During the past ten years some modifications in the practice of marine steam-engineering have been effected, having economy of fuel as their object, for which so large a measure of success has been claimed, that they have attracted much attention among persons to whom steam-engines are at all interesting, and have occasioned a great deal of discussion in the engineering journals.
The machines in which these improvements are exhibited are known as COMPUND ENGINES, and their great popularity, as manifested in their adoption by the most important lines of oceanic steam communication, is a conclusive proof, with many persons, of their decided superiority to the non-compound engines in common use.
There is nothing new in the compound engine, so far as its organs and their arrangement are concerned. Its essential feature, in regard to those points, is borrowed from the precedents of Hornblower and of Woolf, inventors who put forward their systems severally in 1781 and 1804.
The object of these contrivances was the use of a high measure of expansion, with less strain upon the organs of the engine than those encountered in the attempt to realize the anticipated benefits of the practice with engines of the ordinary type. For in the latter, with the same speed and stroke of piston and the same initial pressure, it is plain that, for the development of the same power, the area of the piston must be increased with every augmentation of the measure of expansion, and that at one period of the stroke of the piston a greater strain will be brought upon the engine—greater as the measure of expansion and the area of the piston are greater.
And as the strength that must be provided in any construction is that which shall be adequate to resist the greatest strain brought upon it, the same mean pressure may require greater or less strength in the material, accordingly as the extremes from which that pressure is determined may vary.
The engines of Hornblower and of Woolf were known as double-cylinder engines. Steam was admitted to the lesser of two cylinders, and, having in filling it carried the piston along its length, was exhausted into a second or larger cylinder, where its remaining expansive force was expended in impelling the larger piston.
Some double-cylinder engines have been worked in Cornwall in raising water from mines, but the common pumping-engine of the region has held its own against them.
With the pressures employed in marine service until within a few years, the principle upon which the use of the double-cylinder engine was justified is founded upon a delusion. The advantages of what is called expansion were deduced not from experiment, but from a priori reasoning, based upon the supposed correspondence in the action of steam, and that of gases in general, as enunciated in the law of Marriotte.
It has been found, however, that the advantages anticipated from the use of high measures of expansion with engines of the common type are not to be obtained in practice. Many experiments, among them a great number by the Navy Department, have been made upon the scale of ordinary practice, by which it has been proved that, for the common pressures employed in marine engines, a measure of about twice the initial volume is attended with the most satisfactory economical results.
Any increase of the number of expansions above that measure is attended with the aggravation of several losses, the ratio of which to the whole useful effect obtained is greater with the greater rates of expansion.
For since the cylinder must be larger for the development of the same power with equal initial pressures, it follows, 1st, the direct losses by radiation of heat outward to the surrounding medium will be greater; 2d, the ratio of the whole back-pressure of the condenser vapor to the whole mean pressure of the steam will be increased; 3d, the ratio of the quantity of steam required to fill the space contained in clearance and steam-passages to the whole quantity used will be greater. And besides these losses there are others, not so evident upon mere reflection, but nevertheless decisive as to the claims of the high measures of expansion.
First, there is the loss due to the expansion itself. The work accomplished in the process requires the conversion of heat; and as this must be furnished by the steam, there is condensation as the result. And this condensation must needs engender further expansion, attended by further condensation.
Again, the interior walls of the cylinder and the sides of the piston are chilled during every communication with the condenser, a large quantity of heat being abstracted from them by direct radiation into the mass of vapor in the cylinder about to be ejected towards the condenser, and by the re-evaporation of water that has been deposited upon their surfaces, and this quantity also is augmented with every increase in the size of the cylinder. The water thus re-evaporated has been condensed from steam that has performed no work whatever, and the process gives rise to the &lost important of those effects of high expansion that so greatly modify the value of the principle.
