Warships - A Text Book - Atwood (1917)

WAR-SHIPS
A TEXT-BOOK
ON
THE CONSTRUCTION, PROTECTION, STABILITY,

TURNING, ETC., OF WAR VESSELS
BY
EDWARD L. ATTWOOD, M.lNsr.N.A.

MEMBER OK THE ROYAL CORPS OF NAVAL CONSTRUCTORS;
FORMERLY LKCTUKER IN NAVAL ARCHITECTURE AT THE ROYAL NAVAL COLLEGE, GREENWICH
AUTHOR OF "A TEXT-BOOK OF THEORETICAL NAVAL ARCHITECTURE,"
"THE MODERN WAR-SHIP'';
JOINT AUTHOR OF "LAYING OFF," A TEXT-BOOK ON THE GEOMETRY OF SHIPBUILDING.
WITH NUMEROUS DIAGRAMS
b clitic n
LONGMANS, GREEN AND CO

39 PATERNOSTER ROW, LONDON
FOURTH AVENUE AND SUTH STREET, NEW YORK
BUM BAY, CALCUTTA, AND MADRAS
1917
All rights reserved
.0
BIBLIOGRAPHICAL NOTE
First Printed J ll
ty) 1904
Second Edition November^ 1905
27iird Edition February, iqoS
Fourth Edition February, 1910
Translated into Spanish by D. Juan de
Goytiay Gordia of the Spanish Navy in .

JQII
Fifth Edition J ll
ty* Z 9 12
Sixth Edition January,
Approved for issue to the Officers' Libraries of II. M. ships

PREFACE
THE present work has been prepared in response to suggestions
made by a number of the senior naval officers taking the course
in Naval Architecture at the Boyal Naval College, Greenwich.
An attempt has been made to treat the subject from the naval
officers' standpoint, and certain parts of the subject have been

treated at some length with a view of meeting their special
requirements. I am indebted to several officers for valuable
suggestions as to the subject-matter likely to be of most use.

The blank pages at the end have been inserted in order to provide
space to note particulars and details peculiar to the ship on

which an officer is serving, and to note changes of practice which
may occur. The text is illustrated by a large number of
carefully drawn diagrams. A number of questions have been

prepared as an Appendix. These questions in many cases are
designed to lead to inquiry and discussion, and cannot be directly
answered from the text.
I am indebted to the Controller of the Navy, Rear- Admiral

W. II. May, M.V.O., for his permission to undertake the work,
and to the Director of Naval Construction, Philip Watts, Esq.,
F.E.S., LL.D., for his kindly interest in the undertaking. The

book is, however, not an official publication.
Although prepared primarily for naval officers, yet it is
believed that the work will prove a useful introduction to the
subject of Naval Architecture for apprentices and students at
the Iloyal Dockyards and elsewhere.

E. L. A.
ROYAL NAVAL COLLEGE,

GREENWICH,
1904.
370137
PREFACE TO SIXTH EDITION
IN consequence of the rapid change of practice in recent years
it has been found necessary to embody a considerable number
of modifications in the present edition. The important 1906
Admiralty memorandum concerning the design of H.M.S.
Dreadnought has been retained as an Appendix.
E. L. A.
LONDON,
1916.
CONTENTS
CTI AFTER PAT.E
T. THE STRENGTH OF SHIPS 1
IT. TESTS OF STEEL, ETC., SECTIONS, RIVETS, JOINTS, ETC. . . 10
ITT. FRAMING OF VARIOUS TYPES OF SHIPS 21
TV. BEAMS, PILLARS, AND DECKS 42

V. PLATING OF THE OUTER AND INNER BOTTOMS 54
VT. WATERTIGHT BULKHKALS, POORS, ETC 65
VIT. STEMS, STERNPOSTS, RUDDERS, AND SHAFT BRACKETS ... 80

