzles are in the same vertical line. The spindles of the two steam valves, a and c, are hollow, and admit the spindles of the two condensing valves, b and d, to pass through them. The purpose of this will be explained hereafter. In more ancient forms of the en- gine, the spindles were replaced by a short rack, into the teeth of which the teeth of a circular segment caught. The use of this form will also be stated hereafter. In some engines, the steam and condensing valves of each pair are placed obliquely, instead of being in the same vertical line. In this case each spindle is solid and has its separate steam- tight collar. The most perfect form of valve is that of Trevithick. The 17 130 DOUBLE-ACTING slide valve, if tight, is attended with great friction, and the puppet valve is kept in its seat by a pressure of steam, which, on each square inch of its surface, is equal to that on a similar area of the piston of the engine, and this resistance is estimated at more than the friction of a slide valve. The valve of Trevithick has the form of a cylinder, on the upper end of which, at c c, is a conical ring, and a hollow cone is turned at b 6, on its inner and lower surface. The first of these conical surfaces rests in a hollow frustum e e, the second upon a solid frustum d d. It will there- fore be seen that the resistance to the opening of the valve, is the pressure on a surface equal to the diiference between the areas of the conical surfaces, c c and b 6, instead of that upon a circle whose diameter is c c. The arrows show the directions in which the steam flows. The side pipes sometimes have, for the sake of ornament, the form of pillars, the entablature being extended above to cover the space left vacant by the side pipes. CONDENSING ENGINE. 131 The old rule for the side of these nozzles was, to make their least diameter one-fifth, at least, of the diameter of the Cylinder ; but one-fourth of that diameter is now a more usual dimension. The passages into the boiler must have an equal area, as must the passages of the slide valve that has just been described. 83. The Cylinder of a steam engine has the figure which is denoted by its name ; and, in order to avoid ambiguity, it has been and will always be distinguished by beginning it with a capital, in order to prevent its being confounded with such other parts as have also a cylindrical form. It is represented in the figures on Plate III, by the letter b. This vessel is, in all large engines, made of cast iron, cast with a core, and reamed out to the proper size. This operation requires great care, and should be done in a mill liable to no agitation, for much of the value of the engine will depend upon the interior being as truly a mathematical cylinder as the nature of materials will admit. Near the upper end of the Cylinder is cast a rectangular piece, in which is the passage h ; and at both ends are cast flaunches, to admit the fastening of its lid and bot- tom, by means of screw-bolts and nuts. In the engine on the plate, the lid is screwed to a flaunch on the jacket, and the flaunch of the Cylinder secured to the jacket by packing. 84. The lid of the Cylinder is a circular plate, whose diame- ter is equal to that of the flaunch to which it is adapted. On its lower side, it is turned, so that a circular projection fits the inside of the Cylinder, barely leaving room for the packing. In the middle is an opening to admit the passage of the piston-rod, and around this opening, on the upper side of the lid, is cast a cylindrical stuffing-box, to receive the packing, by which the rod is made to work steam-tight. In small engines the upper part of this stuffing-box is cut into the form of a female screw, to receive the screw that compresses the packing, and the head of the former is turned into the form of a cup, to contain oil. In larger engines the oil cup is connected with the lower part of the stuffing-box by screw-bolts and nuts. 132 DOUBLE-ACTING 85. The bottom plate of the Cylinder is of the same diame- ter with the top, and has a similar projection turned upon it, to fit the Cylinder. The lower steam passage passes through it, and is cast in one piece with it. In the engine before us, the flaunches of the Cylinder and the jacket are united to the bot- tom by the same set of screw-bolts and nuts. The length of the Cylinder must be as much more than the length of the stroke of the piston as is equal to the thickness of the latter, and, in addition, a small space to prevent the piston from striking. In the engine on Plate III, the diameter of the Cylinder is equal to half the length of the piston's stroke. This proportion is not a constant one, but is that sanctioned by the general practice of Watt. In the English steam-boats where the engine is placed beneath the deck, the stroke is ne- cessarily short, and power is gained by increasing the diameter of the Cylinder. We shall have occasion, in speaking of steam-boats, to treat of the proper length of stroke for engines intended to propel them. In those which are applied to manufactures, the propor- tion stated above is perhaps the best. 86. The piston is still usually composed of two pieces of circular section, that are just so much smaller than the internal section of the Cylinder, as to move in it freely without touch- ing. The lower piece is firmly attached to the piston-rod, by making the lower end of this rod of the shape of a truncated cone, of which the lesser base is uppermost, and of the same size with the rod. A key is passed through the rod just above the piston, and unites them firmly. The two pieces of which the piston is composed are connected by screws, and have a semicircular groove cut, as it were, out of their united mass, forming a ring completely around them. This open ring is occupied by the packing. The packing is usually formed of hemp, moistened by an oleaginous substance. This packing is compressed, and made to apply itself closely to the sides to the Cylinder, by the screws which unite the two pieces of which the piston is composed. As the packing wears, the screws are turned, and thus the packing, being again com- CONDENSING ENGINE. 133 pressed, is forced out, and again applies itself to the cylinder. This arrangement may be better understood by the following figure, which represents a section of the Piston : a is the Piston- rod terminating in the truncated cone b\ c c screws to unite the two parts of the piston, d d and e e ; //section of the pack- ing. 87. An ingenious mode of tightening the packing without taking off the lid of the cylinder, was invented by Woolf. The head of each of the screws is cut into the form of a toothed pinion, and the teeth of all these work in a wheel, having a free motion around the piston-rod. It is. therefore, evident that if one of the pinions be turned, not only will the screws attach- ed to it be made to act, but all the others will be equally driven forward. One of the screws has a square head, which can be reached by a key passed through an opening in the lid of the Cylinder, and which is usually closed by an air-tight cap ; it may thus be turned, by removing the cap, and all the others will be turned equally by the wheel and pinions. In the figure annexed a is the piston rod, b b the wheel fit- ted loose upon it, c, c, c, c, c, pinions forming the heads of the screws that compress the packing, d square head formed upon one of the screws, by adapting a key to which, the wheel b b is turned, through the intervention of the pinion to whose screw the key is applied ; the wheel b b turns the remaining pinions, and with them the compressing screws. 134 DOUBLE-ACTING 88. Metallic packing appears likely to supersede all others, and has already done so in many instances. The earliest at- tempt at a substitute of metal for hemp was made by Cart- wright. Two rings of metal, accurately ground to fit the cylinder, are interposed between the upper and lower parts of the piston. Each of these rings is cut into three parts, and they are placed upon each other in such a manner that the joints of the one ring fall half way between the joints of the other, in the same manner as the break-joint of masonry. Three springs are made to act upon each of these rings, one at each joint, and thus to press the two adjacent pieces out- wards. In this manner the springs carry the rings outwards, to replace any diminution by wear, and the breaking of the joints prevents any escape of steam through the apertures that are thus made in either of the rings. The annexed figure shows this arrangement, where a is the piston-rod. 6, b, b springs that press against the joints c, c, c of one of the rings, which is here represented as formed by in- scribing an equilateral triangle in a circle, d, d. d are parts of the three pieces that form the second ring, whose joints fall at the points e, e 7 e, against which springs similar to b, b, b press. CONDENSING ENGINE, 135 In applying a metallic piston, accuracy in the boring is ab- solutely essential, nor can they be introduced except when this part of the workmanship is of the best description. Another form of metallic packing, which is represented be- neath, has been used in locomotive engines. It is composed of 136 DOUBLE-ACTING a screw-formed ring, compressed between the plates of the piston. The several convolutions of the screw are united by solder, un- til they are turned down to the proper dimensions. The sol- der is then melted off. The most perfect form of metallic packing is one in which the elastic force of the steam itself is used as the spring. The piston in this case is a single cylindric plate of cast iron. Two flat rings are turned out off its curved surface, leaving three flaunches. The upper and lower flaunches are pierced by a number of small holes, by which the steam tends to pass into the flat rings. These rings are occupied by a double com- pound ring of bell-metal, the pieces of which are so arranged as to break joint, and thus prevent the steam from passing them and the inner surface of the Cylinder. This packing has the great advantage that its friction is exactly proportioned to the tension of the steam by which the engine is worked ; while in all other methods, if the packing is compressed sufficiently to be tight at the highest tension to which it is subjected, the fric- tion is enormous at lower tensions. 89. The Condenser is a vessel of a cylindric form. It is re- presented in Fig. 2, PI. III. by n. Through the top passes the pipe o, which conveys the steam from the valves of the engine. On the side is an aperture, to which is adapted the valve r, called the Injection-cock, the use of Tvhich is to admit a constant jet of cold water to condense the steam. The capacity of the con- denser, when the engine works with steam of the pressure of 17-*4bs. per inch, is usually one-eighth part of the capacity of the C/linder. Its several dimensions are therefore each one-half of the corresponding measure of the cylinder. But when steam of greater tension is used, the size has been increased to half the capacity of the cylinder, and both have equal diameters. The state of the vacuum in the condenser is ascertained by means of a vacuum guage. This is represented PL I. Fig. 14. a a is an open vessel of mercury, b b a glass tube im- mersed at one end in the mercury, and communicating at the other with the condenser through the tube e. As the vacuum CONDENSING ENGINE. 137 is formed in the condenser, the presure of the external air will force the mercury up the tube b b, and the difference between the height to which it rises, and that at which the mercury of the barometer stands at the time, marks the resistance the gas- eous matter, that cannot be withdrawn from the condenser, of- fers to the descent of the piston. The Condenser communicates with the Air-Pump by a ho- rizontal passage of a rectangular shape. In this passage is si- uated the Foot-valve t. This has usually the form of a shutter hanging by a hinge on its upper side, in a position slightly in- clined from the vertical, and closing by its own weight. The valve is fitted to its seat by grinding or filing. The condenser and air-pump are screwed down to a common base called the bed-plate. 90. The Air-pump q is also a cylindrical vessel, almost identical in figure with the Cylinder. In the engine before us it has half the lineal dimensions of the cylinder, and consequently one-eighth of the capacity, or one just equal to that of the con- denser. The lid of the air-pump is similar to that of the Cylin- der, permitting the passage of the rod through a stuffing-box. 91. The piston of the air-plimp is packed in the same man- ner as that of the cylinder, but, unlike it, is not solid. It con- tains a valve, which is usually of that form called the butterfly valve. In this shape, the Piston-Rod is attached to a bar ex- tending across the piston in the direction of one of its diam- eters ; to this are adapted by hinges, in such a manner as to open upwards, two shutters that fill up the rest of the circular opening of the piston. These shutters, therefore, rise and fall together like wings, whence their name. The piston and its valves are usually called the Bucket of the Air-pump. From the dimensions we have stated above, it will be obvious that the stroke of the Bucket is just half that of the Piston of the Cylin- der. A plan and section of an air-pump bucket are represented on the following page. 18 138 DOUBLE-ACTING 92. On the side of the Air-pump, and near its top, is cast a rectangular passage, which is closed by a valve v ) similar in form and structure to the foot-valve, and which is called the Delivering-door or Clack-valve. 93. Upon the rise of bucket of the Air-pump, the water of condensation is discharged by the delivering-door into a rectan- gular vessel of iron w, called the Hot-water Cistern. The Hot- water Pump, by which the water of condensation, or at least as much of it as is necessary for the supply, is carried to the boiler, is represented at x. It is a common pump, composed of a barrel and two valves. The water converted into steam is, as we have seen on page 116, -nVfth P art of the capacity of the CONDENSING ENGINE. 139 Cylinder for each stroke of the piston ; the pump is made to furnish a greater quantity, or ti^th P art > m order that there may be no risk of a defect in the supply. As the water of condensation is much greater in quantity than this, being twenty-two times as much in weight as the steam that is employed, a large portion of the water must run to waste, which it does by a waste-pipe. The calculation of the size of the hot-water pump may be made as follows, viz. Divide the cubic contents of the cylinder in inches by 900, and this quotient by the length of the stroke of the pump, the quotient will be the area in square inches, whence by the usual geometric rule the area of the valves may be calculated. The stroke of the hot- water pump in the engine on PI. III. is one-third of that of the cylinder. 94. The condenser and air-pump are immersed in a cistern of water, called the cold-water cistern. In some engines this is a basin in the ground, lined with masonry, laid in ce- ment. In steam-boats it is omitted altogether, and its want supplied by increasing the size of the condenser. In other en- gines again, it forms a cast-iron trough or basin, on the sides of which the whole of the apparatus is supported. In the engine represented on PI. III. this is the case, as will be obvious from the several views in which it is represented, a a being this trough. An engine thus supported, and which may therefore be placed upon any solid basis, entirely independent of walls or buildings, is called a portable engine, even when of the largest dimensions. 95. From the cold-water cistern, a pipe passes into the con- denser. The use of this is to admit a jet of water, to condense the steam with greater rapidity, by bringing it in contact with a greater surface. The extremity of this pipe is sometimes co- vered by a nozzle, pierced with holes like that of a watering-pot. The quantity of injection water is regulated by a valve called the Injection-cock, which is to be seen in Fig. 2. 140 DOUBLE-ACTING 96. As the injection-cock is constantly drawing water from this cistern, and as the water it contains is constantly abstract- ing heat from the condenser and air-pump, it requires a con- tan t and regular supply, as well to keep it at a proper temper- ature, as to renew what is actually expended. For this purpose the cold-water pump y is provided. It, like the hot-water pump, is a common pump, communicating with a reservoir of fresh water. We have, upon page 116, stated the quantity of water that is needed to keep the water in this cistern at the pro- per temperature, whence the area may readily be calculated when the length of the stroke is known. The length of the stroke of the cold-water pump, in the engine before us, is the same as that of the air-pump, or half that of the Cylinder. Hall's condenser, which lias recently been introduced, is com- posed of a series of tubes immersed in the cold water cistern. The condensation is effected by the steam coming in contact with the cold surfaces of the tubes, instead of being caused principally by injected water. The tubes are treed from the condensed steam by an air-pump of the same size and structure as that we have described, and this is of sufficient power to diminish the ten- sion of the remaining vapour below that which is due to the temperature of condensation. The vacuum guage, consequent- ly, which in the common condenser does not rise above 26 inches, has been maintained for days together in Hall's Con- denser at 29,5. In the engines to which Hall's Condenser has hitherto been adapted, steam of a tension little greater than a single atmo- sphere has been used. The air-pump has, therefore, sufficient power to pump the condensed water directly into the boiler. This it would not be able to do without much expenditure of the force of the engine, were steam of 2 or 3 atmospheres used, as is frequently the case in our American condensing engines. But their ordinary force-pump working in a hot cistern, would answer the purpose. In the use of this condenser the exact quantity of water which has passed through the engine in the form of steam is returned to the boiler at each stroke of the pump, and being obtained by the condensation of vapour, has the purity of distilled water. If, therefore, a boiler be filled at CONDENSING ENGINE. 141 first with pure water, no inconvenience can possibly arise from the accumulation of solid matter. Nay, even sea-water may be used without its becoming more injurious than at first. As there will be a waste arising from the escape of steam through the safety valves, a small distilling apparatus is added to the ordinary boilers ; so that no other water than what is obtained by condensation need be admitted into the boiler. It is obvious that this condenser is not only convenient, and capable of adding to the power of a given engine, but must be condu- cive to the safety of boilers in which there can be no deficiency of water as long as the engine remains in action. 97. The theory and use of the Parallel Motion, 1, 2, 3. 4, has already been explained — see pages 116, 117. The rule for one of its most usual forms is as follows, viz : The parallel bar is half the length of one arm of the working beam, or one-fourth of the distance between the two glands. The radius-bar is of the same length with the parallel-bar. The two pairs of straps are, of course, equal in length, and are usually three inches less between their centres than the length of the half stroke of the piston-rod. The centre, to which the air-pump is attached, is in the in- ner pair of straps, at the point where a line drawn from the ful- crum of the lever beam, to the upper end of the piston-rod/cuts the inner strap. See page 106. 98. The length of the lever-beam, in Watt's engines, is usually one and a half times the length of the stroke of the pis- ton-rod. The beam is usually cast in one piece. The centres of the parallel motion, pump rods, and connecting rod are turned out of rods of steel, and passed through the beam. In many modern engines the lever beam is a trussed frame of cast- iron, bound by a band of wrought-iron, and this is a most important improvement in the structure of the engine. 99. The air-pump rod u being attached to the inner pair of straps, is at a distance, from the centre of motion of the lever- baem, of one-fourth of the length of the latter. 142 DOUBLE-ACTING The rod of the cold-water pump is attached to the lever- beam, at an equal distance from the fulcrum on the opposite side. The rod of the hot-water pump is at a distance, from the ful- crum, of one-sixth of the length of the working-beam. The length of the connecting rod between the centres, in "Watt's engines, is twice the length of the stroke of the piston. In most American engines it is three times that length ; and less is lost by obliquity of action in the latter case. 100. The arm of the crank z. is half the length of the stroke of the piston in all cases where the arms of the lever-beam are of equal length. 101. The radius of the Fly- "Wheel may be various, accord- ing to the uses to which it is applied ; the motion for driving machinery ought to be taken off at a distance from its centre, equal to that of its centre of gyration. See r page 7. In the engine on PI. III., the fly-wheel has a radius equal to twice the whole length of the cylinder. The weight is calculated by the following rule : Multiply the number of horse's poicers of the engine by 2000, and divide by the square of the velocity of the circum- ference of the wheel per second, the quotient is the weight in cwts. The velocity of the circumference is readily found when the radius is given, for the crank has a velocity as much great- er than that of the piston-rod as the circumference ol a circle is greater than its diameter ; and the circumference of the fly- wheel has a velocity as much greater than the crank as the radius of the former is greater than that of the latter. In the first form of Watt's engines, the valves were opened and shut by apparatus of the same description with that which had been used in the more ancient forms. Tappets were at- tached to the rod of the air-pump, which, during the ascent and descent of the rod, acted upon levers with counterpoising weights. These levers were thus made to give a reciprocat- ing motion to toothed segm-nts, that acted upon racks attach- CONDENSING ENGINE. 143 ed to the valves, and thus opened and shut them. We shall return to this manner of working valves in the history of the engine. When conical valves are used, a spindle is attached to each, and the nozzles are immediately beneath each other. Thus the two spindles of each pair of valves are in the same vertical line ; the upper spindle in each pair is hollow, and the spindle of the lower valve passes through it. The weight of the valves is usually sufficient to close them, and keep them shut j if not, they are loaded until they shut themselves. In order to open them, the following arrangement is employed : — The spindles of the two valves, that are to act simultaneously, as, for instance, the steam valve of the upper pair and the condens- ing valve of the lower, are united by lifting rods, which have consequently the forms of three sides of a rectangle. These rods are pressed upwards at a particular part of the motion of the engine by pieces projecting from a horizontal shaft that has an oscillating motion. These projecting pieces or arms lie in the same plane, and on opposite sides of the shaft ; so that when one of them acts upon the rod that moves one pair of valves, and presses them upwards, the other ceases to act, and permits the other two to fall into their seats by their own weight. The return of this oscillating shaft permits the first pair of valves to shut, and causes the other piece, or cam, to act upon the two that were before shut. 102. This oscillating shaft is called the Tumbling Shaft. It receives its motion by means of a small crank that is attached to it, and moves upon it as an axis. This crank is connected with the axle of the fly-wheel by an apparatus called the Eccentric. This is represented in Figs. 1 and 2, on PI. III., and is shown separately in Fig. 4. A circular plate b has an opening of a circular shape cast in it, but having an eccentric position in respect to it. This last circle just fits the shaft of the fly-wheel, and is wedged firmly to it, so that the latter carries the circular plate b around with it in its revolutions. To the circular plate is fitted a circular ring c, within which 144 DOUBLE-ACTING it can turn, and which, therefore, need not receive any motion from it, but what arises from the eccentricity of the revolution of the plate. To this ring are attached two bars d, e, forming the sides of an isosceles triangle. These are united by frame- work. These bars terminate in a single piece, in the direction of a perpendicular to the base of a triangle, and which has a handle/ turned upon its extremity. The use of this handle is to lift the eccentric from its place, when it is wished to stop the engine, and return it again, when the engine is to be set in mo- tion. A notch of a semicircular figure is cut in the eccentric, which drops upon a pivot, turned upon the crank of the tu fi- ling shaft g. It will be obvious, that while the axis of the fly-wheel is car- ried around, and with it the circular plate, the end of the tri- angular frame will have an oscillating motion communicated to it, which the free motion of the ring c, will allow to be convert- ed into a reciprocating circular motion in the crank of the tum- bling shaft ; it will thus give the latter a motion suited to the opening of the valves, by means of the two arms that have been described. The engine upon the plate (PL III.) has, as has been des- cribed, a slide valve. This is set in motion in a manner differ- ent from the puppet valves. An arm, h, projects from each end of the tumbling shaft, and both are in one plane at right angles to its crank, g, f. To these arms are attached two light lifting rods, that rise above the side-pipe where they are united by a cross head. To the middle of this is attached the rod that moves the slide, and thus the latter is both raised and depressed by the action of the eccentric, while as we have seen, the puppet valves are raised only, and return to their seats by their own weight. 103. When an engine is used for purposes that occasionally require its motion to be reversed, two eccentrics may be employ- ed, that adaptthemselves to cranks situated at the opposite ends of tumbling-shaft, in planes at right angles to each other. Only one of these eccentrics is used at a time ; when it is necessary to reverse the motion of the engine, the piston is stopped at CONDENSING ENGINE. 145 half-stroke, or in the position represented in Figs. 1 and 2, on PL III. The eccentrics are then exchanged ; that before in use being raised, and the notch of the other dropped upon its crank. When the steam again flows, the piston will return in a direc- tion opposite to that in which it was proceeding when stopped. Another method that has been used in some English steam- boats, consists in cutting two notches opposite to each other in the rod of the eccentric ; the rectangular cranks of the tumbling- shaft are at the same end of the shaft ; the eccentric lies between them, and may be made to apply itself to either at pleasure, ^his arrangement has also been so modified as to be placed within the control of the helmsman of a steam-boat. The ne- cessity of communicating with the engineer by conventional signals is thus avoided. 104. It will be obvious that the time at which the valves open and shut may be determined by the position of the eccen- tric upon the shaft of the fly-wheel. This may be done, by this apparatus, far better than it can be by tappets upon the rod of the air-pump, or, as they are usually called, a plug-frame, This determination is of no small importance to the working of an engine. Should the piston be impelled by the steam to the very end of its stroke, a violent blow will take place between it and the head or bottom of the Cylinder ; while, on the other hand, if the steam- valves be opened too soon, a part of it will be expended in diminishing the action of the steam on the oppo- site side of the piston. In both cases power will be wasted, and the lost power will be exerted to injure the apparatus. In put- ting up an engine, the position of the eccentric is determined by actual trial, and the eccentric is left in the position where it is found to tend most to the equable and regular working of the engine. 105. A band passing over a drum on the axis of the fly- wheel turns a second drum, which is upon the axis of a bevel wheel. This bevel wheel gives a motion to another, that car- ries upon its axis the Governor. This arrangement is repre- sented upon the plate, but could not be distinguished by letters. 19 146 DOUBLE-ACTING This governor, as has been stated, is a conical pendulum. The weights revolve in the same plane, which is raised by their centrifugal force when the velocity increases, and falls as the velocity of rotation diminishes. The theory of this instrument shows that its revolutions are half the number that would be performed by a pendulum, the length of which is equal to the distance of the plane in which the centres of the balls revolve, from the point where the bars, by which they are suspended, cross each other. Thus, then, if the least and greatest number of the revolutions that it is intended that the fly-wheel shall perform, in a given time, be known, it will be easy to calculate the length of the conical pendulum. 106. The rods that bear the balls of the governor are united by pivots to two others, also connected by pivots, and sliding at their point of union upon the axis of the governor. The pa- rallelogram that is thus formed, is sometimes above the joint whence the balls hang, as in the horizontal engine on Plate VI, and sometimes below it, as in the high pressure enghr on Plate Vj or the separate figure of the governor on Plate IV, Fig. 2. In either case it gives motion to a lever, which acts at its opposite end upon a rod that moves the handle or lever of the throttle valve. This system of levers is so arranged that the throttle valve is opened to its utmost limit, when the balls of the governor are in their lowest position, and is wholly clos- ed, when they have been thrown out to their greatest extent by the centrifugal force. In the first case, therefore, all the steam that is generated, flows to the engine; in the last, it is wholly cut off. 107. The form of double-acting condensing engine which we have thus described, is that which is most commonly used in manufactures, particularly in Europe. It cannot, however, fail to have been remarked that it contains, at least, one part by no means necessary to its action : this is, the lever beam. The engine, as we shall see hereafter, was originally applied to the single action of pumping water, and in this the pump-rod, or brake, was conceived to be essential j this, when made with CONDENSING ENGINE. 147 equal arms, became the lever beam. Successive advances to- wards perfection in the structure were made, as improvements on the original plan, and not as original inventions. It has thus happened that an unnecessary and cumbrous part of the appa- ratus has been perpetuated. A far more simple form of the engine, and which is in many cases preferable, is that which was used by Fulton in his steam-boats, and of which one is represented on Plate VII It will be at once seen by inspec- tion, that in this engine the beam is suppressed, together with the parallel motion. Asa substitute for these parts, a cross-head a is adapted to the upper extremity of the piston-rod, b ; this works between vertical guides, a, a ; it is connected to the two cranks, c, c, by the two connecting rods, b b, b b, and to these is joined, in the case before us, the axis of the water wheels, d d ; the axis of the fly-wheel might in like manner be turned by these cranks, were it intended to apply the engine to general purposes. The pumps are worked by a beam, e e, far lighter than it need be in the other form of the engine, and but half the length. It is forked at the end nearest the cylinder, which it thus embra- ces, and is connected with the cross-head A of the piston-rod B, by the connecting rods, d d. The peculiarities in this engine, which adapt it to a steam-boat, will be described in another place. 10S. When a steam-engine is to be set in motion, the boiler must first be filled with water by hand, the fire lighted, and the steam raised to the proper tension. The steam and side-pipes, the Cylinder, condenser, and air-pump, will be full of air, and the whole will be cold. The air must be extracted, and the en- gine heated up to the temperature corresponding to the tension of the steam, before it can be set to work. This is done by what is technically called blowing through the engine. All the valves are opened simultaneously by hand, and steam is thus introduced to all the parts. As steam is lighter than air, it will force the air from the cylinder towards the condenser. Hence the air is sometimes allowed to escape by a valve contrived for the purpose ; this is usually adapted to the con- 148 DOUBLE-ACTING CONDENSING ENGINE. denser, by means of a pipe forming an elbow, and bent verti- cally upwards. This pipe is closed by a conical valve opening upwards. So long as air remains in the condenser, and is com- pressd by the steam from above, it is capable of making its way through this valve. The completion of the operation is shown by its beins: followed by steam, which, when this valve is situ- ated beneath the level of the water in the cold water cistern, is known by a slight crackling noise. It is, however, more usual in this country to suppress the valve on the side of the con- denser, or snifting-valve ; in this case the air makes its way through the air-pump, and is discharged at the clack-valve. When the steam thus shows itself, the injection cock is opened, a condensation of the steam in the condenser takes place almost instantly, and the pressure of the steam from the boiler becomes in a short time sufficient to put the engine in motion. The ec- centric is now applied to the crank of the tumbling shaft, and the engine becomes self-acting. In engines with slide valves, a simultaneous communica- tion cannot be made between the boiler and the two sides of the piston. An additional valve is therefore provided, making a communication between the lower end of the side pipe and the boiler. This is called the Blow-valve. It is opened by hand, and closed as soon as the engine is ready to work. This valve is to be seen on PL III. To set a large engine in action has hitherto been a very la- borious operation, whether the slide or puppet valve be used. This difficulty was noticed by Trevithick, who contrived a dou- ble-seated valve, which required much less labour to work it. The most perfect construction of this kind is that brought into use by Mr. Adam Hall of New- York. In this the valves are so nicely balanced that a single man is able to blow through the most powerful engine. CHAPTER VI. GENERAL VIEW OF CONDENSING ENGINES ACTING EXPAN- SIVELY, OP HIGH- PRESSURE, SINGLE-ACTING, AND ATMOS- PHERIC ENGINES, PARTICULAR DESCRIPTION OP HIGH PRES8URE ENGINES. Regulation of steam by the valves of Condensing Engines. — Expansive force of steam, supposing the temperature to remain constant. — Expansive force of steam of a given tension, and in a given engine, on the same hypothesis. — Expansive action of steam of a given tension and constant temperature, when the friction and resistance is taken into view. — Expansive action at increasing tensions, and icith temperatures varying according to the laxo of specific heat. — Effects of steam acting expansively, as usually employ- ed. — Action of high pressure steam when not condensed. — Cases in which high pressure engines are useful. — Recon- sideration of the precautions to be used in boilers generat- ing high steam. — General view of the high pressure en- gine, its steam pipes, side pipes, and valves. — Calculation of the poiver of high pressure engines, their working beam, parallel motion, throttle-valve, governor, and forcing pump. — General view of the single-acting condensing atmosphe- ric engines. — Particular description of a high pressure en- gine, with a beam, and of long and short slide valves. — Particular description of a horizontal high pressure en- gine. — Description of a rotary engine. 109. To set the Double Condensing Engine into motion, two of its valves must be opened. One of these admits steam 150 CONDENSING ENGINES from the boiler, to act upon one side of the piston, while the other lets the steam from the opposite side pass into the con- denser. These two valves are united so as to open and shut together, as are the two which, alternating with them, give mo- tion to the piston in the opposite direction. These valves re- quire a certain space of time to open to their full extent, and thus the motion of the piston in the first instance, and the change at each successive alternation, are effected gradually. So also the valves are permitted to close before the engine has reached the limits of its stroke, and thus the shock the engine would sustain, and the consequent loss of power, are in some measure obviated. This may, obviously, be effected still more certainly, by cut- ting off the steam at an earlier period of the motion of the pis- ton, while the communication with the condenser is still left open. 110. "When the steam is cut off, it does not lose its whole power, nor does it lessen suddenly in force ; for, being elastic, and acting against a partial vacuum in the condenser, it will expand, until it either fill the Cylinder, or until the friction and the resistance of the partial vacuum in the condenser, be- come equivalent to its own expansive force. "Watt, to whom we owe the double-condensing engine, was the first to remark that advantage might be taken of this to increase the effect of a given quantity of steam. Thus, if the Cylinder be but par- tially filled, and the steam then cut off, it will still act expan- sively, and all the force that it continues to exert is so much gained. Were the decrease of the temperature, arising from the change in the relation of the steam to specific heat, left out of view, the force of the expanding steam would decrease in a geometric progression, and might be calculated by means of ta- bles of hyperbolic logarithms. Calculated in this way. the power of a given quantity of steam would be increased in the ratios given on next page. ACTING EXPANSIVELY. 151 Cylinder filled. Power of Steam. Wholly 1. One-half 1.69 One-third 2.10 One- fourth 2.39 One-fifth .... 2.61 One- sixth 2.79 One-seventh - 2.95 One-eighth 3.08 111. The advantages of using the steam expansively would, therefore, according to this hypothesis, be very remarkable ; but to obtain them would require an entire remodelling of the engine and the alteration of its proportions. To use the same quantity of steam, it would be necessary that the steam pipes, the nozzles, and the Cylinder itself, should all be increas- ed in the ratio which the part of the Cylinder filled bears to the whole. If these remain unchanged, the consumption of steam, (supposing the temperature to remain constant,) would be lessened in the same ratio inverted, and the force with which the steam would act upon the piston, would have the following ratio : Cylinder filled. Wholly .... One-half One-third - One-fourth - One-fifth .... One-sixth ... One-seventh - One-eisihth - 3 Force. 1.00 Steam expended. 1 0.84 JL !2 0.70 JL 3 0.57 JL 4 0.52 JL 5 0.46 JL 6 0.42 JL 7 0.39 JL 8 These calculations are, as has been stated, made upon the hy- pothesis, that the temperature continues invariable, which is far from being the case ; for steam, like all other substances, has its capacity for specific heat increased during its expan- sion, and its temperature and consequent elasticity are dimin- ished. 152 CONDENSING ENGINES 112. It must next be taken into view, that the absolute power of the steam is not all exerted ; for steam, as has been seen, act- ing with an expansive force of 17 jr lbs. per square inch, is only capable of overcoming a resistance equivalent to lOlbs. Hence, in an engine working at low pressure, the advantage gained by- making it act expansively, would cease if the steam were cut off earlier than that at half the stroke, for at -ffa the resistan- ces would be equal to the expansive force, even if the tempera- ture remained constant, which, as we have seen, it does not. The motion might, indeed, be kept up for a time by the fly- wheel ; but even then, without taking into view the irregulari- ties that would ensue, the effective action would diminish most rapidly, as will appear from the following calculated results : Cylinder filled with steam Mean effective Force, of 17 l-21bs. Wholly 1.00 One-half - - - - - 0.72 One-third - - - - 0.48 One-fourth 0.26 Ooe-fifth 0.17 One-sixth 0.06 One-seventh - 0.00 We therefore conceive ourselves warranted in the conclusion, that when an engine acts expansively, the steam should never be permitted to expand itself to more than twice the bulk it oc- cupies under the atmospheric pressure. Working at low pressure, in order to produce an equal effect, the engine should be nearly one-half larger in its capacity, and the expense of fuel would be three-fourths of what it would be if the steam were employed in the usual manner. Unless, therefore, in cases where fuel is extremely scarce, there is pro- bably no real advantage to be gained in making low pressure steam act expansively. . There is another point of view in which the expansive action of steam may be investigated, for the steam may be used at an increased pressure. If it have an expansive force of an 113 action ACTING EXPANSIVELY. 153 atmosphere and a half, it would, if cut off at one-third of the stroke, expand, in filling the Cylinder, to the assumed limit of pressure, of half an atmosphere. The original effective force, after allowing for resistance, would be l5lbs. per square inch, and the mean action would be two-thirds of that amount, or lOlbs. per inch during the whole stroke. Hence the engine would now work up to its nominal power. Steam, under a pressure of 1J atmospheres, has, if we leave out of view the temperature, a density one and a half times as great as under atmospheric pressure simply ; hence, to fill one- third of the Cylinder would require the evaporation of as much water as would fill half the cylinder with steam of 212°. An en- gine, therefore, acting expansively with steam of the elasticity of H atmospheres, would, on this hypothesis, do the same work as when acting in the common manner, and consume but half the quantity of fuel. For, as the sum of the latent and sensible heat is the same, both in high and low steam, the quantity of water converted into steam is the same, whatever be its temperature. Let us next suppose the steam to have an elastic force equal to two atmospheres. It might, on the same hypothesis, expand to four times its original bulk before its elasticity became less than half an atmosphere. Hence it might be cut off at one- fourth of the stroke. Its original effective force, after deducting the constant resist- ance, will be 22^ lbs. per square inch, and it will act with a mean force of -f- 6 \ of that amount, or upwards of 12£lbs. Hence the engine will, under such circumstances, work with one-fourth more than the power at which it would be estimat- ed according to the common rule. The steam would in this case also fill half the cylinder be- fore it reached the density of steam of 212°, and hence the quantity of water used, and fuel expended, would be the same as in the former. And, in all cases where the limit of the expansion of the steam is an elasticity equal to half an atmos- phere, the quantity of water evaporated and fuel expended would be constant. But the effective power would go on in- creasing with the elasticity of the steam, according to the fol- io wing table : 20 154 CONDENSING ENGINE Relative power of the same engine acting in the ordinary manner, or expansively. The te in perature being si not to vary on expansion. | Cylinder Fuel .-..:- 1 r-s:-i:e_ :llv 1 -L 3 :..? A - ).5 A :.5 A f ).s - u 4. ..5 E^-:::ve ::ce. 114. In order to cut off the a valve of ihe figare of a throttle valve is placed iu the steam pipe. A weight, or strong spring closes it and keeps it shut, except when the one is lifted, or the other forced back, by the action of the engine. The :s usually performed by placing two teeth or cams, of proper form and size, upon the axis : : the crank. A plan of this l-:ind may be seen on PI. IV. Fig'. 4. where cis the axis of the crank. a and b two cams, or teeth, that act upon the sprit g g d. which is connected with the handle cfoi the expansion valve, by the rod d e. F is a portion of the steam pipe. A very ingenious and simple mode of working acut-orT valve has been invented by Perkins, and applied to his expensive en- gine. He places an additional eccentric upon the shaft of his rlv-wheel. to this is attached a jointed rod directed The end of this rod acts upon the lever of the valve, and by the adjustment of the length of the rod it may be made to act for a longer or shorter time. When the rod ceases to press the valve, a strong spring applied to it causes it to closr. An addition has been made to the short slide valve, by which the steam may be cut off at any part of the stroke of the en- gine. A valve of the usual form is surrounded by a frame com- posed of two plates, each of sufficient surface to cover the ACTING EXPANSIVELY. 155 steam passages. These plates are united by bars, which are pressed down by two strong springs. The eccentric is made to act upon a rod passing through a collar in an axle. To this collar it is adjusted by a screw in such a manner that the two arms of the rod may be made at pleasure to have different relations to each other. The space through which the eccentric moves one end of the rod being constant, the opposite end may be made to pass through differ- ent spaces according to the position of the point in the rod, which is made by the screw the axis of motion. It will there- fore be easily seen that the valve may in its motion be either made to strike the frame or not, at pleasure. In the latter case the steam will not be cut off, and by varying the motion in the former case, it may be cut off at any required point in the mo- tion of the piston. 