In the extensive experiments of the Navy Department, to which allusion has been made, the cost of the power developed was expressed in pounds of water evaporated from 212° Fahrenheit per total horsepower per hour. These quantities were ascertained by the actual measurement, by means of tanks, of the water evaporated, and, being compared with the quantities of steam discharged from the cylinders, as measured by the indicator, the sum of the condensations I have noted was ascertained.
The data may be relied upon to determine the performance of any engines of the kind experimented with, by applying them in correction of the results obtained from indicator measurement alone.
In any investigation of the claims made in behalf of the compound engine, it will be necessary to compare its performance with that of ordinary engines that have, in common with their rival, the latest improvements hitherto devised.
Such a comparison has been lately made by a Board of Naval Engineers, and the results embodied in a report to the Navy Department.
It is upon the materials of that report that I shall proceed to draw in the following discussion of the comparative merits of the two types of engines.
Among the various types of steam machinery employed in the naval service are three, known respectively as the 60" x 36" engines of the Guerriere class; the 50" x 42" engines of the Benicia class; and the 36" x 36" engines of the Swatara class. They may be taken as excellent examples of the most approved type of non-compound engine, as generally employed in screw-propulsion in our own and other naval services.
In common with the modern compound engine, they work surface-condensers.
Apart from the difference in the arrangement of the cylinders, the disposition of the organs of these engines is very similar to that adopted for engines of the compound type in the British Navy.
The remarkable functional difference is in the pressure of steam employed.
The engines of the three classes I have particularized are simple, durable, substantial in construction, and convenient of management. In these respects they are not inferior to any engines now on board naval vessels, whether of the compound or any other type. They are designed to work steam of a boiler-pressure of forty pounds to the square inch, and their valve gear is arranged so as to admit of the use of steam expansively at measures ranging from one and one-half to two and one-half times the initial volume. They are quite as economical in fuel as any other non-compound marine engines now in use.
The type of compound engine most generally adopted abroad is that in which two cylinders are employed, both working through cranks set at right angles to each other upon the same shaft, their axes being parallel and lying in the same plane. The cylinders are of different diameters, but of the same stroke of piston, and the larger is connected with the condenser. The smaller of the cylinders alone has any direct connection with the boilers. It is surrounded by a cylindrical shell, having the same outside diameter as the larger cylinder, and the wide annular space between the outer and the inner cylindrical shells is termed the receiver. Into this receiver the steam from the smaller or high-pressure cylinder is delivered at every stroke of the piston, in expanded volume. From the receiver the steam passes to the larger or low-pressure cylinder, and is therein worked as in common condensing engines.
Exhibiting for comparison the cost of the power, in pounds of steam per horse-power per hour, of a number of compound and non-compound two-cylinder engines; the quantities, as ascertained by indicator measurement, being corrected by adding, in the case of the non-compound engines, the known condensations in the cylinders for their several measures of expansion, as determined by the experiments of the Navy Department; and in the ease of the compound engines, the quantity condensed in the steam-jackets, as estimated upon the basis of an experiment made with the pumping-engine of the Brooklyn Water-Works in 1860.
The admission of the steam from the boiler to the high-pressure cylinder is usually governed by an adjustable expansion valve. A similar valve is sometimes employed with the low-pressure cylinder, not in order to effect expansion—for that results from the relative capacities of the cylinders—but for the purpose of properly distributing the power between them.
The high-pressure cylinder, and commonly the low-pressure cylinder also, is furnished with a jacket, surrounding it at the ends as well as at the sides. The jacket is kept filled with steam of the boiler-pressure. Heat is conducted from the jacket inward .to the cylinder and outward through the inner shell towards the receiver. The steam contained in the receiver, therefore, is superheated by the jacket. The same effect is sometimes enhanced by the use of a steam-coil within the receiver.
When the low-pressure cylinder is provided with a jacket, the latter is also kept full of steam of the boiler-pressure. But in many good examples this cylinder is not jacketed.
Figure I. roughly represents a compound engine such as I have just described.
