TABLE I.--TENSILE STRENGTH OF GUN METAL AT HIGH TEMPERATURES._Scientific American Supplement, No. 821, September 26, 1891

--------------+------------+-------------+-------------+------------+

| | | | |

--------------+------------+-------------+-------------+------------+

Copper 87 /| 50° | 12.34 | 8.38 | 14.64 |

Tin 8 / | | | | |

Lead 1½ \| 400° | 10.83 | 6.30 | 11.79 |

--------------+------------+-------------+-------------+------------+

Copper 87 /| 50° | 13.86 | 8.33 | 20.30 |

Tin 8 { | | | | |

Zinc 5 \| 458° | 10.70 | 7.43 | 12.42 |

--------------+------------+-------------+-------------+------------+

Cast steel has been freely used by the writer for bends, junction

pieces, etc., of steam pipes, as well as for steam valve chests; and

except for the fact that steel makers' promises of delivery are

generally better than their performance, the result has thus far been

satisfactory in all respects. These were adopted because there existed

some doubt as to the strength of gun metal under a high temperature; and

as the data respecting its strength appeared of a doubtful character, a

series of careful tests were made to determine the tensile strength of

gun metal when at atmospheric and higher temperatures. The test bars

were all 0.75 in diameter, or 0.4417 square inch sectional area; and

those tested at the higher temperatures were broken while immersed in a

bath of oil at the temperature here stated, each line being the mean of

four experiments. The result of these experiments was to give somewhat

greater faith in gun metal as a material to be used under a higher

temperature; but as steel is much stronger, it is probably the most

advisable material to use, when the time necessary to procure it can be

allowed.

_Feed Heating_.--With the double object of obviating strain on the

boiler through the introduction of the feed water at a low temperature,

and also of securing a greater economy of fuel, the principle of

previously heating the feed water by auxiliary means has received

considerable attention, and the ingenious method introduced by Mr. James

Weir has been widely adopted. It is founded on the fact that, if the

feed water as it is drawn from the hot well be raised in temperature by

the heat of a portion of steam introduced into it from one of the steam

receivers, the decrease of the coal necessary to generate steam from the

water of the higher temperature bears a greater ratio to the coal

required without feed heating than the power which would be developed in

the cylinder by that portion of steam would bear to the whole power

developed when passing all the steam through all the cylinders. The

temperature of the feed is of course limited by the temperature of the

steam in the receiver from which the supply for heating is drawn.

Supposing, for example, a triple expansion engine were working under the

following conditions without feed heating: Boiler pressure, 150

lb.;--indicated horse power in high pressure cylinder 398, in

intermediate and low pressure cylinders together 790, total, 1,188; and

temperature of hot well 100° Fahr. Then with feed heating the same

engine might work as follows: The feed might be heated to 220° Fahr.,

and the percentage of steam from the first receiver required to heat it

would be 12.2 per cent.; the indicated horse power in the high pressure

cylinder would be as before 398, and in the intermediate and low

pressure cylinders it would be 12.2 per cent, less than before, or 694,

and the total would be 1,092, or 92 per cent. of the power developed

without feed heating. Meanwhile the heat to be added to each pound of

the feed water at 220° Fahr. for converting it into steam would be 1,005

units against 1,125 units with feed at 100° Fahr., equivalent to an

expenditure of only 89.4 per cent. of the heat required without feed

heating. Hence the expenditure of heat in relation to power would be

89.4 + 92.0 = 97.2 per cent., equivalent to a heat economy of 2.8 per

cent. If the steam for heating can be taken from the low pressure

receiver, the economy is about doubled. Other feed heaters, more or less

upon the same principle, have been introduced. Also others which heat

the feed in a series of pipes within the boiler, so that it is

introduced into the water in the boiler practically at boiling

temperature; this is economical, however, only in the sense that wear

and tear of the boiler is saved; in principle the plan does not involve

economy of fuel.