VI IT. STEERING GEARS 98
IX. PUMPING, FLOODING, AND DRAINAGE 109
X. VENTILATION 126
XI. CORROSION AND FOULING . . . 138
XII. ARMOUR AND DECK PROTECTION 148
XIII. RULES OF MENSURATION FOR THE CALCULATION OF AREAS
AND VOLUMES 185
XIV. NAVY LIST DISPLACEMENT, TONNAGE, ETC 192
XV. BUOYANCY, DISPLACEMENT, TONS PER INCH, ETC 196
XVI. INITIAL STABILITY, METACENTRIC HEIGHT, ETC 206
XVII. TRIM, MOMENT TO CHANGE TRIM ONE INCH, ETC 230
XVIII. STABILITY AT LARGE ANGLES OF INCLINATION 242
XIX. THE ROLLING OF SHIPS 252
XX. THE TURNING OF SHIPS 265
XXI. THE RESISTANCE AND PROPULSION OF SHIPS 278

XXII. THE DESIGN OF WAR-SHIPS 298
XXIII NOTES ON THE Loss OF H.M.S. VICTORIA 309
APPENDIX A. QUESTIONS 315
APPENDIX B. ADMIRALTY MEMORANDUM ON H.M.S. DREAD-

NOUGHT 332
INDEX .
335
WAR-SHIPS
CHAPTER I.
THE STRENGTH OF SHIPS
BEFORE considering in some detail the structural arrangements of

various classes of ships in the Royal Navy, it will be of use to
consider briefly the general nature of the strains to which a ship's
structure is likely to be subject.
These strains may be divided broadly into two classes, viz.
1. Structural strains, coming on a ship regarded as a complete
structure, and
2. Local strains, which only affect a particular portion of
a ship.
A ship may have
great structural strength with small local
strength ;example, a small torpedo-boat is strong enough to
for
stand being lifted bodily out of the water on to a ship's deck, while
the plating is very thin and easily damaged. A
full-sized ship
could not be thus lifted (supposing it to be feasible) without the
probability of serious damage to the structure.
I. Structural Strains. The most important of the structural
strains to which ships are subject are those strains tending to bend
them in a fore-and-aft direction. The ship may be regarded as a
huge beam or girder. In order to bring out the various points

in connection with this fore-and-aft bending and the proper
arrangement of the material to make the ship strong enough, we

shall first consider some properties of beams.
Beams. If we 12 in. wide and 2 in,
take a plank, say
thick, and place it on supports 10 ft. apart, a ton weight placed
at the middle would be as much as the plank could bear without
1 B
70
2 WAR-SHIPS.
If, now, we placed six such planks as in Fig. I,

the
breaking.
planks could stand 6 tons. If instead of simply resting on one
another the planks are tightly clamped or bolted together, we
should find that the beam thus made would stand far more than
the six unconnected planks. If we carried the process a step
further, we should havea log 12 in. wide and 12 in. deep of
homogeneous material, and we should find that such a beam

would stand 36 tons before it would be on the point of breaking,
or six times the weight that could be carried by the six unconnected
planks.
We notice that the upper layers of such a beam are being
FIG. 1.
compressed and the lower layers are being stretched, and it is

the resistance offered by the fibres of the beam, to this compression
and this stretching, which enables the beam to withstand the
bending. The middle layer of the beam is unaltered in length, and
none of the upper layers are compressed to the extent that the top
layer is, and none of the lower layers are stretched to the extent
that the bottom layer is. So that if the timber of the beam is just
strong enough to stand the tension at the bottom and the com-
pression at the top without rupture, none of the layers between the
top and the bottom are being used to their full strength. Thus the
portions of the beam furthest away from the centre of the section
contribute most to the strength of the beam.
The following will illustrate the same principle. Suppose it is
desired to make an iron beam 12 ft. long, having a sectional area
of 16 square inches. How should this area be disposed to admit
THE STRENGTH OF SHITS. 3
of the beam standing thegreatest weight at the centre, without

exceeding a stress of 10 tons per square inch on the material ? If
the beam is, as (a) Fig. 2, 8 in. wide 2 in. deep, it will stand
a weight of about 1'5 tons with a section, as
; (6), 4 in. wide
4 in. deep, it will stand about 3 tons; as (c), 2 in. wide 8 in.
deep, about 6 tons, and as (d), 1 in. thick with flanges at top
and bottom 5 in. wide, 10 tons. In the last case the bulk oi
the material composing the section is disposed most effectively in
being away from the centre of the section, and although it has the
same sectional area as (a), the beam can stand between six and
seven times the weight. 1 A
familiar instance of this principle is
seen in the construction of many bridges. The upper and lower
flanges are made exceedingly strong, and the web is often formed
of lattice-work. (Charing Cross Eailway Bridge is an instance of