Figure II. represents an arrangement of cylinders often resorted to, in which both pistons are fixed upon the same rod, the smaller cylinder being at the end of the larger. This is the type employed on board some of the steamers of the. White Star Line, and also by the Morgan Iron Works, of New York, for the new engines of the Tennessee. It will be observed that no receiver is provided in this type, the steam expanding directly from one cylinder to the other. When four cylinders are provided, as in the engines for the Tennessee, no special provision is needed for equalizing the distribution of the power upon the two cranks.
Examples of the economical performance of several English compound engines of the latest and most approved construction, are given in the table exhibited on the blackboard, in which the cost of the power developed by them, expressed in pounds of steam per horse-power per hour, as measured by the indicator, is collated with the cost, ascertained in like manner, and corrected for the known condensations in the cylinder, due to the production of the power, and to other causes, of the several examples of the navy engines already specified.
The former is the sum of the following quantities:
1. The number of pounds of steam per horse-power per hour discharged from the high-pressure cylinder, as measured by the indicator.
2. The number of pounds of steam per horse-power per hour condensed in the high-pressure cylinder in the production of the power, calculated upon the basis of Joule's equivalent.
3. The number of pounds of steam per horse-power per hour condensed in the steam-jackets, estimated upon the basis of an experiment made in 1860 with the pumping-engine of the Brooklyn Water-Works.
I may here remark that the expression of the cost of the horse-power in pounds of coal, in comparing the performance of engines and boilers, is often impracticable, and, in a question of engines alone, it is entirely unsatisfactory.
It is better to accept the quantity of steam as the criterion. Computed, as it is, from diagrams taken from the cylinders, under the ordinary circumstances of practice, it is independent of all conditions that depend upon the boilers alone. In computing the quantity of steam consumed by the compound engines under discussion from that discharged from the high-pressure cylinder, as per indicator, it is assumed that the condensations in that cylinder, due to causes other than the production of the power, are covered by the condensations in the steam-jackets.
Whatever condensations occur, due to the difference of temperature between the interior surfaces of that cylinder and the entering steam, are counter-veiled by the re-evaporation of the water thereby precipitated upon those surfaces, or suspended in the: steam; and the water thus re-evaporated becomes available as steam for the development of power in the low-pressure cylinder.
The mean cost of the horse-power developed in the non-compound engines will compare with the mean cost of the horse-power in the compound engines as follows:
1. Cost of the total horse-power of the non-compound engines, 28.46 pounds of steam; of the total horse-power of the compound engines, 19.05 pounds; the difference in favor of the latter is
(28.36-19.05x100)/(28.46) = 33.06
per centum of the former.
2. Cost of the indicated horse-power of the non-compound engines, 81.75 pounds of steam; of the indicated horse-power of the compound engines, 22.46 pounds; the difference in favor of the latter is
(31.75-22.46x100)/(31.75) = 29.26
per centum of the former.
3. Cost of the net horse-power of the non-compound engines, 40.83
pounds of steam; of the net horse-power of the compound engines, 27.18 pounds; the difference in favor of the latter is
(40.83-27.18x100)/(40.83) = 33.43
per centum of the former.
The total power is that computed from the mean gross pressure upon the piston.
The indicated power is computed from the mean unbalanced pressure.
The net power is computed from the gross mean pressure, minus that required to overcome the resistance of the back pressure and the friction of the engine.
The indicated horse-power is the standard of comparison commonly employed in current discussions of the performance of compound engines, although the net power affords the more rational standard. The gain of 29.26 per centum in the cost of the indicated power is much less than that usually claimed for the compound engines by persons interested in their manufacture.
