The velocity of heat-exchange in this man ner is many times greater than in transfers across the boiler heating surfaces. It is the most rapid known form of condensation of steam, and is often 10 times as rapid as the pro duction of steam in the boiler supplying it.
The conditions of maximum efficiency are mainly two: the reduction to the practicable minimum of the thermodynamic waste by in creasing in all possible ways the area of the in dicator diagram per unit of steam supplied, and by minimizing the wastes of heat between boiler and engine-piston. The first includes the in crease of the initial pressure, with decrease of back-pressure and adjustment of the ratio of expansion of the steam to the range thus se cured; the second involves reduction of conduc tion and radiation by use of suitable non-con ducting coverings of heated surfaces, protection from cooling influences and reducing condensation' by drying and superheating the steam, by increasing the speed of engine and by diminishing the heat-exchanges between metal and steam by fine finish of surfaces, and, where practicable, by interposition of non-conducting material, as was done by Smeaton and at tempted by later inventors.
°Mechanical efficiency,' the ratio of work transmitted from the piston to the point of useful application, ranges from 95 per cent in direct-acting engines as a maximum, to 85 per cent with the older non-condensing engines. It is made a maximum and friction reduced to a minimum by careful design, and especially by securing constant, complete and free lubrication; usually, in the best cases, by a circulatory flow of oil, flooding the bearings and returned by pumps to the source, through a filter, to be again distributed to the rubbing surfaces of the engine. The lost work is the less, as the pres sures are higher within limits determined by the nature of the materials, as the lubricant is bet ter adapted to its intended purpose, and as the flow is more liberal where reaching the rub bing parts. The highest values of the coeffi cient of friction are often 10 and sometimes 20 times the lowest and the careful attention of the engineer to this detail is always well com pensated.
The ultimate limit of economy in operation, with any class of engine, is fixed by financial considerations, and the principle involved in de termining the limit may be thus expressed: That engine is most perfectly adapted to its place and purpose of which the type is such that no practicable substitution will permit the supply of the demanded power at lower total operative costs, including interest on first cost, a sinking fund to provide for replacement at the end of its period of use and annual operating expense; and that size of engine is on the whole best, variation from which in the direc tion of either increased or lessened size will in crease that total expense of operation. In the latter case, the gain by reduction of size will be more than compensated by the loss due to its reduced efficiency. The best engine is that which will give largest returns on the capital invested, adding most effectively during its life to the dividends obtainable from the plant' of which it forms a part.
The adjustment of the ratio of expansion of the steam to the requirements of maximum effi ciency is the vital problem of the designing en gineer and the purchaser of the engine. In the ideal case of the purely thermodynamic ma chine, this ratio is that of the initial to the back-pressure, very nearly, and the terminal pressure on the expansion-line should coincide with the back-pressure. For maximum economy of fuel, this expansion ratio should be reduced in proportion, closely, to the loss by heat-wastes between boiler and piston. For maximum efp ciency from the financial point of view, a still further reduction is required in proportion to the relation of the operating costs apart from those of steammaking to those of engine-opera tion proper. Thus, in the thermodynamic case, with initial and final pressures, respectively, 10 atmospheres and one, the ratio is reduced, often from about 10 to seven or eight by initial con densation and minor wastes, and to six by ad justment to that value, departure from which, in either direction, would increase total costs of the horse-power-hour.