115. The estimate that has been given of the powers of steam acting expansively, is, as has been seen, formed upon the hypo- thesis that it expands to bulks that are inversely as the pressures. This is not the case, in consequence of the change of tempera- ture that the very act of expansion produces. Thus, the steam of a tension equivalent to half an atmosphere, has a tempera- ture of 180° and a density of 0.00032 ; while, with a tension of 2, 3, and 4 atmospheres, it has the following densities : 2 Atmospheres 0,00111 3 do. - 0.00160 4 do. - 0.00210 Steam of 2 Amospheres expanding to 4 times its bulk, has a density of 0.00028 3 do. to 6 times its bulk, - 0.00027 4 do. to 8 times its bulk, - 0.00026 In a vessel which would neither give nor abstract heat, the tension and temperatures would be diminished, in the three several cases, in the ratios of g-f or f , ff, and ff or ff . But in an expansive engine, the cylinder may be readily kept up to the temperature of the steam before it begins to expand, and the 156 CONDENSING ENGINE steam in expanding will derive heat from it. The method, which is occasionlly adopted in a low pressure engine, of enclo- sing the cylinder in an outer case, called a jacket, will be far more beneficial in an engine acting expansively, and the dimi- nution in tension, arising from diminished density, will be counteracted by increased heat. This, however, will be attend- ed with a loss of heat in the surrounding steam, and will require the capacity of the boiler to be increased in proportion. It is, therefore, to be taken into view, that the comparison of engines acting expansively, as given on page 154, is not absolutely true ; but that, in order to make it so, the fire surface of the boiler should be increased ^th at the pressure of two atmospheres, and £th at the pressure of four, and the safety valve loaded with ad- ditional weight in the same proportion. The expenditure of fuel will also be increased in the same degree. The advanta- ges derived from making engines act expansively are still great, notwithstanding this increase in the expenditure of fuel ; for an engine receiving steam of the tension of 4.f atmospheres, cut off at |th of the stroke, will do twice as much work as one re- ceiving low steam, with but six-tenths of the fuel it expends. If we correct our previous calculations upon these principles, the results will be as follows, which will give the actual effect which may be produced by the same engine, acting at low pres- sure or expansively, with different loads on the safety valve. Relative powers of the same engine acting at low pressure or expansively, the change in the relations of the expanding steam to temperature being taken into account. Ir^ Fuel Effective Expended. Force. 1 10 0.55 10 0.56 12* 0.57 15£ 0.58 18 0.59 19 0.60 20 ACTING EXPANSIVELY. 157 This table is, however, far from exhibiting the whole advan- tage of which the method of cutting off the steam before it has filled the cylinder is capable. It will be easily seen that our calculations would be only adapted to the case of a diminu- tion in the fire surface of the boiler, and that in practice a dif- ferent course would be pursued ; — the same quantity of water would still continue to be evaporated, and the same amount of fuel expended, unless the whole system were changed. Let us then suppose that with a given engine and boiler, the steam is cut off at different portions of the stroke. If cut off at half stroke, the density of the steam will be doubled ; and if at one third, tripled, and so on. The tension of the steam will be in- creased even in a higher ratio, for steam of two atmospheres has a density of no more than 0.00110, while twice the density of steam of the tension of a single atmosphere is 0.00118. The usual pressure in the condensing engine also exceeds an atmos- phere by one-sixth, and the tension obtained by cutting off will be a multiple of this instead of one of a single atmosphere. We shall, however, neglect this in our view of the comparative effects of an engine working expansively,Vith steam of different tensions. Relative powers of an engine using the same quantity of fuel, and acting expansively at different tensions. Force in Atmoi spheres. Cylinder filled. Effective force. 1* - - - wholly - - - 10 2 - - - JL 2 - - 10.75 3 _ - - JL 3 - - 27.5 4 m - - JL 4 - - 35.6 5 . - . _L 5 - • 43.5 6 - - - JL 6 - - 51. It will therefore appear that, without any change in the ge- neral distribution and plan of an engine, provided the boiler be strong enough to bear the increased force of the steam, its pow- er may be readily increased, five-fold. This will be done with- out using steam of a temperature higher than is frequently employed in engines of a different structure. 15S CONDENSING ENGINE # It is more usual to cut off the steam at half stroke, and to de- pend, for au increase of force, upon an increased capacity of the boiler to generate steam. This method is, however, disad- vantageous, as it will require au alteration in the boiler, other than an increase of its strength ; and will, besides, demand a more powerful apparatus, and larger supply of cold water for keeping up the vacuum of the condenser. Nor does it give re- sults near as satisfactory as the mode to which we have just re- ferred, if the expenditure of fuel be taken into account, as will be perceived from the following table : Relative force of steam used expansively in a cylinder of constant dimensions, and always cut off at half-stroke. e in atmo 5|* eriis. Fue: [ expended. Effective force. Force with the eame fuel. 2 - - 1 - - IS. lb - 18.75 3 - - H - - 32 - - 21.67 4 - - 2 - - 45 - - 22.5 5 - - H - - 53 - - 23 6 a . 3 . . 72 . . 24 116. It may therefore be inferred, that the best mode of using the double acting condensing-engine, is to make it of the usual form and dimensions, and give it a boiler of sufficient strength, with a fire surface of the usual extent ; but to cut off the steam at as early a period of the stroke as may be considered safe. This method has been brought to the test of actual experiment in the pumping engines employed in the mines of Cornwall, and by its use, the power of an engine of a certain nominal horse power has been increased five-fold. The method of cutting off at half-stroke has been more es- pecially used in the steam-boats of this country, and the tension of the steam has been raised by increasing the fire surface of the boiler. The last object has been effected by a variety of artifices. It may, however, be fairly inferred, that the method of cutting off at such part of the stroke as corresponds to the desired increase in the tension of the steam is much preferable. High as our estimate of the advantages of using steam expan- sively may appear, it is, notwithstanding, far less than those of ACTING EXPANSIVELY. 159 Watt and Woolf. The former hazarded the opinion that steam of 4lbs. was capable of expanding itself to 4 times its bulk, and still retaining the tension of an atmosphere. Woolf seized this expression as the basis of his calculations, and inferred that steam of 5, 6, 7, 8, &c. lbs, was capable of expanding as many times as the unit of measure of the safety valve was loaded with pounds. These views are wholly erroneous, and are contrary to the physical and mechanical properties of steam. Our own reduced estimates are more to be relied upon, and offer sufficient inducements for the employ of the expansive action of steam. Our calculations in respect to the increase of power gained by expansive action have reference, as will be at once seen, to a constant velocity in the working point of the engine. It may, however, happen that the resistance is constant, or increases with the velocity only ; and that the increase in the power arising from expansive action, is applied to an increase in the velocity. Analogous advantages will be gained in this case, which is that of steam navigation. 116 b. It will easily be seen, from what has been stated in relation to the expansive action of steam in the condensing engine, and from what we shall in relation to the increase obtained in the force by using steam of great elastic force in the high pressure engine, that a given engine may be made to work far beyond its nominal power. The horse power is, in either case, estimated from the area of the piston, the height and velocity of its stroke, and the pressure taken at the amount which has hitherto been most frequently used. Thus, in the condensing engine, the pressure is usually estimated, after the resistances are allowed for, at lOlbs. per square inch ; and in high pressure engines, at 401bs. In the former engines, by increasing the tension of the steam in the boiler, and cutting it off in such a manner as to allow it to act by its expansive force, we have seen that the force given by the combustion of a given quantity of fuel, may be increased more than three-fold, and the action of a given engine doubled. A still greater effect may be produced; by using steam of higher tension than such as, in its expansion, will diminish to the limit 160 CONDENSING ENGINE we have assumed in Chap. VI. In addition, then, to the esti- mate in horse powers, which has now become of no other use than a mode of describing the size of an engine, in contracts between the maker and purchaser, it has become customary to compare the work of engines with each other, by a mode of es- timate which is called their Duty. The mode in which the duty of a steam engine is estimated is in the numbers of pounds which can be raised 1 foot high by the combustion of a single bushel of coals. We have seen that this quantity of coal is ca- pable of evaporating 12 cubic feet of water, and therefore of keeping an engine of twelve horsepower in action for an hour. It ought, therefore,, according to the estimate we have just made, raise to a height of 1 foot 2-iu00x 60 x 12=17. 250000ibs. or upwards of seventeen millions of pounds. Watt and Boul- ton constructed an engine,, whose duty reached as high as 19 millions ; and it was said that their own engine at Soho, did work equivalent to a duty of 21. 600000. bs. ; but. on an exami- nation, in legal form, of all the engines they had put up in Corn- wall, two years before the expiration of their patent, it was found that the average duty was no more than 17 millions, or in strict conformity with our estimate. Many of these engines acted expansively ; and one performed a duty of 27 millions, in spite of which the average fell to the limit we have stated. The expiration of Watts' patent left engineers free to make such improvments as experience or science might suggest. The expansive action of steam was the improvement which was prin- cipally relied upon : and, in order to obtain from it the greatest practicable advantage, for the old boilers of Watt, such as are figured on Pi. I. were gradually substituted cylindric boilers ca- pable of bearing steam of o;reat tension. In this way the force of the steam has been gradually raised from little more than a single atmosphere to 10, and an intelligent Cornish engineer states that he has seen it raised as high as 20 or 30 atmospheres. In this way the average duty has been regularly on the increase, being in 1S33, 19£ millions ; in 1814, 20£ millions ;. in 1515. the same : in 15L6. nearly 23 millions ; in 1817, 26£ millions ; in ISIS. 25J millions ; in 1819, 26* millions ; in 1820, 28| ACTING EXPANSIVELY. 161 millions ; in 1825,32 millions; in 1828, 37 millions; in 1829 41 millions ; in 1830, 43^ millions. During this time, single engines have performed far more than the average, and in the year 1835, one has reached a duty of 94 millions. 117. When steam of high pressure is used to propel engines, it is frequently made to act without the aid of a condenser, and consequently in opposition to the whole pressure of an atmos- phere. The engine, in this case, becomes much more simple, inas- much as the condenser and air-pump may be dispensed with, as well as the cold and hot water pumps ; but for the latter is substituted a forcing pump to feed the boiler, and in most cases a common pump will be needed, to raise the supply. The cold water cistern, and the water for condensation are no longer necessary, and thus a very great weight may be saved, which in some cases is of great importance. In estimating the resistances which the action of the steam meets with, it is to be considered, that the imperfection of the vacuum of a condensing engine merges in the pressure of the atmosphere, in one where the steam is not condensed ; and that thus the resistances, which, in the former, were estimated at 7^-lbs. per square inch, may be diminished, as well as by the power required to work the air and cold-water pumps. The resistances, other than the pressure of the atmosphere, need not therefore be taken at more than 5 lbs. per square inch, which, added to the pressure of the atmosphere, makes a constant re- sistance to the action of the steam, in a high pressure engine of 201bs. per square inch. Hence, steam of an expansive force of two atmospheres will work in a given cylinder with the same force that steam of 17^-lbs. would work in a condensing engine. But steam under a pressure of two atmospheres has rather less than two-thirds of the density of steam of 17^-lbs. per inch ; and hence it would require more than l£ times as much water to be evaporated in order to fill the cylinder, and l£ times as much fuel. In this case, therefore, there would be a loss of 50 per cent, in using a high pressure engine. . If the steam had a pressure of 2$ atmospheres, its effective 21 162 HIGH PRESSURE ENGINES. force would be 17flbs. per square inch, or would bear, to that in a low pressure condensing engine, the ratio of 7 to 1. But, to fill the Cylinder with steam of corresponding density, would require nearly twice as much fuel. At a pressure of three atmospheres the effective power of the steam becomes 25lbs. per square inch, or bears to that of a low pressure engine the ratio of 5 : 2. To fill the cylinder with steam of this density, requires fuel in about the same ratio ; and hence, at this limit, the power of high and low pressure en- gines, consuming the same quantity of fuel, becomes nearly equal. With four atmospheres of steam, the effective pressure be- comes 401bs., the consumption of fuel is about three to one ; and here the high pressure engine has an advantage in the ra- tio of four to three. At five atmospheres, the effective pressure is 551bs. per inch, the ratio of water evaporated or fuel consumed as 34-ths to 1. Arranging these and similar calculations in a table, we have as follows : Effect of High Pressure Steam to work Engines. P.'Tfsupe in | Atmospheres. Fuel in toe same Engine. Force in the same Engine. Force with the same Fuel. 2 H 1 75 H 2 1.75 0.S75 3 2£ 2.5 1.000 4 3" 4 1.333 5 34 5.5 1.46 6 4J 7 1.55 10 7 13 1.S6 20 14 25 2.00 30 20 43 2.15 40 26 55 2.23 US. It thus appears that the useful effect of high pressure engines increases far more slowly than the increase of the elastic force of the steam. This arises from the fact, that the density of steam increases nearly as fast as the pressure under which it HIGH PRESSURE ENGINES. 163 is generated. Did both increase in the same ratio, there would be nothing gained by the use of high steam. A high pressure engine is, therefore, far inferior in power to one in which the steam acts expansively, and is subsequently condensed ; as will appear from a comparison of the preceding table with that on page 154. There are, however, cases in which the high pressure en- gine is preferable to any other. Thus, when water is scarce, the high pressure engine dispenses with the use of that em- ployed in condensing the steam, which, as we have seen, is 22 times as great as that which is evaporated from the boiler. The weight of the air-pump and condenser, of the cold and hot water cisterns, as well as of the water they contain, are all saved. Hence, where locomotion is important, as where steam is employed to propel carriages upon railways, high pressure engines can alone be used. These engines are also much sim- pler in their construction, being composed of fewer parts ; and they occupy far less room than condensing engines, whether the latter act expansively or not. Advantages similar to those obtained in the condensing en- gine may be obtained by permitting the steam to act expan- sively in the high pressure engine. The obstacle to be sur- mounted in this case, is the danger which may be feared, from increasing the tension of the steam to so a high degree as would be necessary to obtain important results. 119. Whether an engine be constructed to receive the most important advantages from the expansion of the steam, or be a simple high pressure engine, in which the steam, after it has caused the piston to perform its motion, is permitted to escape into the open air, the boiler must be so constructed as to con- tain and generate steam of high elastic force. Common high pressure engines work usually with steam of from five to six atmospheres, and there is no doubt that expansive engines might be constructed in such a manner as to be worked advan- tageously with a little less than five atmospheres. The load of the safety valve is at this latter limit 571bs., while to contain steam of six atmospheres, requires a load of 751bs. per square 164 HIGH PRESSURE ENGINES. inch. It remains to inquire, how far it may be consistent with safety to employ steam of such expansive force ? The prin- ciples on which the strength of boilers depends, have been fully illustrated in Chapter III. From what has there been stated, it wilj appear that the cylinder is the best form for boilers, and that a boiler of this shape, and of small diameter, may be made to resist the regular pressure of far more than six atmos- pheres ; that by diminishing the diameter of the cylinder the strength is increased in the inverse ratio of the squares of the diameters ; and that by a reduction in this dimension, any re- quired strength may be obtained. The application of the Hy- drostatic press furnishes a proof, in the first instance, of the co- hesive force of the material and the joints, to resist any given pressure ; and the proof may be finally completed by subjecting the boiler to the action of steam of more elastic force than it is ever likely to be compelled to bear in practice. The steam- guage will enable the engineer to know that the pressure is kept below the desired limit, and the safety valve opens as soon as that limit is reached. In case of a deficiency in the supply of water, or obstruction in the feeding apparatus, a thermometer will show the increased heat that is the consequence, and plates of fusible metal will melt as soon as a safe limit of heat is passed. A self-acting feeding apparatus will generally furnish a regular supply, for the failure of which the last-mentioned apparatus affords a safeguard. Registers and dampers will allow the fire to be moderated, and almost extinguished, whenever it becomes necessary. Next, the tendency of solid matter to collect and be deposited on the bottom of the boiler, may be lessened by mixing vegetable feculse with the water ; but careful cleansing will be required, at proper intervals, to obviate all danger from this cause. If an engine must be in constant operation, it ought never to have less than two boilers, one of which can be employed while the other is under repair ; and in all cases of constant employment, there should be one boiler more than is necessary to supply the engine. If the work be of such a nature, that the persons employed about the engine may have a tempta- tion to increase the force of the steam beyond the proper de- HIGH PRESSURE ENGINES. 165 gree, there should be two safety valves, one of which should be beyond their control. No one of these precautions should be omitted when high steam is used, unless they are impossible from circumstances. In locomotive engines, and in steam-boats, spare boilers cannot be introduced ; the necessity for them, may, however, be done away, by prescribing stated periods of inactivity, when the boil- ers may be cleansed. With proper precautions, we do not hesitate to say that boilers in which steam is generated of no greater tension than is ne- cessary for giving the condensing engine its full power by ex- pansive action, may be rendered as little liable to accident as low pressure boilers ; and, indeed, the more common cause of explo- sion, namely, the exposure of the metallic flues, or the sides of boilers, to the fire, when not covered with water, is as likely to affect low pressure boilers as high. Of the two fatal explosions that have occurred in the harbour of New-York, one was a cop- per boiler containing low, the other an iron one containing high steam. But although, with proper precautions, high pressure boilers may be rendered as little liable to burst as those planned for generating low steam ; the explosions of the former, when they do take place, are more likely to produce dangerous con- sequences than the latter. We have ourselves been in two instances in steam-boats when low pressure boilers have given way, and the fact was only known by the stopping of the en- gine. This will always be the case when they give way un- der the ordinary pressure of the steam, for, supposing the safety valve to be loaded with 31bs. per inch, the escape of no more than a fifth part of the steam will restore the equilibrium be- tween the outer and inner sides of the boiler. Even if the boiler burst at the limit of its proof, the quantity of steam that can escape is little more than the half of that it contains in it. In boilers containing steam of four or five atmospheres, a rent will allow the steam to expand itself to four or five times its original bulk, even when it takes place under ordinary cir- cumstances ; while if it occur at the limit of the proof, in con- sequence of the safety valve ceasing to act, the steam may have a tendency to expand itself to ten or twelve times its original 166 HIGH PRESSURE ENGINES. bulk, and even in the former case the explosion may be dan- gerous. In a low pressure engine, then, any dangerous explo- sion that can occur, grows out of the subsidence of the water below its proper level, and the very weakness of its material is a cause of safety. While in a high pressure boiler, the same risk is incurred, and, in addition, a giving way, even under the usual state of the steam, may sometimes be dangerous. A boiler, however, that has been properly proved, and is ex- amined at regular stated periods, cannot well burst, except by the clogging of its safety valve, or by the uncovering of its sides and flues. The former accident may be considered as hardly within the limit of possibility, if the guages of the engine are in order, and the engineer attentive to his duty ; and as both species of boiler are equally liable to the latter, we conceive that it may be considered as certain that no more risk is now incurred by using high pressure steam than by using low. In expressing this opinion, it may be repeated that proper safe- ty apparatus must be applied to the high pressure boiler, and that in steam-boats and locomotive engines there should be an ad- ditional safety valve, beyond the control of any person on board the former, or entrusted with the management of the lat- ter. Moreover, the practice which is said to prevail on the Mississippi, of using steam of 10, 15, or even 25 atmospheres, is to be reprobated. 120. Such being our views of the possibility of using high steam with safety, cylindrical boilers, generating high steam, are rapidly superseding all others. They are applied in most cases to condensing engines acting expansively ; but where sav- ing of weight or of room, and even of original cost, is an object ; in locomotive carriages ; in boats navigating shallow rivers ; and wherever water is scarce : high pressure engines will be employed. Being less complex, they will also be preferred wherever good workmen to perform the repairs, or intelligent engineers, are not to be obtained. Upon the laud, such boilers should be simple cylinders, having the fire-place and flue be- neath them. But in steam- boats and locomotive engines, inter- nal furnaces and flues are indispensable. HIGH PRESSURE ENGINES. 167 121. The parts of a high pressure engine are the following : A steam-pipe through which the steam passes from the boiler ; the area of this is calculated upon the principles laid down on page 89. Side pipes connected at one extremity with the steam-pipe, and at the other with the open air. In these are situated the valves which admit steam alternately to the opposite sides of the piston, or permit its escape. The original form of the valve was a cock with two passages and four openings, proposed at first by Leupold, and adopted both by Trevithick and Evans. This valve is represented on the next page. M N is the cylinder of the engine ; a b the side-pipe, in the middle of which is a conical socket, to which is adapted a frustum of a cone, with two passages d e, and consequently four openings ; c is the steam pipe, and fg the exhaust-pipe, or in a condensing engine the communication with the condenser ; h is the lever by which the valve is turned through a quadrant at each stroke of the piston ; and i is the other position of the lever. The steam, in the position of the valve that is represented on the drawing, flows from the pipe c through the passage e into the lower part of the side pipe b, and thence enters beneath the piston ; while the steam above the piston flows out at a through the passage d into the pipe/g*; in the other position of the valve, the motion of the steam is obviously reversed. , 168 HIGH PRESSURE ENGINES. For this, in almost all high pressure engines, has been substi- tuted a short slide valve, which has been found by experience to be more advantageous. This valve is worked by an Eccen- tric placed upon the axis of the crank; a convenient mode of doing this is represented upon PL IV. at Fig. 3, where, A is the axis of the crank ; B the circular plate, a a a a triangular frame ; b rod and adjusting screw ; c handle, d arm of tumbling shaft ; e axis of tumbling shaft ; //spindle of slide valve; HIGH PRESSURE ENGINES. 169 g g steam chest ; H steam pipe ; I eduction, or exhaust pipe ; k k bottom of cylinder, on which is cast a piece 1 1 containing one of the steam passages. A plan and section of this valve are represented on PI. II. Fig. 5, and will be described hereafter. The Cylinder resembles in form that of a condensing engine, fitted with a piston, and piston-rod passing steam-tight, through the cover of the engine. 122. The power of the engine is calculated by taking the continued product of : the effective pressure of the steam per square inch in lbs., the area of the piston, the length of stroke in feet, and the number of strokes per minute ; which product divided by 33000 gives the number of horse powers. The effective pressure of the steam is 201bs. less than its absolute expansive force, or 51bs. less per inch than the load of the safety valve. It is, however, usually calculated at two-thirds of the pressure of the steam, which is the true amount when the steam is equivalent to 4 atmospheres, or the safety valve is loaded with 451bs. per square inch. 123. The action of the piston may be conveyed to the work- ing points of the engine in the same manner as in the condens- ing engine. All, therefore, that has been said on pages 104, et seq., is applicable here. So also, when the machinery requires regulation, a fly-wheel is adapted to the crank, and where the work must be performed at a constant velocity, a Governor, acting upon a throttle-valve, is adapted. A parallel motion for a high pressure engine is represented on PL IV. Fig. 1. In this, c d represents the lever-beam ; b the piston-rod ; c and d the pivots or centres of the parallel motion ; e the pivot to which the piston rod is attached ; g g a part of a fixed frame that bears the pivot h of the radius bar ; 22 1T0 ATMOSPHERIC ENGINE. c e and dfare the straps ; h f the radius bar ; e/the parallel bar. The governor of a high pressure engine is represented on the same plate at Fig. 2 ; a and b are the two bevel wheels ; c d the axis of the governor; e e spherical weights, suspended by rods from joints on a fixed collar/; g is a collar sliding on the axis c d. by the motion of the bars g h. g h j i i is a circular arc forming a loop at each end, in which the bars/ h.fh play; g h k is a lever moving on the pivot k ; / I a connecting rod that unites the end I of the lever to the handle or lever of the throttle valve. 124. The forcing pump that feeds the boiler, and the lift- pump that supplies the water which the former injects, are worked by rods from the lever beam, when the engine has one. In other cases, motions are taken off from the piston-red in such a way as to answer the same purpose. 125. The high pressure engine is thus fitted to subserve all the objects that can be fulfilled by the double-acting condensing engine. There are, however, engines which are not suited to produce more than a reciprocating action, and which were of older date, although of less value, being applicable to but few purposes. Such is the single-acting condensing engine. In this engine, the lever-beam is loaded with a weight, at the end opposite to that to which the piston is attached, and the latter rests, when the engine is not in action, at the top of the Cylin- der. The Cylinder and steam passages are then filled with steam ; a communication is now opened, from the lower side of the piston, with the condenser, and the pressure of the steam on the upper side forces the piston down to the bottom of the Cvlinder : the steam and condensing valves are next closed, and the third valve, which forms a communication between the opposite sides of the piston, is opened ;. there being now no ATMOSPHERIC ENGINE. 171 resistance to the motion of the piston, other than the friction, the weight at the opposite end of the beam again preponderates, and the piston is drawn back to its pristine position, the steam flowing through the open valve and steam pipe from the upper to the lower side of the piston. In this engine the steam acts only during the downward stroke of the piston, and during its return no force is exerted upon the piston. The effort is, therefore, directed to raise a weight, which may, in its descent, perform a work of such nature as is suited to this peculiar species of alternating motion. The raising of water, by a forcing pump, is the most usual purpose to which such an engine is applied. A parallel motion is unnecessary in this engine, and the piston-rod is connected with the beam by a chain, that applies itself to a circular arc on the end of the beam. The pump-rod, loaded with a weight, is attached in a similar manner to the opposite end of the beam. The air-pump, the cold and hot water pumps, are attached to rods worked by the beam. The power of this engine being exerted during one motion of the piston only, is obviously no more than half of that of a double-acting condensing engine of the same dimensions. It is, besides, applicable to but very few purposes : and as these very purposes may be accomplished by a double-acting condensing engine of half the size, this form has gradually fallen into disuse. It was, however, at one time much in use for draining the water of mines, and raising that fluid for the supply of cities ; and it has not wholly gone out of use for the former purpose, in its application to which its powers have been increased by making the steam act expansively. 1 26. At a still earlier period in the history of the Steam En- gine, an engine was employed, in which the air of the atmos- phere acting upon the piston was the prime mover. The vacuum on the lower side of the piston is caused by a conden- sation, effected in the cylinder itself. This engine is, for the reasons mentioned on page 102, far inferior to those in which a separate condenser is employed. It is also inferior in effect to 172 HIGH PRESSURE ENGINES. an engine in which steam is the moving power, for the latter is always made to act with a force a little superior to simple at- mospheric pressure. ]27. The form of engine last mentioned is now obsolete, and the preceding one nearly so. The expansive engine does not differ in form from the double-acting condensing en- gine: we shall therefore restrict ourselves to the description of the high pressure engine, and leave the detail o( the others until we treat of the history of the invention. On PI. V. is represented a high pressure engine of 30 horse power, manufactured by the \\ est Point Foundry Association. a is the cylinder, the stroke oi whose piston is about three and a half times the diameter. It stands upon a rectangular vessel i, through which the waste steam passes, heating water that is raised to it by a lift pump, not represented in the plate. The piston-rod />, is seen only in the end view, and is hidden in the other by the sides c c, in which its cross-head moves. This rod is attached to a lever-beam n, by straps a a, whose length from centre to centre is half the stroke of the piston- rod. d is the lever-beam, the length of each of whose arms is ra- ther more than three times as much as the stroke of the engine. e the Connecting-rod or Shackle-bar. f the Crank. g g. the Fly-wheel. h, a reservoir of water, through which the Waste steam passes bv a pipe, until it finally escapes by the tube r r. f t\ is the eccentric, which moves the tumbling shaft A\ to which is attached, by connecting rods, a cross-head /, which gives motion to the slide valve contained in the side pipe b. o- gp, is an endless chain passing over drums, one on the axis of the crank, the other on that of the vertical bevel-wheel. h, are two bevel-wheels that give motion to the axis of the governor k. "While the balls of the governor diverge, they raise one end of the lever /', the opposite end is pressed down and closes the throttle valve, situated at c, in the steam-pipe. HIGH PRESSURE ENGINES. 173 d is the rod of the forcing pump that conveys water from the reservoir h, to the boiler. A section of the Cylinder of this engine, and of its slide valve, is represented on PL II. Fig. 4. /, lower passage for the steam. e, upper passage for do. hi, openings in the sliding pipe, adapting themselves alter- nately to the passages e and/. gi third opening in the steam pipe, represented as applying itself to the eduction passage m. m, eduction passage to which the openings Zand g apply themselves alternately. ij i, interior of the side pipe, in which the slide is worked by the spindle. k, spindle, connected by rods with the eccentric. Another slide valve for a high pressure engine is represented in its connexion with the Cylinder at Fig, 5, on the same plate. a, b, c, d, is a rectangular bar of cast-iron, called the steam- chest : it is constantly receiving steam from the boiler. The lower side of this has three openings, represented at g t c,/, in the ground plan. Within the steam-chest, is a septum g t being a cup, or trough of a rectangular shape, whose open face is downwards, and is ground to apply itself closely to the lower plate of the steam- chest, against which it is firmly pressed by the steam. This septum is of such a size as to cover two of the openings of the plate, and to exclude the third. Hence, one or other of the lateral openings always communicates with the middle opening, through the septum, while the remaining one receives steam from the steam-chest. This septum is drawn backwards and forwards by the spindle A, which is worked by an eccentric. The central opening c, corresponds to the eduction pipe by which steam escapes ; the other two communicate, one with the upper part of the Cylinder, the other with the lower, and in the varying positions of the sliding septum, steam is alternately admitted from the steam-chest, and allowed to escape by the eduction pipe through these apertures. 174 HIGH PRESSURE ENGINES. A Horizontal high pressure Engine, manufactured by the West Point Foundry, is shewn on PI. TL. together with its two cylindrical boilers. A, ashpit. B, B, furnace doors. C, C, boilers. D, cylinder. E, piston-rod. F, connecting rod. G, G, G, fly wheel. H. governor. /, reservoir of cold water. K, forcing pump. L, cistern of water to be heated by waste steam. a, pipe forming communication between the water in the two boilers. b, b, b, steam pipe. c, safety valve. d, lever and weight of safety valve. e, pipe for waste steam from safety valve. /, pipe for waste steam after it has passed the valves and been used in the cylinders. g, steam chest containing slide valve. A, pipe by which the cold water is conveyed to the reservoir 7". i, pipe by which water passes from the reservoir to the cold water cistern. k, continuation of pipe /. L I. I, parallel motion for vertical forcing pump. n, 7i, 7i, eccentric. o. tumbling shaft. A section of the cylinder of this engine, with its side pipe, is to be found on Pi. II. Fig. 6th. It has been objected to the horizontal form of Steam Engines, that the packing wears unequally, being first abraded on the lower side of the piston, and that the Cylinder itself must final- ly be worn into an elliptical shape. With proper precautions in the use, however, no practical difficulty need arise. Engines of this form have advantage, in various cases, that will hereaf- HIGH PRESSURE ENGINES. 175 ter be enumerated, particularly in their application to steam- boats. High pressure engines are also occasionally constructed without the lever beam ; the form and distribution of the parts resemble, in this case, the condensing engine figured on PL VII. In consequence of the loss arising in the conversion of the reciprocating rectilinear motion of the piston of the usual forms of steam engines into one which is circular, by means of the crank, a loss which is supposed by many persons to be much greater than it really is, it has been frequently attempted to ob- tain a rotary motion directly. For this purpose a very great number of engines have been planned, and many of them have been constructed and actually tested. In four of them have the views of the projectors been realized ; nay, it might at one time have been safely stated that all attempts at the construction of a rotary engine had resulted in failure. From this general censure might perhaps be excepted the engine of James. Still, its action has not been found to be as efficient as that of the more usual forms of engine, using the same quantity of fuel. At the present moment, however, (1836,) an engine in which a rotary motion is produced by the reaction of steam, is in the course of experiment, and there is great reason to hope that it will be successful. It has been tried in several instances with such results as to make it certain that as great a power can be ob- tained from it in many cases by a given quantity of fuel than in any other mode in which steam has been applied. Practical difficulties exist in applying it to all the various objects for which the steam engine is used, particularly when it may be necessary to change the direction of the motion. This engine is the invention of — Avery, of Syracuse, N. Y. CHAPTER VII. EARLY HISTORY OF THE STEAM ENGINE. Introduction. — Statue of Memnon. — Hero of Alexandria. — Eolipyle. — Anthemius and Zeno. — Cardan. — Mathesius. — Baptista de Porta. — De Causs. — Brancas. — Wilkins and Kircher. — Marquis of Worcester. — Haatefeuille. — Papirts first plan. — Savary. — -Papirfs Engine for the Elector of Hesse. — Newcomen and Cawley. — Potter's Scog- gan. — Beightorfs Hand- Gear. — Smeaton. — Leupold. 128. The description of the steam engine, given in the pre- vious chapters, has been limited to the three more important varieties : the double-acting condensing engine impelled by low steam ; the double engine acting expansively ; the high pres- sure engine. These alone are in general actual use, and the consideration of their theory is all that is directly valuable to the maker or user of steam engines. The various other forms that the engine has assumed, in its progress from rude begin- nings to its present improved state, the several projects that have been brought forward and successively abandoned, may be best treated of in the historical form. In this manner also, may the relative merit of the inventors and improvers be best set forth. The steam engine, as it is the most powerful agent by which the power of man has been extended, so also has it employed the labour, ingenuity, and talent of more individuals than any other human invention. To give the true history of the steam engine, as indeed of HISTORY OF THE STEAM ENGINE. 177 most of the discoveries which have conferred important benefits on mankind, would be, in fact, to enter into the annals of near- ly all the arts and sciences. Instruments have been frequently contrived, and principles stated, for which the world was not at the time prepared. Centuries sometimes elapse before the period arrives at which the wants of society call for their appli- cation, or the intelligence of the age can appreciate their merit. Then some more fortunate genius recalls the forgotten plan from oblivion, or, unconcious of the labours of his predeces- sors, derives from his own resources, inventions, not perhaps more meritorious than theirs in the abstract, but suited to the condition and wants of his cotemporaries. To the last then is the world really indebted, and to him is the gratitude due. His predecessors may have even gone beyond him in actual progress, his cotemporaries may have been upon the eve of the same discovery, and may have been so far advanced, that a few years, months, or even days, would have placed them by his side. But the good fortune, and it may perhaps be little more, of him who first reaches the useful result, must eclipse the merit of all others. Priority in the application of an inven- tion to practical purposes, if associated with originality, or even with the calling up of forgotten projects, that were impractica- ble or useless at the moment of their first conception, is the point on which a claim to high distinction in the annals of the useful arts must depend. In the history of the steam engine then, a few names stand prominent, in consequence of the immediate advantages to the world with which their labours were followed. Savary, who first successfully substituted steam for the labour of animals ; Newcomen, who first succeeded in applying it to move a solid body, through whose intervention the work might be perform- ed ; Watt, who called in physical science, to discover and reme- dy the defects of his predecessors, and made the steam-engine an instrument of universal application ; Fulton, who performed the first successful voyage by the impulse of steam ; and Evans and Trevithick, who cotemporaneously gave to the engine such a form as suited it for locomotion on land, and ascertained that such locomotion was practicable. 23 178 HISTORY OF THE STEAM ENGINE. It is necessary, before we enter into the history, to particular- ize these authors of the great steps the steam engine has made, in principle or in application ; for the more minute our inqui- ries become, the more will the real and vastly superior merit of these parties seem to descend to the level of others, whcse ingenuity either formed the basis of the strides made by those we have named ; who had previously neariy reached the same results ; or were on the very eve of attaining them. Thus Savary and Newcomen, united, did but little more than the Marquis of Worcester had done before them, but had not ap- plied to purposes of real utility ; Watt found a competitor in the person of Gainsborough ; and but a few weeks would have plac- ed Stevens on the very eminence where Fulton now stands. The fitness of the time at which these several inventors suc- ceeded in their projects, it it be rather to be ascribed to good fortune than to pre-eminent merit, tends still more than any other cause to separate them from their competitors. Had the mines of Cornwall been still wrought near the surface, Savary or New- comen would hardly have found a vent for their engines. Had the manufactures of England been wanting in labour-saving machinery, the double-acting engine of Watt would have been suited to no useful application ; a very few years earlier than the voyage of Fulton, the Hudson could not have furnished trade or travel to support a steam-boat, and the Mississippi was in possession of dispersed hordes of savages. But if good fortune in the circumstances of the time, or in the success of the enterprize, were to be arguments against the honours that history assigns, we should sink its greatest names to the level of the most obscure, and those who have changed the face of the earth, to those who prepared, by gradual steps, the means by which the changes were effected. In the history of a mechanical invention too, it may frequently appear, on a cursory examination, that those who, by unsuccessful, although ingenious efforts, actually retarded the progress of discovery, are as meritorious as those who convinced the world of the value and practical merit of their inventions. So soon, how- ever, as success is attained, jealousy calls up all analogous pro- jects, however far from being adapted to the times at which HERO OF ALEXANDRIA. 179 they were proposed, or which some simple but undiscovered step prevented from being introduced into practice, and ranges them on an equal level with, or even exalts them beyond, those to whom the world owes the perfected invention. Conflicting national pride too comes in aid of individual jea- lousy, and the writers of one nation often claim for their own vain and inefficient projectors the honours due to the success- ful enterprize of a foreigner. If success be a title to honour in general history, it ought to be still more so in mechanical inventions. They require, to bring them into use, an union of practical and theoretic attainments, the want of either of which may render them abortive. Few projectors are thus doubly qualified, at the commencement of their career; and few or none have been successful, until by long and costly experience, they have added practice to theoretic knowledge, or have, by laborious study, brought science to the aid of mere mechanical skill. That steam is capable of exerting a mechanical force must have been obvious from the most remote antiquity, for we have no reason to believe that man was ever ignorant of the use of fire. But to apply steam to any useful purpose is an idea com- paratively recent. Still, however, the remotest antiquity that can be reached by profane history, has been quoted as affording an instance of the employment of steam, if not for a useful pur- pose, at least for one that produced no unimportant effect at the time, and excited the curiosity of mankind for centuries. 129. The elder Hero of Alexandria, who lived about 130 years before the Christian era, is the first author who gives any ac- count of the application of the vapour of water. We are una- ble to quote his work in the original, but are indebted for a no- tice of it to the beautiful little treatise of Stuart, to which we, once for all. acknowledge our obligations.* In this work it is stated that Hero expressly ascribes the sounds produced by the statue of Memnon to steam generated in the pedestal, and * Historical and Descriptive Anecdotes of Steam-Engines, and of their Inven- tors and Improvements, by Robert Stuart, Civil Engineer* — London, 1829. ISO HERO OF ALEXANDRIA. issuing from its mouth. Now, by the researches of Champol- lion, who is the highest authority on this point, the Memnon of the Greeks is identified with Amenophis II., a prince of the 17th Egyptian dynasty, who reigned at Thebes 16C0 years before Christ. Here. then, we have an application of steam, if the surmise of Hero be true, before the date of the Exodus of the Israelites. "We must, however, express our opinion, that this is rather an ingenious explanation of the philosopher him- self of the mode in which he could have effected the same ob- ject, than an account of what was really performed by the Egyptian priests. 