If, as is often asserted, the indicated horse-power is obtained at a cost of only two pounds of coal per hour, the boilers must evaporate 11.23 pounds of water per pound of coal. This quantity is much greater than has ever been evaporated by boilers of the type employed with the compound engines under consideration. The quantity evaporated in such boilers per pound of coal, at the high rates of combustion generally employed in English practice, will be found not to exceed eight pounds of water from a temperature of 100° Fahrenheit. When the apparent evaporation is greater, the increase may be due to superheating the steam, the results of which practice would be equally advantageous in the case of engines of either type. The cost of the indicated horsepower, then, in pounds of coal per hour, would be
22.46/8 = 2.81
Taking the evaporation of the boilers used with the non-compound engines at their maximum rate of combustion to be nine pounds, the coat of the indicated horse-power in pounds of coal will be
31.75/9 = 3.53
This quantity corresponds very nearly with the results recorded in the Neam-logs of the engines in question when burning anthracite of the best quality.
The difference in favor of the compound engine is therefore
(3.53–281x100)/(3.53) = 20.39
per centum of the cost of the indicated horse-power of the non-compound engines in pounds of coal per hour.
The boiler-pressure employed with the compound engines in question is sixty pounds per square inch above that of the atmosphere.
The employment of so high a pressure has occasioned the adoption of a type of boiler, cylindrical in form, constructed of thicker plates -than have been commonly used for marine purposes. This type is thought to promise some advantages over that hitherto preferred for the naval service. The latter is for the most part of the vertical water-tube variety, quadrangular in form, and unrivalled in economy of fuel. It has been found lacking in durability since the use of surface-condensers has become general.
It is thought that from the greater facility with which means for the prevention of corrosion, both internal and external, may be applied, due to the greater simplicity of their construction in the reduction of the bracing and otherwise, the cylindrical boilers may be made to render service for a longer period than those they will replace.
For equal areas of grate surface, however, the space occupied upon the floors of the vessels by cylindrical boilers of diameters practicable in naval vessels of the lesser rates, exceeds that occupied by boilers of the quadrangular form, by about thirty per centum of the latter. Reducing the grate surface in proportion to the expected gain in net power developed, the space required for the cylindrical will be nearly the same as for the quadrangular form for the development of the same power with equal rates of combustion.
A great deal has been written in the current engineering journals in discussion of the causes to which the gains accomplished are ascribed. The explanation most generally accepted is that which refers the improved results to the greater facility with which steam of high pressure can be employed at high measures of expansion. But this does not suffice. The gain in economy does not increase in the ratio of the augmentation of the measure of expansion, or nearly so even.
There is really no correspondence between the gains that should result according to the law of the expansion of gases, and the gains accomplished in the use of compound engines.
The practical measure of expansion that has been found most economical in cylinders working steam of low pressure into a condenser, is not greater than two times the initial volume.
The low-pressure cylinder of the compound engine works under precisely the same conditions as the cylinders of ordinary condensing engines. The initial volume of the steam received by it per stroke of piston is governed by the capacity of the high-pressure cylinder. If, therefore, a measure of two times the initial volume be employed in the low-pressure cylinder, that should have twice the capacity of the high-pressure cylinder.
The latter works under nearly the same conditions as the cylinders of ordinary non-condensing engines. The practical measure of expansion that has been found most economical in working steam of sixty pounds pressure in such cylinders is not more than four times the initial volume.
Now, the compound engine is essentially an arrangement by which two engines—a non-condensing and a condensing engine—are conjoined, and the best results will be obtained from it when the steam is worked in its several cylinders with that measure of expansion which would be appropriate to either, were it detached and worked by itself.
The several measures of expansion in the two cylinders are, then, the factors which make up the total measure effected, and this practical measure is (4 x 2) = 8 times the initial volume. To increase either of these factors by any considerable amount, for the sake of effecting a higher total rate for the development of the same power, would result in a direct increase of the cost of the power.
With a given capacity of high-pressure cylinder, therefore, it would seem useless to employ a low-pressure cylinder of a relative capacity greater than two or two and one-half times the former.
The best examples of English compound engines have cylinders whose relative capacity is as one to three, and there are many in which the proportion is as one to four. That the low-pressure cylinders of the compound engines whose performance is given in the table, and the measures of expansion employed in them, are too large for economy, will appear if we compare the cost of the total horse-power developed in them alone, with the cost of the total power developed in the cylinders of the navy engines.