this construction.)
Longitudinal Strains. In arranging the structure of a ship

so that it shall efficiently resist the bending in a fore-and-aft
direction, we arrange the material on the aboveprinciples. Special
attention is paid to the sufficiency of the strength of the upper and
lower parts of the ship, as the uppermost decks and the side
plating adjacent, and the bottom plating, keel, longitudinals, etc.,
at the lower part.
The longer the ship is in proportion to her depth, the more
necessary does it become to pay attention to these portions of the
structure. Thus the upper-deck plating and side plating adjacent,
and the structure at the keel, were made much stronger in a cruiser
of 14,000 tons than in a battleship of 15,000 tons of the same date,
because the proportion length -r depth was so much greater in the
former case than in the latter. Fig. 3 shows in comparison the
keel structure in these two cases.
1
The thinaing down of the web cannot, of course, be carried on indefinitely.
WAR-SHIPS.
In attempting to make these strains the subject of calculation

we are confronted with the difficulty that it is not possible to
2S.L8S:-
KEEL OF BATTLESHIP KEEL or CRUISER
FIG. 3.
accurately determine the maximum strains the ship may reasonably

be expected to withstand. An extreme case would be to assume
that the ship is caught amidships on a rock, with the ends
unsupported, or, again, if the ends only were supported. It would,

however, be impossible, and indeed not necessary, to so construct
a ship that the structure could stand anything so extreme as either
of these conditions. The weight involved for the hull structure
would be prohibitive.
For purposes of calculation, the following assumptions are
made (a) the ship is supposed to be momentarily poised on the
crest of a wave of the same length as the ship, Fig. 4, and (b) astride
ON WAVE CREST
ACROSS WAVE TROUGH

FIG. 4.
the trough of a wave of the same length, Fig. 4. Under the first
assumption, the ends of the ship would tend to drop relative to the
middle, and we should have the upper works tending to tear apart,
and the lower works to buckle up. Under the second, the reverse
would be the case, viz. the keel and lower parts of the structure
would tend to tear apart, and the deck and upper parts to buckle.
The first is termed hogging, and the second sagging.
THE STRENGTH OF SHIPS, 5
In a well-constructed ship these are only tendencies, and the

material is able to withstand the strains thus
brought to bear upon
it. If, however, the ship is not strong enough, these tendencies
will show themselves by the giving way of the structure at some
point. In some destroyers, for instance, the compressive strain on
the upper deck has caused the plating to buckle between the
beams, and in these vessels it is most important to provide fore-and-
aft stiffening to the deck, so that the plating will be able tostand
up without deformation.
to the strain
Even in still water there are strains on a ship, because the
weight and the support of the buoyancy vary along the length.
At the ends, for instance, there would be an excess of weight over
buoyancy, because of the fineness of the ends, and at other portions
of the length the reverse may be the case. These strains, however,
are quite small in amount compared with those which might come
on a ship in a seaway.
If the results of calculation on a certain ship on the above
assumptions give a certain stress on the material, and the ship is

found to show no signs of straining on service, it is safe to proceed
with another ship, which by a similar process of calculation is
found to be equally strong'. The stresses on the material thus
found are not regarded as absolute values. There are many con-
ditions in the problem which cannot be taken into account, but
the stresses thus calculated form a valuable means of comparison
from one ship to another.
One feature in the construction of recent cruisers has been the
adoption of special steel of high tensile strength. Instead of the

tests specified for ordinary mild steel, viz. 26 to 30 tons per square
inch, this special steel has to stand a tensile test of between 34

and 38 tons per square inch. This steel is somewhat expensive,
and is only used in special places, as e.g. for the topmost deck,
and the upper and lower portions of the outer bottom plating.
These are portions which, as seen above, are the most severely
strained when
the ship is hogging or sagging. Torpedo-boat
destroyers of recent construction are built with decks and outer
bottom plating of a steel of still higher tensile strength, viz.
between 37 and 43 tons per square inch.
In deck-protected cruisers, like Diadem and previous ships,
and second and third class cruisers (see Figs. 21, 22, 24, 27), the
heavy protective deck fitted in the region of the waterline, although
valuable as a means of stiffening the ship, is not in a position to
6 WAR-SHIPS.
contribute anything like its fair share to the structural strength,