130. Hero constructed or described more than one instru- ment, entitled to the epithet of steam engine. Two of them, of which one would have answered to raise water and the other would have produced a rotary motion, are figured below, In the figure marked b, a is a vessel in which water is boil- ed : the pipe c proceeds nearly to its bottom : the steam will therefore accumulate in the upper part of the vessel, and force the water in a jet through the pipe c. A fountain may thus be formed, on which may be supported the ball o. In the figure marked c. o is a similar vessel ; two pipes, a and c, proceed from it : these are bent towards each other, and serve as pivots to the sphere i, in which there are openings cor- responding to those in the pipes a and c. From points in the HERO. 181 sphere diametrically opposite to each other, proceed the pipes m and w, which are bent towards the end at right angles, and direct- ed to opposite sides of the apparatus. The steam generated in the vessel o passes through the pipes a and c, into the sphere i, and thence into the pipes m and n, issuing from which in op- posite directions, it, by its reaction, gives a rotary motion to the sphere. Hero does not give the slighest hint that his invention was capable of any useful application, nor does he appear to have imagined that he was in possession of an instrument that was in future ages to produce such important results. The Greek philosophers, however, seem rarely to have at- tended to the practical value of their investigations ; it was suf- ficient for them to discover and to astonish ; and even when they mention arts and instruments that seem to have been ac- tually introduced, they avoid contemptuously all notice of their uses in the arts. " The ancient philosophers," says an ingenious author, " esteemed it an essential part of learning to conceal their knowledge from the uninitiated ; and a consequence of their opinion that its dignity was lessened by its being shared with common minds, was their considering the introduction of mecha- nical subjects into the regions of philosophy, as a degradation of its noble profession, insomuch that those very authors among them, who were the most eminent for their own inventions, and were willing by their own practice to manifest unto the world these artificial wonders, were, notwithstanding, so infected by this blind superstition, as not to leave any thing in writing concerning the grounds and matters of these operations ; by which means it is that posterity hath unhappily lost, not only the benefit of these particular discoveries, but also the proficien- cy of these arts in general. For when once learned men did forbid the reducing them to vulgar use and vulgar experiment, others did thereupon refuse those studies as being but empty and idle speculations ; and the divine Plato would rather choose to deprive mankind of those useful and excellent inventions than expose the profession to the ignorant vulgar." We are luckily fallen upon happier times. The student and the profi- cient in science no longer shut themselves up from the busy 182 EOLTPYLE. world, or hide their acquisitions like mysteries from the public ; but their whole endeavour is to bring their learning into such a form as may calculate it for the most wide dissemination, and enable it to produce the most extensive usefulness. 132. The Eolipyle, however, was an instrument well known to the ancients. It was applied by them to but one single ob- ject, that of exciting the energy of combustion. It is mentioned by Yitruvius, Ltb. I. Cap. VI., as an illustration of the causes of the winds. It was supposed that the blast actually proceed- ed from the Eolipyle, but as^steam would not support combus- tion, we must look to some other cause for its effects in this res- pect. We find it in the lateral communication of motion that takes place among fluids, by which a current of air is made to follow the course of the steam that issues from the neck of the Eolipyle. We give a figure of this instrument. It is composed of a globe or other hollow vessel A, to which a pipe B, is adapted. If a portion of water be introduced, and the vessel placed over a fire, steam will be generated, and issue forcibly from the narrow aperture. If it be mounted on wheels, it will recoil by the reaction of the escaping vapour ; and a ro- tary motion may be produced, by two pipes, but in opposite directions, as in the machine of Hero. 133. A knowledge of some of the properties of steam seems CARDAN — BAPTISTA PORTA. 183 to have been retained daring the flourishing periods, and even to the decline, of the Roman empire. In the reign of Justinian a dispute occurred between Anthemius, the Architect of that Emperor, and the Orator Zeno, which shows this fact. Yet the knowledge was here applied to mere purposes of private malice, while it might, by the exercise of no greater ingenuity, have produced important and useful consequences. From this pe- riod until the revival of learning, we find no record of any use of steam, either for useful or entertaining purposes. 134. Cardan is the earliest modern author in whom we de- tect any hint of a knowledge of the mechanical properties of steam. This extraordinary man, who united all the learning of his age to even more than all its superstition, appears to have known, not only the expansive force of steam, but the fact that a vacuum could be produced by its condensation ;. a fact so im- portant in the action of the steam engine. Among his propo- sals is one for the use of the current of rarified air in a chim- ney, to produce a rotary motion. He, first of the moderns, gives adescription of the Eolipyle. The work which contains the former of these plans is dated 1571.* 135. A German of the name of Mathesius, in 1571, to bor- row the words of Stuart, " displayed almost as much ingenuity in contriving to introduce so untoward a subject into a sermon as a description of an apparatus, answering to a steam engine, as would be required to invent the machine itself, and which he gives as an illustration of what mighty efforts could be produc- ed by the volcanic force of a little imprisoned vapour." 136. The researches of modern writers, among whom we may note with the highest praise him that we have just men- tioned, has disclosed various persons, who seem to have had ideas more or less just of the mechanical power of steam. The only one that we consider worthy of notice is Baptista Porta, a Neapolitan, who lived towards the close of the sixteenth centu- * Stuart's Historical Anecdotes of the Steam Engine, page 19. 184 DE CAUSS. ry. His machine, which is the germ of several that have been noted as original, is figured below. Water is boiled in a vessel A, placed upon a furnace. The steam rises through the pipe b into the upper part of the box or vessel C, the lower part of which is filled with water. The pressure of the steam on the surface of the water forces it up the rising pipe D. 137. Next in the order of time is De Causs. engines contrived by him is the following: Among various BRANCAS — WORCESTER. 185 A spherical vessel A has a pipe b b inserted, until it nearly reaches its lower part. The vessel is partly filled with water, which is boiled, and the steam accumulating in the upper part, forces the water up the pipe. Here it will be observed that the heated water is itself raised, and the powers and utility of the engine are evidently far less than those of the machine of Por- ta, 138. The first person who seems to have had an idea that the power of steam was capable of being applied to any other useful purpose than that of raising water, was Brancas, an Ita- lian, who proposed to direct the blast issuing from an Eolipyle upon the leaves of a wheel, which, being set in motion by its impetus, might serve to move machinery. This method is un- luckily imperfect and wasteful, yet the attempt is deserving the highest praise, inasmuch as he is the only person, who, in the infancy of these investigations, entertained any hope of realiz- ing the vast benefits that steam has since conferred upon the world. Had steam been confined in its action to the single ob- ject of raising water, it might have been of notable use in a few cases ; but its great and important value, as a prime-mover, has been only realized since methods of applying it, to any species of work whatsoever, have been discovered. 139. This plan of Brancas was repeated by Bishop Wilkins j and Kircher proposed to apply two Eolipyles to concur in the same effect. The last-named author also proposed an engine similar in principle to that of Porta. 140. Of all those who attempted to apply steam to useful pur- poses, without being successful in introducing his engine into general practice, the Marquis of Worcester fills the greatest space. He has been claimed by English authors as the first who made any experiments of importance upon steam ; and it has been asserted that the next of their countrymen, who un- dertook the investigation, did no more than copy, without ac- knowledgment, the plans of Worcester. Even the first truly successful form the steam engine assumed has been shown to be 24 186 WORCESTER. consistent, in many respects, with the description of one of the engines of this nobleman. It is yet a disputed point, what was actually the form of the engine of Worcester. His description is at best vague, and is without any figure ; various authors have exercised their inge- nuity in framing plans of a machine that should be consistent with the expressions of his work. We do not consider it im- portant to do so, but shall content ourselves with quoting his own words. They are to be found in a little treatise, entitled " A century of the names and scantlings of such inventions, as at present I can call to mind to have tried and perfected, which, my former notes being lost, I have at the instance of a powerful friend endeavoured, now in the year 1655, to set down in such a way as may sufficiently instruct one to put the whole of them into practice? This work was originally printed in London in 1663, and has been six times reprinted ; the reprint of 1813 has been consulted for the following, being the 68th Proposition. " An admirable and most forcible way to drive up water by fire, not drawing or sucking it upwards, for that must be, as a philosopher calleth it, infra spheram activitatis, which is but at such distance, but this way hath no bounder, if the vessels be strong enough ; for 1 have taken a piece of a whole cannon, whereof the end was burst, and filled it three quarters full, stopping and screwing up the broken end, as also the touch- hole, and making a constant fire under it ; within twenty-four hours it burst, and made a great crack ; so that, having found a way to make my vessels so that they are strengthened by the force within them, and the one to fill after the other, have seen the water to run like a constant fountain forty feet high ; one vessel of water, rarified by fire, driveth up forty of cold water ; and a man that attends the work is but to turn two cocks, that one vessel of water being consumed, another begins to force and refill with water, and so successively." Vague as this description is, it would still be possible to con- struct an engine that would perform a similar work by the ex- pansive force of steam. It would be very inferior to modern engines, but would yet be effectual. WORCESTER. 187 It has generally been imagined that this is the sole reference to steam in the Century. But two others certainly correspond so closely to the character of our modern high pressure engines, that it may not be amiss to quote them also. They are the ninety-eighth and hundredth propositions of his work. " An engine so contrived that working primum mobile backward or forward, upward or downward, circularly or con- trariwise, to and fro, upright or downright, yet the pretended operation eontinueth and advanceth, none of the motions above mentioned hindering, much less stopping the other ; but unani- mously agreeing, they all augment and contribute strength to the intended work and operation ; and therefore I call this a semi-omnipotent engine, and do intend that a model thereof be buried with me." " How to make one pound weight to raise an hundred as high as one pound falleth, and yet the hundred pound descend- ing, doth what nothing less than one hundred pounds can effect. Upon so important a help as these two last-mentioned inven- tions, a waterwork is, by many years' experience and labour, so advantageously by me contrived, that a child's force bringeth up an hundred feet high, an incredible quantity of water, even two feet diameter, so naturally that the work will not be heard into the next room; and with so great ease and geometrical sym- metry, though it work day and night from one year's end to the other, it will not require forty shillings reparation to the whole engine, nor hinder one day's work ; and I may boldly call it the most stupendous work in the whole world ; and not only with little charge to drain all sorts of mines, and furnish cities with water, though never so high seated, as well as to keep them sweet, running through several streets, and so per- forming the work of scavengers, as well as furnishing the in- habitants with water enough for their private occasions ; but likewise supplying rivers with sufficient water to maintain and make them portable from town to town, and for bettering of lands all the way it runs. With many more advantageous and yet greater effects of profits, admiration, and consequence ; so that, deservedly, I deem this invention to crown my labours, to 18S WORCESTER. reward my expenses, and make my thoughts acquiesce in the way of further inventions." In the first of these steam obviously meets the description of his primum mobile, for in whatever direction it proceeds, it is still capable of exerting the same mechanical force. The sin- gle pound raising one hundred, in the second, meets the con- ditions under which the piston of a steam engine acts, for its weight bears even a less proportion to the power of the engine. The following is an extract from a manuscript left by the Marquis of Worcester. " By this I can make a vessel of as great burthen as the ri- ver can bear to go against the stream. * * ******** " And this engine is applicable to any vessel or boat whatso- ever, without being therefore made on purpose ; and worketh these effects. It roweth. it draweth, it driveth, (if need be) to pass London Bridge, against the stream at low water." It is to be remarked, that Worcester claims, on his title-page, the merit of having actually completed, and used all the inven- tions he describes in the work : in support of this assertion va- rious evidence has recently been adduced. He employed a mechanic for thirty-five years, under his di- rections, in the manufacture of models ; and many of his projects that appear, in his manner of announcing them, absolutely im- possible, have been unexpectedly realized by modern inven- tions. That the steam engine of Worcester was no vague concep- tion, but was actually put into operation, a recent discovery has settled, upon testimony the most convincing. The Grand Duke of Tuscany, Cosmo de Medicis, travelled in England in 1656. His manuscript account of his journey remained un- published until 1S18, when a translation was made and printed. The following is an extract from this translation : " His highness, that he might not lose the day uselessly, went again after dinner to the other side of the city, extending his excursions as far as Yauxhall, beyond the palace of the Arch- bishop of Canterbury, to see an hydraulic machine, invented by my Lord Somerset, Marquis of Worcester. It raises water HAUTEFEUILLE — MORLAND. 189 more than forty geometrical feet by the power of one man only ; and in a very short space of time will draw up four vessels of water, through a tube or channel not more than a span in width." Here, then, is a description of an engine in actual operation, and corresponding in terms with that referred to in the century of inventions. 141. In the several projects of which we have hitherto spo- ken, the expansive power of steam was used alone. It was made to act directly upon the surface of water to raise it ; or, issuing from the orifice of an Eolipyle, set a wheel in motion ; or again, issuing from two tubes attached to an Eolipyle, caus- ed that instrument to revolve upon an axis, by the reaction of the vapour. In each of these ways the use of high steam is essen- tial to success, and this upon a large scale is attended with dan- ger, particularly in the low state of mechanic arts, and before the various contrivances we have mentioned in chapter III. were invented. The action of steam of a force no more than equal to the pressure of the atmosphere, against a vacuum formed by its own condensation, is a far more safe, and, as we have seen, more useful application of its energy. The researches of Stuart seem to show that this was first proposed by a Frenchman of the name of Hautefeuille. He, in the year 1678,* published a work, in which he intimates that the alternate generation and condensation of the vapour of alcohol might be applied, without waste, to the production of mechanical effects. We have, however, no proof that this project ever went farther than the mere proposal. It is, notwithstanding, to be consi- dered as one far beyond the knowledge of that age, of the nature and properties of steam. 142. Sir Samuel Morland, who was cotemporary, appears, from the very words he employs, to have been merely an imita- tor of the Marquis of Worcester, and therefore claims no notice among those who aided in the progress of the steam-engine. * Stuart's " Anecdotes." 190 MORLAND PAPIN. 143. In 1680, the year previous to that in which Sir S. Morland visited France, Dr. Denys Papin,a French Protestant, invented the safety valve, which has since been of such impor- tant service in the construction of the steam engine. It was first employed by him in an apparatus called the digester. This apparatus is a boiler, within which water is retained under pressure, in order that it may be heated beyond the tempera- ture at which it boils in the open air. The original object was to extract the crelatinous matter from bones, in order to apply its solution as food. To prevent any risk of danger, a conical aperture was left in the lid of the vessel, and to this was adapt- ed a conical stopper, pressed by a weight suspended at the end of a lever. It was, in short, identical with the most usual form of safety valves at the present day. Although he thus employed water at a high temperature, and had discovered one of the methods that are still in use, of ren- dering the boiler safe, still it was long before he attempted to apply the power of steam. The motion of a piston in a cylin- der was suggested by him, as a method of adapting the expan- sive force of an elastic fluid to produce mechanical effects. In this apparatus he at first proposed to employ air rariried by heat ; he next attempted to exhaust the space beneath the pis- ton, and make use of the pressure of the atmosphere ; and final- ly, to raise the piston by the inflammation of gunpowder. In a letter to Count Zinzendorf, however, he proposes to use steam for the same purposes. In the Leipzig Transactions, also, for the year 1690, he repeats the proposition, and explains the prin- ciple upon which he founds the application of this substance, both to raise a piston, and to produce a vacuum by its conden- sation. There is, however, no evidence that a separate boiler ever entered into his views, without which it would have been impossible to make any useful application of his principle. Nothing, then, had been actually effected by Papin in this earlier stage of his researches, and he did not extend them far- ther until steam had actually been successfully employed in raising water ; if we can indeed say that he ever was success- ful in pointing out a mode in which it could be rendered of practical value. SAVARY. 191 144. The history of steam, applied to purposes of acknow- ledged utility, commences then with Savary. It has been much debated whether this person were in reality an inven- tor, or had merely the judgment to perceive an opening for the introduction and adaption of. previous discoveries. His own statement is, however, clear, distinct, and worthy of cre- dit. Having been, in the early part of his life, employed in the mines of Cornwall, he was aware of the vast expenditure in- curred in keeping them free of water ; and an accidental obser- vation appeared to point out to him a simple and easy mode, in which a substitute could be found for the expensive labour of animals. Being at a tavern in London, he threw upon the fire a Florence flask containing a small quantity of wine ; he observed the wine to boil, and a cloud of vapour to issue from the neck, while the interior remained transparent. Struck with the appearance, he seized the flask and inverted the neck in a basin of water ; after a short time he found the flask filled with liquid, in consequence of the condensation of the steam forming a vacuum, into which the water was raised by the pressure of the atmosphere. The very form and arrangement of his apparatus is a proof of the truth of his story, for it is no more than a flask-shaped vessel of iron, in which a vacuum is formed by the conden- sation of steam. This part of its principle had not before been acted upon, nor even thought of, except in the suggestion of Hautefeuille, which we have before spoken of. The action of the vessel is in this respect identical in principle with that of the common pump. He did not, however, limit his views to this single action, but proceeded to add to it the action of the forcing-pump. For this purpose, so soon as the flask was fill- ed with water, steam proceeding from the boiler, of a high tem- perature and corresponding tension, was admitted into the flask, after the communication with the water beneath was closed, which, acting on the surface of the water contained in the ves- sel, forced it up a lateral pipe. As it was impossible to obtain a perfect vacuum by the condensation of the steam, the first part of the action of Savary's engine was limited to the height of 25 feet ; the second part has no limit, but in the tension of the 192 SAVARY. steam, and the strength of the materials, of which the vessel and the rising-pipe were composed. This however. was. from the imperfect state of materials and workmanship, limited to less than 70 feet ; so that the two different actions of the engine. working in succession, raised the water to little more than 90 feet. Even at this comparatively small limit, the danger attend- ing the use of this engine became excessive, while the height, to which it was capable of raising water, was entirely too small for the purpose of draining mines. Several different engines, placed at different levels, would have remedied the last defect, but the cost of attendance would have been enhanced in proportion. Such defects were obvious ; there were, however, others which could not be accounted for until the doctrine of latent heat was discovered, and which we shall return to on a subsequent page. The cause which affects the action of high pressure engines, and prevents them from working with the power that migh: first sight, have been anticipated, is also to be found in opera- tion in this engine. The force required to raise water from sixty-five to seventy feet, is equivalent to steam of a tension of not less than three atmospheres : now. as the density of steam increases nearly as rapidly as its tension, it is obvious that to obtain this in quantity sufficient to fill the vessel, would require the evaporation of nearly three times as much water as would fill it. were the tension no more than a single atmosphere. Hence, in fact, little is gained by the second part of the action of this rioe : for water may be raised, with an equal expenditure of fuel, nearly as high by condensation, in vesse' :d at differ- ent levels, as it is by direct pressure of any intensity, however great. We have placed on the opposite page a section of the en^iDe of Savary, which in its complete form was double, one vessel receiv- ing the water in consequence of the condensation of the steam, while from the other it was forced up by direct pressure: these vessels alternated with each other in their operation. SAVARY. 193 I is the boiler in which the steam is generated. O, steam pipe, by which steam is conveyed to the vessel P. q q, pipe communicating with a reservoir beneath ; through this pipe the water is raised to the vessel P, where the steam is condensed by the pressure of the atmosphere. S, rising-pipe through which water is forced when the steam flows from the boiler through a valve on the steam-pipe O, which is manoeuvred by the lever z m. R R, valves opening alternately. x ) reservoir to supply water of condensation ; it receives water from the rising-pipe S, through a pipe governed by a float and stop-cock. y, pipe through which the cold water falls on the outside of the vessel P. n, gauge-cock to show the height of water in the boiler. 25 194 SAVARY — PAPIN. We have stated the more obvious defects of Savary's engine^ as well as one which is not usually quoted. There is, how- ever, another of far greater importance, but which rests upon a physical principle, entirely unknown at the period in which he lived. This grows out of the necessity of filling the vessel al- ternately, with steam of high tension, and water of a low tem- perature. When the steam is first admitted into the vessel, it will be con- densed against its sides and upon the surface of the water ; nor will it begin to act mechanically until both be heated to the temperature of 212°. Its full effect will not take place until both are heated to such a degree as will maintain the steam at a temperature, and consequent tension, appropriate to the height of the place of discharge. As the water is forced out of the ves- sel, fresh cold surfaces are exposed, and must be heated in their turn ; and when the vacuum is to be formed, the outside of the vessel is cooled by the affusion of water, while the inside is far- ther cooled by the rise of water from the reservoir beneath. In these different ways it has been found, by experiments carefully conducted, that iiths of the steam is condensed without acting at all, and that, of course, a similar proportion of fuel is wasted. The engine of Savary, therefore, is confined to a single ob- ject, namely, that of raising water ; and even this, for the rea- sons we have stated, it does to great disadvantage. Still, how- ever, the introduction of this engine was not only important as a step to the construction of more perfect ones, but it was of it- self of some value when compared with the methods for raising water that were at that period in use. 145. An apparatus which, at first sight, bears a strong simi- larity to Savary's, was constructed by Papin, for the Elector of Hesse, in 1707. It differs in having a piston, working in the vessel into which the steam is alternately admitted and con- densed, and makes no important use of the pressure of the at- mosphere. It appears that, even with the aid of the celebrated Leibnitz, he had been unable to bring his cylinder engine to perfection, and had abandoned his researches until again sti- mulated by the success of Savary. PAPIN NEWCOMEN AND CAWLEY. We give a figure of this last engine of Papin. 195 a is the boiler, furnished with a safety valve b, pressed down by a weight c, suspended from a lever. Water is introduced into the boiler through this valve, fis the forcing vessel, hav- ing an aperture at the top closed by the valve g. A piston is placed in this vessel, having a socket into which a cylinder of iron z, heated red hot, is introduced to keep up the tempera- ture of the steam ; water is admitted into the forcing vessel through the funnel x, and valve h. The rising pipe k enters an air vessel. The action of the steam in the forcing vessel raises the water into the air vessel, whence, by the pressure of the condensed air, it runs in a continual stream ; when the piston has descended to the bottom of the vessel/, the valve d is closed, and no more steam flows over ; the valves e and g are opened ; through the former, the steam that has been used escapes, and through the latter the forcing vessel is again filled. 146. The time had now arrived in which the world was to derive essential advantages from the employment of steam as a moving power. Even Savary's engine, although more valuable than any other we have hitherto spoken of, had obvious defects, which prevented its coming into general use. These obvious defects were remedied by the engine of Newcomen and Caw- ley, their patent for which issued in 1705. Departing from the idea entertained by all former inventors, except in the abortive proposition of Papin, of making the steam act directly to raise water, either by pressing upon its surface or by forming a va- cuum on its condensation, Newcomen and Cawley sought the 196 NEWCOMEN AND CAWLEY. means of working the brake of a forcing pump. With this view, the pump-rod being loaded with a weight sufficient to bring it to rest in its lowest position, the brake or lever of the pump was made with equal arms, and resting on a pivot in the middle of its length. The pump-rod being thus loaded, and attached in this manner to one end of the beam, a piston, of size considerably larger than that of the pump, was attached to the other, and made to fit a Cylinder, at the upper end of which it rested, under the pre- ponderating weight of the pump-rod and its load. The Cylin- der had in its bottom a valve opening upwards, by which steam could be at pleasure admitted or cut off. To the side of the Cylinder and near its bottom was attached a horizontal pipe, bent upward at the open end : in this was placed a valve open- ing to the air, which is called the snifting valve. Steam of the temperature of 212° being admitted into the Cylinder, would, from its levity, rise to the upper part of that vessel, displace the air previously contained therein, which flows out through the latter valve, making a sound which has given this valve its name. If the steam that thus enters the Cylinder have its com- munication with the boiler closed, it may readily be condensed, and a partial vacuum formed, beneath the piston. The pres- sure of the atmosphere will now act, and force the piston down- wards to the bottom of the Cylinder ; the opposite end of the lever-beam will be raised, and with it the pump-rod, and the weight with which it is loaded. If the communication with the boiler be again opened, the pressure on the opposite sides of the piston will again become equal, and the preponderating weight of the pump-rod will cause it to descend, and draw up the piston to its primitive position. A second condensation will cause the piston again to descend, and the process may thus be kept up so long as the boiler continues to supply steam. The condensation in the Cylinder was at first produced by cooling the outside, by the affusion of cold water ; and, when the action was required to be rapid, by placing the cylinder in an external cylindrical space. A hole having been accidentally made near the bottom of the Cylinder, the water spouted into it, and the condensation was found to be much more rapid. NEWCOMEN AND CAWLEY. 197 This was then imitated, by adapting a pipe to the Cylinder, through which a jet was made to flow as often as it was ne- cessary to condense the steam. This pipe and injection appa- ratus, were governed by a stop cock or valve placed upon it. Thus there were two valves necessary to the action of this en- gine, and these were to act alternately, the one opening as the other closed, and vice versa. 147. In the original form of the engine these were worked by hand, a boy being placed within reach of the levers that opened and shut them, to perform that operation as often as necessary. This employment being excessively irksome, one of the persons was not slow to perceive that it might be per- formed, even better than it could be by any personal attention, by the alternating motion of the lever beam itself. This im- portant step towards the perfection of the engine was made by a boy of the name of Potter, and was immediately adapted to all the engines of Newcomen and Cawley. It will be at once obvious, that the steam in this engine was employed solely to form a vacuum by its condensation, and that the pressure of the atmosphere was the efficient agent. Hence, as Savary's patent comprized the use of steam for this purpose, he was associated in the profits of Newcomen and Cawley. As this was the sole use that was made of the steam, it was unnecessary to generate it of a tension greater than that of the atmosphere; hence its use became perfectly safe, while the height to which it was capable of raising water was as great as could be effected by a forcing pump worked by any agent whatsoever. This engine, therefore, far exceeded that of Sa- vary, both in its ease of application and its power. On the other hand, the principal physical defects, noted as affecting Savary's engine, were still inherent in Newcomen's, and its mechanical execution became far more difficult. So great, indeed, was the latter difficulty, in the then imperfect state of the arts, that it was found impossible to keep the piston tight, except by covering its surface with a mass of water, whose presence still further enhanced the physical imperfections. 198 NEWCOMEN AND CAWLEY, The steam being condensed within the Cylinder, the whole was cooled down at each stroke to the temperature of conden- sation ; while the part of the Cylinder above the piston in its lowest position, was still further cooled by the mass of water employed to render it tight. On the re-admission of the steam, the whole was again to be heated up to the boiling point ; thus the waste of fuel was quite as great as in the engine of Savary. Another imperfection grew out of the partial nature of the vacuum that it was possible to produce in the cylinder. Water which boils under the ordinary mean pressure of the atmos- phere at 212°, rises into vapour at all temperatures whatsoever, and boils at lower temperatures under diminished pressure. Hence, so soon as the piston began to descend, the action of at- mospheric pressure was lessened by the generation of fresh steam, and although this was in its turn condensed, its place would be occupied by new steam of a lower temperature, and a resistance would be opposed to the descent of the piston. In consequence of this retarding force, it was found in prac- tice impossible to make the pressure of the atmosphere, which is, at a mean, 15lbs. per square inch, act upon the piston with a mean force of more than 17^-lbs., and from this, in estimating the action of the machine, the friction, and other retarding forces, are to be deducted. This engine, therefore, consumed about twelve times as much fuel as would have generated steam suf- ficient to fill the Cylinder, and worked with but half the force the moving agent was capable of exerting. The rectilineal motion of the pump and piston rods was, in this engine, accommodated to the circular motion of the ends of the lever-beam, in a very simple and ingenious manner. The ends of the beam were made in the form of arcs of circles, and the rods were suspended from them by chains, attached to the highest point of each arc. Thus, as the active pressure of each of these, in the performance of its share of the work was vertically downwards, it was always applied directly to the beam, the two rods being respectively always in the direction of tangents to the circular arcs formed upon the working beam. 148. The valve apparatus of Potter, called by him the Scog- NEWCOMEN AND CAWLEY, 199 gan, was, in 1718, superseded by a more perfect arrangement invented by Beighton. A frame or bar was attached by a chain, working also over a circular arc, to the lever beam ; projecting pieces or pins, forming a rack, were attached to the frame ; the valves were moved by quadrants cut into teeth, and acting up- on a rack connected with the spindle of the valve ; to each of these quadrants was attached a lever, which was pressed by the pins upon the frame through a circular arc, until it passed the line of motion of the frame, and was disengaged ; from this position, the lever was made instantly to return to its original place, by the action of a weight. These levers being also fur- nished with handles, to enable the valves to be open and shut by hand, the apparatus was called the Hand Gear, the frame and pins, the Plug Frame. This mode of working the valves continued to be used up to the beginning of the present centu- ry, with but little improvement ; nor has it yet fallen wholly into disuse. The engine of Newcomen is exhibited in the annexed draw- 200 SMEATON — LEUPOLD. ing, by which its mode of action and the uses of its several parts may be better understood. a is the boiler. t, the steam pipe. e, the steam valve. c, the Cylinder, into which the injection water is seen play- ing through the valve and pipe p. r, the piston. Sj the snifting valve. m, a reservoir of water, whence the injection pipe is supplied and water flows, through the pipe w, to keep the piston tight. The injection water is discharged through the pipe i, and the excess of that floating on the piston by the pipe h. w is the weight attached to the pump-rod, by the action of which the piston is returned to its highest position. The lever beam and pump are too obvious to need descrip- tion. 149. The engine of Newcomen and Cawley was improved in its mechanical structure by Smeaton, and derived additional force from the general improvement of the mechanic arts. Smeaton also formed tables of the dimensions of the several parts. With these improvements it is still occasionally used, particularly in places where fuel is cheap and abundant ; where its small cost, and its safety, are considered as more than coun- terbalancing the great waste of fuel with which it is attended. 150. In 1718, a German engineer, of the name of Leupold, published a work containing a description of two engines, the merit of which he ascribes to Papin. They are, however, ra- ther to be considered as ingenious applications of his own, of the principle of that inventor, aided by the knowledge of what had been effected by Savary and Newcomen. In one of these the steam was made to act alternately upon the surface of water in two vessels, and it is so similar in every thing, but the form and position of its parts, to the engine of Savary, that we do not conceive it necessary to describe it minutely. The second is a high pressure engine with pistons, and is extremely ingenious. LEUPOLD. 201 besides being remarkable as the first in which steam of high tension was made to act upon a piston. The first of these is liable to all the objections that we stated in speaking of Savary's engine. The second is far better, and even preferable in many- respects to the engine of Newcomen. It is, besides, applicable to the production of a continuous rotary motion, and is there- fore the first that could have been applied to general purposes in the arts. Of this last engine we have in consequence given a figure. Steam is generated in the boiler c, and flows thence through the pipe d ; it is represented as passing through one of the pas- sages in a four-way cock, beneath the piston a, while the steam which had filled the other cylinder is escaping into the air through the passage e. The piston a works a lever and the pump-rod g-, while b works another lever and the pump-rod/. h is the fire-place. The two pumps force water alternately into the rising pipei 26 202 LEUPOLD. Did the levers act upon eranks situated upon the same axis, a continuous rotary motion might be produced. Steam in this case is the moving power, and is not condens- ed as in the engine of Savary. It therefore, is constantly retard- ed by the pressure of an atmosphere, and must have a tension of at least two atmospheres in order to work to advantage. The history of the steam engine is thus brought down, from the most remote period at which the power of that agent was first suspected, until it assumed a definite form, and became ca- pable of useful application. Hitherto, however, but one spe- cies of work came directly within its scope. In the succeeding chapter we shall find its physical defects remedied or removed, and its application finally made universal to every species of manufacturing industry. CHAPTER VIII. CONCLUSION OF THE HISTORY OF THE STEAM ENGINE. Power and Dejects of Newcomers Engine. — Birth and Edu- cation of Watt. — Professor Robison. — Waffs first experi- ment. — Professor Anderson. — Watts second experiment. — Inferences. — Separate Condenser. — Steam applied as the moving poioer. — Packing. — Jacket and Air-pump. — Working Model. — Dr. Roebuck. — Experimental Engine. — Watt's first patent. — GainsborougKs claim. — Boring apparatus. — Form of Watts first Engine. — Saving of Fuel. — Projects for rotary motion. — Fiztgerald, Steioart, and Clarke. — Double-acting Engine of Watt. — Wash- borough and Pickard. — Crank. — Sun and Planet Wheel. Other Inventions and Improvements of Watt. — Hornblower. — Watts patent extended. — Governor. — Introduction of steam into various mechanic arts. — Expiration of Watts patent. — Cartioright and Sadler. — Murray, Maudslay, and Fulton. — Woolfe. — Oliver Evans. — Trevithick and Vivian. — Rotary Engines. — Conclusion. 151. In the preceding chapter the prominent defects of the engine of Newcomen and Cawley have been pointed out. In spite of these, it had been found of immense value in practice, in raising water for the supply of cities, and more particularly in draining mines. So great, indeed, had been the advantages derived from it in these cases, that hopes had been from time to time entertained that this engine might be rendered efficient in performing work of other descriptions ; and it had even been thought of as the prime means of propelling boats. That the energy of the prime mover was adequate to any of these pur- 204 "WATT. poses was certain, but mechanical difficulties opposed its appli- cation. Even had these been overcome, the engine was liable to physical imperfections, which had not at this period been sus- pected, far more formidable than those which are merely me- chanical. The latter, we now know, are so great in amount, as to have prevented the atmospheric engine from competing with almost any other prime mover, except in a few particular cases. These objections, whether physical or mechanical, might have been gradually removed ; the former by the gene- ral progress of the arts, the latter by the discoveries in physical sciences with which the close of the last century teemed. The steam engine, however, was not destined to wait for the slow changes which follow the application of purely theoretic prin- ciples to practical purposes. A single individual was found, who, by his own researches and unaided efforts, reached the law of the relations of steam to heat, which was, about the same time, discovered in its more general form by Dr. Black. This illustrious individual was James Watt, 152. Watt was the son of respectable, but poor parents. His grandfather exercised the profession of a schoolmaster, his fa- ther that of a merchant in Greenock, in Scotland. Having re- ceived the elements of a liberal education, which the excellent school system of Scotland places within the reach of all, Watt, at the early age of sixteen, became the apprentice of a maker of optical instruments in the city of Glasgow. Two years after- wards he removed to London, and obtained employment from a maker of mathematical and philosophical instruments. In this employment, his health became affected, and he was com- pelled to return to his native district. In undertaking business on his own account, he would have preferred Glasgow, as offering far greater prospects of success than Greenock, and hence became anxious to settle in the for- mer place. To this plan, however, obstacles presented them- selves, in the form of the laws of the corporation, by which the exercise of a trade was restricted to those entitled to the privi- leges of a burgess, to which Watt had no claim. From this state of difficulty he was fortunately relieved by the interposi- ROBISON ANDERSON. 205 tion of the professors of the University. This institution pos- sessed, as a remnant of ancient privileges, the right of claiming immunity from the corporate restrictions, and Watt was furnish- ed by them with apartments within the college buildings, in which he pursued his trade. 153. His attention was first called to the subject of steam by Professor Robison, then a student of the University of Glasgow, at a date as early as the year 1759 ; but their researches were attended with no important advances. 154. In 1761, Watt made experiments with an apparatus re- sembling the engine of Leupold ; but becoming aware of the dan- ger attending the use of high steam on a large scale, he ceased from any farther pursuit in that direction. 155. In 1764 he was employed by Professor Anderson, then holding the chair of mechanical philosophy in that institution, to repair a working model of Newcomen's engine. The obvi- ous waste of steam that he found to attend the action of this model, and the great quantity of injection water it required, struck him as facts unaccounted for by any previous scientific reason. Suspecting that the first defect might arise from an er- roneous estimate of the comparative densities of steam and wa- ter, he, by a few simple experiments, endeavoured to ascertain the true relation, and found that water in becoming steam ex- pands itself, under ordinary pressures, to 17 or 1800 times the bulk it had previously occupied. This is not far from the truth, as we now know from more accurate experiments, and corresponded with the estimate of Smeaton. At this density for steam, his experiments shewed that six times as much steam, as was simply sufficient to fill the Cylinder, was expend- ed at each stroke of the piston. He at once attributed this increased expenditure to the cooling of the Cylinder. The great quantity of injection water next engaged his attention, and the high heat the Cylinder retained in spite of the large quantity admitted. By adapting a bent tube to a common tea- kettle, and immersing the end of the tube in a vessel of cold wa- 206 BLACK. ter, he passed steam from the kettle into the cold water, by which the steam was condensed. The temperature of the wa- ter was increased by the heat of the condensed steam ; and by inquiring into the gain of weight which had taken place when the water reached the boiling point, he inferred that it required six times the weight of the steam simply to effect its condensa- tion, without lowering its temperature, or. that 1S00 'measures of steam were capable of heating six measures of water to their own temperature, although the 1S00 measures are derived from no more than one measure of water. Thus he reached experi- mentally one of the most important facts of the doctrine of la- tent heat, a doctrine which had been that very year taught in the same institution, for the first time, by Dr. Black. On com- municating the result of his observation to that distinguished chemist, he received from him an explanation of that doctrine, which furnished the confirmation and rationale of the pheno- menon he had observed. His experiments also shewed him that the pressure of steam increased nearly in geometric progression, while its tempera- ture was raised in arithmetic. The decrease of tension at low- er temperatures follows a similar law ; and hence the pressure of the atmosphere on the piston never acted with a force great- er than eight pounds per square inch. 156. Thus, then, the cause of the imperfections of Newco- mems engine became apparent at one and the same time, by the aid of actual experiment, and by the application of the ge- neral theory of Black. But it was far less easy to point out the remedy for these defects than to discover the cause. It was evident, that to obtain all the power the steam was capable of exerting, the Cylinder should not be colder than the steam which entered it ; while, on the other hand, the condensed steam should not raise the injection water above the temperature of 100° at the very outside, while a lower temperature would be preferable. The mode which he adopted, of meeting these two requisites, is as simple as it is ingenious ; yet it was not attain- ed without great study and reflection on his part. WATT. 207 157. It was not until a year after his performance of the ex- periments we have spoken of, that it occurred to him that if a communication were opened between the Cylinder of the steam engine, and another vessel exhausted of air, the steam would rush suddenly into the empty vessel ; and that, provided the latter were kept cool by being immersed in water, or by injec- tion, the steam would continue to flow until the whole were condensed. 