The mean quantity representing that cost is, for the compound engines, inclusive of the quantity condensed in the steam-jackets, the benefit of which is obtained in the low-pressure cylinder, 31.02 pounds of steam. The difference is
(31.02 – 28.46 x 100)/(28.46) = 9
per centum of the latter in favor of the non-compound cylinders. This comparison is made with those compound engines only in which the powers developed in the several cylinders are nearest equality.
Had the low-pressure cylinders been smaller, the ratio of the back to the effective pressure in them would have been reduced—and reduced in a greater proportion than the ratio of the back-pressure in the high-pressure cylinder to its effective pressure would increase—and a direct gain accomplished by a decided reduction of the total measure of expansion.
Notwithstanding the defective proportions of the cylinders and the apparent inferiority of the boilers employed in the steamers whose engines we have taken as examples of the new system, it appears that there is a considerable gain in economy to be expected from the adoption of the compound engine for the naval service. The gain in power for the same quantity of coal consumed, expressed in per centums of the power obtained by the non-compound engines, is 25.6.
Whether they will fulfill the requirements of the service in other respects as well as is done by the engines now in use, is a question that can be determined only by experience. They are more complex in arrangement, and the details include a greater number of parts; at least one of the cylinders must be fitted with a separate cut-off valve and gear.
There is a greater number of joints, and hence, as well as from the higher pressures employed, a greater liability to leakage.
The action of the cylinders depends each upon that of the other, and neither can be made to act by itself without the employment of special appliances of a complicated and normally useless character, in the absence of which the disabling of one cylinder means the disabling of both.
Against these apparently not insuperable disadvantages we have to set the promised gain of 25.6 per centum in the indicated horse-power obtained from the same weight of fuel consumed, and the probable greater longevity of the boiler.
In adapting the system to the circumstances of American practice, the English precedents cannot be exactly followed. Designs of machinery for vessels of war are made under conditions that differ greatly from those that determine its arrangement and disposition in vessels of commerce. And the difference in the character of the fuel commonly used in this country from that used in Great Britain, enforces a variation in practice, so far as the boilers are concerned, entailing a necessity for the occupation of more space in the vessels for that portion of the motive power.
It is only by experience, as has been said already, that the actual value of the improved machinery, and its merits for adoption in the naval service, to the exclusion of other types, can be determined. In the pursuit of economy the conspicuous modification of engineering practice is in the pressures of steam employed. There is nothing new in the compound engine except this feature. The mere combination of mechanical devices that distinguish it from engines of the common type has never availed for the reduction of the cost of steam-power, although the attempt to compass that object by their use has been made again and again.
The new system can only be judged after competition with such modifications of that now in vogue as may result from the employment of the steam-jacket and of the higher pressures and speeds of piston to which the superiority of the former in economy is chiefly due.
Let the compound engine compete with the older type, so modified as to increase the speed of the piston to that employed in the former; let the cylinder be reduced in diameter to suit the increase of the steam-pressure; let the practical point of cutting off—in other words, the practical measure of expansion—be ascertained for the higher pressures, just as it has been determined for the pressure of forty pounds; let the steam-jacket be applied, as it may be with great increase of efficiency, to the reduced cylinders, and we shall have a formidable competitor with the compound engine.
A very considerable gain in economy of fuel is certain to result from such modifications, and it is also certain that they will produce engines convenient of management, simple in construction, not specially liable to derangement, and capable of operating singly by simple disengagement, in case of injury to either.
If justice has been done the compound engine in this discussion of its merits, it will be plain to all who have noted the claims put forward in its behalf, that some of its advocates have permitted themselves greatly to exaggerate its performances.
And that no injustice has been done in this discussion is clear enough when we reflect that all the possible errors in the method of investigation panned are in favor of the compound engine. If there is error, it may be that the cost of the power is greater than has been stated; it cannot possibly be less.