1
because of its position near the neutral axis of the section. In
the large cruisers subsequent to the Diadem, armour protection
was adopted at the side, and the main deck was made a protective
deck as well as the middle deck (see Fig. 23). This is a much
better distribution of the material as regards its usefulness in the
structure of the ship, as well as for purposes of protection. In the
design of more recent large cruisers there was the adoption of a

battery on the main deck instead of casemates, and the upper
deck over this battery was made a thick deck. This deck is still
better adapted to assist in the structural strength (see Fig. 138).
In the most recent ships, both battle-ships and cruisers, the
armour is carried to the forecastle deck, and this deck is made
"
a " protective deck. Here again this material is most effectively
disposed to contribute to the structural strength.
In large ships in the Eoyal Navy it is the established practice
to build the structure mainly on the longitudinal system. That
is, the longitudinal portions of the structure are made continuous,
and the transverse portions are made in short pieces between, or
intercostal. This system is carried out from the keel to the lower
edge of armour or the protective deck, over the length of the
double bottom (about two-thirds the length in a battle-ship).
This system is admirably adapted to the formation of a double
bottom. At the ends of the ship the vertical keel is still con-
tinuous, but the transverse framing is made the continuous part
either side of the keel. The longitudinal strength is not so

important at the ends, and is obtained by intercostal girders, the
various platforms and bulkheads as well as the outer bottom plating.
For smaller ships the transverse framing is made continuous
and more closely spaced, and the fore-and-aft framing is mostly
intercostal. The close spacing of the framing is necessary in order
that the outside plating shall be well stiffened to hold it up to
itswork, as it is necessarily thin in a small ship. In these vessels
a double bottom is not fitted. These features of construction will
be fully dealt with when we consider in detail the structural
more
arrangements of various classes of ships.
If we compared two ships of the same size, one a war-ship and
the other a merchant ship built to the rules of a registration
society, we should find that the scantlings (or sizes of the steel
used) of the former would be considerably less than those adopted
1
The neutral axis is a horizontal line through the C.G. of the section.
THE STRENGTH OF SHIPS. ^
in the latter case. There are several reasons which contribute to

this. Merchant ships carry heavy weights of cargo in large holds,
and the transverse framing especially has to be very massive to
take the weight of the cargo. Again, merchant ships in some
trades frequently ground when loading or discharging their cargo,
and this again necessitates a good margin of strength as compared
with war-ships, which are more carefully handled in this respect.
Also, in war-ships a very extensive system of watertight subdivision
is adopted ; the large number of bulkheads and flats thus obtained
assistvery materially in the structural strength. Most merchant

ships are built with the transverse framing for the most part
continuous, with fore-and-aft girders intercostal. This system,
although a very convenient method as regards economy in con-
struction, is not so efficient from a structural point of view for

large ships as the method adopted for war-ships.
An important point in connection with the scantlings adopted
in a ship is the subsequent maintenance. Inspection and care of
the structure of war-ships is carried out in a most careful and
thorough manner, and in this way the margin for deterioration
may be made considerably less than in ships not so carefully
looked after. The regulations for inspection of the structure of
II. M. ships will be referred to later.
A very important
principle in ship construction to be borne in
mind the necessity for avoiding any discontinuity in strength.
is
.Any part of the fore-and-aft structure which has to be ended,

must be tapered down over several frame spaces, and a deep
girder which has to be continued to the end of a vessel in a
smaller form must be tapered down gradually to avoid any dis-
continuity of strength.
A superstructure like a boat deck or a flying deck is very high
up, and if made a continuous portion of the structure is likely to
be severely strained, and

it is not sufficiently
strong to stand any
severe strains. On this
FlG< 5t
account the plating of
this deck is deliberately cut through, and a sliding joint is made as
Fig. 5 to save the rupture of the deck that would otherwise occur.
The side plating of the superstructure adjacent is cut by the gangway
ports. In the large cruisers of Invincible and Lion classes, the fore-
"
castle deck goes right across the ship and is made a "strength deck.
8 WAR-SHIPS.
In Neptune, Hercules, and Colossus the boats are carried on