158. With this idea he appears to have entertained, at the same time, the intention of using steam itself as the moving power instead of atmospheric pressure, and his experiments on its elasticity had shown him that a small increase in its tempe- rature would probably give a very considerable addition of pow- er. To effect the latter part of his plan, it would be necessary to make the piston-rod work air-tight, through a lid or cover adapted to the cylinder. A modification of the common air- pump had an arrangement that served as a model of the latter method, the barrel being covered by a lid, having at its centre a collar of leathers, by which the passage for the pump-rod was rendered air-tight. 159. It next became obvious that the piston could not be rendered tight in this case by keeping a mass of water floating upon its surface, for this liquid would have been speedily eva- porated, and would have wasted rnlich heat. Hence, more per- fect workmanship would be required, and the packing must be moistened with a liquid that did not boil, except at a tempera- ture higher than that to which the steam was ever raised. Oil is a liquid of this description ; but tallow, which becomes fluid at a temperature below that of ordinary steam, is still better. It is said that he originally proposed a packing of leather, but as this substance chars and cranks at a comparatively low temper- ature, it is unfit for the purpose, and bands of hemp were em- ployed in its stead. 160. To keep the cylinder from losing heat too rapidly, he conceived the idea of enclosing it in the Jacket. 20S WATT — ROEBUCK. Two methods occurred to him of keeping up a vacuum in the condenser. The first was that of adapting a pipe thirty- four feet in length, plunging at its lower end into a reservoir of water ; the second, that of exhausting the vessel by a pump. The former being applicable in but few cases, he chose the lat- ter for general use ; and we have, indeed, no instance of the first being applied in practice. 161. These views were submitted to the test of experiment, first in an apparatus of small size, and finally in a working model, whose Cylinder was nine inches in diameter. The re- sults were as satisfactory as his most sanguine expectations could have anticipated, and convinced him that he had dis- covered the means by which all the physical defects of the ancient engines could be remedied ; the steam no longer wasted by admission into a cylinder cooled by injection ; and a va- cuum far more perfect obtained, than had ever been before reached. The expense of constructing a steam engine, and the difficul- ty of inducing capitalists to embark in an untried scheme, seems to have deterred him from bringing it forward : and he devoted himself, for upwards of three years more, to pursuits far beneath the powers of his mind. 162. It is, even at the present day, rare to find in men whol- ly devoted to business pursuits that acquaintance with phvsi- cal principles which will enable them to judge of the merits of an improvement in the arts that rests wholly on those princi- ples. And such was the invention of Watt, which differed, as far as superficial examination could reach, from the engine of Newcomen, only in the addition of a cumbrous appendage, of which the light of physical science alone could exhibit the appropriate use, and manifest all the importance. For informa- tion of that description it was in vain to seek among the tra- ders of Glasgow at that early period, and Watt wisely determin- ed to keep his discovery to himself until he could meet with a person qualified to appreciate its merits. Such a coadjutor he at last found in the celebrated Dr. Roebuck, a person to whom ROEBUCK BOLTON. 209 Great Britain is under great obligations as the founder of the Carron works, in which the manufacture and application of cast iron was brought to that degree of perfection, which has added so much to the wealth of that country. Educated in the most liberal manner, and for a learned pro- fession, he became an adept in all the chemical and physical sciences of the day, and had applied his knowledge to the es- tablishment of a chemical manufacture whence he was deriving enormous profits. These he undertook to apply to mining for coal and iron at Kinneil, and to the establishment of the cele- brated manufactory of iron, of which we have spoken. 163. His scientific intelligence at once appreciated the whole merit of Watt's improvement, and he gladly furnished the funds for constructing an experimental engine, which was tried in the drawing of water from one of the mines at Kinneil. This engine worked as well as had been anticipated, and Watt was furnished by Roebuck with the means of securing his inven- tion from piracy, in the form of a patent. In return for his advances. Roebuck became joint proprietor of the patent, and from his capital and influence, Watt had rea- son to hope for the speedy introduction of his invention into general use. But Roebuck had embarked in schemes be- yond the reach of his finances, and of these, one, so far from being profitable, was ruinous to his fortune. The Carron works indeed flourished, but the mining speculation made no re- turns ; and so far from being able to assist Watt any further, he was himself compelled to abandon, not only his least promising scheme, but also that which was going on successfully, and thus to leave to others to profit by the fruits of his intelligence and enterprize. In the wreck of his affairs, his share of Watt's pa- tent passed into the hands of his friend Bolton of Birmingham. This skilful and enterprising merchant was not only well qualified to appreciate the merit of Watt's invention, but pos- sessed the capital, by the aid of which alone it could be brought into successful operation. 164. The first patent of Watt is dated in March, 1769, and 27 210 GAINSBOROUGH — WATT. when an application was made to Parliament for its extension-; opposition was made, on the plea that the most important part of his invention, the separate condenser, had been invented at a period at least as early as the date of the patent, by a person of the name of Gainsborough. This claim was, however, set aside in consequence of clear and decided proof that Watt's method was not only original with him, but earlier in date. 165. It does not appear that any other of the methods by which Watt had rendered his engine so superior to all former ones, had occurred to Gainsborough or any other person. Still we have no reason to doubt that Gainsborough had actually, and by investigations of his own, reached the plan of a separate condenser; and we cannot but believe that the study of the doctrine of latent heat might have led others, at a date not much later, to a similar discovery. That the improvements in physical science had rendered the world ripe for the introduc- tion of Watt's invention, need be no diminution of his great merit ; for, if not the only one who thought of the remedies we have mentioned above, he was undoubtedly the first, and prior by several years to any other person. Nor is it merely by pri- ority of invention that Watt is to be distinguished ; there was a finish and completeness about every plan that emanated from his mind, which suited it at once for practical usefulness. This, indeed, was a peculiar trait of the genius of Watt ; in- vention was with him so much a habit, that he rarely examin- ed any project without suggesting improvements ; while caution, almost amounting to timidity, led him to keep back from the world his own discoveries until he felt assured of their suc- cess. This excess of caution would probably have retarded, if not wholly prevented his success, had he not been fortunate in his connexion with a partner, who possessed the boldest spirit of mercantile enterprize, united to the most consummate judg- ment and prudence. In this point of view we may consider the world as being almost as much indebted to the intelligence and business ability of Bolton as to the genius of Watt. English writers have spoken of Watt as illiterate and defi- cient in education. If he is to be judged by the standard of WATT. 211 classical knowledge, to which alone the name of learning is given in that country, the assertion might be true. But we should rather be inclined to cite him as an instance of an edu- cation exactly directed to the purpose of rendering him useful in the important career to which he was called. The public schools of Scotland, if the mere pedantry of classical knowledge be neglected, give that species of instruction which extends the power of using the vernacular tongue for all business and prac- tical purposes ; Watt, besides, became a good practical geometer, and his early pursuits compelled him to be acquainted with all the physical science that was then known. When to this was added practical skill in mechanical operations, we cannot but think that Watt had derived from education that knowledge which was exactly suited to render him eminent. 166. We have stated that the engine of Watt dispensed with the use of water for the purpose of keeping the piston tight, and that a greater degree of accuracy was in consequence re- quired in the workmanship. The Cylinders of steam engines are cast hollow, by means of a core that fills up a part of the mould. They are then reamed out, and brought to the proper size by means of a borer, or tool affixed to a revolving axis. Great improvements in the boring of Cylinders were introduc- ed by Smeaton at the Carron works, but the method was not, at the date of Watt's patent, so perfect, as to give the interior a form wholly independent of the original shape of the cavity. But Watt had the advantage of receiving, just as he was about establishing a manufacture of steam engines at Bolton's works of Soho, near Birmingham, a new method of boring. This was the invention of Mr. John Wilkinson, a proprietor of iron works, at Birsham, near Chester. Watt immediately availed himself of this improvement, and was, by means of it, enabled to furnish Cylinders of a perfection of workmanship that had previously been despaired of. In a Cylinder of fifty inches diameter, con- structed by him at an early date, the greatest error was less than the sixteenth part of an inch. This perfection of workmanship was almost essential to the introduction of steam into general use, as a prime-mover of machinery ; for, although the process 212 WATT. of raising water had been, and might still have been, effected with advantage by Cylinders of less accuracy, the durability of the engine would, even in this very limited application of its powers, have been lessened, while the delicate operations it now performs would have been impracticable. This method of boring was still further improved, until a six foot Cylinder could be bored, with no error greater than the fortieth part of an inch. 167. In the first engine of Watt, the piston was attached by chains to the lever beam, from the opposite end of which the pump-rod, loaded with a weight, hung. The primitive posi- tion of the instrument is therefore the same as in that of New- comen, namely, the pump-rod preponderates, and holds the pis- ton in its highest attainable position. The engine has three valves, by one of which steam is admitted beneath the piston into the Cylinder, where it displaces the air, and fills its cavity wholly. So soon as the Cylinder is thus filled with steam, this valve is shut, and the remaining two are opened ; through one of these the steam passes into the condenser, which acts to convert the steam into water, partly in consequence of the cold- ness of its sides kept constantly immersed in a cistern of water, and partly by the aid of a jet of injection water ; through the other valve, steam flows from the boiler and presses down the piston, thus causing a motion at the opposite end of the beam, and raising the pump-rod. The piston having reached its lowest position, these two valves are closed. It thus becomes neces- sary that the piston should be again raised to its original position by the weight attached to the pump-rod. This might have been done by allowing the steam situated above the piston to escape into the air, and permitting steam to flow into the lower part of the Cylinder. The pressure on both sides of the piston being thus nearly equalized, the weight would have preponderated and raised the piston. But in this case a Cylinder full of steam would be condensed at each descent of the piston, and another allow- ed to escape, without effect, at each ascent. The waste of the last-mentioned quantity of steam was thus obviated by Watt ; a pipe was adapted to the side of the Cylinder ; in this the two watt's single engine. 213 steam valves were placed so that the communication, from the lower of these valves with the boiler, could only be effected by- opening the upper one. Hence, in first filling the Cylinder, both of these valves must be opened at once ; in all other cases they act alternately. The upper valve was placed above the passage by which steam enters into the upper end of the Cylin- der, and thus only the lower steam valve intervened between the steam acting on the piston from above, and that rushing into the condenser from below. The piston having reached its lowest position, the steam and condensing valves are shut, and the valve in the side pipe is opened ; a communication is thus made between the steam above the piston, and the part of the cylinder beneath it, the weight of the pump-rod then acting upon the piston, will meet no other opposition than the friction of the piston itself, and the resistance which the steam experi- ences in passing through a pipe ; the weight will therefore pre- ponderate, the piston will be drawn up, and the steam will cir- culate from the upper side of the piston through the side pipe, and fill the space beneath the piston. The upper valve alone is a steam- valve, except when the en- gine is to be set in motion ; at which time the second valve is opened with it, and admits steam to the lower part of the cylin- der. In all other cases the latter is merely a valve of commu- nication, and may be called the equilibrium valve. The third valve may be called the condensing valve. This arrangement may be better understood by the inspec- tion of the following figure, and the uses of the parts will be understood by reference to the description of the double-acting engine. 214 watt's single engine. watt's single engine. 215 a, Cylinder. 6, Piston represented in its primitive position. c, Steam-valve. d, Steam-pipe. e, Equilibrium-valve. f, Condensing-valve. g, Pipe leading to condenser. h, Condenser. i, Air-pump. k, Hot-water cistern. I, Air-pump rod. m, m, m, Hand Gear. n. ?i, n, Tappets of the Plug-frame. o, Side-pipe. /?, Hot-water pump, q, Cold-water pump. r, Cold-water cistern. s, Foot-valve. w y x, Piston-rod, working in a stuffing-box at w. y, Working-beam. 168. Still further to diminish the loss of heat, Watt pumped back the water of condensation into the boiler ; by these several improvements so great a saving of fuel was obtained, that the pa^ tentees asked no other remuneration for the use of the invention, except one-third part of the value of this saving. In a single mine in Cornwall, where three of their engines were employed, this compensation was commuted for £8000 sterling per an- num. The valves still continued to be opened and shut by an ap- paratus similar to the plug-frame and hand-gear of Beighton, but improved, and rendered more easy in its action ; the former became a part of the rod of the pump by which the vacuum of the condenser was maintained. The pump and piston-rods were still suspended by chains from circular arcs forming parts of the lever-beams, and the air-pump rod was suspended in the same manner; the latter, therefore, required a weight to return it downwards, after it had been raised by the beam. 216 STEWART — CLARKE. The condenser and pump underwent various modifications before Watt was satisfied with their action, and finally assum- ed the form described in treating of the double-acting engine. 169. Previous to the time of Watt, there had been but little demand for steam, as a moving power, for any other purpose than that of raising water : and for several years after his first researches, the state of the manufactures of England was not such as to require powers beyond what could be obtained from natural waterfalls or the action of the wind. Savary had. in- deed, proposed to make the water raised by his apparatus fall upon an over-shot wheel, and thus to apply the power of steam to any manufacturing purpose whatever. Similar projects had been entertained in relation to the engine of Newcomen. Neither of these, however, could have been applied to any ad- vantage, in consequence of the great cost of fuel they must have occasioned, particularly as the effective power of a wheel is considerably less than the absolute mechanical force of the water employed. The several projects of steam-boats that we shall hereafter speak of, necessarily required rotary motions : but these were all imperfect and abortive. Leupold's engine alone, had the two pistons been applied to cranks situated upon the same axis, could have produced a rotary motion ; but the inventor does not appear to have been aware of the value of this part of his own invention, or at least did not consider this application of it of importance. 170. In 1757, a person of the name of Fitzgerald attempted to take off a rotary motion from the piston of .Newcomems en- gine by means of ratchet-wheels, that could be forced forwards during: the descent of the piston, but would remain fixed dur- ing its ascent. The continuity of the motion was to be kept up by a fly-wheel. This was too imperfect a method to be suc- cesful, and had no result. A similar project was entertained by persons of the names of Stewart and Clarke, who attempted to apply it to sugar mills in Jamaica ; but this, like the other, was abandoned as impracticable or useless. Still later, at a colliery watt's double engine. 217 in England, a drum for raising coal had been worked by an at- mospheric engine, but even this rude apparatus was but imper- fectly driven. Hence it was left for Watt to fit the steam en- gine for general use, as well as to improve it in its application to the sole purpose in which its energies had been successful before his day. 171. The first step towards making the steam engine capable of producing a continuous motion in machinery, was to make the piston work during its ascent as well as its descent; for both in the atmospheric engine, and the first engine of Watt, the moving power is exerted only to press down the piston, which is afterwards returned to its original position by the action of a counterpoise. Watt, whose caution was equal to his genius, proceeded in his inventions by slow and gradual steps. His first engine hardly varied from Newcomen's in external form, and was, in truth, rather a great and all-important improvement upon that imperfect apparatus, than an invention absolutely original. In the same manner his double-acting engine was obtained by a slight and simple alteration of his former one. It was required that the piston should be forced upwards as well as downwards by the steam; he effected this by adding one additional valve to the three employed in his single-acting engine. The equili- brium valve of the single engine became a steam valve, instead of serving as a mere communication between the opposite sides of the piston, and the valve that was added was one forming a communication between the condenser and the upper part of the cylinder. Hence it was necessary that the steam-pipe should extend to the lower pair of valves, and the condensing pipe to the upper ; the side pipe was thus doubled. The improved hand gear of Beighton was still retained, to open and shut the valves. The mode of operation of this engine has already been fully explained, it is therefore unnecessary to repeat it here. But it did not at first assume the perfect form in which it has been re- presented in chapter IV. The steps by which Watt proceeded were as follows. The rectilineal motion of the piston-rod hav- 28 21S WASHBOBOUGH — PICKARD. ins: been rendered capable of exerting an equal force, both dur- ing its ascent and descent, a connexion with the beams by chains was no longer sufficient : for although they would be efficient in drawing the beam downwards, their flexibility would not admit of their forcing it upwards. It hence became necessary that their connexion should be made of a rigid mate- rial, and yet in such a manner as to permit the rectilineal motion of the one to accommodate itself to the circular motion of the other. True to his general system of slow and cautious im- provement. Watt attempted at first no violent alteration. The circular end of the lever beam was merely cut into teeth, or ra- ther had a toothed segment bolted to it. and for the chain a rack was substituted, which caught into the teeth of the seg- ment. Thus the stroke of the piston became effective, both during its ascent and descent. It will be obvious, however, that this is a rude and imperfect method, and he speedily con- trived a better in the shape of the parallel motion. 172. About the time that Watt undertook to adapt his prin- ciple to general purposes, an engineer of the name of Washbo- rough attempted to attain a similar end, by means of the atmos- pheric engine. His plan was very similar to Fitzgerald's, and was improved by Pickard, Engines of their joint construction came into use in Gloucestershire, and at the block manufactory of Mr. Taylor at Southampton. The first actual application of the crank to the steam engine seems to have been due to Pickard. 173. Watt however, at an earlier date, conceived the idea of making two single-acting cylinders act upon cranks situated upon the same axis, and thus produce a continuous rotary mo- tion. In the attention which the introduction of his engine into use for raising water required, this idea was suffered to re- main unimproved until he had completed the plan of making the engine double-acting, and of communicating the motion of its piston to the beam, by the rack and toothed segment. To ap- ply the motion thus obtained to general purposes, it became ne- cessary to convert the reciprocating motion of the beam into CRANK. 219 one continuous and rotary. That a crank was the most sim- ple and obvious means of performing this has already been shown, and Watt recurred at once to the idea we have stated above, as having suggested itself to him, except that in the double-acting engine but one crank would be necessary. Va- rious simple and familiar instruments have rotary motions, that are produced by this instrument. Among these may be men- tioned, as of most frequent occurrence, the Potter's wheel ; the turning lathe; and a variety of the spinning-wheel, then in constant use, although now nearly obsolete. The friends of Watt assert that his views were communicated by a workman, who passed from his employ to that of Pickard ; but there is no reason why both may have not fallen upon the same simple and obvious plan of producing the same kind of motion. Be this as it may, Pickard took out a patent for the application of the crank to produce a rotary motion in the Steam Engine, and Watt, satisfied that his own ingenuity could provide a substi- tute, did not attempt to contest it, but left it as an obstacle in the way of his other competitors. 174. To produce the effect that the crank was intended to perform, he adapted to his engines an apparatus called by him the sun-planet wheel. This is represented on the following page. 220 SUN-PLANET WHEEL. z, Connecting Rod. z Zy Fly-wheel. a, Wheel fixed upon the axis of the fly- wheel, b, Wheel revolving upon a pivot at the extremity of the con- necting-rod. The wheels a and b, having equal radii, whose sum is equal to the length of the stroke of the engine, the teeth of the wheel b will apply themselves to those of the wheel a, during the whole motion of the engine. The wheel b will turn the wheel a around, and cause the axis of the wheel, to which the latter is attached, to revolve. It will be obvious that the axis of the fixed wheel must re- volve twice as fast when driven in this manner as it does when propelled by means of a crank ; there are, in consequence, cases where it may be better suited to the required work than the WATT. 221 crank ; such was the case in the earlier adaptations of Watt's engine to manufacturing- purposes. When, however, in his sub- sequent engines, a crank furnished a better and more suitable speed, he did not hesitate to employ it, confident in the priority of his own claim to its application. 175. We have thus seen Watt taking up the engine in a very imperfect state, gradually perfecting it in its application to its ancient purpose, and finally rendering it universal in its uses. He also made many accessory improvements, by which its use was rendered more easy and certain. Of these we may parti- cularize the steam-guage, the barometer guage for the vacuum of the condenser, the self-acting feeding apparatus, the self-regu- lating damper, and the form of boiler which is yet most general- ly employed with double-acting condensing engines. Under his directions the hand-gear of Beighton was first im- proved, and finally superseded by the eccentric, and the long slide valve was in troduced. The eccentric and slide valve were claimed also by Murray of Leeds; but although he may, perhaps, be entitled to the merits of a separate discovery, Watt was suc- cessful in showing the priority of his own claim to them. After the parallel motion was added to the engine, the beam still continued to be of wood, as was the connecting rod. Subse- quent steps led to the substitution of the more inflexible materi- al cast-iron, and the pivots of the parallel motion, instead of be- ing, as before, placed beneath the beam, were now cast upon it, and turned down to the proper size and shape. Frames and pillars of iron to support the beam were gradually substituted for the floors of buildings and walls, that were at first used ; brass boxes forming the sockets for all the circular motions were introduced ; and the external beauty of the machinery improv- ed by perfection of finish, that added equally to the power and durability of the engine. It was in 1778 that Watt made the piston act during both its motions, and he did not cease, to the very end of his life, to extend its usefulness and improve its structure. No valuable addition to the condensing engine was made ex- cept by himself, or under his direction, if we leave out those of 222 HORNBLOWER. Murray, which we have mentioned, and to which Watt was able to substantiate an earlier and more authentic claim. 1 76. The application of steam acting expansively is also due to Watt. One of his single engines employed it in this man- ner at Soho as early as 1776 : and he used it also in his double- acting engines almost from their first construction. We have seen, in another place, that he did not reap all the advantages of which this method is capable, nor was he permitted to hold it. as an invention of his own. without contest. Two brothers of the name of Hornblower, in 1752 took out a patent for the use of the same principle, but by means of two Cvlinders. In the first of these the steam acts by its tension, and produces an effect equal to that which it does in the high pressure engine. On escaping from this it enters a second and larger Cylinder, in which it expands, and from the opposite side of whose piston the steam flows alternately to the condenser. Thus. then, the resistance which the atmosphere opposes to high steam, is re- moved, and the effect of its expansion brought into play. As the separate condenser interfered with the patent n^ht of Watt, this plan could not be brought into use. nor was it desirable that it should: for an engine of equal power on this construction is more costly than that of Watt, and it is difficult to make the two Cylinders employed, one containing high the other expan- sive steam, act in such harmony, that one of them shall not be retarded by the other. 177. Five years of Watt's patent had run out before he had fairlv introduced his single engine into use. He therefore made application, in 1775, for an extension of the usual period, and the application was. after much opposition, granted. Thus, by a noble effort of national generosity the profits of his discovery were secured to him for a term of years sufficient to remune- rate him for his labours and sacrifices. The patent-right thusex- tended became the object of a series of attacks, leading to judi- cial investigation ; but in spite of the interested and continual opposition, the patent was in every case maintained. It is, indeed, highly to the credit of the institutions of Great CONICAL PENDULUM. 223 Britain that this long contest should, in all its points, have been constantly decided in favour of him to whom the world, after an interval that has deadened all partial feeling, assigns unani- mously the merit of discovery. Such an honourable result, we fear, couid hardly have been attained in our own country, in which the most carefully guarded patent-rights are proverbially insecure, and those inventions w T hich have added most to the national wealth, have been those that have been of least pecuni- ary value to the inventors. A conical pendulum had been applied to mills of various des- criptions before the time of Watt. The suggestion of the valu- able use to which it might be applied in the steam engine, is said to be due to a Mr. Clarke of Manchester ; we do not, how- ever, know whether he ever applied it in practice. It was, whe- ther as an original invention of his own or not, speedily adopt- ed by Watt, and adapted to all his engines where regularity of motion is needed. 178. The patent for Watt's double-acting engine is dated in 1782, and in the same year one was erected at the Bradley Iron Works on this principle. It had the toothed segment on the lever beam, and a rack attached to the piston-rod. Since that period the improved engine has been introduced to a very great extent in the manufacture of iron; for impelling the blowing apparatus in blast furnaces, and for rolling, hammering, and slitting wrought iron. The patent for the parallel motion was issued in 1784, and in that year the Albion Flour Mills were erected in London. Two of Watt's double-acting engines, of 50 horse power each, were applied to drive twenty run of mill stones ; the establish- ment was conducted with great profit until the year 1791, when the building, with all the machinery and stock, was consumed by fire. It was suspected at the time to be the work of an incen- diary, instigated by those who, by the aid of other prime mo- vers, were unable to compete with the improved agency of steam. The experiment, however, was so far successful, as to satisfy all that the engine might be advantageously adapted to almost every species of manufacturing industry. 224 CARTWR1GHT SADLER. In 17S5 the first cotton mill moved by steam was erected by Messrs. Robinson and Papplewick, in Nottinghamshire. In 17SS, a coining apparatus for copper was erected at Soho, and driven by a steam engine ; the machinery there applied has been imitated at the Royal Mint of Great Britain and the Impe- rial Mint at St. Petersburgh, and all were set in motion by dou- ble-acting engines on Watt's construction. In 1793, cotton was first spun at Glasgow by steam. In the year 1 793, it was introduced into the woollen, worsted, and flax manufactures; and in 1797 was employed at Sheffield for grinding cutlery. 179. In the yearlSOO, the extended term of Watt's patent ex- pired. Up to this time the introduction of his engine into use had been slow. This has been ascribed to the prejudice enter- tained against the monopoly, but probably is in some measure due to the fact that the arts did not keep up with the rapid im- provement of the steam engine. At this date the steam engines in London did not exceed 650 horse powers ; in Manchester, 450 were in use : and about 300 at Leeds ; while upon our own continent but four engines of any importance were to be found, two of them at Philadelphia and one at New- York, all employ- ed for raising water. ISO. During the continuance of Watt's patent, various plans were proposed, which were rendered abortive in consequence of his being in possession of the sole right of using the only plan by which low pressure engines could be rendered efficient, the separate condenser. Hence, with the exception of Hornblower, against whom Watt and Bolton obtained a verdict, there are no important names to be mentioned, except those of Cartwright and Sadler ; these two engines are chiefly remarkable for the suppression of the lever-beam. Watt, as we have already stated, proceeded in his improve- ments slowly and gradually, and they were all applied to the form in which he found the engine existing. Hence the beam, a heavy and cumbrous appendage, formed a constant part of all his engines, as it had done of the original pumping engine of CARTWR1GHT SADLER. 225 Newcomen. The parallel motion, the connecting-rod, the sun and planet wheel, the rods of his air, cold, and hot water pumps, were all adapted to this part of the ancient apparatus ; and from the use made of it in working the latter, it appeared to be almost indispensable. In Cartwright's engine the beam was suppress- ed altogether, and with it the separate pumps. A cross-head, was placed upon the piston-rod, bearing two short connecting rods, that turned the cranks of two wheels of equal diameter catching into each other, and a pinion attached to the axis of the crank was driven by one of them. As this engine did not work as well as the double-acting engine of Watt, it merits no further notice at this stage of the history ; although, had it ap- peared before the improvements of Watt, it would have been of great value. Cartwright is, however, to be mentioned with high praise as the inventor of the metallic packing for pistons, which, as has been stated on a former page, promises to super- sede all others. Sadler's engine was a single-acting engine, differing from Watt's principally in the position of the equilibrium valve, which was situated in the piston itself. A wheel was fixed to an axis passing at right angles through the top of the piston- rod ; this worked between guides, and on the opposite end of the axis was placed the connecting-rod that turned the crank of the fly-wheel. The rod of the air-pump was worked by a short lever, the centre of whose motion was at the end, instead of the middle, as in the ancient beam, and which, having no other work to do but that of pumping, was much lighter than the latter. From the date of the expiration of Watt's patent, the use of the double-acting condensing engine has been extended in a rapidly increasing ratio, insomuch that far more engines are now made at Soho, in spite of the ardent competition of various manufacturers, than were ordered while Watt was possessed of the sole right of making engines, as well as using his principle. 181. The improvements made in the condensing engine since that period have principally consisted in the finish and perfec- tion of the parts, and in this Murray of Leeds has been most distinguished, his engines having a beauty of proportion and 29 226 HURRAY WOOLFI accuracy of workmanship exceeding most others. In England, the beam has been continued in almost all cases e xcept in the en- gine of Mandslay. while in this country it has. in many instan : been laid aside. The fast engines constructed in America for Pulton's steam-boats have the form represented in PI. TIL. which is superior to that of either Sadler or Maudslay. Where the beam is retained, the parallel motion has been supe everal American engin ss, : simple slide to guide the con- necting strap. The adaptatiou of this to the lever beam is the invention of Mr. R. L. Stevens, whose name we shall have occa- sion to quote here; .::-.. iccessml constructor of steam-boats. The eccentric and slide valve belong to this last period of the history of the double-acting engine, but their invention has been already referred to. 182. The use of the expansion :: s:eam has been stated to have originated with Watt, and we hav r mentioned the attempt of Hornblower to adapt the same principle to an engiue com- posed of two Cylinders. This, which was defeated in con se- quence of its interfering with Watt's patent, was irived in 1804 by Woolfe, with a boiler for generatin g r higii ste am . T i s n_:ne has been found to work to great advantage: but for reasons mentioned in speaking of Homblower's engine, it will be obvious that steam of equal tension would act to still greater advantage in an engine composed of but a single Cylinder. This last method has received great extension in several Amer- ican engines, and in the pumping e nwall. I: has been seen that the plans proposed during the term of s patent were either such direct infringements :s to be prohibited by legal proceedings, or wholly inferior in utility to his inventions. The very year, however, that saw his patent expire, also saw the introduction into use of: vines, that, had they been brought into perfection bef: v.: have :~>m- peted with that of Watt upon equai the history of one of these we are compelled to go back almost to the date of arlierc ?s 153. Oliver Evans, well known in this counrrv as an excel- EVANS. 227 lent mill-wright, and as the inventor of the labour-saving ma- chinery in grist mills, entertained the idea of the possibility of propelling wagons by the action of high steam as early as 1 772. Soon after, he ascertained, by experiment upon a small scale, the practicability of so doing ; and in 1786 # applied to the State of Pennsylvania, which (under the old Confederation) had not parted with this attribute of sovereignty, for an exclusive privi- lege. It is well to remark that his engine was from the first intended to be double-acting, and that even the last-mentioned date is but little later than the construction of the engines for the Albion mills, whose principle was long kept secret by Watt. His applications, both for private and public patronage, were treated as the reveries of insanity, and it was not until 1801 that success in his profession enabled him to raise the funds for e r ecting an experimental engine. This was first applied to grind gypsum, and afterwards used in sawing marble. It was publicly exhibited in Philadelphia in that year. In 1804 he was employed by the corporation of Philadel- phia, to construct a dredging machine to be worked by steam ; with this he made successful experiments, both on locomotion and navigation by steam, that will be mentioned in their proper place. Not the least of the improvements of Evans lies in the form of his boilers, which he was the first to make in the form of a cylinder ; a form that we have already shown to be prefer- able to any other yet proposed. His first experiments were made with a gun barrel, and he steadily adhered to that form in his subsequent operations. The engine of Eyans retained the lever beam of Newcomen, and has been copied in this respect in many American engines, of which the beautiful one figured on PI. V., is a specimen. In others, the arrangement in PI. VII. has been adopted, and others again are horizontal, as represented on PL VI. The latter form has hitherto been principally used on the Mississippi and its branches. The high pressure engine came more early into general use in the United States than it did in Europe, and long experience has rendered its proportions better understood in our country than they are in England. It is with us the favourite form, except in the steam-boats of the Atlantic coast. In these, a fear of the greater danger, with which it was thought 228 WATT. to be attended, has prevented its introduction. It seems, how- ever, to be now almost conceded, that, with proper precautions, boilers generating high steam may be rendered as safe as any others ; and hence the conclusion has been drawn that high steam, acting expansively, as it is the most powerful application of steam, will, wherever circumstances will admit, supersede all other methods. 184. The year 1S01 also witnessed the construction of the high pressure engine of Trevithick and Vivian. The boiler in this case was a Cylinder of cast iron ; the fire was made within it, and hence it is less safe than the boiler of Evans. The Cy- linder was immersed in the boiler, in order to retain the heat of the steam. In the first engines it was attempted to condense the steam ; but this is always attended with disadvantage, un- less when the steam has had an opportunity of cooling itself by expansion. The fly-wheel, connecting rod, and crank were above the Cylinder, and no parallel motion or beam was need- ed. In the application of the engine to locomotion, a plan of connecting rods like those on PL VII. was finally adopted, but, so far as we can learn, only one connecting rod was used at first, even in this case, as in the engines of Sadler and Maudslay. Such is the history of the engines that are now in actual use, or have served as steps to the present state of the art. Another class remains to be mentioned. 1S5. Watt had included in his first patent a method of pro- ducing a rotary motion by the direct action of steam, but had with sound judgment abandoned it in favour of a double-acting Cylinder engine. In spite of this virtual acknowledgment of the inferiority of this principle, innumerable projects have since been entertained of rotary engines. The result of these exper- iments may be summed up in a few words. The advantage to be derived is, in fact, of but little moment, while the mechanical difficulties that lie in the way are such as have hitherto prevent- ed any engine having a rotary motion, produced by the direct action of steam, from coming into general use. Among the various rotary engines which have been proposed, we may mention one by the Hon. R. Sherman of Connecticut, ROTARY ENGINES. 229 which was exhibited recently in New- York, and which per- formed well. That which is exhibited in the annexed draught, has been Fig. 1. Fig. 2. 230 WATT. for several years in actual use in the western part of the state of New- York, and has been employed to propel a boat on the Morris Canal. In this engine the gates marked 1 and 1 receive the pressure of the steam which enters through the passage 5, and is discharged at 6. These passages are so situated that one of the gates will be always receiving the pressure, and dur- ing its action the other gate is lifted over a partition or diaphragm by means of a curved wheel 7, which runs between friction rollers. Of this engine Fig. 1. is a section, and Fig. 2. a plan. The principle of reaction, which has been attempted in Bar- ker's mill, where water produces a rotary motion by issuing from holes placed near the ends of a moveable arc, has also been pro- posed as a mode of using steam. We have described an in- stance of this kind in the engine of Avery. The Cylinders of engines have occasionally been suspended on trunnions; in this case the piston-rod may be applied directly to the crank. The earliest of these was one constructed by French in 1S08, in a steam-boat in the Harbour of New- York, a model of which of the same date is among the apparatus of Columbia College. 1S6. In the brief sketch we have thus given of the History of the Steam Engine, many ingenious contrivances and inven- tions have been passed over. These have been omitted for want of space, and because few of them, however ingenious, have had any prominent effect in introducing the steam engine into more general use. The different forms of boilers that have been pro- posed, or even actually used, would occupy no small room. The object of our essay is, however, accomplished when the engine has been traced, from its first rude beginning to those forms in which it is found in most common use, and when we have notic- ed those different inventions that have tended to facilitate its pro- gress, or by which it has been fitted the better to subserve the pur- poses for which it was invented. The most important step is un- doubtedly that made by Watt, and it is remarkable in the history of the arts, not more from the immense value that it has had in its practical application, than for being the result of scientific research and the study of physical principles, by the most ele- gant and accurate processes of induction. CHAPTER IX. APPLICATIONS OF THE STEAM ENGINE. General view of the applications of the Steam Engine. — Rais- ing water. — Grinding corn. — Cotton Spinning. — Naviga- tion. — BossuPs laics of the impact of fluids. — Principles of the action of Paddles. — JuanJs laws of the action of fluids on solids moving in them. — Maximum speed of vessels. — Poioer required to propel paddles. — Relation between the power and the surface of the Paddles. — Laws of the motion of Steam-boats. — Theory of paddle wheels. — Comparison between theory and observation. — Practical Rules. — Sug- gestions for the improvement of Steam Navigation.- — Steam-boat engines. — History of Steam Navigation. — Na- vigation of the Ocean by Steam. — Rides for Boilers of Steam-boats. — Application of Steam to Locomotion. — His- tory of the Steam carriage. — Conclusion. 187. The steam engine is now applied to almost every spe- cies of manufacturing industry, and as a substitute for the la- bour of men and animals in almost every art, and in many of the other cases in which they were formerly employed. In its earliest forms it was used to raise water, and still in its more perfect shapes fulfils the same object ; it performs almost every variety of manufacturing manipulation ; propels vessels through the water ; and drags carriages upon railways, and even upon common roads. 