_Auxiliary Supply of Fresh Water_.--Intimately associated with the feed

is the means adopted for making up the losses of fresh water due to

leakage of steam from safety valves, glands, joints, etc., and of water

discharged from the air pumps. A few years ago this loss was regularly

made up from the sea, with the result that the water in the boilers was

gradually increased in density; whence followed deposit on the internal

surfaces, and consequent loss of efficiency, and danger of accident

through overheating the plates. With the higher pressures now adopted,

the danger arising from overheating is much more serious, and the

necessity is absolute of maintaining the heating surfaces free from

deposit. This can be done only by filling the boiler with fresh water in

the first instance, and maintaining it in that condition. To do this two

methods are adopted, either separately or in conjunction. Either a

reserve supply of fresh water is carried in tanks or the supplementary

feed is distilled from sea water by special apparatus provided for the

purpose. In the construction of the distilling or evaporating apparatus

advantage has been taken of two important physical facts, namely, that,

if water be heated to a temperature higher than that corresponding with

the pressure on its surface, evaporation will take place; and that the

passage of heat from steam at one side of a plate to water at the other

is very rapid. In practice the distillation is effected by passing

steam, say from the first receiver, through a nest of tubes inside a

still or evaporator, of which the steam space is connected either with

the second receiver or with the condenser. The temperature of the steam

inside the tubes being higher than that of the steam either in the

second receiver or in the condenser, the result is that the water inside

the still is evaporated, and passes with the rest of the steam into the

condenser, where it is condensed, and serves to make up the loss. This

plan localizes the trouble of deposit, and frees it from its dangerous

character, because an evaporator cannot become overheated like a boiler,

even though it be neglected until it salts up solid; and if the same

precautions are taken in working the evaporator which used to be adopted

with low pressure boilers when they were fed with salt water, no serious

trouble should result. When the tubes do become incrusted with deposit,

they can be either withdrawn or exposed, as the apparatus is generally

so arranged; and they can then be cleaned.

_Screw Propeller_.--In Mr. Marshall's paper of 1881 it was said that

"the screw propeller is still to a great extent an unsolved problem."

This was at the time a fairly true remark. It was true the problem had

been made the subject of general theoretical investigation by various

eminent mathematicians, notably by Professor Rankine and Mr. William

Froude, and of special experimental investigation by various engineers.

As examples of the latter may be mentioned the extended series of

investigations in the French vessel Pelican, and the series made by Mr.

Isherwood on a steam launch about 1874. These experiments, however, such

as they were, did little to bring out general facts and to reduce the

subject to a practical analysis. Since the date of Mr. Marshall's paper,

the literature on this subject has grown rapidly, and, has been almost

entirely of a practical character. The screw has been made the subject

of most careful experiments. One of the earliest extensive series of

experiments was made under the writer's direction in 1881, with a large

number of models, the primary object being to determine what value there

was in a few of the various twists which inventive ingenuity can give to

a screw blade. The results led the experimenters to the conclusion that

in free water such twists and curves are valueless as serving to augment

efficiency. The experiments were then carried further with a view to

determine quantitative moduli for the resistance of screws with

different ratios of pitch to diameter, or "pitch ratios," and afterward

with different ratios of surface to the area of the circle described by

the tips of the blades, or "surface ratios." As these results have to

some extent been analyzed and published, no further reference need be

made to them now.

In 1886, Mr. R.E. Froude published in the Transactions of the

Institution of Naval Architects the deductions drawn from an extensive

series of trials made with four models of similar form and equal

diameter, but having different pitch ratios. Mr. S.W. Barnaby has

published some of the results of experiments made under the direction of

Mr. J.I. Thornycroft; and in his paper read before the Institution of

Civil Engineers in 1890 he has also put Mr. R.E. Froude's results into a

shape more suitable for comparison with practice. Nor ought Mr. G.A.