heavy girders between the superstructures. The ends of these
girders are bolted to the superstructure supports with oval holes in
order that they may not take up any structural strains for which
they are not adapted, being so high up in the ship.
Transverse Strains. The above remarks apply to the fore-
and-aft structural strength; the transverse strength is also of
importance. The rolling of the ship will tend to rack or distort
the section, and this tendency is resisted by the transverse bulk-
heads, which are numerous in a war-ship because of the W.T.
subdivision, and also by the connections of the beam arms to the
frames, which are made specially strong for this purpose.
A ship when placed in dry dock, especially one with heavy
weights of armour at the side, will have severe strains tending to
tear the decks. Such a ship, however, is carefully shored as the
water leaves so as to obtain plenty of support. It should be noted
that the shores must be placed on transverse or longitudinal
frames, or at bulkheads, in order to prevent the plating being
injured. Recent ships have been built with the bottom amidships
quite flat, which is of material assistance when docking, as three
rows of blocks can be arranged for the vessel to ground upon (see
Fig. 13A). Other ships not flat bottomed have been provided with
side docking keels as Fig. 13, to take the side blocks over the
midship portion of the length.

2. Local Strains. The above strains are termed "structural,"
because they come on the ship regarded as a complete structure,
but there are a number of isolated strains that are brought to bear
only on a portion of the vessel. Such strains are termed local.
Panting. This term is used to denote the in-and-out working
of the plating. It is especially found at the forward end of a
ship which subject to blows from the sea.

is If necessary, extra
stiffening is added forward. In the large ships of the Navy

panting is not experienced. Formerly the forward plating was
thickened up to support the stem, and in some cases the armour
has been continued right forward. Eecent vessels have thick
plating forward to take the rub of the stockless anchors. The
plating at the forward end is also well stiffened by the more
closely spaced frames, and by the flats and platforms.
Severe local strains exist in the neighbourhood of heavy
weights. Thus in way of an armoured barbette for a pair of
heavy guns, a tremendous weight is localized over a short
THE STRENGTH OF SHIPS. 9
portion of the length. Beneath such places the framing of the

ship is made very strong, and bulkheads and pillars are arranged
to distribute the strain throughout the structure.
Under theengines, where we have machinery in motion and
not simply a dead weight, extra strength is necessary, and the
framing, both longitudinal and transverse, is made very massive.
The thrust from propellers has to be transmitted to the
structure of the ship by means of the thrust block. The fore-
and-aft framing under the thrust is constructed very strong.
is necessary to support decks over which
Special strength heavy
guns have to fire. Cabin bulkheads are utilized for this purpose
wherever possible. At other places a girder is worked beneath the
deck formed of strong continuous angles and plate beneath the beams,
connected to the beams by intercostal plates and bars as shown in
Fig. 5A. This girder is well supported by pillars to the deck below.
CE.CM PLATING
1
CHAPTER II.
TESTS OF STEEL, ETC., SECTIONS, RIVETS, JOINTS, ETC.
AT the present time steel is the material almost universally

employed in the building of ships. The steel employed is known
as mild steel, which is a very pure form of iron, containing less
than per cent, of carbon. Mild steel is quite a different
material from the steel that is used for knives and tools. This
tool steel is capable of taking a temper and hardening. Mild
steel is produced of uniform quality and very reliable, and it is
admirably adapted to stand the rather rough treatment in the

shipyard, necessary in the formation of a ship's structure. A
property of mild steel, which has frequently been found of
immense service, is its capability of bending without fracture.
There are many cases on record of steel ships having received
severe injuries to the skin plating, but remaining quite intact.
With wrought iron as formerly used, which is not so ductile,
fracture is much more easily obtained. High tensile steel is being
ased in certain parts of cruisers and destroyers instead of mild
steel. It is of greater strength, but is more expensive than mild
steel, and it has to be carefully treated in working it into the
ship to prevent reduction of the strength.
Tests of Material. It is obviously of the greatest importance
that the steel, employed in the construction of the ships of the
etc.,
Royal Navy first-class uniform quality, and in order