188. In raising water, pumps may be adapted to the beam of the engine, and the useful effect may be safely taken, at the raising of 24000ibs, one foot high per minute, for every horse 232 STEAM-BOATS power of the engine, estimated in the manner that has been pointed out. In pumping by the steam engine, it would appear from ex- perience that the stroke of the pump-rod should not exceed eight feet. The number of cubic feet of water which a pump will deliver per minute, may be found by multiplying together half the velocity of the pump-rod, the square of the diameter of the barrel, and the constant fraction 0.00518. The velo- city at which the maximum work is performed, and which is to be used in the foregoing calculation, is found by multiplying the square root of the length of stroke by the constant number 98. The proper velocity for a pump of 8 feet stroke is there- fore about 270 feet per minute, and as the pump-rod is suspend- ed from the end of a lever of equal arms, the velocity of the piston of the engine is the same. The friction of the pumps is estimated as equal to a column of water whose height is the sum of If feet for each separate pump or lift, and of ^,-th of the whole height to which the water is to be raised. These quantities must be added, therefore, to the height, in order to obtain the efficient resistance. Calling this H ; the number of cubic feet of water to be raised per minute, W ; the pressure of the steam on each circular inch p • and as- suming the velocity of the piston to be 180ft : D, the diameter of the cylinder of the steam engine is found by the formula, HxWX0.7332v p / The diameter of the barrel of the pump may be found by the formula,* d=V3.15W These rules are independent of the conventional estimate of horse power, and are therefore well adapted to practical appli- cation. The estimate of the quantity raised per horse power, which has been given above, has been far exceeded in practice by the expansive action of steam. The duty of the pumping engines in Cornwall has been raised in this way from an average of 17 * Grier's Dictionary. d=v( ] STEAM-BOATS. 233 millions of pounds for each bushel of coals to one of 4.3^ mil- lions, and one engine has raised more than 90 millions. 189. In Grist mills, it is estimated that the power of five horses is necessary for every run of stones, and for performing the work necessary to supply them, moving the whole of the usual labour-saving machinery. In applying the engine to this pur- pose, rotary motions of the proper velocity are taken off from the axis of the crank by systems of wheels and pinions. The dimensions of engines to perform this description of work are given in the following table :* ishels Ground per hour. Diameter of Cylinder in inches. Eushels Ground per hour. Diameter of Cylinder in inches. 4 12.5 26 29 6 14.6 28 29.8 8 16.75 30 31.1 10 18.5 32 32 12 20.2 34 33.3 14 21.75 36 34.2 16 23.25 38 35.2 18 24.75 40 36 20 26.25 42 37.3 22 27.25 44 38 24 28.1 48 39.5 The engine is supposed to be double-acting, condensing, and using steam without expansion. By employing the expansive action of the steam, an increase in the quantity of work which may be performed will be obtained, in ratios which may be inferred from our previous investigation of that method of action. 190. In manufacturing machinery, the motions are taken off in the same manner. It would be tedious, nay impossible, to recite every particular case of this sort ; we shall therefore limit ourselves to the spinning of cotton. In this branch of manufac- ture, it is estimated that each horse power will drive 200 thros- * Grier's Dictionary. 30 234 STEAM-BOATS. tie spindles, or 1000 mule spindles, and perform all the work of preparing the cotton for them. Those subjects which appear to require more full illustra- tion in this work, are the propulsion of vessels, and the motion of carriages upon rail-roads. 191. Vessels in which steam is used as the moving power, are generally propelled by the action of paddle wheels. These receive a continuous rotary motion from the steam engine, and the paddles tend to impel the vessel in consequence of the re- sistance which opposes their passage through the water. This method, which is the most simple and perhaps the most obvious, has in practice been found preferable to any other which has yet been proposed. In order to give motion to a paddle-wheel, it is only necessary that it should be fastened to the axle of the crank of any of the usual forms of engine. The wheel will thus be caused to perform a complete revolution in the same time that the piston takes to perform a stroke, estimating un- der that term its whole motion, from the time it leaves either ex- tremity of the cylinder, until it return again to the place whence it set out. The force which the wheels exert in propelling the vessel depends upon : the velocity with which they strike and move through the water ; the immersed area of the paddle ; and the fluid resistance of the water. The velocity of the vessel will depend upon the force with which the wheel tends to impel it on the one hand, and the resistance the water opposes to its progressive motion on the other. To determine the velocity a given engine will give to a vessel, and the conditions under which a maximum effect may be produced, is evidently a problem of great complexity. It does not appear yet to have been solved in a satisfactory manner, nor does it seem to be within the strict limits of analysis, in con- sequence of the great number of circumstances which must be taken into account. It is, however, possible, by reference to scientific principles, and comparing their results with the facts observed in practice, to form rules which may be of value in the construction of steam vessels. NAVIGATION BY STEAM. 235 The action of a paddle-wheel upon the water must be gov- erned nearly, if not exactly, by the laws which fluids follow in impinging upon solid bodies. These may be stated as fol- lows : 1. With equal surfaces equally inclined to the fluid, the re- sistances are nearly proportioned to the squares of the veloci- ties ; 2. With equal velocities, and equal inclinations of the sur- faces to the fluid, the resistances are proportioned to the areas of the surfaces ; 3. With equal velocities and equal surfaces, the resistances are nearly proportioned to the squares of the angles of inclina- tion, until the angle of incidence diminish to 50° ; beyond this, the resistance decreases more rapidly. 4. The measure of the action of a fluid upon a plane surface, to which it acts at right angles, is equal to the weight of a co- lumn of the fluid whose height is that whence a heavy body must fall to acquire the velocity, and whose base is the area of the surface on which the fluid acts. 5. As the resistance increases in the ratio of the squares of the velocities, there must be a maximum beyond which a given force cannot propel a plane surface through a fluid. 6. This maximum velocity being given, the maximum of effect will be produced by a paddle-wheel when it moves through the water with one third of that maximum velocity. 192. Experiments appear to be wholly wanting by which the maximum velocity of a plane surface, moving through a fluid, can be determined. It, however, happens that this may be deduced from observation of the rate at which the wheels of steam vessels move ; for there is in most cases a very consider- able excess of power, and hence the relative velocity of the wheel necessarily becomes that at which the work is most efficiently done, and is therefore one third of the maximum ve- locity with which the paddle might be impelled had it no work to perform. This relative velocity of the circumference of the paddle-wheel has been found in American steam vessels to be not far from 6£ feet per second. 896 NAVIGATION BY ;7I^M. The maximum velocity of a paddle-wheel when it he: work to perform, might therefore be taken at 19f feet per second. or 13 2 English miles per honr. Were the laws of the resistance of fluids to boilers moving through them identical with those of impact it might be infer- red that the last-named Telocity is also that of the maximum speed of steam vessels. If we were to take this as the limit, it might be inferred that the proper velocity to be given by the engine to the circumference of a paddle-wheel should be 26 feet per second in order to give to the vessel a velocity of 1 1 ± English miles per hour, and retain for the wheel a relative velocity or rate of motion through the water of 6£ feet per se- cond. Although this is by no means true, and although there are now many instances in which boats have been driven at a higher velocity than 13.2 miles per hour. We shall rest at this point for the present. i T he greatest s:eed of vessels, -ni consequently the maximum velocity of paddle-wheels, may be examined, either by the aid of theory, or ascertained by actual experiment. F : : the theoretic investigation we may have recourse to the princi- ples of Don George Juan. I: is s.eied by this author, that the resistance which opposes the motion of bodies moving in fluids, may be divided into three parts : — 1 A resistance growing out of the disturbance of the condi- tions of equilibrium, and arising from the friction of the fluid. This is a constant force- El The fluid resistance proper, which varies with the square of the velocities. This has so small a co-efficient, that it is in- sensible at small velocities, but increasing with their squares, j| speedily becomes the most important, and the first bears then so small a relation to it, that at mean velocities this need alone be taken into view. 3. The wwre raised in front of the moving body, and a want I support from behind, growing out of the space which the body leaves unfilled behind it when the velocity becomes NAVIGATION BY STEAM. 237 great. The mere resistance, growing out of these causes, in- creases with the fourth power of the velocities, but, in addition, the weight of the body must be raised upon an inclined plane, and hence will arise a limit beyond which the velocity of a ves- sel cannot be carried. At small and mean velocities, this species of resistance is wholly insensible, but it finally becomes, in consquence of its rapid increase, the most important of the three. It is when this occurs that we would fix the limit of the speed that can be ad- vantageously given to a vessel, or with which a paddle can be impelled. 194. The exact limit of speed at which the wave raised in front of a vessel becomes an insuperable obstacle to an increase of velocity, depends upon circumstances for which no theory can account. In our former edition we ventured to place this limit at 12 nautical miles per hour, and this was the greatest speed which had been reached in American steam-boats. In all which then existed, a prodigious wave was raised in front at rapid motions. This important resistance led our naval architects to modify the figure of the prows, and they have thus proceeded by successive steps, until, in the case of the steam-boat New- York, no perceptible wave is formed. It ap- pears, therefore, almost impossible in the present state of our knowledge to limit the speed which may be attained by steam- boats. The experiments of Juan were made at velocities less than the least which are now given to steam-boats, and there is good reason to believe that his theory ceases to be true at the higher velocities. The result of the practice upon the Hudson seems to prove, that at velocities exceeding 10 nautical miles per hour, the resistance, so far from varying with the squares of the velo- locity, becomes almost constant. This at least is certain ; every increase in the rotary velocity of the paddle-wheel, has been at- tended with an equal increase in the progressive velocity of the vessel. The expenditure of steam is, however, in a greater ra- tio than the velocity ; but this is easily accounted for when it is considered that, in order that it exert a given pressure on a pis- 23S NAVIGATION BY STEAM. ton in more rapid motion, it must have a greater tension in the boiler. Experiments have been made on a large scale upon the mo- tion of boats on the Firth and Clyde, and on the Monkland Canal in Scotland, by Mr. McNeill. His conclusions as are fol- lows : 1. In a wide and deep canal the resistance was observed to increase with the velocity, but not in any uniform ratio. 2. In a shallow and narrow canal the resistance had a limit at a certain velocity, and thereafter decreased with an increase of the velocity. 3. That the resistance bore a relation to the inclination of the keel. 4. That the boat rises in rapid motion, being in some cases, on an average, four inches less immersed in the water than when at rest. This rise was greatest at the bow, and least at the stern. Mr. Russell, who had observed similar facts, comes to the conclusion that at a velocity of 43.8 miles per hour the vessel would no longer be immersed, but would skim along the sur- face. This corresponds with what is observed in the ricochet of cannon balls, which occurs only so long as they retain con- siderable velocities. 195. To determine the power required to propel a paddle- wheel through the water with a given velocity, we must consi- der that the measure of force depends not only on the resistance overcome, but the velocity with which it is conquered. Hence it would appear that the resistance estimated as being equal to the weight of a column of water whose base is the area of the paddle, and whose height is that whence a heavy body must fall to acquire the velocity, is to be multiplied by the velocity per minute, and the product divided by the weight raised by the unit of power in the same space of time. The height due to a given velocity is found by dividing the square of the velocity per second by the constant number 64. The unit of power is 32,0001bs. raised one foot high, but this is probably reduced in the machine itself, as we have here- NAVIGATION BY STEAM. 239 tofore seen, to 24,0001bs. Hence we would have the following rule : Multiply together the cube of the relative velocity, the num- ber of seconds in a minute, the weight of a cubic foot of water in 62£lbs., and the area of the paddle in feet, divide the pro- duct by the constant number 24,000 multiplied by the constant number 64, the quotient is the horse power. This rule, however, is obtained by neglecting very many of the circumstances which ought to be taken into account, and must therefore be very wide of the truth. It is also difficult to determine the mean relative velocity, for this varies at every possible inclination of the paddle. A very beautiful investiga- tion is given in one of the appendices to the new edition of " Tredgold on the Steam Engine," by Mr. Morway. In this, all the circumstances seem to betaken into account, and the results of the theories correspond very closely with observation in British steamers. We fear that this elegant investigation gives formulae altogether too complex for the use of practical men. We shall, in consequence, prefer to deduce rules from experience. The other rules, which have been deduced from theory, are as follows : In the same vessel, and with a constant relation between the area of the paddle-wheels and the transverse section of the vessel, the velocities are as the cube roots of the powers of the engines.* The relation between the velocity of the wheels and of the vessel is constant so long as the ratio between their surfaces remains the same. It is obvious that this last rule cannot be correct, if it be true, as we have inferred, and shall hereafter show by a comparison of various observations, that the relative velocity of the circum- ference of paddle-wheels is a constant quantity. 196. If we apply the rule for the area of paddle-wheels to the case of a. paddle-wheel working at a maximum, or with a rela- tive velocity for its vertical paddle of 6.5 feet per second, we shall find that each horsepower of the engine should be capable of impelling a paddle of half a square foot. But a paddle does 240 NAVIGATION BY STEAM. not act during the time of its immersion in the water with equal intensity ; and although no loss of power might arise from this cause, the obliquity of its action applies a part of the force to resistances that do not assist in propelling the vessel. Thus, on entering the water, a part of the force is applied to lift the wheel from its axis, and on quitting the water to press it down. In addition, a quantity of water is raised upon the paddle, apart of whose weight acts in direct opposition to the moving power. The loss growing out of these causes can only be investigated experimentally. We shall attempt this from a comparison of the circumstances of the steam-boats North-America and Presi- dent. The former navigates the Hudson River, and is remark- able for a speed that has hitherto never been equalled by any other steam vessel ; the latter plying between New-York and Providence, it has been found, in her construction, necessary to preserve stability as well as to obtain speed, and if her velocity be less than that of the former, she still combines the two quali- ties of speed and safety in a degree superior to any vessel we are acquainted with. The particulars necessary for our purpose in relation to the President, are as follows, viz : Breadth of beam, 32 ft. 6 in. Draught of water, 9 ft. Diameter of Water-wheels, . . 22 ft. Length of bucket, 10 ft. Depth of do 3 ft. 6 in. She has two engines of the following dimensions : Diameter of Cylinder, .... 4 ft. Length of Stroke, 7 ft. Number of double strokes or complete revolutions of paddle- wheel per minute, 21 . When but one engine works, the number of revolutions of the single wheel on which it acts are reduced to 17i. The average passages to Providence have been performed, when both engines acted, in 15£ hours ; when but one was used, in 19£. The distance between New- York and Providence is usually estimated at 210 miles ; carefully measured, however, upon a NAVIGATION BY STEAM. 241 map, it is found to amount to no more than 160 nautical, or 184.3 English miles. From these data, the average velocity of the boat through the water is very nearly 12 miles per hour, or 17.6 feet per second ; the average relative velocity of the wheel 6.5 feet per second, when both wheels and engines were in motion ; the average velocity of the boat when but one en- gine worked, becomes 9.45 miles per hour, or 13.86 feet per second ; the relative velocity of the wheel 6.3 feet per second. The wheels, therefore, move with a relative velocity almost identical with that which our hypothesis has assumed to cor^ respond to a maximum effect. But the actual effect is far be- neath the rule we have laid down. Estimated from a compa- rison with other condensing engines, those of the President would have each a nominal power of about 110 horses, which, in consequence of the rapidity of their action, is increased to about double ; but by the rules on page 121, the power of each of the engines is that of 160 horses. As each paddle has a surface of no more than 35 square feet, each horse power drives no more than 0.22 feet of paddle, or less than one-fourth of a square foot. And it would appear, from comparing the relative velocities in these two cases, as if, in this vessel, the proper ratio between the moving power and the paddle had been at- tained. The North-America has the following dimensions : J3readth of beam, 30 ft. Draught of water, ----- 5 ft. Diameter of Water-wheel, 21 ft. Length of bucket, 13 ft. Depth of do. ----- 2 ft. 6 in. She has two engines of the following dimensions : Diameter of Cylinder, - - - - 44^- in. Length of Stroke, ----- 8 ft. Double Strokes per minute, 24. The estimate of her speed furnished by her owners, is 19.8 feet per second. The relative velocity of the wheel is 6.6 feet per second, exceeding our theoretic limit one-tenth of a foot. The relation between the velocities of the boat and the wheel is as 3 to 4. 31 242 NAVIGATION BY STEAM. The power of each of the engines, estimated by the rule on page 121 j is 186 horses, the area of each paddle 32^ ; and hence each horse power propels no more than 16 hundredth parts of a square foot through the water. The velocity of the wheel is, however, greater than that of the President in the ratio of 6.3 : 6.6, or of 21 to 22 ; which makes the comparison more favourable than would at first ap- pear, to the North-America. For, conceiving the wheel of the former to work to the greatest possible advantage, each horse power would, at the increased relative velocity the latter has, propel no more than one-fifth of a square foot of paddle. The powers of the engines of both boats, as estimated by us, far exceed what would usually be ascribed to them from a mere consideration of their dimensions. Those of the President being of the size of condensing engines usually estimated at 110 horse powers ; those of the North-America, at 98 horse powers. This difference arises from the great speed with which they are driven ; it being usual to give no more velocity to the piston of a condensing engine than about 200 feet per minute, while that of the North -America has 384 feet, and that of the President 336 feet. The near coincidence of the actual performance of these boats with our theory, except in one respect, is a tolerable warrant for its accuracy. The point in which this difference occurs, is the area of paddle that can be driven by each horse power of engine. The rule on page 239 would make this force equivalent to move half a square foot with the velocity of 6.5 feet per second ; while in the case of the President, the actual performance, reduced to that velocity, is no more than one-fifth of a square foot ; while in the case of the North-America it falls as low as one-sixth. The disturbing causes that effect this variation from the theory are obvious, and have been explained, but it is not easy to reduce them to calculation. In the new class of vessels which has come into use in the neighborhood of New- York similar results have been obtain- ed, as will appear from the following facts : NAVIGATION BY STEAM. 243 Steam-boat Cleopatra. Diameter of wheel, - - - - 23 ft. Length of bucket, - - - - ll£ ft. Breadth of do. 2fft. Revolutions per minute, 24 Velocity of wheel per second, - - 28.8 ft. of vessel, ----- 22.6 Relative velocity of wheel, - 6.2 Steam-boat Lexington. Diameter of wheel, - - - - 24 ft. Length of bucket, - - - - lift. Breadth of do. - - - - - 2f ft. Revolutions per minute, 23 Velocity of wheel per second, - - 28.8 ft. of vessel, - 22.5 ft. Relative velocity of wheel, 6.3 ft. Steam-boat Massachusetts. Diameter of wheel, - - - - 22 ft. Length of bucket, - - - - 10 ft. Breadth of do, - ' - - - 2^- ft. Revolutions per minute, 26 Velocity of vessel per second, - - 19.95 ft. of wheel, - 26.25 ft. Relative velocity of wheel, - 6.3 ft. The same general fact, that the velocity of a paddle-wheel through the water is a constant quantity in a wheel of given diameter and dip, and does not vary in different wheels materi- ally from 6.3 ft. per second, has been observed in the high pres- sure steam-boats which navigate the Mississippi and its branch- es by Professor Locke. 197. We have seen that, according to the usual theory, the resistance sustained by a body moving in a fluid is proportion- ed to the square of its velocity and the area of its section. The moving force, necessary to give a vessel a given velo- 244 NAVIGATION BY STEAM. city, should therefore, as has also been stated, be equal to this resistance multiplied by the velocity, or proportioned to the cube of the velocity; and in similar vessels the resistance is proportioned to the square of one of the homologous dimen- sions. Thus, to obtain double the velocity in a given vessel, and with given wheels, it might appear eight times the force should be employed, and so on. But. as the space passed over in a given time is proportioned to the velocity, the actual expenditure of power, in performing a given distance, is proportioned to the squares of the veloci- ties. These laws would be true only when the weight of the en- gine is considered as # constant. but as this increases in a greater ratio than the power, it would make the acquisition of great ve- locities still less advantageous. If, however, we hare recourse to facts instead of theory, we find that the resistance never increases in the cases which occur in practice, in a ratio as great as the square of the velocity ; and it appears probable, that at the higher velocities it becomes al- most constant. Supposing the resistance to vary with the ve- locity simply, we obtain the following rules, which are more con- sistent with experience : (1). To obtain double the velocity in a given vessel, ice must ernploy an engine of A times the power. 2 . To obtain an equal velocity in similar vessels of differ- ent dimensions : we must employ engines varying in force with the areas of their transverse sections, or with the squares of their homologous lineal dimensions : or. to express the same fact in another manner, with the squares of the cube roots of their respective tonnages. (3). The actual expenditure of power in passing through a distance with different velocities is as the velocities. An obvious advantage will be gained by increasing the size of the vessels. lor the resistances varv as the square of similar dimensions, while the tonnage increases with their cubes. It has been imagined by some, that the motion of steam-boats was different in a current from what it is in still water. This, NAVIGATION BY STEAM. 245 however, cannot be the case, unless it be so rapid that the slope becomes an important element, forming an inclined plane up which the weight is to be lifted. This conclusion is obvious from the following considerations. When a vessel is in a cur- rent, and the propelling force ceases to act, she speedily acquires the velocity of the fluid, and is relatively at rest in respect to it. If the propelling force be steam applied to wheels, and they be set in motion, the action upon the fluid is precisely the same as if no current existed, and hence the velocity through the water will be the same as when the fluid is at rest. Thus, then, the ve- locity, in respect to the shore, will be the sum or difference of the velocity the vessel would have in still water, and that of the stream. The case is, of course, widely different when the force is applied by chains, or other connexion, with a fixed point up- on the shore. 198. The next consideration in respect to paddle-wheels is their diameter. This is determined by means of the velocity that it is intended to give the circumference, compared with the velocity of the piston ; the wheel being attached to the crank will make half a revolution for each stroke of the piston. Hence, in this mode of gearing the wheels, great velocities can only be at- tained by a proportionate increase of the diameters. This is at- tended wiih several practical inconveniences, first in a great in- crease of the weight, and secondly in raising the height of the centre of gravity. The action of a paddle depends, as has been shown, upon its relative velocity, but when the paddles of a wheel act in succession, the water will follow in the wake of the first which acts ; and the second may, if it succeed at too short an interval, impinge upon water that is already in motion. For this cause, the paddies upon the wheels should not be more numerous than is just sufficient to keep up a continuous action. The proper arrangement, for this reason, is such, that when one paddle is vertical, the preceding one shall be just issuing from, and the succeeding one just entering the water. When paddle-wheels impinge against the water in an oblique direction, they sustain a sudden shock, when the interval is as great as we have pointed out ; the reaction of this sudden re- 24G NAVIGATION BY STEAM. sistance upon the engine is injurious, and it checks and destroys the accumulation of power which the water-wheel might other- wise attain, and distribute, upon the principle of the fly-wheel. Hence in the early steamboats, fly-wheels driven with greater velccity than the paddle-wheels, were found of great value ; and even in the rapidly moving boats of the present day, where the velocity and weight of the paddle-wheels enable them to answer the purpose of a fly-wheel, these shocks are not without an injurious effect. Various methods have been proposed to remedy this defect. In some English steam-boats, the paddles have been placed obliquely upon the circumference of the wheel, but still perpen- dicular to a plane tangent to it ; their inclination to the vertical plane, therefore, remains the same as in the usual form, but they enter the water by an angle, instead of striking with one side, and hence do not experience the shock of which we have spoken. There is, however, a defect which more than compensates any advantage to be derived from this arrangement. The wheels act in directions inclined to the plane of the vessel's keel, and thus a part of their power is exerted to press the vessel in a lateral direction ; and although the two wheels mutually neu- tralize this part of each other's action, the whole of the force exerted in this direction is wasted. A better arrangement has been introduced into his steam- boats by Mr. R. L. Stevens. The wheel is triple, and may be described, by supposing a common paddle-wheel to be sawn into three parts, in planes perpendicular to its axis. Each of the two additional wheels, that are thus formed, is then moved back, until their paddles divide the interval of the paddles on the original wheel into three equal parts. In this form, the shock of each paddle is diminished to one- third of what it is in the usual shape of the wheel ; they are separated by less intervals of time, and hence approach more nearly to a constant resistance; while each paddle, following in the wake of those belonging to its own system, strikes upon water that has been but little disturbed. Believing this oblique action to be a great defect, many at- NAVIGATION BY STEAM. 247 I tempts have been made to construct wheels in such manner that their paddles might dip and rise from the water in a vertical position. The first attempt of this sort was made twenty years since by a civil engineer of the name of Busby, who applied his wheel to one of the Jersey City ferry-boats. When in action it was found inferior in power of propulsion to the common pad- dle-wheel, and is, of course, still more so to that of Stevens. Within a few years several attempts of the same kind have been made in England ; but after impartial investigation by Barlow, it seems to be conclusively proved, that, except when there may be a great variation in the dip of the paddles, they are inferior to the common wheel. Still more recently, a paddle to which the name of cycloidal is given, has been introduced in England. The form of this may be conceived by imagining the paddles of a common wheel to be sawn each into three parts by cuts parallel to the axis of the wheel, and that two of the parts are each moved backwards one third of the distance between two contiguous paddles. This me- thod, however is not original, for it was tried some years since on the Hudson, and after a fair trial, abandoned. It is unquestionably inferior to the wheel of Stevens in river navigation, but possesses advantages similar to those with vertical paddles in navigations when the dip of the paddle may be subject to variation. This variation often occurs in the navigation of the ocean, as the stock of the fuel must be great at setting out, and will be ex- hausted before the passage is completed. It cannot, however, be questioned that, should it be found practicable to reef the paddles of Stevens's wheel in such manner that their dip may be con- stant while the draught of water of the vessel changes, this method would be superior for the navigation of the ocean to any other. The objection usually made to the common paddle-wheel, and, in consequence, to Stevens's, is, that there is a loss of power arising from the oblique action. This loss, it has been shown by Barlow, is more than compensated by the increase in the relative velocity of the paddle when in its oblique position. It thus happens, that with a given expenditure of steam the wheel with vertical paddles revolves more rapidly than the common 248 NAVIGATION BY STEAM. » wheel, but at the same time gives a less progressive motion to the vessel. Another loss in the application of the power, arises, as we have seen, from the water that is lifted by the paddles as they pass out of the water. But the loss is not equal to the whole weight lifted, for this water will already have acquired a veloci- ty of rotation that will diminish the pressure on the paddle, and as the paddle is oblique, the actual pressure may be resolved into two forces, one of which retards the motion of the wheel, and is lost, while the other acts horizontally to propel the vessel. This loss will, of course, be least in large wheels, if the immer- sion of the bucket be constant. The larger the wheel, the less will be the weight of the water lifted. The paddle is not immersed wholly in the water, except when nearly in its vertical position. Hence it does not exert a con- stant force to propel the vessel, but as the expenditure of power from the engine will follow the law of the area, no loss arises from this cause. But the inclination of the paddle is, besides, constantly varying ; and while the water opposes a resistance perpendicular to the surface of the paddle, depending upon the area immersed and the square of the velocity ; only that part of this resistance, which, when decomposed, is parallel to the surface of the water, acts to propel the vessel, the rest is wasted upon the vessel, whose weight is alternately lifted and forced downwards. A steam vessel is set in motion with a velocity that gradually increases until it becomes uniform. At this time the resistance of the water to the motion of the wheels exactly balances the progressive motion of the vessel. Hence, if we knew the rela- tion between the laws by which the resistance to plane surfa- ces, and to those of the figure of a vessel, are governed, we might determine the proportion which ought to exist between the area of the paddle and that of the midship frame of the vessel. Experiments by the Society of Arts in London appear to show, that when a solid of small size is fashioned into the figure of a vessel, the resistance was not more than ^th of that which opposes a plane surface. Other observations make the resistance to good models vary from ^th to -p^th. But observation on a large scale gives far more favourable re- NAVIGATION BY STEAM. 249 suits. In the case we have above quoted of the steam-boat Pre- sident, the resistance to the tranverse section of the vessel is no more than^th part of that incurred by the wheels when both en- gines act ; while, when but one acts, it falls as low as-^-th. In the North- America, it appears to be no more than - 2 -\d. This in- ference has recently been confirmed by a series of experiments recorded in the Transactions of the Royal Society by P. W. Barlow. In eleven boats the resistance varied from -^ -th to ^th, and was at a mean -fa. In the great improvements which ex- perience has suggested in the figure of our more modern boats, and in the false prows which have been adapted to our older ves- sels, the resistance has been diminished still further ; and we need no longer hesitate to allow a higher limit than even • 2 - L 7 -. The im- provements have with sound judgment been directed to the ob- ject of preventing the formation of the wave, and thus to get rid of the most important retarding cause altogether : this has in some instances been almost completely attained. In the steam frigate Fulton, the ratio of these resistances has been reduced below ^V. The table, therefore, which was given in our first edition, may be considered as obsolete. No possible danger, then, can arise, in assuming" that in a vessel of a o-ood model, the resistance to the progressive motion falls as low as^th part of that which acts upon the paddles. If, then, we assume for the relation between the absolute velocities of the boat, and the wheel, when work- ing to the greatest advantage, the proportion already stated of 3:4; the most advantageous size of the paddles will be such that the area of each should be one-sixth of that of the midship frame of the vessel, or the sum of those which act at a time, on both wheels, one-third of that quantity. If the engine be so constructed as to give the paddle-wheel a rotary velocity of 26 feet per second, the boat will acquire a velocity of 13.2 miles per hour. Besides the paddle-wheel, various other plans for propelling vessels by steam have been suggested. Some of these will be referred to in the history of Steam Navigation. In addition to these, it has been proposed to drag vessels by means of a chain resting on the bottom of a canal. This method is to be prefer- 32 250 NAVIGATION BY STEAM. red in this case, inasmuch as the wave which is raised by the wheels is destructive to the banks of the canal. This proposal has recently been carried into effect with success on a Ferry in England. In long canals, the expense of a continuous chain is objection- able ; but it has recently been discovered that the friction of a chain on the bottom of a canal is sufficient to propel a vessel. A very ingenious arrangement for this purpose, in which an endless chain, of which a part lies on the bed of the canal, and is set in motion by a steam engine in the vessel, has been con- trived by Mr. Leavenworth of New- York. 199. Our practical rules may now be summed up and reca- pitulated, as follows, viz : 1. The relative velocity of the circumference of the wheel appears to be, in all cases, about six and a half feet per second. 2. Each horse power of engine, calculated according to the rules on page 121, will drive a paddle of the area of one-fifth part of a square foot with this velocity. 3. The maximum absolute velocity of a paddle, to give the greatest velocity a boat has usually attained, is 26 feet per second ; but there are recent instances where as much as 29| feet per se- cond has been reached ; and if the wave can be avoided, it would be unsafe, without a better theory of the motion of bodies in flu- ids than has yet been investigated, to name a limit which new improvements may not exceed. 4. In a vessel of good model, these velocities may certainly be attained, when the relation between the area of the midship frame of the vessel, and that of the paddles of both wheels, is as three to one ; and they have been attained when the relation has been as small as 10 : 1. 200. We cannot quit this subject without suggesting some views, which it is hoped may still farther improve steam navi- gation. It has been seen that if the power of the engine be cal- culated according to the usual rule, each horse power ought to propel a surface of paddle equal to half a square foot ; while in the two instances which we have cited the performances have NAVIGATION BY STEAM. 251 been no more than .16 and .22 ft. In the one case, therefore, at least two thirds, and in the other more than one half the force estimated appears to be useless. It is obvious that this arises from the great velocity with which the piston of the engine is driven, in order to acquire the great velocities now usual in na- vigation. The whole tension of the steam acts as a pressure on the piston, only when that is at rest, and with every increase of velocity the pressure must be diminished. Experience de- rived from the action of engines employed in manufactures and pumping, would seem to show, that the maximum perform- ance of a condensing engine takes place, when the velocity of the piston is from 250 to 280 ft. per minute. It is now usual to drive those of steam-boats at rates between 400 and 600 ft. per minute. If the former numbers express one-third of the velocity which the piston would assume had it no work to perform, the defect in the performance of the engines of the North-America and President is almost exactly consistent with theory. On the other hand, it is absolutely necessary that the rotary velocity of the circumference of the paddle-wheel should not be less than 26 feet per second, if the vessel is to move with the velocity which is now demanded. It would therefore appear, that a great saving of steam must accrue from such a disposition of the engine, as would allow its piston to move with no greater velo- city than 280 feet per minute, and should, notwithstanding, give one of 26 feet per second to the wheel. Two modes suggest themselves at once for accomplishing this object, namely, to in- crease the number of revolutions of the wheel by gearing, or to substitute for the crank the original contrivance of Watt, the Sun and Planet wheel. There is, however, an objection to the use of toothed wheels in steam-boats which is not unfounded, and if this be as positive as is usually thought, these methods are not likely to come into practical use. There remains, how- ever, another, which we have not seen suggested, namely, to make the arms of the lever beam of unequal lengths. In this way an engine of comparatively short stroke, and whose piston would in consequence move with a proportionally less velocity, would give the necessary speed to the paddle-wheel, while the 252 NAVIGATION BY STEAM. crank would still be applied at a favourable distance from the axis. The increase which has been given to the speed of the pad- dle-wheel is partly gained by an increase in the velocity of the piston, and partly by enlarging the diameter of the wheel itself. Adopting the method just suggested, the number of strokes of the piston might be increased, and the diameter of the wheel diminished. A similar advantage would be gained by the use of the Sun and Planet wheel. In steam vessels intended for the navigation of the ocean, the British nation has as yet been more successful than ourselves. They have, however, failed in giving them any thing like the velocity which we are in the habit of using. It w r ould seem to be easy to give the American rate of motion to steam vessels, without impairing the good and sea-worthy qualities of the British steamers. In this event the mast and sails which they yet rind of use, would be no more than a useless incumbrance, as our vessels outstrip even brisk gales of wind, and would, in consequence, find it always acting in opposition to them. More particularly is the bowsprit objectionable. The motion of a steam vessel, in her pitching is not, like that of an ordinary vessel, derived from the motion of the waves alone, but is, in addition, influenced by the action of the engine. It therefore will be of- ten struck by the waves in a high sea, and it is not wonderful that several accidents have already happened to this spar, in the few voyages that have been made between New- York and Great Britain. Not only is the bowsprit an incumbrance, and exposed to danger, but it is wholly unnecessary, even if it be admitted that steam vessels must be equipped with sails. The length of steam vessels is so great, that by lessening the after-sail, all the functions of the jib will be fulfilled by stay-sails spread between the cutwater and the foremast. If the speed of steam vessels on the ocean be increased to the American rate, sails may still be necessary in order to provide for accidents to the engine, or to spare the expenditure of fuel. It will not, however, be necessary to make any other provision for spreading them, than to step the lower masts, and reeve their NAVIGATION BY STEAM. 253 standing rigging; the topmasts and top-gallant masts, with the yards, should be stowed away, and only sent aloft when the ne- cessity for spreading sail arises. To one acquainted with the manner in which steam is used in our river steam-boats, and in which it may doubtless be applied on the ocean, the heavy masts, yards, and sails, with which the British steamers are incumber- ed, are not less offensive, than are the structures which we pile on the decks of our vessels, to the nautical eye. In the steam vessels for ocean navigation, the timbers ouo-ht to be carried up to the level of the upper deck, the whole plank- ed in, and strengthened by ceiling plank. Instead of the par- tial coverings of cabins and wheel-houses, the whole should be closed in by a spar deck extending from stem to stern, and, if possible, flush. The great relative length which must be given to steam ves- sels, renders them more liable to the usual tendency to the change of figure called hogging than other ships. They, in con- sequence, require to be proportionably stronger. This increase of strength is usually sought, by increasing the number and the scantling of the timbers. We conceive that this is wrono- in principle, inasmuch as the weight of the vessel is the cause of the change of figure, and to increase it in adding to the strength of the material, may in the end render the evil which is to be con- quered greater. We should, in consequence, prefer to dimin- ish the scantling, and even lessen the number of timbers, taking care that each shall constitute a frame ; and would propose to meet the tendency to change of figure by laying the ceilino- plank diagonally, and by a system of lattice-work plankino- reaching from the keelson to the main-deck. In addition, a series of iron stays, extending from the stem to the stern-post over the lower mast-heads, might be employed. The several frames ought to be united by bolts or rods of iron extending through three continguous timbers, and in such number, that every timber should be connected on each side with the second in distance from it. The engines employed should be of the character to which the name of portable is given ; that is to say, the whole of the parts which compose each of them should be united in a frame 254 NAVIGATION BY STElil. of iron. In this way the engine will act upon itself, and not upon the vessel. A horizontal engine is also to be preferred when it can be used, as a vessel is much less racked by its mo- tion than by one whose cylinder is vertical. The boilers should be of such a form as carry the least weight of water in proportion to their fire surface which is consistent with safety. For this reason, those with tubular flues are to be preferred. The only fuel which can be employed is coal, for the weight and hulk of wood would be an insuperable objection to its use on long voyages. Of the different kinds of coal, the bi- tuminous, under equal weights, gives the greatest quantity of heat, but generates so much smoke as to render the vessel un- comfortable for passengers. It is therefore probable that the mode of burning anthracite coal in boilers, with tubular flues, like those used in locomotive engines, and in which the igni- tion is promoted by a blowing engine, will be preferred. None of the vessels, which have yet been constructed for navigating the ocean, appear to us to be worthy of being cited as models to be copied. The English steamers are objectionable from the excessive and useless weight of their engines, and the great space they occupy ; and from the great size, and the weak- ness of their boilers. The steam frigate Fulton can hardlv be considered as intended for ocean navigation. The Natchez, which runs as a packet between New- York and New Orleans, appears to unite a greater number of advantages than anv other vessel, but has not sufficient tonnage to unable her to carry fuel for crossing the ocean. I. The steam engine, such as has been described in Chapter V., requires several modifications to suit it for the pur- pose of propelling boats. When placed in the middle of the vessel, that form represented in PI. TIL. in which the great working-beam is suppressed, and two connecting-rods adapted to the piston by a cross-head, is often used. But when two en- gines are employed, the beam must be retained. The cold- water cistern would load the vessel with an enormous weight, and hence the condenser is not immersed in water : the hot- water cistern is. generally speaking, set upon the top of the air- NAVIGATION BY STEAM. 