It has been often urged that misrepresentation of such facts as have to do with the performance of machinery is without sufficient motive, and that what so many unite in declaring must needs be true.
But a very close observer of human conduct has felt himself moved to say that "where personal interests come into play, there must be, even in men intending to be truthful, a great readiness to see the facts which it is convenient to see, and such reluctance to see opposite facts as will prevent much activity in seeking for them."
Discussion After Chief-Engineer Baker's Paper.
The reading of the paper concluded, upon motion of Commodore Foxhall A. Parker the thanks of the Institute were voted to Chief-Engineer Baker.
Several diagrams, representing some new forms of engines and boilers, were examined and discussed.
Commodore Ammen remarked that he had witnessed the application of the principles of the compound engine on the Ohio River twenty years ago. He stated that there was said to be considerable economy as the result.
Mr. Kafer said: "Compound engines were used on the lakes many years ago, but soon abandoned on account of the increased consumption of fuel, chiefly due to the low pressure of steam used. If there is any economy in the compound engine of the present day, it is mainly due to the use of high-pressure steam when compared to the single-cylinder or non-compound engine using steam of a lower initial pressure; but comparing them with reference to economy, using the same initial pressure and the same rate of expansion, the advantage would seem to be in favor of the non-compound. If the same quantity of steam be used during each stroke from the boiler in the high-pressure cylinder of the compound as in the non-compound engine, the work performed by the steam before the cut-off valve on either cylinder closes will be the same, the work-performed during expansion in the non-compound will be greater than in the compound, as in the former the steam expands from the initial pressure to the terminal doing work, while in the latter the steam expands from the initial to the terminal pressure in the high-pressure, and then escaping to the receiver or steam-ports leading to the low-pressure cylinder, decreasing the pressure without doing work; and when the piston starts in the low-pressure cylinders it is with a much less initial pressure than the terminal pressure of the high-pressure cylinder, while the rate of expansion and the terminal pressure before the steam escapes to the condenser is the same in the compound as in the non-compound; the loss in the compound engine being due to the free expansion without doing work.
"The clearance space in screw-engines is approximately ten percent of the stroke displacement. This clearance space at the beginning of the stroke is filled with steam, compressed by the piston after the exhaust-valve is closed, of a pressure less than the boiler-pressure. The nearer the pressure in the clearance space at the end of the stroke approaches the initial or boiler pressure, the less steam is required from the boiler to fill this space ; with a single-cylinder engine it is practically impossible to compress the steam from the pressure in the cylinder at the time of closing the exhaust-valve (about four pounds per square inch, absolute pressure) to the initial pressure commonly used in compound engines — say seventy-five pounds per square inch, absolute pressure; while in the high-pressure cylinder of the compound engine, with the back pressure fifteen pounds per square inch, absolute, at the moment of closing the exhaust-valve, the steam remaining in the cylinder can be compressed to seventy-five pounds per square inch at the end of the stroke as readily as to twenty pounds in the non-compound engine, which is about the pressure at the end of the stroke in a well-designed non-compound engine.
"This gain in filling the clearance spaces by compression does not lessen the power of the engine materially, and is beneficial in its working, as it brings the piston and rods to rest at the end of the stroke without straining action on the crank-pin.
“There are many other points to be considered, which apparently are against the compound engine, such as increased external radiating surface, complexity of mechanism, in addition to the loss by free expansion; but the gain by compression, and possibly the smaller difference in the temperature of the steam, and consequently of the walls of the cylinder that are in direct communication with the condenser, seems to more than offset all its disadvantages, if the statements of compound-engine builders are to be relied on.
"What is needed is a set of reliable experiments with a non-compound and a compound engine, using the same pressure of steam and same piston-speed; but the non-compound engine, to compete favorably, must have a minimum of clearance space, and be well jacketed."
The Institute then proceeded to executive session.