should be of
to ensure this, a complete system of testing and inspection of
material is arranged for at the works at which it is made. Only
firms of the highest standing and having works of sufficient capacity
are dealt with, these firms being on the Admiralty List. The
process of manufacture is carefully watched by a resident
Admiralty Overseer. The following is an abstract of the tests
carried out by the firm under his supervision.
Tests of Steel. (a) Tensile test of mild steel to determine the
10
TESTS OF STEEL, SECTIONS, RIVETS, JOINTS, ETC. 11
ultimate strength. A
strip is cut from the plate or bar selected by the
Overseer either lengthwise or crosswise of the plate, and it is planed
to the sha pe shown in (a)
Fig. 6, having a parallel _^_ _

width of 1 J in. to 2J in.
over a length of 8 in.
It is necessary that the
edges should be planed

and not sheared, as
shearing makes the |
edges brittle. Such a FIG. 6.

test piece is placed in a
testing machine, and should have a tensile strength of between

26 and 30 tons per square inch before fracture, and the stretching
or elongation in the length of 8 in. should be 20 per cent., i.e. when
the parts are put together, as in (&) Fig. 6, the original 8 in. should
be 9 1 in.
(b) Sending test of mild steel to determine the ductility. This

property of ductility is of importance for the reasons already
stated. Strips 1J- in. wide, heated to a blood red and cooled in
water of 80 Fahr., must stand bending double to a curve of which
the inner diameter is three times the thickness. This test can be
done cold if preferred by the Overseer.
The Overseer selects one plate or bar for testing purposes out
of every batch of fifty, or less than fifty.
For the high tensile steel H. T. used in cruisers, a tensile test of
33 to 38 tons per square inch is specified. Similar results as to
elongation and ductility are required as for mild steel.
For the high tensile steel ff.H.T. used in destroyers, a tensile
test of between 37 and 43 tons per square inch is specified for
plates 3J- Ib. per square foot and over, and 35 to 45 tons under
3J Ibs. per square foot. An elastic limit is required not less than
1
one half the ultimate tensile strength with a minimum of 20 tons.

The elongation and ductility required are rather under those given
above for mild steel, varying from 8 to 15 per cent., the upper
limit being for plates 10 Ib. per square foot and over.
For H.T. plates one plate in 10 is tested, and there are
also chemical tests to determine the percentage of silicon and
1
Below the elastic limit the material is elastic, i.e. it will resume its original
length the strain be removed.

if Above the elastic limit this is not so, and
" "
permanent set takes place.
12 WAR- SHIPS.
carbon. For H.H.T. plates tests may be taken from one plate in
every batch of 20 or less.
Nickel plates have been used in some ships for protective

steel
purposes, and have to show a tensile strength of 36 to 40 tons per

square inch with an elongation of 13 per cent, in a length of 8 in.
Also a chemical analysis is taken to determine the proportion of
nickel which must not be less than 3 per cent.
In addition to the above specified tests, the Overseer makes
an examination of the plates and bars to see that there is no
lamination or flakiness, and that there are no objectionable hollows
or other surface defects. For angle bars, etc., the Overseer can
also test the bars at his discretion to ascertain the ductility under
ordinary working conditions.

It is the Admiralty practice to specify plates by weight per
square foot, and angles, etc., by weight per lineal foot. Thus a
steel plate J in. thick weighs 20*4 Ibs. per square foot. plate A
if ordered as 20 Ibs. per square foot is thus really rather under
J in. thick (0*49 in.). A
steel angle bar specified as 3 in. X 3 in.
of 7 Ibs. per foot is rather under f in. There is a distinct advantage
in ordering steel by weight in this way. We
have an exact check
on the thickness supplied, because by a simple calculation of the
area or measurement of the length we can tell what a plate or bar
ought to weigh of the specified weight per square or lineal foot.

Comparing this with the actual weight supplied, any excess or
defect of thickness is at once apparent. Such variations of thick-
ness are difficult to determine by actual measurement. Small
excesses of thickness would amount in the aggregate to a con-
siderable weight in a ship's structure. With plates 20 Ibs. per
square foot and over, and also bars, the manufacturer is allowed a
latitude of 5 per cent, under the specified weight, and no latitude

above. For plates under 20 Ibs., a latitude of 5 per cent, below
and above is allowed.
Thin steel sheets under 2 '5 Ibs. per square foot, and steel bars
under 3 Ibs. per foot run, are not tested to a special requirement,
but are examined for surface defects and flaws, and have bending
and other tests at discretion.
Treatment of Mild and High Tensile Steel. There are
certain precautions necessary in the treatment of steel. Plates
and bars should be heated as little as possible when heated, no
;
work must be done when the temperature has fallen to a Hue heat
(600-400 Fahr.). At this temperature the steel is very brittle,
and the plate or bar must be reheated to complete the work.