255 pump ; and the delivering- door is a conical valve surrounding the air-pump rod. Water for condensation is supplied by a standing-pipe, passing through the bottom of the boat and rising above the level of the external mass of fluid ; the injection-cock is below this level, and the water is forced into the condenser, by virtue of the difference of level. The waste hot water passes out by a similar pipe. These pipes are called standing pipes, and they are represented on PL VII., at h h and l\ h h being the pipe adapted to the condenser, which it supplies through the injection-cock z, and I being the pipe through which the waste hot water is discharged from the cistern on the top of the air-pump H. A hand force-pump K, is employed to fill the boiler at first, and it is afterwards supplied by a force-pump I. The hand pump may also be employed to keep up the water in the boiler, when the engine is not in action. The condenser is increased to half, and the air-pump to one- third of the capacity of the cylinder. These standing pipes are exposed to danger, and the openings through which they pass are of such size that the slightest acci- dent, or even overloading, may be followed by the sinking of the vessel. It has, in consequence, been proposed to adapt valves to these openings, which might be shut in case of an accident hap- pening to the steam pipe, or of the vessel being loaded beyond her proper depth. Valves for this purpose have been invented in England by Kingston, and in this country by Mr. Haswell, the engineer of the U. S. Steam frigate Fulton. Another very important modification consists in the size of the valves and steam-pipes. It has already been seen, that in some cases, when steam acts expansively, the area of the nozzle should be increased ; but in steam-boats the great velocity re- quired for the wheels being usually gained, not by gearing, but by increasing the velocity of the piston, this can only be at- tained by affording a passage for an increased flow of steam. This method of increasing the speed has this advantage, that velocity is gained without increasing the weight of the engine, by merely adding to the fire surface of the boiler. In the steam-boats on the Hudson, not only has the velocity of the piston been increased by increasing the number of strokes, 256 NAVIGATION BY STEAM. but by adding, at the same time, to the length of the cylinder. And, although it is obvious that in this way the pressure of steam of a given tension in the boiler, upon the piston, must be lessen- ed, an equal area of paddle-wheel is driven. This we ascribe, in opposition to a high authority, to the fact that the crank of the engine acts in a more favourable point in the wheel. It would appear to us, that the true position of the extremity of the crank would be in the circle described by the centre of resistance of the paddle, and that it is only when applied to this circle, that all the force of the steam is applied to propel the vessel. Now, as it would be impossible to give the area of the crank so great a length as this, the nearer it approaches to it, the better. Hence, the method used in the American steam-boats has not only been successful in practice, but is founded upon true mechanical principles. The great length which is given to the cylinders of Ameri- can engines, and the supposed necessity of placing their bed plates so high that the axle of the wheels may lie below them, is attended with the disadvantage of impairing the stability of the vessel. Now, although a vessel ought not to be too stiff, because in that case the motion of rolling is violent, it seems probable that in our vessels there is not that degree of stability which is necessary for perfect safety. We have, therefore, to mention with approbation a very ingenious form of engine planned by Mr. Lighthall. In this, the cylinder lies in a hori- zontal position near the keelson of the vessel, and has the long stroke of the American boat engines. The motion of the piston is communicated to the wheels by straps working in guides, a lever beam, connecting-rod, and crank. The form is therefore similar to that of the engine on PI. III. provided it were laid upon its side. The manner in which Mr. Lighthall has pro- vided for the working of the pumps and valves of his engine is simple and sufficient. By means of this form of engine, all the advantages to be derived from length of stroke are secured, without any of the defects of the usual methods. The general form of the engine on PI. III. is now more used in steam-boats in the U. S. than that on PL VII. The stroke of the piston, and the length of the connecting-rod and crank, being NAVIGATION BY STEAM. 257 greater in proportion to the height of the lever beam than in the first of these engines. The lever beam is not, as in PI. III., a solid mass of cast iron, but is an open frame of that material surrounded by a strap of vvrought-iron. 202. The application of steam to the propulsion of vessels appears to have been among the very first ideas that suggested themselves to the inventors or improvers of the engine. Wor- cester, in the quotation that we have made from the " Century of Inventions," speaks of the capacity of his invention for row- ing. Savary proposed to make the water raised by his engine turn a water-wheel within a vessel, which should carry paddle- wheels acting on the outside ; and Watt, as we are well assur- ed by a personal auditor, stated in conversation, that, had he not been prevented by the pressure of other business, he would have attempted the invention of the steam-boat. Newcomen alone gave, as far as we can learn, no intimation of any such design ; and this we are rather to take as an evidence of his correct ap- preciation of the powers of his engine, than as arising from any want of ingenuity. In truth, before the time of Watt, no modi- fication under which steam was applied to useful purposes would have been able to propel vessels successfully. Even with all his improvements, the fuel is a great load, and its car- riage no small difficulty ; but, before he lessened its consumption so materially, it would have been hardly possible for a vessel to carry enough of combustible matter, except for very short voyages. Previous in date to all these persons, recent discoveries have brought to light an ancient record in which we have the de- scription of a vessel propelled by steam in a manner that ob- tained the suffrages of the witnesses. Blasco de Garay, an officer in the service of the Emperor Charles V., made, at Barcelona, in the year 1543, an experi- ment on a vessel, which he forced through the water by appar- atus, of which a large kettle, filled with boiling water, was a conspicuous part. If this be true, and we have no reason to doubt the authenticity of the records, De Garay was not only the first projector of the steam-boat, but among the first who 33 258 HISTORY OF conceived the idea of applying a steam engine to useful purposes. He was. however, too far in advance of the spirit of his age to be able to introduce his invention into practice, and even the recollection of his experiment had been lost, until the record was accidentally detected among the ancient archives of the province of Catalonia. This experiment was, therefore, with- out any direct practical result ; neither did it produce any effect in facilitating the researches of subsequent inquirers, and may therefore be considered rather as a matter of curious antiqua- rian research, than as deservedly filling any space in the history of the steam-boat. English authors have also raked up from oblivion a patent granted in the year 1736, to a person of the name of Jonathan Hulls. He, however, never made even an acting model of his invention, and the prime-moveT itself was at the time in a state far too imperfect to have permitted its being successfully used in the manner proposed by Hulls. So far, then, from classing this among ingenious and profitable improvements, we should ra- ther be inclined to rank it among those which, from their ob- vious impracticability, merit the oblivion into which they in- stantly fall. The paddle-wheel, it has been stated, is the only apparatus that, when worked by steam, has been found completely success- ful in propelling vessels. The use of this for such a purpose, but set in motion by other prime-movers, is of remote antiquity, and was from time to time again brought forward, used for a season, and again abandoned. Among these attempts may be mentioned a boat constructed on the Thames by Prince Rupert, whose action was witnessed by Papin, by Savary, and probably by Worcester. So far as re- gards the antiquity of the method, Stuart quotes manuscripts from the library of the King of France, from which he states it was ascertained, that during one of the Punic wars, a Ro- man army was transported to Sicily upon vessels moved by wheels worked by oxen. The use of a water-wheel, in a man- ner the reverse of that in which it was employed to propel machinery, is almost too obvious to be entitled to the character of invention ; it was therefore only necessary that the necessity STEAM NAVIGATION. 259 for their use should exist, and their introduction would have followed as a matter of course. It was, however, long questionable whether they could be used to advantage when attached to a steam engine, and in the earlier experiments, the blame appeared to fall upon them, ra- ther than upon the imperfections of the engine, or the unskil- ful and unartist-like manner in which they, and the rest of the apparatus, were adapted to the vessels. We have stated that Watt's engine was the first possessed of sufficient powers to be used to advantage in vessels. This is not merely an inference from what can be observed in the prac- tice of the present age, but was, in 1753, made a matter of ma- thematical proof by Bornouilli, in a memoir which gained a prize offered by the French Academy of Sciences. He, how- ever, expresses his opinion too broadly, applying his inference rather to the power of steam itself, than the mode in which it was then commonly applied. Still there were some who, not aware of the defects of the prime-mover, continued to seek for the means of applying it to vessels. Among these may be named Genevois and the Comte d'Auxiron. The former, whose attempt dates as early as 1759, is chiefly remarkable for the peculiarity of his appar- atus, which resembled in principle the feet of aquatic birds, opening when moving through the water in one direction, and closing on its return. The latter made an experiment in 1774, but his boat moved so slowly and irregularly, that the parties at whose expense the trial was made, at once abandoned all hopes of success. In 1775 the elder Perrier, afterwards so celebrated as the in- troducer of the manufacture of steam engines into France, made a similar attempt, which was equally unsuccessful. But, not discouraged, and ascribing his failure to the use of paddle- wheels, he applied himself for some years afterwards to the search for other substitutes for oars. It does not appear, how- ever, that he made any valuable discovery. The Marquis de Jouffroy continued the pursuit of the same object. His first attempts were made in 1778, at Baume les Dames, and in 1781 he built upon the Saone a steam-vessel 260 WATT. 150 feet in length and 15 in breadth. In 1783 his experiment became the subject of a report made to the French Academy of ScienceSj by Borda and Perrier. The report is said to have been favourable. We have seen that the double-acting engine of Watt was not made public before 1781, and that it was not until 1784 that it received those improvements by which it was fitted to keep up a continuous and regular rotary motion. No pre- vious engine having the necessary properties, we feel warrant- ed in rejecting all attempts prior to the former date as prema- ture, in attempting to perform that to which the means in the possession of the projectors were inadequate. We are to look to our own country, not only for the first successful steam-boat, but for the very earliest researches into the subject, after the improvement of the engine by Watt had rendered success attainable. The very nature and circum- stances of the United States appeared to call for means of con- veyance different from those which are employed in other countries. Our whole coast is lined by bays and rivers, by the aid of which a safe parallel navigation, might, at small expense, be extended from one extremity of the Union to the other ; but which, land-locked, and protected from the winds, is at some seasons tedious to the ordinary methods. Still more recently, the Mississippi and its innumerable branches have become the seat of flourishing settlements, separated from the Atlantic coast by ridges of barren mountains, and almost inaccessible from the Gulf of Mexico by either sails or oars, in consequence of the rapidity of the stream. Our population, with the wants and curiosity of the highest civilization, is still scattered over so vast a region, as to demand rapid means of communication and great foreign importations. These wants could not have been satisfied, nor this active curiosity gratified, by any means yet discovered, except the steam-boat. The earlier projectors appear, however, rather to have reference to the prospective state of our country than to circumstances which existed at the moment of their attempts. Hence we shall find that they sought in foreign countries the encouragement, the wealth of their native land was inadequate to afford. RUMSEY — FITCH. 261. Rumsey and Fitch were cotemporaneous in their researches. Both attempted to construct steam-boats as early as the year 1783, and modes of both their contrivances were exhibited in 1784 to General Washington. Rumsey's was the first in date of exhibition, but Fitch was first enabled to try his plan upon a scale of sufficient magnitude ; for, in 1785, he succeeded in moving a boat upon the Delaware, while Rumsey had not a boat in motion upon the Potomac before 1786. Fitch's apparatus was a system of paddles ; Rumsey at first used a pump, which drew in water at the bow and forced it out at the stern of his boat. The latter afterwards employed poles, set in motion by cranks on the axis of the fly-whsel of his engine, which were intended to be pressed against the bottom of the river. About the date of these experiments Fitch sent drawings of his apparatus to Watt and Bolton, for the purpose of obtaining an English patent ; and in 1789 Rumsey visited England upon the same errand. The former was not success- ful in obtaining patronage; but the latter, by the aid of some enterprising individuals, procured the means to build a vessel on the Thames, which, however, was not set in motion until after his death, in 1793. Fitch's boat was propelled through the water at the rate of four miles per hour. We may now reasonably doubt whether pad- dles would have answered the purpose upon a large scale, for more than one experiment on this principle has since been tried, and without success. The method of Rumsey is more obvious- ly defective, and we need not wonder that it was followed by no valuable results. Next in order ot time to Fitch and Rumsey, we find Miller, of Dalswinton in Scotland. This ingenious gentleman had, as early as 1787, turned his attention to substitutes for the common oar, and had planned a triple vessel propelled by wheels. Find- ing that wheels could not be made to revolve wjth sufficient rapidity by men working upon a crank, the idea of applying a steam engine was suggested by one of his friends, and an engi- neer of the name of Symington employed by him to put the idea into practice. The vessel was double, being an experi- mental pleasure-boat on the lake in his grounds at Dalswinton. 262 MILLER. The trial was so satisfactory, that Miller was induced to build a vessel sixty feet in length. This was also double, and it is asserted that it was moved by its engines along the Forth and Clyde canal at the rate of seven miles per hour. The boat, the wheels, and the engine, were, however, so badly proportioned to each other, that the paddles were continually breaking, and the vessel suffered so much by the strain of the machinery as to be in danger of sinking, and Miller found it unsafe to venture into any navigation of greater depth than the canal. The appara- tus was therefore removed and laid up, and here the experi- ments of Miller ceased. He himself appears evidently to have considered this experiment an absolute failure, and ascribed the blame to the engineer. We have to remark that the double boat used by Miller, was a form ill suited to the purpose ; in the ferry boats of that structure, introduced by Fulton into this country, the resistance growing out of the dead water included between the two hulls, has been found such, that they have been gradually abandoned, and single vessels substituted. John Stevens, of Hoboken, commenced his experiments on steam navigation in 1791. Possessed of a patrimonial fortune, and well versed in science, he was at the time wanting in the practical mechanical skill that was necessary to success ; he was hence compelled, at first, to employ men of far less talent than himself, but who had been educated as practical machinists. His first engineer turned out an incorrigible sot ; his second be- came consumptive, and died before the experiment was completed. Stevens then resolved to depend upon his own resources, and built a workshop on his own estate, where he employed work- men under his own superintendence. In this shop he brought up his son, Robert L. Stevens, as a practical engineer, to whom many important improvements in steam navigation, and the most perfect boats that have hitherto been constructed, are due. During these experiments, Stevens invented the first tubular boiler ; and his first attempts were made with a rotary engine, for which, however, he speedily substituted one of Watt's. With various forms of vessels, and different modifications of propel- ling apparatus, he impelled boats at the rate of five or six miles per hour. They were, in truth, more perfect than any of his STEVENS STANHOPE. 263 predecessors', but did not satisfy his own high-raised hopes and sanguine expectations. These experiments were conducted at intervals up to the year 1807, and much diminished his fortune. We must, however, pass from the detail of them, and the notice of the parties who became concerned with him. in order to speak of what was doing in Europe in the meantime. The Earl of Stanhope, in 1793, revived the project of Gene- vois, for an apparatus similar to the feet of a duck. It was placed, in 1 795, in a boat furnished with a powerful engine. He was, however, unable to obtain a velocity greater than three miles per hour. While engaged in these experiments, he re- ceived a letter from Fulton, who proposed the use of pad- dle-wheels ; and it is probable that his neglect to listen to this suggestion caused a delay in the introduction of the steam-boat of at least twelve years ; for we cannot doubt that the ingenui- ty of Fulton, backed by the capital and influence of Lord Stan- hope, would have been as successful then as it was on a subse- quent occasion. In the year 1797 Chancellor Livingston, of the state of New- York, built a steam-boat on the Hudson River. He was asso- ciated in this enterprize with a person of the name of Nisbett, a native of England. Brunei, since distinguished for the block ma- chinery, and as engineer of the London Tunnel, acted as their en- gineer. In the full confidence of success, Livingston applied to the legislature of the state of New- York for an exclusive privilege, which was granted, on condition that he should, with- in ayear,produce a vessel impelled by steam at the rate of three miles per hour. This they were unable to effect, and the pro- ject was dropped for the moment. In the year 1800 Livingston and Stevens united their efforts, and were aided by Mr. Nicholas Roosevelt. Their apparatus was a system of paddles resembling a horizontal chain pump, and set in motion by an engine of Watt's construction. We now know that such a plan, if inferior to the paddle-wheel, might answer the purpose ; it, however, failed in consequence of the weakness of the vessel, which, changing its figure, dis- located the parts of the engine. One of the workmen in their employ suggested the use of the paddle-wheel in preference, 264 LIVINGSTON — EVANS. but, as Stevens candidly states, their'minds were not prepared to expect success from so simple a method. Their joint proceedings were interrupted by the appointment of Chancellor Livingston to represent the American govern- ment in France, but neither he nor Stevens were yet discour- aged ; the latter continued to pursue his experiments at Ho« boken, while the former carried to Europe high-raised expec- tations of success. It has been stated that Symington was employed by Miller of Dalswinton as his engineer ; we have now to record an attempt made by him under the patronage of Lord Dundas of Kerse. Miller's views appear to have been directed to the navigation of estuaries and rivers, if not to that of the sea itself. Symington, on the present occasion, limited himself to the drawing of boats upon a canal. The experiment was made upon the Forth and Clyde canal, but the boats were drawn at the rate of no more than three and a half miles per hour, which did not answer the expectations of his patron, and the attempt was abandoned. During this attempt, Symington asserts that he was visited by Fulton, who stated to him the great value such an invention would have in America, and by his account, took full and am- ple notes. In the attempt he thus makes to claim for himself the merit of Fulton's subsequent success, he is defeated by the clear and conclusive evidence that Fulton exhibited in a court of law, of his having submitted a plan analogous to that after- wards carried into effect, to Lord Stanhope, in 1795, six years prior to the experiment of Symington. That Fulton, whose thoughts had continued to dwell upon steam navigation, and who saw with prophetic eye, the vast space for this development afforded by the Mississippi and its branches, should have visited all the places where steam-boats were to be seen, was natural ; but a comparison of the draught of Symington's boat, which is still extant, with the boats constructed by Fulton, furnishes conclusive evidence that the latter borrowed no valuable ideas from the former. In the same year, 1801, Evans made, at Philadelphia, an ex- periment of a most remarkable character. Being employed by the Corporation of that city to construct a dredging machine, FULTON. 265 he built both the vessel and the engine at his works, a mile and a half from the water. The whole, weighing 42,000 lbs., was mounted upon wheels, to which motion was given by the en- gine, and thus conveyed to the river. A wheel was then fixed to the stern of the vessel, and being again set in motion by the en- gine, she was conveyed to her destined position. Evans, how- ever, appears long to have abandoned the hopes of exciting his countrymen to enter into his projects of locomotion, and content with his steady business as a millwright, and the proof he had thus given of the soundness of his ancient projects, pursued the matter no farther. We have thus completed the review of those attempts at na- vigation by steam which were abortive, either from absolute de- ficiency, or from their not fulfilling the expectations of the par- ties interested. It is now our more gratifying task to record in- stances of complete success. Livingston, who, as we have stated, carried with him to France a sanguine belief that steam naviga- tion was practicable, met Fulton at Paris. They were imme- diately drawn to each other by similarity of views, and the latter undertook to make those investigations which the avocations of the other prevented him from doing. It occurred to Fulton that the first step towards success was to investigate fully the capa- bilities of different apparatus for propulsion. These preliminary experiments were made at Plombieres, and led to the conviction that of all methods hitherto proposed, the paddle-wheel possess- ed the greatest advantages. He next planned a mode of attach- ing wheels to the engine of Watt, ingenious in itself, but com- plicated, and which he afterwards simplified extremely. Up to this time the relation of the force of the engine to the velocity of the wheels and the resistance of the water to the motion of the vessel, had never been made a matter of preli- minary calculation. Aware, however, that upon a proper com- bination of these elements all positive hopes of success must depend, he had recourse to the recorded experiments of the So- ciety of Arts, and limiting his proposed speed to four miles per hour, planned his machinery and boat in conformity. The ex- perimental vessel was then constructed at Paris, and being launched upon the Seine, performed its task in exact conformity 34 266 FULTON. to his anticipations. It was then, as afterwards, remarkable, that by a sound view of theoretic principles, the single boats of Fulton always possessed the speed which he predicted at the moment of planning them. This was not the case when he attempted double vessels, in consequence of his leaving out of view that important resistance which was mentioned in speak- ing of Miller's vessel. This preliminary experiment was performed in 1S03. While Fulton was engaged in preparing for it, a person of the name of Des Blancs, who was possessed of a patent for apparatus for steam navigation, endeavoured to interrupt it as an infringe- ment on his rights. Fulton, however, communicated to him his preliminary experiments, in which he had found paddle- wheels superior to the chain of floats proposed by Des Blancs, and the opposition ceased. The trial on the Seine having proved successful, it was resolved to take immediate measures to have a boat of large size constructed in the United States ; but as at that time the work-shops in America were incapable of fur- nishing a steam engine, it became necessary to order one from Watt and Bolton. This was done, and Fulton proceeded, to England to superintend its construction. In the meantime Livingston was sufficiently fortunate to obtain a renewal of the exclusive grant from the state of New- York. We here remark an anachronism in the work of Stuart. Sy- mington's own narrative, as given by that author, seems to place the interview with Fulton in 1S01. Stuart, in a subsequent place, refers it to the date of this visit of Fulton's to Eng- land. We have previously stated it as happening at the former date upon Symington's authority, as this is alone consistent with the expression of astonishment that he records. For this could hardly have been uttered subsequent to the trial made upon the Seine. Each of the dates, however, causes a dilemma. If he saw Symington's boat in 1801, he returned to France with his previous impression in favour of paddle-wheels very much weakened ; if not until 1804, he had already performed more than Symington. In like manner the claim of Henry Bell, so pertinaciously maintained by British authors, falls to the ground. Bell claims FULTON — BELL. 267 the merit of having furnished Fulton with the plan of his suc- cessful steam-boat on the ground of his having furnished plans and drawings, which he heard, two years afterwards from Fulton, were likely to answer this end. On receiving this letter, he states that " he was led to consider the folly of sending his opinions on these matters to other countries, and not putting them into practice in his own." Now, as Bell did not build his first boat until 1812, we cannot place the date of Fulton's second letter earlier than his return to America in 1806, and that it was written from America Bell's expressions render evident. Fulton, therefore, could have derived no bene- fit from his advice, for his experiment in France was in 1803, and the engine of Watt and Bolton, which was first used on the Hudson, must have been ordered at least a year before the al- leged date of Bell's communications. Neither can we reconcile his claims with the statement made by his friends, that he was several years in bringing his plans to perfection, and his boat was, after all, very inferior to those constructed by Fulton seve- ral years earlier. The anxiety of the British public to transfer the honours of Fulton to Bell, is manifest from a report of a Committee of Parliament, where it is stated that Bell came to this country to construct boats for Fulton, while it is now ad- mitted that he never was on this side of the Atlantic. We ap- prehend, however, that the correspondence with Bell took place on a different occasion. When Fulton planned his ferry-boats for the East River (New-York), he proposed to make them double ; he therefore naturally desired to know something of Miller's vessel which he had never seen, and, by Bell's own statement, the request of Fulton for information was limited to that single object. Bell asserts that he furnished, in addition, views and plans of his own, but long before this time Fulton's boats were in successful operation, and many competitors had already appeared, not only in those places where an exclusive grant existed, but even within the waters of the state of New- York. The engine ordered from Watt and Bolton reached New- York towards the close of the year 1806, and the vessel built 268 FULTON — STEVENS. to receive it was set in motion in the summer of 1807. The success that attended it is well known. In the mean time Livingston's former associate, the elder Stevens, had persevered in his attempts to construct steam-boats. In his enterprize he now received the aid of his son, and his prospects of success had become so flattering, that he refused to renew his partnership with Livingston, and resolved to trust to his own exertions. Fulton's boat, however, was first ready, and secured the grantof the exclusive privilege of the State of New- York. The Stevens's were but a few days later in moving a boat with the required velocity, and as their experiments were con- ducted separately, have an equal right to the honours of inven- tion with Fulton. Being shut out of the waters of the State of New-York by the monopoly of Livingston and Fulton, Stevens conceived the bold design of conveying his boat to the Delaware by sea, and this boat, which was so near reaping the honour of firstsuccess, was the first to navigate the ocean by the power of steam. From that time until the death of Fulton, the steam-boats of the Atlantic coast were gradually improved until their speed amounted to eight or nine miles per hour, a velocity that Ful- ton conceived to be the greatest that could be given to a steam- boat. To this inference he was probably led by the observa- tion of the increased resistance growing out of the wave raised in their front. His three earlier boats, the Clermont, the Car of Neptune, and the Paragon, were flat bottomed, their bows form- ing acute curved wedges, the several horizontal sections of which were similar. His last boats had keels, but they were in- troduced for no other purpose than to increase their strength. In the boats constructed by his successors after his death, a nearer approach was made to the usual figure of a ship, but the waves still formed an important obstacle. In the mean time the younger Stevens was steadily engaged in improving steam navigation, each successive boat constructed under his direction possessing better properties than the former. The view he took of the subject was different from that of Fulton ; believing that the great size of the wave was owing to defective form, he instituted experiments, both on a large and small scale, to determine the STEAM NAVIGATION. 269 figure in which this obstacle is of least magnitude. On the set- ting aside of the exclusive grant of the State of New-York to Liv- ingston and Fulton, he prepared a boat for navigation of the Hudson, which performed its voyage at the rate of 13 and a half English miles per hour. Steam-boats were not introduced into Great Britain until 1812, five years later than the successful voyage of Fulton. Bell, whose name has already been mentioned, built the first upon the river Clyde at Glasgow. In March, 1816, the first steam-boat crossed the British Channel from Brighton to Havre. Since that period their use has been much extended and their structure improved ; but, until lately, no European steam-boat had attained a speed of more than 9 miles per hour. In 1815 steam-boats, previously constructed by Fulton for the purpose, commenced to run as packets between New- York and Providence, Rhode Island, a part of which passage is per- formed in the open sea. One of these vessels had been intend- ed to make a voyage to Russia, but the greatness of the expense deterred the proprietors from undertaking it. This voyage was performed in 1817 by the Savannah, and in 1818 a steam-ship plied from New-York to New-Orleans as a packet, touching at Charleston and the Havana. In 1815 also, a steam-boat made a passage from Glasgow to London, under the direction of Mr. George Dodd ; but it was not until 1820 that steam-packets were established between Holyhead and Dublin. In 1825 a passage was made, by the steam- ship Enterprise, from London to Calcutta. All doubts, therefore, in respect to the practicability of navigating the ocean by steam might have been considered as settled. In point of economy, however, it can never compete with sails, and hence probably can only be used to advantage for conveying passen- gers, or for purposes of war. In the steam-boats of the Ohio and Mississippi, high-pressure engines are now in the most general use. The boilers are usually cylindrical, with internal flues ; and the favourite posi- tion of the cylinder is horizontal, resembling the engine on PL IV. Many of them, however, have conical valves, which are necessarily placed in vertical boxes ; this has demanded a novel 270 STEAM CARRIAGES. arrangement of the steam and eduction pipes, and of the appar- atus for working the valves. In France, Steam navigation has been of even more recent in- troduction than in England. Five years, as we have seen, elapsed from the time of Fulton's successful voyage until Bell navigated the Clyde, four more passed before a boat, built in England, crossed the Channel, and proceeded up the Seine to Paris. As steam navigation took its rise on the Hudson, so the steam-boats navigating that river have uniformly been before all others in point of speed. The passage to Albany does not at present (1839) average more than 10 hours, which is at the rate of nearly fifteen miles per hour. It is stated by Mr. Red- field, that the maximum velocities are 16 miles per hour, that 15 miles per hour is no unusual rate, and 14 may be considered as an ordinary performance on the Hudson river. The first boats which approached to this degree of speed were constructed under the direction of R. L. Stevens. Others, however, speedily follow- ed ; and the attainment of such velocities, which European wri- ters even at the present moment declare to be impossible, is due to the competition which has existed upon the Hudson. The speed of which we speak, has been obtained by increas- ing the length of the stroke of the piston, the area of the steam- pipes and valves, the diameters of the wheels, and by changes in the form of the vessels, to which false prows have been adapt- ed as experiments, until the figure of least resistance seems in some cases to have been reached. In some of the newer vessels the model has reached such a degree of perfection that no wave is raised at the bow, and no depression caused at the stern of the vessel. Above all, the expansive action of the steam has been employed, by means of which a given engine can be driven with greater velocity, and at a diminished cost. Others have approached this same speed so nearly, that the difference of passage has not been many minutes in the distance of nearly 150 miles. In a passage made by the author, on the Hudson, in 1829, the wheels of the New-Philadelphia averaged 2§J revolutions per minute ; and the piston moved with a velo- city of 405 feet per minute, being 21 feet more than has been STEAM CARRIAGES. 271 stated on a former page as the velocity of those of the North- America. Since that time the velocities of the pistons of steam- boats have been still further increased, and have in some cases amounted to as much as 600 ft. per minute. 204. The first attempt to navigate the ocean by steam was made, as we have seen, by John Stevens of Hoboken, in the year 1809, when he sent a vessel, originally constructed for a ferry-boat, from New- York to Philadelphia, around the capes of the Delaware. In the summer of 1815, the first steam vessel built on the Clyde by Bell made a passage from Glasgow to Liverpool, and during the autumn of the same year several other vessels, also built on the Clyde, were sent to different parts of England. During the equinoctial storm of 1816 one of these crossed from Brighton to Havre, in a gale which the cutter packets employed at that time on the station were unable to weather. The practicability and safety of navigating the stormy seas which surround the British Islands being thus demonstrated, the British Government was not long in undertaking to esta- blish lines of packets for the conveyance of its mails. The first line was established between Holyhead and Dublin, and has been in successful operation for twenty years. It is said that they have rarely failed in sailing at the appointed time, and have met with few or no accidents. Before the death of Fulton, he had planned a vessel which was intended to be used on the Baltic. This vessel was in a state of forwardness at time of his death. Circumstances pre- vented his successors from sending this vessel on her destin- ed voyage, but she was placed as a packet between New- York and Newport, R. I. in which passage the open sea is navigated for a short distance. The very voyage contemplated by Fulton was effected in 1818 by a vessel built in New- York, called the Savannah. The Savannah made her passage from New- York to Liverpool, partly by steam and partly by the aid of sails, in 26 days. From Liverpool this vessel proceeded around Scotland to the Baltic, and up that sea to St. Petersburgh. In returning thence she touched at Arendahl in Norway, and, without ma- 272 STEAM CARRIAGES. king any other intermediate port, reached New- York in 25 days. Daring the year 1819, a vessel rigged as a ship, but furnish- ed also with a steam engine, was built at New- York, for the purpose of plying as a packet between that port and Charleston, Cuba, and New-Orleans. So far as safety and speed were con- cerned, the experiment was successful ; but after several passa- ges it was found that the number of passengers was not suffi- cient to defray the expense, and the scheme was abandoned. The vessel was of such excellent model and construction, that she was purchased by the Brazilian government for a cruizer, and was as late as 1838 still in existence in that service. Be- fore this, however, the engine was taken out, and no other mode of propulsion employed except her sails. This vessel was con- structed under the direction of Mr. Jasper Lynch, who had ac- quired his knowledge of the use of the steam-engine from Ful- ton. The experiment, although a failure in point of profit, was worthy of the most complete success. The vessel had admir- able properties both as a sea-boat and a sailer, and the speed was not less than that which the best English steamers have reached up to the present time. Nothing was wanting except a sufficient tonnage to have enabled this vessel to cross the At- lantic in a time as short as that employed by the Great Western and Liverpool. The regularity and safety with which the passages between Holy-Head and Dublin were performed, established the fact of the superior safety of steamers in stormy and dangerous seas. Lines of packets were, in consequence, speedily established be- tween different points of the British Islands, and from Great Britain to the continent. Communications by steam have lont£ existed to Hamburgh, Rotterdam, Antwerp, Calais, and Havre ; and there are numerous steam packets plying between different ports of England and Ireland. The most important line is that between London and Leith, in which the largest steam vessels built before those intended for the navy or for crossing the Atlantic, were employed. The British Government has gradually extended its lines of communication to Lisbon, Gibraltar, Malta, and Corfu. It has STEAM CARRIAGES. 273 had it also in contemplation to extend them to Syria, in order to reach the Euphrates by land, and thence to establish steam- packets to Bombay. A company has also been formed for build- ing steamers to proceed to India by the way of the Cape of Good Hope. The first voyage to India by steam was performed in 1825, by the Enterprize. This vessel took her departure from Fal- mouth, and was 47 days between the Cape of Good Hope and Calcutta. As in the passage of the Savannah, the voyage was performed by the alternate aid of wind and steam. In spite of these experiments, of greater or less promise, it was seriously maintained by no mean authority, as late as Au- gust 1838, that the passage of the ocean, as a regular business by steam vessels, was impracticable. The most that could be hoped, as was alleged, would be to pass from the most western ports of Europe to the Azores or Newfoundland, and then take in a fresh supply of fuel. In the face of these discouraging predictions, the direct pas- sage from a port in Great Britain to New- York was made al- most simultaneously by two steamers before the end of the year in which the argument was held. Of these vessels, one (the Great Western) had been built for the express purpose, and had a tonnage adequate to the great probable consumption of fuel ; the other (the Sirius) was of the very class which had furnished the basis of the opinion ; and yet the fuel which could be carried was not entirely exhausted. It is therefore establish- ed beyond all possibility of doubt, that steam vessels, if they have the capacity of 12 to 1400 tons, may perform the direct passage from England to New -York by steam alone. It would also appear that no difficulty need exist in combining the sea- worthy qualities of the English steamers with the rapid motion of the American steam-boats ; and this may be effected, along with a considerable saving in fuel, and a great reduction of the weight of engine, boiler, and water. With such reduc- tions the carriage of many tons of cargo, as well as of passen- gers, will become possible, and the profits of the speculation will be placed upon a secure basis. The form of the engines and boilers of the British steamers 35 274 STEAM CARRIAGES. which have crossed the Atlantic, does not materially vary from that given in PL Till. The required increase of power has been given by enlarging the diameter of the cylinders be- yond the proportion which is there exhibited : and the extent of iron frame-work in which the engine is supported and kept to- gether, has been enlarged. The proportions of the cylinder, and the manner in which two working beams are suspended from the piston rods in each engine, have been adopted with a view to ensure the stability of the vessel by placing the weight as low as possible. So long as the masts and sails of steamers approach in weight and extent to those of ordinary vessels, this is no un- wise precaution : but as we firmly believe that sails might be dispensed with, this reason will no longer exist. The weight of these beams in particular is much greater than is admitted in American engines of equal power, where, instead of solid masses of cast-iron, a light frame-work of that material, surrounded by a strap of wrought iron, has been substituted, with a positive gain of strength. The boilers of the English vessels are of a form which is very weak, the flues are of great size, and the quantity of water is much greater in relation to the fire surface than is admitted in the American practice. While, therefore, we have to admire the sagacious views with which a sufficient capital to build such noble vessels has been contributed, and contrast it with the limited scale on which the navigation of the ocean has been attempted in this country, we believe that great improve- ments remain to bs made, by the introduction of the methods which we have cited as having contributed to give the great speed, which has been attained in the river boats of the United States. This is nearly one half more than has yet been reach- ed in Europe, and with it there can be no doubt that the pas- sage may be accomplished in 12 days. 205. The subject of the explosion of steam boilers has recently attracted a great share of public attention. A vast number of facts, and a great variety of written opinions, have been collect- ed by the Secretary of the Treasury, and published by order of Congress. Among these papers we may quote, for the infor- STEAM CARRIAGES. 275 mation of our readers, one by Mr. Red field of New- York. This gentleman adopts a different view of the subject from that given by us in Chap. II. Still the results at which he arrives are in strict conformity with those derived from the other theory, and are therefore to be implicitly relied on. " If high-pressure engines must continue to be used, (of which I see not the utility or necessity,) the working pressure should never exceed fifty pounds to the square inch ; and this may be easily effected by increasing the size and stroke of the working cylinders and piston. The forms of the boilers should be cylindrical, and their diameters from 36 to 42 inches, supported by their centres as well as at their terminations. Flues, if of a size affording but one or two in each boiler, are always dangerous ; they displace too much water, and also obstruct the proper cleaning. Flues, however, are not to be dispensed with, but their number ought to be increased and their size diminish- ed. An upper tier of four flues, and a lower tier of two, (the latter somewhat larger than the former,) are not too many for boilers of 42 inches in diameter ; or 44 to 48 inches, if low pres- sure. These smaller flues, if properly arranged, will greatly fa- cilitate the cleaning, and displace but little water ; but their length should not usually exceed ten or twelve feet, as they ab- stract heat very rapidly. They will be better if made perfectly smooth on their inner surface, from a single long sheet of iron, lighter than the shell ; and are not often liable to leaks or acci- dents. The outer shell should never be less in thickness than a full quarter of an inch ; and a thickness much exceeding this, it is well known, cannot be used with advantage. "In condensing engines which work expansively, called low- pressure, when working with ordinary speed, the pressure of the steam should usually range between one and one and a half atmospheres above the boiling point. But on emergencies the pressure may be increased to two atmospheres. The boilers should have a range of strength falling but little short of those used for high pressure. They may be constructed in the common wagon top form, provided that they are properly braced in their flat sides and arches, and have as many as four or six flue- arches for a boiler of eight or ten feet in width. The returning 276 STEAM CARRIAGES. flues should be cylindrical, and of smaller diameter. The water-sides, water-bottoms, bridge-walls, and other flat surfaces, should, however, be brace-bolted at intervals of six inches ; and the arches, shell, and all other portions, secured in a propor- tionate manner. If a steam- chimney is used, even of the cir- cular form, it should be brace-bolted at smaller intervals than any part of the flat surfaces which are covered by water." 