Calvert's carefully planned experiments to pass unnoticed, of which an

account was given in the Transactions of the Institution of Naval

Architects in 1887. These experiments were made on rectangular bodies

with sections of propeller blade form, moved through the water at

various velocities in straight lines, in directions oblique to their

plane faces; and from their results an estimate was formed of the

resistance of a screw.

One of the most important results deduced from experiments on model

screws is that they appear to have practically equal efficiencies

throughout a wide range both in pitch ratio and in surface ratio; so

that great latitude is left to the designer in regard to the form of the

propeller. Another important feature is that, although these experiments

are not a direct guide to the selection of the most efficient propeller

for a particular ship, they supply the means of analyzing the

performances of screws fitted to vessels, and of thus indirectly

determining what are likely to be the best dimensions of screw for a

vessel of a class whose results are known. Thus a great advance has been

made on the old method of trial upon the ship itself, which was the

origin of almost every conceivable erroneous view respecting the screw

propeller. The fact was lost sight of that any modification in form,

dimensions, or proportions referred only to that particular combination

of ship and propeller, or to one similar thereto; so something like

chaos was the result. This, however, need not be the case much longer.

In regard to the materials used for propellers, steel has been largely

adopted for both solid and loose-bladed screws; but unless protected in

some way, the tips of the blades are apt to corrode rapidly and become

unserviceable. One of the stronger kinds of bronze is often judiciously

employed for the blades, in conjunction with a steel boss. Where the

first extra expense can be afforded, bronze seems the preferable

material; the castings are of a reliable character, and the metal does

not rapidly corrode; the bronze blades can therefore with safety be made

lighter than steel blades, which favors their springing and

accommodating themselves more readily to the various speeds of the

different parts of the wake. This might be expected to result in some

slight increase of efficiency; of which, however, the writer has never

had the opportunity of satisfactorily determining the exact extent.

Instances can be brought forward where bronze blades have been

substituted for steel or iron with markedly improved results; but in

cases of this kind which the writer has had the opportunity of

analyzing, the whole improvement might be accounted for by the modified

proportions of the screw when in working condition. In other words, both

experiment and practical working alike go to show that, although cast

iron and steel blades as usually proportioned are sufficiently stiff to

retain their form while at work, bronze blades, being made much lighter,

are not; and the result is that the measured or set pitch is less than

that which the blades assume while at work. Some facts relative to this

subject have already been given in a recent paper by the author.

_Twin Screws_.--The great question of twin screw propulsion has been put

to the test upon a large scale in the mercantile marine, or rather in

what would usually be termed the passenger service. While engineers,

however, are prepared to admit its advantages so far as greater security

from total breakdown is concerned, there is by no means thorough

agreement as to whether single or twin screws have the greater

propulsive efficiency. What is required to form a sound judgment upon

the whole question is a series of examples of twin and single screw

vessels, each of which is known to be fitted with the most suitable

propeller for the type of vessel and speed; and until this information

is available, little can be said upon the subject with any certainty. So

far the following large passenger steamers, particulars of which are

given in table II., have been fitted with twin screws. It appears t be a

current opinion that the twin screw arrangement necessitates a greater

weight of machinery. This is not necessarily so, however; on the

contrary, the opportunity is offered for reducing the weight of all that

part of the machinery of which the weight relatively to power is

inversely proportional to the revolutions for a given power. This can be

reduced in the proportion of 1 to the square root of 2, that is 71 per

cent. of its weight in the single screw engine; for since approximately

the same total disk area is required in both cases with similar

proportioned propellers, the twins will work at a greater speed of

revolution than the single screw. From a commercial point of view there

ought to be little disagreement as to the advantage of twin screws, so

long as the loss of space incurred by the necessity for double tunnels

is not important; and for the larger passenger vessels now built for

ocean service the disadvantage should not be great. Besides their

superiority in the matter of immunity from total breakdown, and in

greatly diminished weight of machinery, they also offer the opportunity

of reducing to some extent the cost of machinery. A slightly greater

engine room staff is necessary; but this seems of little importance

compared with the foregoing advantages.