206. Steam is also employed to move carriages upon the land. For this purpose, the wheels of the carriage are set in motion by the engine, in the same manner that the paddle- wheels of a steam-boat are caused to turn ; the friction which they experience upon their track causes them to move forward, unless they meet a resistance to their progressive motion equal to this friction. The experiments of Coulomb and Yince show that, under the circumstances in which wheels act, the friction of their circumference will depend upon the weight with which they are loaded, and the nature of the rubbing surface, but not in the least upon the velocity. The tire of wheels is made of iron, and steam-carriages usually run upon tracks, also of iron, forming what is styled a rail-road. Rail-roads are parallel bars of iron, laid either level, or with a gentle and uniform slope ; and steam has, as yet, only been usefully applied to locomotion up- on roads of this character. The reasons why they should be superior in this respect to a common road are obvious. The resistance is not only regular and uniform, but equal upon every wheel ; while on a common road there is a constant vari- ation in slope, and in the nature of the surface ; and besides, obstacles are frequently met that affect but one of the wheels, and thus tend to turn the carriage to one side. There is thus a want of continuity in the motion of the carriage, a lateral sliding friction of the wheels upon the road, and one arising from penetration into the materials of which the road is made. In addition, the friction of the wheel upon the shoulder of the axle and on the linen pin, is of great amount on a common road. In spite of these difficulties, some tolerably successful experiments have been performed with steam-carriages upon common roads. The case, however, that is most usual as well as most advan- STEAM CARRIAGES. 277 tageous, is motion upon rail-roads. Here the friction is that of iron against iron. We cannot anticipate that the wheels will be prevented from sliding upon a rail- road by the maximum friction that takes place between two pieces of iron in experi- ments ; dust, moisture, and other circumstances interfere to lessen the adhesion. It cannot, therefore, be safely taken at more than |th part of the weight. If there be a force applied, sufficient to cause the wheels of a carriage to turn around, it will continue to go forward until the resistance becomes equal to -J-th of the weight of the carriage. The carriage is, therefore, under the same circumstances as if it were drawn forward by a cord capable of bearing a strain of ^-th part of its weight. The resistances to the progressive motion are the friction upon the axis of the wheels, and the disturbances growing out of lateral shocks. The friction of steel axles upon brass boxes, well coated with oil, is -j^-th part of the weight ; and the force applied to overcome it has its intensity increased in the ratio of the radius of the crank to the radius of the axle. As the radius of the crank of an engine of a given power cannot be increased without diminishing the area of the piston or its own velocity, there is no gain of force by simply varying the propor- tions of its engine. On the other hand, as with an equal num- ber of revolutions, points will move faster on the circumference of a larger wheel than they will on a smaller one, and the pro- gressive motion will depend on the velocity of the circumference, there is a constant and regular gain in velocity, by increasing the diameter of the wheels. This, however, has its limit in practice, for, by increasing the diameter of the wheels, the cen- tre of gravity is raised, and the machine becomes unstable. According to the best experiments and observations, the fric- tion of carriages upon rail-roads has been in some cases dimin- ished to -fo th ; and may be safely taken as not more than g-MJi. A locomotive carriage, therefore, all of whose wheels are driven by the engine, may move forward if it drag behind it any weight less than thirty-two times its own. It might, at first sight, appear that, as the friction which causes the carriage to go forward increases with its weight, heavy carriages and engines were the best for locomotion ; but 278 STEAM CARRIAGES. the resistances increase also with the weight, and thus all weights, not absolutely essential to the structure of the engine, are disadvantageous. Hence, for locomotion, no other en- gine but that of high pressure can be admitted : for condensing engines of equal power are not only heavier in themselves, but require a quantity of cold water for condensation, that would, of itself, furnish a load for the engine. So also the boiler and the load of water, should be the smallest that is consistent with the generation of the necessary quantity of steam. The workmanship of the carriages used on rail-ways has been regularly improved for several years past, and probably has not attained perfection. The want of perfection in the workmanship, and perhaps the absolute impossibility which exists of making all the wheels of equal diameter, has led to the practice, in rapid motions, of giving no more than one pair of wheels a motion from the engine. This pair bears little more than half the weight, and hence the propulsive power is appar- ently less than if all the wheels were driven. This loss, how- ever, is not real ; for the sliding of wheels, not absolutely equal in diameter, will consume more power than is apparently lost. On the other hand, in slow motions, and in the ascent of in- clined planes, heavy engines, of which all the wheels are driven by the engine, are employed. A vast improvement has taken place in the performance of locomotive engines since the publication of our first edition. At that time we did not venture to state the actual draught of a locomotive at more than seven times its own weight. We are now enabled to rate it as high as thirty-two times as much as rests on the driving wheels. With an engine of the weight of 8 tons, the load has been as great as 175 tons, or more than 40 times the weight which rests on the active wheels ; and the velocity with this load is 12.^ miles per hour. In doubling the load, the velocity is diminished to ^th, while in a given dis- tance the expenditure of fuel is diminished one half. An engine constructed by H. R. Dunham & Co. of New-York for theHarlsem Rail Road weighed 20,400 lbs. or about 9 tons ; the boiler being full of water, and the engine in working order. Of this weight 10,680 lbs. bore on the driving wheels. The STEAM CARRIAGES. 279 load drawn was 105 tons upon 35 cars, whose weight is not given. The road was not level, and the slopes were from 25 to 30 feet per mile. A locomotive engine is propelled in all cases by steam of high pressure. This mode of employing steam is rendered necessary by the great quantity of water required in condensa- tion, which would of itself furnish a large part of the load which can be drawn. The cylinder of the engine has been usually placed horizontally, or but little inclined. Some of those on the Baltimore and Ohio Rail Road have been placed vertically. Two cylinders are generally used, acting upon cranks on the axle of the same pair of wheels, at right angles to each other. In this way one piston is at its maximum ac- tion while the crank of the other is passing the centres, and greater regularity of motion is ensured. When the other wheels are to be set in motion, they are united with the first pair by means of connecting rods. We have already stated in what cases all the wheels are to be driven, and when no more than one pair. In the former case no more than four wheels are used. In the latter case, after trying curricle engines, those with six wheels have been found most serviceable. The English en- gineers place the driving wheels, which are of greater diameter than the remaining four, between the other two pairs. In the American engines the driving wheels are at one end of the carriage, and the four others are united in the same frame on which the opposite end bears. Engines of this form, of great perfection of workmanship, have been constructed by various artists, of whom the most celebrated are Baldwin and Norris. We have obtained, as an illustration of this part of the subject, a draught of a locomotive by Dunham of New- York. This is represented on PL IX., and is a specimen of the form now con- sidered as most advantageous. An engine with six wheels was first planned in the year 1826 for the Mohawk and Hudson Rail Road, by Mr. J. B. Jervis. In order to compare the action of steam upon rail-roads, with its performance in propelling boats, we have the following principles : — 280 STEAM CARRIAGES. Friction opposes a resistance which has a constant measure at all velocities ; but the measure of the power required to over- come it ; will depend both on the resistance and the velocity. Hence the powers of engines, by which different velocities are obtained in the same carriage, are proportioned to the velocities. But as the time for passing over a given space is inversely as the velocity with which the distance is performed, a given dis- tance should be performed, with a constant load, at any velocity whatever, with a constant expenditure of fuel. If the same locomotive engine have its velocity increased by lessening the loads it drags or diminishing the friction, by both of which methods a limited change in velocity maybe attained, the expenditure of steam has been found to increase in a higher ratio than the velocities. This arises from the fact, to which we have more than once referred, that the action of steam of a given tension on the piston of an engine is diminished, when the velocity is increased. Were it not so, the expenditure of steam should be in this case proportioned to the velocities. It is therefore obvious, that when speed is the sole object in view, locomotion on land soon becomes more advantageous than steam navigation, for the power in the latter case increases, according to the received theory, as the cubes of the velocities ; and the expenditures of fuel as the square. Even if the view which we have presented as more consistent with the facts, be true, the power must be increased as the squares, and the ex- penditure of fuel with the first power of the velocities. On the other hand, friction on rail-roads has not yet been so much diminished as to enable them to compete either with steam or canal navigation, in the conveyance of heavy loads at small velocities. The friends of rail-roads have anticipated that they will soon be enabled to lessen the friction so much as to place them, in all respects, on a par with either of the other species of transportation. It would, however, appear, both from theory and experience, that unless when a saving of time is the principal object, the application of steam to navigation is more advantageous than to the rail-road. 207. Evans, as has been already mentioned, was the first STEAM CARRIAGES. 281 who entertained rational hopes of being able to move carriages by steam, for we must reject the views of Robison and Watt as wholly impracticable ; and indeed the impossibility of using the condensing engine was ascertained and admitted by Watt. Evans not only was the first to entertain correct views, but was also the first to submit them to practice, in the removal of his dredging machine, which has been before referred to in the pre- sent chapter. In 1802 Trevithick and Vivian took out a patent for the application of their engine to propel carriages upon rail-roads. In 1804 they published a description of a carriage intended for common roads, but it was not until 1806 that an actual ex- periment was made. This was performed upon the Merthr Tydvil Rail-Road, in Wales. The performance of the appar- atus was, however, far less than might have been anticipated from its power, and this was ascribed to a want of sufficient adhesion of the wheels to the rail. We recollect having heard this failure ascribed to the circumstance that but one of the wheels was set in motion by the engine ; but all the authorities that we have consulted seem to agree, that all the four wheels were made to revolve. The failure, which, had the first statement been true, is at once to be accounted for, becomes difficult to explain if these authorities state the real circumstances. A difficulty, however, in the use did occur, and being ascrib- ed to the cause that has been mentioned, a person of the name of Blenkinsop undertook to obviate it. For this purpose he laid a rack, or rail cut into teeth, between the other two rails, along the whole extent of road : into this a pinion, set in mo- tion by the engine, caught. This method was found effectual at slow velocities, and was used from the year 1801, in which it was invented, nearly up to the present time, at Middleton Colliery, near Leeds in England. It will not admit of great velocities, but is applicable to the rising of ascents far more steep than can be overcome by the mere adhesion of the wheels to the road. In 1812, Messrs. W. & E. Chapman obtained a patent in England for a locomotive engine, the power of which was ap- 36 282 STEAM CARRIAGES. plied by means of a chain fixed at the two ends, and passing over an axle upon the carriage that was caused to revolve by the engine. In 1813 Mr. Brunton, of Batterly Iron Works, proposed a plan for locomotion by steam, in which he employed a system of levers, resembling, in their action, the bones of the human leg. In 1815, Dodd and Stephenson, of Killingworth, in England, returned to the original principle of adhesion, and were com- pletely successful, showing that on rail-roads, absolutely or nearly level, the friction was sufficient to produce progressive motion in all cases except when the rails were covered with snow. Their engine had six wheels, two of which were moved by the engine, and the others connected with them by an endless chain passing over drums. Locomotive engines have received, since that time, continual improvements. Two cylinders have been used, each acting upon a pair of wheels. The next step was to use two cylin- ders acting at right angles to each other upon the same pair of wheels, and to move the others by connecting rods. In these several improvements, the weight of the engine and its parts were gradually increased to an excessive amount. The centre of gravity was also raised so high as to render the carriages unstable. In consequence of this, a search has more recently taken place for engines and carriages of small weight. This has been successful in a remarkable degree, in locomotive engines exhibited upon the Manchester and Liverpool Kail- Roads. The details of these experiments are to be found in the Mechanics' Magazine for November and December, 1829, and in the Quarterly Review for March, 1830, to which we re- fer our readers. The Baltimore and Ohio Rail-Road was projected, and some parts of it finished, as early as the Manchester and Liverpool. It also became the seat of a number of experiments, and these have been continued upon it, and on other more recent rail- roads, until such a degree of perfection has been reached in the structure of locomotive engines in the United States, that they have been made an article of export. CONCLUSION. 283 208. In concluding this work, a few reflections on the impor- tance of the subject may not be irrelevant. The steam engine has been described in its most usual and most perfect forms ; in the historical sketch, it has been traced from the earliest no- tices of the knowledge of the mechanical power of steam, down to the present time, when it occupies so important a space among the productions of human skill. Feeble and imper- fect in its first beginnings, and limited, for nearly a century after its introduction, to a single, and by no means important object, it became in the hands of Watt an instrument of universal application. It is now equally subservient to those purposes which require the greatest delicacy of manipulation, and those which demand the most intense exertions of power. Its introduction and gradual improvement have required in- ventive talents of the highest order, and the exertions of genius the most sublime ; in its uses we see developed and realized, not only the brilliant conceptions of poetry, but the wildest fa- bles of romance ; it has already changed the state of the world, and altered the relations of civilized society ; and in its farther progress it seems to promise to perform even more inportant services, and to fulfil yet higher destinies. APPENDIX. ANALYSIS OF A NEW THEORY OF THE STEAM ENGINE BY THE CH. G. DE PAMBOUR. The following analysis of a new theory of the Steam Engine, made by its author, from the full exposition which he has laid before the French Institute, will be found to possess much interest. Before it was received, the second edition had been prepared for the press ; and, even had it been judged expedient, it would have been too late to adopt it as the basis of practical rules. It may, however, be stated, that however fully we concur in the views of the Chev. de Pambour, it would have been premature to adopt it until it had received a more general sanction ; and that its assumption might have for a time un- fitted our own work for the use of practical men. It is therefore an- nexed as an Appendix, for the purpose of giving it circulation, and preparing the public mind for its reception in the place of that of Robinson, which has hitherto formed the basis of all treatises on the Steam Engine, and from which, although aware of its defects, we have not ventured to deviate. 288 PART L PROOFS OF THE INEXACTITUDE OF THE ORDINARY METHODS^ AND EXPOSITION OF THE ONE PROPOSED. § 1. JYIode of calculation hitherto in use. — All the problems in the application of steam-engines merge into these three — The velocity of the motion being given, to fuad the load the en- gine will move at that velocity. The load being given, to find the velocity at which the engine will move that load ; And, the load and the velocity being given, to find the vaporiza- tion necessary, and consequently the area of heating surface re- quisite for the boiler, in order that the given load be set in mo- tion at the given velocity. The problem, which consists in determining the useful effect to be expected from an engine of which the number of strokes of the pis- ton per minute is counted, that is, whose velocity is known, evident- ly amounts to determining the effective load corresponding to that ve- locity ; for that load being once known, by multiplying it by the ve- locity we have the useful effect required. According to the mode of calculation hitherto admitted, when it is wanted to know the useful effect an engine will produce at a given velocity, or, in other words, the effective load that it will set in mo- tion at that velocity, the area of the cylinder is multiplied by the ve- locity of the piston, and that product by the pressure of steam in the boiler ; this gives, in the first place, what is called the theoretical ef- fect of the engine. Then, as experience has shown that steam-en- gines can never completely produce this theoretical effect, it is re- duced in a certain proportion, indicated by a constant number, which is the result of a comparison between the theoretical and practical ef- fects of some engines previously put to trial ; and thus is obtained the number which is regarded as the practical effect of the engine, or the work it really ought to execute. A mode perfectly similar is followed, for determining the vapori- zation which an engine ought to produce in order to produce a de- sired effect ; that is to say, for resolving the third of the problems which we have presented above. As to the second of these problems, 289 that which consists in determining the velocity the engine will as* sume under a given load, no solution of it has been proposed in this way, and we shall expose, farther on, some fruitless essays that have been made to resolve it in another way. As in the above-mentioned calculation no account is taken of fric- tion, nor of some other circumstances which appear likely to dimin- ish the power of the engine, the difference observed between the theoretical and the practical result excites no surprise, and is readi- ly attributed to the circumstances neglected in the calculation. § 2. First objection against this method of calculation. — This mode of calculation is liable to many objections, but for the sake of brevity we limit ourselves to the following : — The coefficient adopted to represent the ratio of the practical effects to the theoretical, varies from \ to f, according to the various systems of steam-engines ; that is to say, that from f to ^ of the power exerted by the machine is considered to be absorbed by friction and divers losses. Not that this friction and these losses have been measured and found to be so much, but merely because the calculation that had been made, and which might have been inexact in principle, wanted so much of coinciding with experience. Now it is easy to demonstrate that the friction and losses which take place in a steam-engine can never amount to f, nor to -*- of the total force it developes. It will suffice to cast an eye on the explana- tion attempted, on this point, by Tredgold, who follows this method in his Treatise on Steam-Engines.* He says (art. 367,) that, for high pressure engines, a deduction of -~^- must be made from the to- tal pressure of the steam, which amounts to a deduction of -^ on the ordinary effective pressure of such engines ; and to justify this de- duction, which, however, is still not enough to harmonize the theore- tical and practical results in many circumstances, he is obliged to es- timate the friction of the piston, with the losses or waste, at -fa of the power, and the force requisite for opening the valves and over- coming the friction of the parts of machine, at -j-J-j- of that power. Reflecting that these numbers express fractions of the gross power of the engine, we must readily be convinced that they cannot be cor- rect ; for, in supposing the engine had a useful effect of 100 horses, * The author here refers to the first edition of ' Tredgold on the Steam-En- gine :' in the new edition just published, the algebraic parts are transformed by the editor into easy practical rules, accompanied by examples familiarly explain- ed for the working engineer. 37 290 which, from the reduction or coefficient employed; supposes a gross effect of 200 horses, 12 would be necessary to move the machinery, 40 to draw the piston, &c. ! The exaggeration is evident. Besides, iu applying this evaluation of the friction to a locomotive engine, which is also a high pressure steam-engine, and supposing it to have 2 cylinders of 12 inches diameter, and to work at 75 lbs. total pressure, which amounts to 60 lbs. effective pressure, per square inch, we find that from the preceding estimate, the force necessary to draw the piston would be 5650 lbs., whereas our own experiments on the locomotive engine, the Atlas, which is of these dimensions, and works at that pressure, demonstrate that the force necessary to move, not only the two pistons, but all the rest of the machinery, including the waste, &c, is but 48 lbs. applied to the wheel, or 2831bs. applied on the piston. It is then impossible to admit, that in steam-engines the friction and losses can absorb the half, nor the third, much less the-f of the total power developed ; and yet there do occur cases wherein, to re- concile the practical effects with the theoretical ones thus calculated, ' it would be necessary to reduce the latter to the fourth part, and even to less ; and, what is more, it often happens that the same engine which in one case requires a reduction of f^ will not in other cases need a reduction of more than about -£-. This is observed in calcu- lating the effects of locomotive engines at very great velocities, and afterwards at very small ones. There is no doubt, then, that the difference observed between the theoretical effect of an engine and the work which it really performs, does not arise from so considerable a part of the applied force being absorbed by friction and losses, but rather from the error of calculating in this manner the theoretical effect of the machine. In effect, this calculation supposes that the motive force, that is, the pressure of the steam against the piston or in the cylinder, is the same as the pres- sure of the steam in the boiler ; whereas we shall presently see, that the pressure in the cylinder may be sometimes equal to that of the boiler, sometimes not the half nor even the third of it, and that it depends on the resistance overcome by the engine. § 3. Formula, proposed by divers authors to determine the velocity of the piston under a given load, and proofs of their inexactitude. — We have said that this problem was net resolved by the foregoing method. The following are the attempts made to that end by ano- ther way. Tredgold, in his Treatise on Steam-Engines (art. 127 291 and following), undertakes to calculate the velocity of the piston from considerations deduced from the velocity of the flowing of a gas, supposed under a pressure equal to that of the boiler, into a gas sup- posed at the pressure of the resistance. He concludes from thence, that the velocity of the piston would be expressed by this formula, V = 6-5 Vh, in which V is the velocity in feet per second, and h stands for the difference between the heights of two homogeneous columns of vapour, one representing the pressure in the boiler, the other that of the resis- tance. But it is easily seen that this calculation supposes the boiler filled with an inexhaustible quantity of vapour, since the effluent gas is supposed to rush into the other with all the velocity it is susceptible of acquiring, in consequence of the difference of pressure. Now, such an effect cannot be produced, unless the boiler be capable of supplying the expenditure, however enormous it might be. This amounts, conse- quently, to supposing that the production of steam in the boiler is unlimited. But, in reality, this is far from being the case. It is evident that the velocity of the piston will soon be limited by the quantity of steam producible by the boiler in a minute. If that pro- duction suffice to fill the cylinder 200 times in a minute, there will be 200 strokes of the piston per minute; if it suffice to fill it 300 times, there will be 300 strokes. It is then the vaporization of the boiler which must regulate the velocity, and no calculation which shall exclude that element can possibly lead to the true result ; consequent- ly the preceding formula cannot be exact. This is why, in applying this formula to the case of an ordinary lo- comotive engine of the Liverpool Railway with a train of 100 tons, the velocity the engine ought to assume is found to be 734 feet per second, instead of twenty miles an hour, or five feet per second, which is its real velocity. Again, in his Treatise on Railways (page 83), Tredgold proposes the following formula, without in any way founding it on reasoning or on fact : v = 24o— , in which V is the velocity of the piston in feet per minute, I the stroke of the piston, P the effective pressure of the steam in the boil- er, and W the resistance of the load. But as this formula makes no mention either of the diameter of the cylinder, or of the quantity of steam supplied by the boiler in a minute, it clearly cannot give the 292 velocity sought ; for if it could, the velocity of an engine would be the same with a cylinder of one foot diameter as with a cylinder of four feet, which expends sixteen times as much steam. The area of heating surface, or the vaporization of the boiler, would be equally in- different : an engine would not move quicker with a boiler vaporizing a cubic foot of water per minute, than with one that should vaporize but ^ or -^q. Hence this formula is without basis. Wood, in his Treatise on Railways (page 351), proposes the fol- lowing formula also, without discussion, V = 4_, where V is the velocity of the piston in feet per minute, / the length of stroke of the piston, W the resistance of the load, and P the sur- plus of the pressure in the boiler, over and above what is necessary to balance the load W. This formula being liable to the same objections as the preceding, is also demonstrated inadmissible apriori. Consequently, of the three fundamental problems of the calculation of steam-engines, two have received inaccurate solutions by means of the coefficients, and the third, as we have just seen, has received na solution at all. § 4. Succinct exposition of the proposed theory. — After having made known the present state of science, with regard to the theory and estimation of the effective power of steam-engines, it remains to exhibit the theory we apply to them ourselves. It is well known, that in every machine, when the effort of the motive power becomes superior to the resistance, a slow motion is created, which quickens by degrees till the machine has attained a certain velocity, beyond which it does not go, the motive power being incapable of producing greater velocity with the mass it has to move. Once this point attained, which requires but a very short space of time, the velocity continues the same, and the motion remains uni- form as long as the effort lasts. It is from this point only that the effects of engines begin to be reckoned, because they are never em- ployed but in that state of uniform motion ; and it is with reason that the few minutes, during which the velocity regulates itself, and the transitory effects which take place before the uniform velocity is ac- quired, are neglected. Now, in an engine arrived at uniform motion, the force applied by the motive power forms strictly an equilibrium with the resistance ; for if that force were greater or less, the motion would be accelerated 293 or retarded, which is contrary to the hypothesis. In a steam-engine the force applied by the motive agent is nothing more than the pres- sure of the steam against thepiston, or in the cylinder. The pressure therefore in the cylinder is strictly equal to the resistance of the load against the piston. Consequently the steam, in passing from the boiler to the cylinder, may change its pressure, and assume that which is represented by the resistance of the piston. This fact alone exposes all the theory of the steam-engine, and in a manner lays its play open. From what has been said, the force applied on the piston, or the pressure of the steam in the cylinder, is therefore strictly regulated by the resistance of the load against the piston. Consequently calling P' the pressure of the steam in the cylinder and R the resistance of the load against the piston, we have as a first analogy, P' = R. To obtain a second relation between the data and the qusesita of the problem, we shall observe that there is a necessary equality be- tween the quantity of steam produced, and the quantity expended by the machine ; the proposition is self-evident. Now, if we express by S the volume of water vaporized in the boiler per minute, and ef- fectively transmitted to the cylinder, and by m the ratio of the volume of the steam generated under the pressure P of the boiler, to the volume of water which produced it, it is clear that m S will be the volume of steam formed per minute in the boiler. This steam passes into the cylinder, and there assumes the pressure P' ; but if we suppose that, in this motion, the steam preserves its tempera- ture in passing from the boiler to the cylinder, or from the pressure P to the pressure P', its volume increases in the inverse ratio of the pressures. Thus the volume m S of steam furnished per minute by the boiler will, when transmitted to the cylinder, become P m S . p7. On another hand, v being the velocity of the piston, and a the area of the cylinder, a v will be the volume of steam expended by the cylinder in a minute. Wherefore, by reason of the equality which necessarily exists between the production of the stiam and the expen* diture, we shall have the analogy of a P a v = m S . — ; : 294 which is the second relation sought. Consequently, by exterminating P' from the two equations, we shall have as a definitive analytic relation among the different data of the problem : «iS P a it This relation is very simple, and suffices for the solution of all questions regarding the determination of the effects or the proportions of steam-engines. As we shall develope its terms hereafter, in taking it up in a more general manner, we content ourselves to leave it for the present under this form, which will render the discussion of it easier and clearer. The preceding equation gives us the velocity assumed by the pis- ton of an engine under a given resistance R. If, on the contrary, the velocity of the motion be known, and it be required to calculate what resistance the engine will move at that velocity, it will suffice to resolve the same equation with reference to R, which will give v m S P a v Finally, supposing the velocity and the load to be given before- hand, and that it be desired to know what vaporization the boiler should have to set the given load in motion at the prescribed velocity, it will still suffice to draw from that analogy the value of S, which will be c _ a v R ~ln~T' On these three determinations we rest for the moment, because, as will soon appear, they form the basis of all the problems that can be proposed on steam-engines. § 5. New proofs of the exactitude of this theory, and of the inac- curacy of the ordinary mode of caladation. — The theory just develop- ed demonstrates that the steam may be generated in the boiler at a certain pressure P, but that in passing to the cylinder it necessarily assumes the pressure R, strictly determined by the resistance to the piston, whatever the pressure in the boiler may be. Consequently, according to the intensity of that resistance, the pressure in the cy- linder, far from being equal to that in the boiler, or from differing from it in a certain constant ratio, may at times be equal to it, and at other times very considerably different. Hence those who, in per- forming the ordinary calculation, consider the force applied on the 295 piston as indicated by the pressure in the boiler, begin by introduc- ing into their calculation an error altogether independent of the real losses to which the engine is liable. To this cause, then, and not to the friction and losses, which can form but the smallest part of it, must be attributed the enormous difference which, in this mode of calculation, is found between the theoretical effect of the engine, and the work which it really executes. We have already proved the mode of action of the steam in the cy- linder by the consideration of uniform motion ; but in examining what passes in the engine, we shall immediately find many other proofs. 1st. The steam, in effect, being produced at a certain degree of pressure in the boiler, passes into the tube of communication, and thence into the cylinder. It first dilates, because the area of the cy- linder is from ten to twenty-five times that of the tube ; but it would promptly rise to the same degree as in the boiler, were the piston im- moveable. But as the piston, on the contrary, opposes only a cer- tain resistance, determined by the load sustained by the engine, it will yield as soon as the elastic force of the steam in the cylinder shall have attained that point. The piston, in consequence, will be a valve to the cylinder. Hence the pressure in the cylinder can never exceed the resistance of the piston, for that would be supposing a vessel full of steam, in which the pressure of the steam would be greater than that of the safety valve. 2nd. Were it true that the steam flowed into the cylinder, either at the pressure of the boiler, or at any other pressure which were to that of the boiler in any fixed ratio, as the quantity of steam generat- ed per minute in the boiler would then flow at an identical pressure in all cases, and would consequently fill the cylinder an identical number of times per minute ; it would follow, that as long as the en- gine should work with the same pressure in the boiler, it would as- sume the same velocity with all loads. Now, we know that precisely the contrary takes place, the velocity increasing when the load dimin- ishes ; and the reason of it is, that when the load is half, the steam flowing also at a half pressure into the cylinder, and consequently acquiring a volume double what it had before, will serve for double the number of strokes of the piston. 3rd. Applying the same reasoning * inversely, we perceive that were the pressure in the cylinder really bearing a constant ratio to that in the boiler, or if it be preferred, constant so long as that in the 296 boiler did not vary, we should, in calculating the effort of which the engine would be capable, always find it the same, whatever be the velo- city of the piston. Thus, at any velocity whatever, the engine would always be capable of drawing the same load ; which experience again contradicts, for the greater the velocity of the piston, the lower the pressure of the steam in the cylinder, whence results, that the load of the engine lessens at the same time. 4th. Another no less evident proof of this is easily adduced. Were it true that the pressure in the cylinder were to that in the boiler in any fixed proportion, since the same locomotive engine always requires the same number of revolutions of the wheel, or the same number of strokes of the piston to traverse the same distance, it would follow that, as long as those engines worked at the same pressure, they would consume in all cases the same quantity of water for the same distance. Now, the quantity of water, far from remaining constant, decreases on the contrary with the load, as may be seen by the experiments we have published on this subject. Here therefore again it is proved, that, notwithstanding the equality of pressure in the boiler, the densi- ty of the steam expended follows the intensity of the resistance, that is to say, the pressure in the cylinder is regulated by that resistance. 5th. Similarly, the consumption of fuel being in proportion to the vaporization effected, it would follow, if the ordinary theory were ex- act, that the quantity of fuel consumed by a given locomotive, for the same distance, would always be the same, with whatever load. Now we again find by experience that the quantity of fuel diminishes with the load, conformably to the explanation we have given of the effects of the steam in the engine. 6th. It is again clear, that if the pressure in the cylinder were, as it is believed, constant for a given pressure in the boiler, that so soon as it was recognised that an engine could draw a certain load with a certain pressure, and communicate to it a uniform motion, it would follow that the same engine could never draw a less load with the same pressure, without communicating to it a velocity indefinitely accelerated 5 since the power, having been found equal to the resis- tance of the first load, would necessarily be superior to that of the second. Now, experience proves, that in the second case the veloci- ty is greater, but that the motion is no less uniform than in the first ; and the reason of this is, that though the steam may indeed be pro- duced in the boiler at a greater or less pressure, and that it matters little, yet on passing into the cylinder, it always assumes the pressure 297 of the resistance, whence results that the motion must remain uni- form as before. 7th. Finally, in looking over our experiments on locomotives, it will be seen that the same engine will sometimes draw a light load with a very high pressure in the boiler, and sometimes a heavy load with a very low pressure. It is then impossible to admit, as the ordi- nary calculation supposes, that any fixed ratio whatever has existed between the two pressures. Moreover, the effect just cited is easy to explain, for it depends simply on this, that in both cases the pressure in the boiler was superior to the resistance on the piston ; and it needed no more for the steam, generated at that pressure or at any other, satisfying merely that condition, to pass into the cylinder and assume the pressure of the resistance. It is then visible, from these various proofs, that the pressure in the cylinder is strictly regulated by the resistance on the piston, and by nothing else ; and that any method like that of the coefficients in the ordinary calculation, which tends to establish a fixed ratio between the pressure in the cylinder and that of the boiler, must necessarily be inexact. § 6. Verification of the two modes of calculation by particular ex- amples. — We have sufficiently demonstrated the want of basis of the ordinary calculation ; but as the inaccuracy we have just exposed in that method might by some be supposed to be of slight imporlance, and they might conceive that, in practical examples, it amounted to the obtaining of results, which, if not quite exact, were at least very near the truth, we will now attempt to apply it to some particular cases. The coefficient of reduction for high pressure engines, working without expansion and without condensation, not being given by the authors who have treated on these subjects, we propose, in order to determine it, the two following facts, which took place before our eyes : — I. The Leeds locomotive engine, which has two cylinders eleven inches in diameter, stroke of the piston sixteen inches, wheel five feet in diameter, drew a load of 88-34 tons, in ascending a plane in- clined 1 in 1300, at the velocity of 20*34 miles an hour ; the effective pressure in the boiler being 54 lbs. per square inch, or the total pres- sure 68*71 lbs. per square inch. II. The same day, the same engine drew a load of 38*52 tons in descending a plane inclined 1 in 1094, at the velocity of 29*09 ; the 38 298 pressure in the boiler being precisely the same as in the preceding trial, and the regulator open to the same degree. These experiments may be seen in pages 233 and 23-1 of our Treatise on Locomotives. If on one hand be reckoned, according to the ordinary method^ the theoretic effort applied to the piston, and on the other hand the effect really produced, viz., the resistance opposed by the load plus that of the air against the train, we find, on referring the pressure and the area of the pistons to the foot square : — » 1st case. — -Theoretic effort applied on the piston, ac- cording to the ordinary calculation 1-32X (68-71X144) 13,060 lbs. Real effect . . S,846 Coefficient of correction ... 0.68 2nd case. — Theoretic effort, the same as above . . 13,C60 Real effect 6,473 Coefficient of correction . . . 0.50 The mean coefficient, to apply to the total pressure, to convert the theoretic effects to the practical, is then *59. We find, then, three very different coefficients : choose the first case, then an error occurs in the second ; choose the second, and an error must arise in the first ; by taking the third, you will only divide the error between the two. In every way an error is inevitable, and that alone suffices to prove that every method, like the ordinary one, which consists in the use of a constant coefficient, is necessarily in- exact, whatever be the coefficient chosen, and to whatever engine the application be made ; for it is evident that tne same fact would occur in every kind of steam-engine. Only that it might be less marked, if the velocities at which the engine were taken were less different ; and this is what has hitherto prevented the error of this method from being perceived, for all the engines of the same system being imita- ted from each other, and moving nearly at the same velocity, the same coefficient of correction seems tolerably to suit them, from the factitious limit that had been laid down for the speed of the piston. Besides, in stationary engines one cannot, for want of precise de- terminations of the friction, disengage in the result the part which is really attributable ta it from that which constitutes a positive error. 299 But here we may easily be convinced that neither of these coefficients of correction represents, as the ordinary theory would have it, the friction, losses, and various resistances of the machine ; for direct experiments made on the engine under consideration, and noted in our Treatise on Locomotives, enable us to estimate separately all these frictions, losses, and resistances. Reckoning, then, the fric- tion of the engine at 82lbs., taking account besides of its additional friction per ton of load, and adding for each case the pressure sub- sisting on the opposite side of the piston by. the effect of the blast pipe, we find, as the sum of the friction and indirect resistances — 1st case. — Friction 1,257 lbs. or *10 of the theoretic result. 2nd case. — Friction 873 lbs. or *07 of the theoretic result. Thus we see, that in each of the two cases, the friction and indirect resistances, omitted in the calculation, do not in reality amount to more than 10 or 7 hundredths of the theoretic result ; and if we should be disposed to add to that -*$ or *05, for the filling of the vacant spaces of the cylinder, which we could not estimate in lbs., it will be »15 and •12; whereas the coefficients of correction would raise them to -32 in one case, and '50 in the other ; that is, to 2 and 4 times what they really are. If, then, from these coefficients, be deducted the true value of the friction and losses, it will appear that the theoretic error, intro- duced into calculation under the denomination of friction, is 17 per cent, of the total power of the engine in the one case, and 38 per cent, in the other. But it is to be remarked, that, from the preceding evaluations, viz., of the direct resistances first, and theu of the friction and indirect re- sistances, we have, for each of the two cases in question, the sum of the total effects really produced by the machine as follows : — 1st case. — Direct resistances ..... 8,846 lbs. Friction 1,257 10,103 2nd case. — Direct resistances 5,473 Friction . 873 6,346 300 We are therefore enabled now to compare these effects produced with the results either of the ordinary calculation or of our theory. 1°. In applying the ordinary calculation with the mean coefficient •56 determined above, and comparing its result with the real effect, we find — 1st case. — Effort applied on the piston, according to the ordinary calculation, 1-32 x (68-71 X 144) x *59 7,705 lbs. Effect produced, including friction and every resis- tance 10,103 Error over and above the friction and resistances 2,398 2nd case. — Effort applied on the piston, according to the ordinary calculation, the same as above . 7,705 lbs. Effect produced, including friction and every resis^ tance ....... . 7,346 Error over and above the friction and resistances . 359 Mean error of the two cases 1,37S It is then evident what error would have been committed in calcu- lating the effects of this engine from the coefficient -59 ; but it is equally evident, that in applying any other coefficient ivhatever, the error would only transfer itself from one case to the other, without ever disappearing ; and thus it is that the coefficient, -59 has almost annulled the error of the second case, by transferring it to the first. To apply our formula with reference to the same problem, viz. : — t> mS P a v we have nothing more to do than to substitute for the letters their value, taking care to refer all the measures to the same unit. In making then these substitutions, which give P = 68-71 X 144 lbs., m = 411, a = 1-32, and observing that the effective vaporization of the engine has been S = -77 cubic foot of water per minute, we find, — ■ 1st case. — Effort applied by the engine at the given velo- city, according to our 411 X 0-77 X (68-71 X 144) the01 ?' 9§3 ' " * 10 > 5071bs - Effect produced, including friction and resistances, as above 10,103 Difference 404 301 2nd case. — Effort applied by the engine at the given ve- locity, according to our . 411 X 0-77 X (68-71 X 144) theory a — |- - . . . . 7,215 lbs. Effect produced, including friction, &c 7,346 Difference 131 Mean difference of two cases 267. It appears, then, that by this method, the useful effect is found with a difference only of 267 lbs., a very inconsiderable difference in experiments of this kind, wherein so much depends on the manage- ment of the fire. 2°. To continue the same comparison of the two theories, let it be required to calculate what quantity of water per minute the boiler ought to vaporize, to produce either the first effect or the second. The method followed by the ordinary theory, again consists in pre- viously supposing that the volume described by the piston has been filled with steam at the same pressure as in the boiler, and then in applying to it a fractional coefficient to account for the losses. Now, in the first case, the volume described by the piston at the given velocity, is 1*32x298=: 393 cubic feet. Had this volume been filled with steam at the pressure of the boiler, it would have required 393 a vaporization of = '96 cubic foot of water per minute. But the real vaporization was but -77 ; wherefore, in the first case, the coef- ficient necessary to lead from the vaporization indicated by the ordi- •77 nary calculation, to the real vaporization, — = *81. In the second case, we find in the same manner, that the coeffi- cient should be -55 ; whence, in this problem, as in the preceding one, no constant coefficient whatever can suffice. Performing, however, the calculation with the mean coefficient, •68, we find, — 1st case. — Vaporization per minute, calculated by the ordinary 1-32 X 29S theory, with the coefficient, — X *68 . . . '65 Real vaporization '77 Error • • '^ 302 2nd case. — Vaporization per minute, calculated by the ordinary . . 1-32X434 theory, with the coefficient, X *6S . . . *95 Real vaporization »77 Error -18 The mean error committed is then -£- of the vaporization, and be- ing, as it is, a mean, it may, in extreme cases, become -§-, or amount to half of the whole vaporization. This is the error committed in seeking a coefficient expressly for the vaporization. Bnt when the coefficient, determined in the pre- ceding case, that is, by the comparison of the theoretical and practical effects, is used as a divisor, as by many authors it is, much greater errors are induced, which we will show by an example farther on. In our theory, on the contrary, the vaporization necessary to set in motion the resistance a R at the velocity v } is given by the formula m Jf We have then, — 1st case. — Vaporization calculated from our . 10103 X29S the °^ 411 X (68-71X144) * 74 Real vaporization .77 Difference «03 2nd case. — Vaporization calculated from our 7346 X 434 the ° r >"' 411 X (68-71X144) * 78 Real vaporization .77 Difference «01 3°. Lastly, in the case of finding the velocity of the piston, sup- posing the resistance to be given, any method similar to the ordinary one must inevitably lead to errors ; but we must dispense with com- parison, since this problem has never been resolved, and we shall therefore in this case merely show the verification of our own theory. The formula relative to this problem is mSP v = . 303 We find then,- — 1st case. — Velocity of the piston in feet per minute, calcu- iac A 411 X '77 X (68-71 X 144) ted from our theory, , , .. — — — - . . 310 •" lulu3 Real velocity * . . . . r . . 298 Difference * . . . 12 2nd case.' — Velocity of the piston from our ., 411 X '77 X (68-71X144) theol 7> 7^6 - 426 Real velocity ............ 434 Difference ............. 8 It consequently appears, that in each of the three problems in question, our theory leads to the true result ; whereas the ordinary theory, besides that it leaves the third problem unresolved, may, in the other two, lead to very serious errors. Before abandoning this comparison, we request attention to an effect, in calculating by the ordinary theory, which we have already mentioned, but which is here demonstrated, viz., that this calculation gives the same force applied by the engine in both the cases consi- dered, notwithstanding their difference of velocity : and such will always be the result, since the calculation consists merely in multi- plying the area of the piston by the pressure in the boiler, and reduc- ing the product in a constant proportion. This theory therefore maintains, in principle, that the engine can always draw the same load at all imaginable velocities. Again we see, that, in the same calculation of the load or effort applied, the vaporization of the en- gine does not appear, which would imply that the engine would always draw the same load at all velocities, whatever might be the vaporization of the boiler, which is inadmissible. We shall also remark, that in calculating by the ordinary theory the vaporization of the engine, no notice is taken of the resistance which the engine is supposed to move ; so that the vaporization ne- cessary to draw a given load would be independent of that load — another result equally impossible. To these omissions, therefore, or rather to these errors in princi- ple, are to be attributed the variations observable in the results given of the ordinary theory in the examples proposed. 304 PART II. ANALYTIC THEORY OF THE STEAM-ENGINE. ARTICLE. I. CASE . OF A GIVEN EXPANSION WITH ANY VELOCITY OR LOAD WHATEVER. § 1. Of the change of temperature of the steam during its action in the engine. — When an engine is at work, the steam is generated in the boiler at a certain pressure ; it passes from thence into the cy- linder, assuming a different pressure, and, in an expansive engine, the steam, after its separation from the boiler, continues to dilate it- self more and more in the cylinder, till the piston is at the end of the stroke. It is generally supposed, that in all the changes of pressure which the steam may undergo, its temperature remains the same ; and it is consequently concluded, that during the action of the steam in the engine, the density and volume of that steam follow the law of Mariotte, namely, that its volume varies in the inverse ratio of the presure. This supposition greatly simplifies the formulae ; but, as reason and experience prove it to be altogether inexact, we are com- pelled to renounce it, and will substitute in its place another law, de- duced from observation of the facts themselves. We have recognised in a numerous series of experiments, by ap- plying simultaneously a manometer and a thermometer, both to the boiler of a steam-engine, and also to the tube through which the steam, after having terminated its effect, escaped into the atmos- phere, that during all its action in the engine the steam remains in the state denoted by the name of saturated steam, that is, at the max- imum density for its temperature. The steam, in fact, was produced in the boiler at a very high pressure, and escaped from the engine at a very low one ; but on its issuing forth, as well as at the moment of its formation, the thermometer indicated the temperature correspond- 305 ing to the pressure marked by the manometer, as if the steam were immediately generated at the pressure it had at that moment. Thus during its whole action in the engine, the steam remains constantly at the maximum density for its temperature. Now, in all steams, the volume depends at once on the pressure and the temperature ; but in the steam at the maximum density, the tem- perature itself depends on the pressure. It should then be possible to express the volume of steam of maximum density, in terms of the pressure alone. The equation which gives the volume of the steam in any state whatever, in terms of the pressure and temperature, is very simple : it is deduced from Mariotte's law combined with that of M. Gay- Lussac. The equation which gives the temperature in terms of the pressure, for the steam at the maximum density, is also known : it has been deduced from the fine experiments of Messrs. Arago and Dulong on steam at high pressures, and from those of Southern and other experimenters on steam produced under low pressures. By eliminating then the temperature in these two equations, we shall obtain the analogy required, which will give immediately, with re- gard to steam at the maximum density, for its temperature, the vo- lume in terms of the pressure alone. But here arises the difficulty. The equation of the temperatures is not invariable ; or rather, the same equation does not apply to all points of the scale. To be used with accuracy, it requires to be changed according as the pressure is under that of one atmosphere, or comprised between one and four atmospheres, or again if it be above four atmospheres. Now, when steam is acting in an engine, it may happen, according to the load, or to other conditions of its motion, that the steam generated at first at a very high pressure, may act or be expanded in the engine sometimes at a pressure exceeding four atmospheres, sometimes at a pressure less than four atmospheres, but yet exceeding one, and sometimes at a pressure under that of one atmosphere. It is impossible then to know which of the three for- mulre is to be used in the elimination ; and consequently it is impossi- ble by this means to attain a general formula representing the effects of the engine in all cases. Moreover, were either one of these formulae adopted, the high radical quantities they contain would so complicate the calculations as to render them unfit for practical purposes. And it is to be re- marked, that these divers formula, after all. are not the expression of 39 306 the true mathematical law which connects the temperature and the pressure in saturated steam, but merely empirical relations, which ex. periment alone has demonstrated to have a greater or less degree of approximation. A formula of temperatures given by M. Biot is indeed adapted to all points of the scale, and may be useful in a great number of deli- cate researches relative to the effects of steam ; but as it gives only the pressure in terms of the temperature, and is, from its form, inca- pable of the inverse solution, namely, the general determination of temperatures in terms of the pressure, it is unfit for the elimination proposed. Under these circumstances the only resource is to seek a direct relation in terms of the pressure alone, whose results shall represent immediately those of the two preceding formulae combined ; that is^ to calculate first by means of those formulae a table of volumes of the steam, and then to seek a direct and simple relation to represent those results. This we have done. M. Navier had proposed a formula for this purpose. But that for- mula, though sufficiently exact in high pressures, differs widely from experience in pressures below that of the atmosphere, which are useful in condensing engines ; and it is possible to find one much more exact for non-condensing engines, namely, that we are about to offer. We propose then, for this purpose, the following formulae, in which p represents the pressure of the steam expressed in pounds per square foot, and /* the ratio of the volume of the steam to that occupied by the same weight of water : Formula for high or low j 10000 pressure engines with condwatioS r 0-4227 + 0-UU258, Formula for high press- ) 10000 ure non-condensing en- ] = I j ' 1-421 gines j' 1-«1 +U-002Sp The first formula is equally suitable to pressures above and below that of the atmosphere, at least within the limits likely to be consi- dered in applying it to condensing steam-engines. Those limits are eight or ten atmospheres for the highest pressures ; and eight or ten pounds per square inch for the lowest, in consequence of the friction of the engine, the pressure subsisting against the piston after imper- fect condensation in the cylinder, and the resistance of the load. 307 Within these limits then the proposed formula will be found to give very approximate results. This rirst formula might also be applied, without any error worthy of notice, to non- condensing engines. But as, in these, the steam can scarcely operate with a pressure less than two atmopheres, by rea- son of the friction of the engines and the resistance of the load, it is needless to require of the formula exact results of volumes for pres- sures under two atmospheres. In this case then the second formula will be found to give those results with much greater accuracy, and will consequently be preferred in practice. This will be readily recognised in a table annexed to the work, presenting a comparison of the volume of the steam calcu- 4ated by the ordinary formulae in terms of the pressure and tempera- ture, and by the proposed formulae in terms of the pressure alone. We state then generally this analogy : h = — ; . . . . (a) Consequently, if the steam pass in the engine, from a certain volume m' to another known volume m, and thereby abandon its primitive pressure P', to assume an unknown pressure p, it is easy to recognise that the following relation will exist between those two pressures, and will serve to determine the unknown quantity p, viz. : p tnf 1 — n p ii 1 — nm This is the relation which we substitute in lieu of that hitherto em- |>loyed, and according to which the volume appears to vary in the inverse ratio of the pressure. It will be observed that such an hypo- thesis may be deduced from the analogy we have just offered, by mV making n = 0, and q = , m being the volume, and P the pressure V of the steam in the boiler ; for it is plain that we shall then have, WlP <■ = >-• that is to say, the volumes are inversely as the pressures. § 2. Of the divers problems which present themselves in the calcu- lation of steam-engines. — We distinguish three cases in an engine : that wherein it works with a given rate of expansion of the steam, and with a load or a velocity indefinite ; that in which it works with a given rate of expansion, and with the load and velocity proper to 80S produce its maximum of useful effect with that expansion ; and lastly, that wherein, the engine having been previously regulated for the expansion of the steam most favourable in that engine, it bears, more- over, the load most advantageous for that expansion ; which, conse- quently, produces the absolute maximum of useful effect in the engine. We have said that the three fundamental problems of the calculation of steam-engines consist in finding successively the velocity, the load, and the vaporization of the engine. After the solution of these three problems, that which first presents itself, as a corollary to them, consists in determining the useful effect of the engine, which may be expressed under six different forms, viz. : by the work done, or the number of pounds raised one foot high by the engine in a minute ; by the horse power of the engine ; by the actual duty or useful effect of one pound of coal ; by the useful effect of a cubic foot of water converted into steam ; and by the number of pounds of coal, or of cubic feet of water, that are necessary to produce one horse power. Another research, in fine, no less important, is the rate of expan- sion at which the steam must work in an engine, in order that it may produce given effects. We shall present successively the solution of all these questions. The various problems will be resolved in each of the three cases above mentioned. In the two last, the question will be to calculate the rate of expansion, the velocity, the load, and the effects which cor- respond to the maximum of, relative or absolute, useful effect of the engine. In the ordinary calculations of steam-engines, the solution of three questions only had been attempted, viz., — to find the load, the vapo- rization, and the useful effect, under its different forms ; which solu- tion is, as we have seen, faulty. As to the determining of the velo- city for a given load, and that of the rate of expansion for given ef- fects, the calculation of these had not been proposed. Moreover, the very nature of the theory employed in those calculations did not allow of distinguishing, in the machine, the existence of the three cases which are really found in it. The distinction we establish may, therefore, at first appear obscure, expressed, as it is, in general terms, and including relations unusual in the consideration of steam- engines ; but, on a closer view of the question, these relations will be seen to be of indispensable necessity, in order to calculate with 309 exactitude either the effects or the proportions of steam-engines of all systems. § 3. Of the velocity of the piston under a given load. — To em- brace at once the most complete mode of action of the steam, we will suppose an engine working by expansion, by condensation, and with an indefinite pressure in the boiler; and to pass on to unexpansive or uncondensing engines, it will suffice to make the proper suppres- sions or substitutions in the general equations. From what has been already shown of our theory, the relations sought between the various data of the problem are necessarily deduc- ed from two general conditions ; the first expressing that the engine has attained a uniform motion, and consequently that the quantity of labour impressed by the motive power is equal to the quantity of ac- tion developed by the resistance : the second, that there is a neces- sary equality between the emission of steam through the cylinder and the production by the boiler. The limits of this extract will not allow us to develop those calcu- lations, simple as they may be ; but that the proceeding may be un- derstood, we shall state that, expressing by P the pressure of the steam in the boiler, and by P' the pressure of the same steam in the cylinder before the expansion, by L the length of stroke of the piston, and by L' the portion traversed at the moment the expansion begins, by a the area of the piston, and by c the clearance of the cylinder, or the space at each end of the cylinder beyond the portion traversed by the piston, and which necessarily fills with steam at each stroke ; last- ly, by r the resistance of the load, by p the pressure subsisting on the other side of the piston after imperfect condensation, by f the friction of the engine when not loaded, and by 6 the increase of that friction per unit of the load r, these four forces, as well as the pres- sures, being moreover referred to the unit of surface of the piston ; the first of the above conditions produces the following analogy : F«(L' + Q S L' L + c , : — — ■ — - < — — - — 4- loff . T - l na L > = *h'((l + i)r+p+f) (A) This equation expressing that the labour developed by the mover is found entire in the effect produced, be it remarked, that it is not es- sentially necessary for the motion to be strictly uniform. It may equally be composed of equal oscillations, beginning from no velocity, and reiurning to no velocity, provided the change of velocity take 310 place by insensible degrees, so as to avoid the loss of vis viva, and that the successive oscillations be performed in equal times. As to the second condition of the motion ; if we denote by S the volume of water vaporized by the boiler in a unit of time and trans- mitted to the cylinder, by m the volume of the steam formed under the pressure P of the boiler, compared with the volume of the same weight of water unvaporized, and by v the velocity of the piston, the equality between the production of the steam and its consumption will be found to furnish the second general analogy : S v = — a (L' + c) (B) Consequently, by eliminating P' from these two equations, and Writing, for greater simplicity, L ~~. ' naL L'-f-c L , L + c + log- rri »flL L' -f- c L' -j*c we find definitively : L S 1 v _ — ^_ t — ^^, ... (1) L -1- c a n + q * j (1 + i) r +p +f J an equation which gives the velocity of the motion in terms of the load and of the other data of the problem. This formula is quite general, and suits every kind of steam- engine with continued motion. If the engine be expansive, L' will be replaced by its value corresponding to the point of the stroke where the steam begins to be intercepted ; if the engine be unex- pansive, it will suffice to make L' = L, which will give at the same time*=l. If it be a condensing engine, p must stand for the pressure of condensation ; if it be not a condenser, p will represent the atmospheric pressure. And finally, the quantities n and q will have, according to the case considered, the above-mentioned value. § 4. 0/ the load and useful effects of the engine. — If, instead of seeking the velocity in terms of the load it be required, on the con- trary, to know the load suitable to a given velocity, the same equation resolved with reference to r becomes, L — — S — nav ar = T "Vxi • • • • ( 2 ) 3°. To find the vaporization of which the engine ought to be ca- 311 pable, in order to put in motion a resistance r with a known velocity v, the value of S must be drawn from the same analogy, thus : S = h^l av (n + q*\(l +*)r +p+fty .... (3) 4°. The useful effect produced by the machine, in the unit of time, at the velocity », is evidently arv. Hence that useful effect will have for its measure, L a T , , S n(lV I f uE.= "t C , ^~-T^ .'. . .-(4) 5°. If it be desired to know the useful effect, in horse power, of which the engine is capable at the velocity v, or when loaded with the resistance r, it suffices to observe that what is called one horse power represents an effect of 33,000 lbs. raised one foot per minute. All consists then in referring the useful effect produced by the engine in a unit of time, to the new unity just chosen, viz. to one horse power ; and it will consequently suffice to divide the expression already ob- tained in the equation (4) by 33,000. Thus, the useful effect in horse power will be, uE. uHP. = — .... (5) 33U00 v ' 6°. We have just expressed, in the two preceding questions, the effect of the engine by the work which it is capable of performing. We are now on the contrary about to express that effect by the force which the engine expends to produce a given quantity of work. The useful effect of the equation (4) being that which is due to the volume of water S converted into steam, in the unit of time, if we suppose that in the same unit of time N pounds of fuel be consumed, it is clear that the useful effeet produced by each pound of fuel will be the Nth part of the above effect. It will then be, uE. uE. 1 lb. co. = — - (6) To apply this formula, it will suffice to know the quantity of coal consumed in the furnace per minute, that is, during the production of the vaporization S ; and this datum may be deduced from a direct experiment on the engine, or from known experiments on boilers of a similar construction. 7°. The useful effect of the equation (4) being that which proceeds from the vaporization of the volume of water S, if it be required to 312 know the useful effect that will be produced by each cubic foot of water, or by each unit of S, it will be sufficient to divide the total ef- fect uE. by the number of units in S. It will then be, uE. uE. 1 ft. wa. == -— (7) 8°. In the sixth problem we have obtained the useful effect produ- ced bv one pound of fuel. We may then, by a simple proportion, de- duce from thence the quantity of fuel necessary to produce one horse power, viz. 33000 N Q. co. for 1 hp. = (8) 9". And similarly, the quantity or volume of water necessary to produce one horse power will be, 33000 S Q. wa. for 1 hp. = rr— (9) r un. § 5. Of the expansion of steam, to be adopted in an expansive en* gine, in order to produce wanted tjjects. 10c. Finally, if it be required to know what rate of expansion the engine must work at, in order to obtain from it determined effects, the value of L' must be drawn from equation (1). It will be given by the formula, ' V L'+c \-nah .... (10) JL -j- c S — n a v — — — \-i This formula not being of a direct application, we annex to the work a table which gives its solutions for the expansion from hun- dredth to hundredth, with a very short calculation. We confine ourselves to these inquiries as being those which may most commonly be wanted ; but it is clear that by means of the same general analogies, any one whatever of the other quantities •which figure in the problem may be determined, as the case may require. Thus, for instance, may be determined the area of the piston, or the pressure in the boiler, or the pressure in the condenser, correspond- ing to determined effects of the machine, as has been done for loco- motives in our work on that subject. 313 ART. II. CASE OF THE MAXIMUM USEFUL EFFECT, WITH A GIVEN RATE OF EXPANSION. § 1. Of the velocity of the maximum useful effect. We have resolv- ed the above problems in all their generality, that is, supposing the engine to move any load whatever with any velocity whatever, under this single condition, that the load and the velocity be compatible with the capability of the machine. The question is now to find what velocity and what load are most advantageous for the working of the engine, and what are the effects which, in this case, may be expected from it ; that is to say, its maxima effects for a given rate of expansion. 1°. In examining the general expression of the useful effect pro- duced by the engine at a given velocity, we perceive that the expres- sion attains its maximum for a given rate of expansion when the velocity is a minimum ; now from the equation (B) the smallest value of v will be given by P' = P. The velocity corresponding to the maximum useful effect will therefore be, S L 11 = ^TfT^P) * L'+c ' ' ' ' ( U ) Let us however remark, that, mathematically speaking, the pressure P' of the steam in the cylinder can never be quite equal to P, which is the pressure in the boiler ; because there exist between the boiler and the cylinder conduits through which the steam has to pass, and the passage of these conduits offers a certain resistance to the motion of the steam ; whence results that there must exist, on the side of the boiler, a trifling surplus of pressure equivalent to the overcoming of the obstacle. But as we have proved elsewhere, that, with the usual dimensions of engines, this difference of pressure is not appreciable by the instruments used to measure the pressure in the boiler, the introduction of it into the calculations would render the formulas more complicated without making them more exact. For this reason we neglect that difference here. The velocity given by the preceding equation is, then, that at which the engine will produce its maximum effect for a given expan- sion. This velocity will result from the condition P' = P, or reci- procally, when this velocity takes place in the engine, the steam enters 40 314 the cylinder with full pressure, that is, with the same pressure it has in the boiler. It is necessary to remark that the velocity of full pres- sure will not be the same for all engines ; on the contrary, it will vary in direct ratio with the vaporization, and in the. inverse ratio of the area of the cylinder. It may then occur to be, in one engine, the half or the double of what it would be in another: which shows that it is an error to believe that, because the piston of stationary en- gines does not in general exceed a certain velocity of from 150 to 250 English feet per minute, the steam of the boiler necessarily reaches the cylinder with no change of pressure. It is easy to be seen that a fixed limit, whatever it may be, cannot in this respect suit all engines ; and that the only means of knowing the velocity of the maximum effect, or of full pressure of an engine, is to calculate it directly for that engine. Such is the object of the formula we have just given. This formula, moreover, is of a remark- able simplicity, and requires no other experimental knowledge than that of the production of steam of which the boiler is capable. § 2. Of the load and maximum useful effect of the engine — 2°. The useful resistance which the machine is capable of putting in mo- tion at its velocity of the maximum effect above, is to be drawn from equation (2), substituting for v the value just obtained. Calling the load r we shall find it expressed by • aP P+f ar=±=- — : — a-— — • .... (12) (l-f<5)* 1-}-** V ' and it is at the same time visible that this load is the greatest the en- gine can put in motion with the given expansion L , for it corres- ponds to the lowest value of v in equation (2). Thus, the greatest effect of the machine, with a given rate of expansion, is attainable by working the machine at its smallest velocity and with its maximum load. It will be observed that this equation may be used to determine the friction of the engine without a load, and its additional friction per unit of the load, upon the same principles that we have employed in our Treatise of Locomotive Engines for similar determinations. This is also the mode we propose for steam-engines of every system. 3°. The vaporization necessary to an engine, in order to exert a Certain maximum effort r at its minimum velocity v, will be given by equation (3), by substituting in itr' and v', or will be drawn more simply from equation (11), thus : — 315 L' + c S ==(»•'+ q¥)av'.—. p— .... (13) JLi 4°. The maximum of useful effect producible in the unit of time, by an engine working with a given expansion, will be known by for- mula (4), by introducing for v the velocity proper to produce that ef- fect. Thus is found, L S cP „ > max. uE. =— . ■ —] (p+f) [ . • • ■ (14) It will be observed that this maximum useful effect depends partic- ularly on the quantity of water S, evaporated per minute in the boil- er. Hence we see plainly the error of those who pretend to calcu- late the useful effect or the power of engines from the area and the velocity of the piston, which they set in the place of the vaporization produced : this vaporization not only entering not into their calcula- tion, but forming no part of their observations. 5°. The useful effect, in horse power, of the engine will be ex- pressed by max. uE. uHP. = — - — — (15) 6°. 7°. 8°. 9°. The various measures of the useful effect will here be deduced from equations similar to those (6), (7), (8), and (9). 10°. The expansion at which the engine ought to be regulated, in order to draw a given load at the most advantageous velocity, or pro- ducing the maximum of useful effect with that load, will be derived from equation (12), which gives, LM-cc L' . . L+c ? ti + i )r+p+f l L' + c — i I TTl T l°g- TT~, — I = =F7 r na — t LcL+c lo L+c> P ' L ?L — i / • } ^ V1 ^ J i . . ; . . (20) and the solutions of this formula will be found immediately, and without calculation, by means of the table given above, as suggested by equation (10). 316 ART. III. CASE OF THE ABSOLUTE MAXIMUM OF USEFUL EFFECT. The preceding inquiries suffice for engines working without ex- pansion, merely by making L' = L ; because those engines fail un- der the case of expansion fixed a 'priori. But it is otherwise with engines in which the rate of expansion may be varied at will. We have seen that, for a given expansion, the most advantageous- way of working the engine is to give it the maximum load, which is cal- culated a priori from equation (12). Hence we know what load is to be preferred for every rate of expansion. But the question now is to determine, among the various rates of expansion of which the engine is susceptible, each accompained by its corresponding load, which will produce the greatest useful effect. For this purpose we must recur to equation (14), which gives the useful effect produced with a maximum load r' s and seek among all the values assignable to L', that which will raise the useful effect to a maximum. Now, by making the differential coefficient of that ex- pression, taken with reference to L', equal to nothing, we find as the condition of the maximum sought : L + c / L'i L p +f r ]0 ^i7+- c - na H 1 -j:) E P +- L -—T — 77 ••••( 30) Vl7+7-- noL ) This equation will be resolved in the same manner as the equa- tions (10) and (20), by means of the table already given ; and after having found the value of — , it will be introduced in the equations of Article II. ; and the corresponding velocity, load, and useful effects, will be determined. 1 However, as the supposition of n = o, q = — p, that is to say, the supposition that the steam preserves its temperature during its action in the engine, will give a sufficient approximation in a great many cases, we present here the corresponding results of all the for- 317 mulae. They will show, already to a very near degree, the maximum absolute effects which it is possible to obtain from an engine, in adopting simultaneously the most advantageous rate of expansion and the most advantageous load. (21V " — — ?L? Velocity of the absolute maximum { )V ~ a ' L(p +/) + P c useful effect. L(p+/)+Pc (22) ar =a — Load of the piston corresponding }j _ p to the absolute maximum useful low - — - effect. g L(jp+/) + Pc _ av" L(p + f) + ¥c (23) S = — . =— =- Vaporization. m LP MSP (24)a b -max.u.E = ar"«" = — : — 1 -f- S Absolute maximum of useful ef- P (L + c) feet. ° g * L (p+/) + Pc- ab - max - u - E Absolute maximum of useful force (25) u nr = g^jjj^ in horse power. L(p-f-/) Rate of expansion which produces (30) L = these e ff ects . The four determinations of the useful effects of a given quantity of fuel or water will be furnished by equations similar to those (6), (7), (8), and (9). The only remark we shall make on the subject of these formulae is, that the load suitable to the producing of the absolute maximum use-* ful effect is not the maximum load that may be imposed on the en-, gine. In effect, from equation (12), we know that the maximum, load for the engine takes place when L' = L 5 and not when Thus the greatest possible load of the engine is that of the maxi* mum useful effect without expansion ; but by applying a lighter load, that of equation (22), and at the same time the expansion of equation (30), a still greater useful effect will be obtained. 318 PART III. APPLICATION OF THE FORMUL.E TO THE VARIOUS SYSTEMS OF STEAM ENGINES. We shall not give here the applications to different systems of steam- engines, which are developed in this part of the work. We shall confine ourselves to what concerns Watt's steam-engines, because they are the most generally employed in the arts. WatVs rotative double-acting steam-engine. — These engines being without expansion, the proper formulse for calculating their effects will be deduced from the general formulse by making L = L, which will give also K = 1, and by replacing the quantity p by the pressure of condensation. We see, moreover, that for these engines, the ex- pansion being susceptible of no variation, since that detent does not exist, the third case, considered as to engines in general, cannot oc- cur. There will be then but two circumstances to consider in their working, viz., the case wherein they operate icith their maximum load, or load of greatest useful effect, and the case in which they ope- rate with any load whatever. The effects therefore of these engines will visibly be determined by the following equations : e co + + * + p 3 ft £> hi II 1 1 1 i i H 00 CO 00 OS oo C o p o p 3 p 00 H o o &5 o o CO k w o o CO 3 ° p c p p H H ffl u i-> ^d tr 1 + + + CO + + 11 CO + + + ^ + ©J CO + + + •a + Co ^ o CO 1 1! II 1 1 II 2 p H 00 5 p 00 3 t 3 00 O 00 o 00 00 p o o p o o CO p 2! P c o fcJ CO ts 3 H ft c ft t— •] O I CO + o-, + ►d 3 + "0 ^ ^ + + 1H? CO ^ Is ^s 320 Athough these formulae may at first sight appear complicated, they will nevertheless be found very simple in the calculation. It is only necessary to fix attention to refer all the measures to the same unit, as will be seen in the following example. It must be remarked also, that as soon as the velocity and load of the eugine are determin- ed, the useful effect will be known immediately, being their produce. To apply, however, these formulae, some previous observations are necessary. In good engines of that system the pressure in the condenser is usually 1-5 lb. per square inch, but the pressure in the cylinder itself, and under the piston, is in general 2-5 lbs. more, which gives ^ = 4 X144lbs. It has been deduced, moreover, from a great number of trials made on Watt's engines, that their friction, when working with a moderate load, varies from 2*5 lbs. per square inch of the piston, in engines of smaller dimensions, to 1-5 lb. in the more powerful ones ; which includes the friction of the parts of the machinery and the force necessary for the action of the feeding and discharging pumps, &c. By moderate load in these engines is meant about 8 lbs. per square inch of the piston. Now, our experiments on locomotives, showing the additional friction of an engine to be -f- of the resistance, give room to think that the additional friction caused in the engine by that load may be about 1 lb. per square inch. The above information attributes then to Watt's engines, working unloaded, a friction of from 1*5 lb. to *51b. per square inch, according to their dimensions, which would give 1 lb. for engines of a medium size : this information, agreeing with what we have deduced from our inquires on locomotives, as has been said above, we shall continue to admit, in this place, respecting the friction, the data already indicated in this respect, viz. : — f=l X 144 lbs. o*='14. As an application of these formulae, we will submit to calculation an engine constructed by Watt at the Albion JMills near London. The following were its dimensions : — Diameter of the cylinder, 34 inches, or a = 6-287 square feet ; Stroke of the piston 8 feet, or L = 8 feet; Clearance of the cylinder, ji- of the stroke, or c=-4 foot; Effective vaporization, -927 cubic foot of water per minute, or S= •927 cubic foot: Consumption of coal in the same time, 6*71 lbs. or X=6'7] lbs. ; Pressure in the boiler, 16-5 lbs. per square inch, or P=-16-5 X144 lbs.; 32 i Mean pressure of condensation, 4 lbs. per square inch, orp=4x 144 lbs. And finally, the engine being a condensing one, we have n= *4227 and q — -000000258. The engine had been constructed to work at the velocity of 256 feet per minute, which was considered its normal velocity ; but when put to trial by Watt himself, shortly after its construction, it assumed, in performing its regular work, esteemed 50 horse-power, the velocity of 286 feet per minute, consuming at the same time the quantity of water and fuel which we have just reported. If then we seek the effects it was capable of producing at its velo- city of maximum effect, and then at those of 256 and 286 feet per minute, we shall find, by the formulae already exposed i M laximum useful effect V = 286 256 v' = 214 Velocity of the piston in feet per minute ; ar = 5,621 6,850 9,133 Total load of the pis- ton in lbs* ; r 144 6-21 7-57 10*09 Load of the piston in lbs. per square inch ; S = •927 •927 *927 Vaporization in cubic feet of water per minute ; *-E = 1,607,610 1,753,600 1,957,180 Useful effect in lbs. * raised to one foot per minute. "HP = 49 53 59 Useful effect in horse power* u-E i "». co. _. 239,585 261,340 291,680 Useful effect of 1 lb* of coal, in lbs. raised to one foot per mi- nute. u-Eipe. =1,734,200 1,891,700 2,111,300 Useful effect due to the vaporization of one cubic foot of water, in lbs. raised to one foot per mi- nute. 41 322 q co. for i h. _ . 133 . 12 6 -113 Quantity of coal in lbs., producing the effect of one horse power. Qwa.forih.— . 019 -017 -016 Quantity of water, in cubic feet, produc- ing the effect of one horse power. Such are the effects that this engine should produce, and we se, in consequence, that in performing a labour estimated at fifty horses, it was to be expected the engine would acquire the velocity which in fact it did, viz., that of 256 feet per minute. Let us now see to what results we should have been led, had we applied the ordinary calculations to the experiment of Watt, which we have just reported. In this experiment, the engine vaporizing •927 cubic foot of water, and exerting the force of fifty horses, as- sumed a velocity of 286 feet per minute. We then find that, since the engine had a useful effect of no more thau fifty horses, and that the theoretical force, calculated according to that method, from the area of the cylinder, the effective pressure in the boiler, and the velocity of the piston, was, 6-2S7 X (16-5— 4) X 144 X 256 — = 9S horses. 33UUU It resulted that, to pass from the theoretical effects to the practical, it was necessary to use the coefficient -SI. Consequently, bv fol- lowing the reasonings of that theorv. the following conclusions were to be drawn : — 1-. The observed velocity being 2S6 feet per minute, the vapori- zation calculated on the quantity of water, which reduced to steam at the pressure of the boiler, might occupy the volume described by the piston, and afterwards divided, as is done, by the coefficient, to take the losses into account, would have been : -j-^Vrr X 6 257 X 256 . = 2-305 cubic ieet per minute, instead of -927. •oi 2 : . The engine having vaporized only «927 cubic foot of water per minute, the velocity calculated on the volume of steam formed, at the pressure of the boiler, and afterwards reduced by the coefficient, not as has been done, since this problem was not resolved, but as must naturallv be concluded from the signification attributed to that 323 1530 X *927 'coefficient, could but be ^-r^r X '51 = 115 per minute, m- 0*287 stead of 286. 3°. The coefficient found by the comparison of the theoretical ef- fects to the practical being -51, the various frictions, losses, and re- sistances of the engine would amount to *49 of the effective power; whereas these frictions, losses, and resistances, consisting merely of the friction of the engine and the clearance of the cylinder, could be estimated only as follows : — Total friction (including the additional friction) 2 lbs. per square inch, or as a fraction of the effectual pres. sure, * . -17 Clearance of the cylinder, ± of the effective force, or * . *05 •22 Some authors also employ constant coefficients, not however using the same to determine the vaporization as to find the useful effect. This manner of calculating has arisen from those authors having recognised from experience, that the steam has in the cylinder a less pressure and density than in the boiler; but as they cannot settle a priori what is that pressure in the cylinder, and that they always seek to deduce it from that of the boiler, instead of concluding it directly and in principle, from the resistance on the piston, as we do ; the diminution of pressure observed by them could not be defined in its limits, and it remained simply a practical fact which they used to ex- plain the coefficient. This change in the coefficient employed, avoids the first and second of the contradictions we have just indicated ; but the third, as well as all the objections we have developed in the first part against the use of any constant coefficient, remain in full force ; that is to say, that in this method, the power of the engine is calcu- lated independently of the vaporizing force of the boiler, and the va- porization independently of the resistance to be moved ; that the ef- fort exerted by the machine is found always the same at all velocities ; that no account can be taken of the opening of the regulator, unless a new series of coefficients be introduced to that end, as well as for all the changes of velocity, &c. In consequence, we conclude from this comparison, as well as from what precedes, that the theory in general use for calculating the effects or the proportions of steam-engines, cannot lead to any sure 324 results ; while the one, which we have deducted from the best known principles in mechanics, and from the direct observation of what takes place in the engines, represents their effects with accurancy. S25 HASWELL'S VALVE FOR- INJECTION, BLOW-OFF, AND DISCHARGE-PIPES OF STEAM VESSELS. In the body of the work, a valve invented by Mr. Haswell, of the United States Steam-frigate Fulton, has been referred to. A draught and description has, since that part was printed, been furnished by that gentleman. It is intended to be applied to the injection, blow- off, and delivering-pipes in steam vessels, for the purpose of cutting off at pleasure all communication between them and the water in which the vessel floats. These pipes may, in consequence, be re- moved and repaired without the necessity of going into dock ; and the vessel may be prevented from being filled with water, should they be injured by violence or burst by the frost. The annexed Fig. 1. represents a half-breadth plan, and Fig. 2. a vertical section of the valve and fixtures, as applied to an injection pipe. Note. When applied to blow-off pipes, one valve will answer for any number of boilers, by giving the top of the valve chamber a coni- cal or hemispherical form, with flanges for the connecting of branch pipes from the different boilers. S26 DESCRIPTION OF PLATE. F?V. 1. -£>' A, Opening through the bottom or side of the vessel. a, Lead pipe to shield the opening. B, Oak planking, to which the valve is first fitted and bolted, and then firmly secured to the shin of the vessel by the copper screws b. C, C, C, Valve chamber, of brass. D, Yalve, sliding in grooves z, planed in the sides of the cham- ber. d, Yalve stem, of copper. e, Stuffing box, for valve stem. f, Coupling, connecting valve stem and iron screw g, by which the valve is thrown forward to close the opening, or drawn back to admit of the water flowing through it. h, Standard and Binder, in which is placed the nut i. K, The Pipe, secured to the valve chamber by the Flange /. m, A Thumb screw, by which, when the valve closes the pipe, the water is drawn off in cold weather, to prevent its freeziug and bursting the pipe. . ///-^ Scale. '5*hm* inches to a foot. 327 Fig. 1. Fig. 2. /'/../. - — / i, . a . FZ. JJT. - 2 Fie, 1 a u j¥ fig 5 ; J!- JLs o U -u- >x PZ.Y1 Pi, , •J) \)6l<)6 ubrarVofTongress 021 213 141 6