Plates or bars which have had a large amount of work put upon
them while hot and have to be reheated should be subsequently
annealed (i.e. heated to redness and allowed to cool slowly under
a heap of ashes).
When holes are found that the strength
punched in steel, it is
per square inch of the material considerably less than the

left is
original strength. The forcible entry of the punch makes the

steel round the hole brittle. It is found, however, that the process
of riveting partially restores the strength, the influence of the hot
rivet and the subsequent hammering must alter the structure of
the steel round the hole. In some large ships it has been laid
down that the whole of the holes in sheer strake and the strake
below, and also the upper deck, over about /o^18 ^ ne length, shall
be drilled.
The high tensile steel mentioned above as used in cruisers and
destroyers has all the holes drilled. Punching is found to cause
serious depreciation of the strength.
There are many holes in a ship's structure which have to be
countersunk, i.e. the hole is made conical to take the point of the
D, Fig. 9). In this case the holes are punched about
rivet (see
J in.small and the injured material round the punched hole is

removed by the countersinking drill, and the strength of the plate
is not deteriorated.
Pickling. found during the process of manufacturing
It is
steel plates that there isformed on the surface a black oxide or
scale, called mill scale, which clings very tenaciously to the
surface, and if not removed will subsequently peel off and take
with it the paint, leaving the bare steel underneath liable to
x
corrosion. This scale also
electro-negative to the steel, so
is
that if moisture gets in under the scale, corrosion of the steel will
go on very rapidly. It is, therefore, most important to remove all

traces of the mill scale before the steel is painted. In Admiralty
work it is the practice to immerse the plates on edge for a few
hours in a bath containing dilute hydrochloric acid (1 part acid to

19 parts water). This loosens the scale, and the plates when
removed are brushed with wire brushes and washed with a hose to
remove the scale. The portions of the structure thus pickled are
those liable to come into contact with sea-water or water in the
1
I.e. if an electric current is set up between the scale and the steel, the steel
will corrode, like the zinc does in the zinc-copper cell.
i
4 WAR-SHIPS.
bilges. These are the plates of the outer and inner bottoms, lower
plates of freshwater tanks,
and the lower
plates of bulkheads,
plates of frames.
Rivet Steel and Rivets, Tests of. The efficiency of the
structure of a ship must ultimately depend on the quality of the

rivets used to connect the various portions of the hull together.
) COLD. (ft) HOT.
FIG. 7. Rivet tests.
The following is a summary of the tests made to ensure having

rivets of first-class quality :
The steel bars for mild steel rivets must stand a tensile test of between 26
and 30 tons per square inch, with an elongation of not less than 25 per cent, in 8
diameters. These tests are seen to be rather more severe than those for mild steel.
Eivets for use with high tensile steel are made from steel bars
of special quality, so as to utilize the greater strength of the steel.
H. T. Rivets have 3 ribs on the head as P, Fig. 9, the bars for rivets up to
and including in.have a tensile test of 34 to 38 tons per square inch with
elongation of 20 per cent, in a length of 8 diameters. Over ^ in. the tensile
test is32 to 36 tons with same elongation.
H. H. T. Rivets have 4 ribs on the head. The tensile strength is 37 to 43
tons per square inch with elongation of 18 per cent, in a length of 8 diameters.
With bars less than | in. diameter the elongation is 25 per cent, in a length of
4 diameters.
Nickel Steel Rivets have 1 rib on the head. The tensile strength is 36 to 40
tons per square inch with elongation of 20 per cent, in a length of 8 diameters,
and a chemical analysis must show not less than 3 per cent, of nickel.
The rivets themselves, when made of all the above qualities, must cool
slowly, and samples taken at random must stand the following tests :
(i.)
A cold test as (a), Fig. 7.
(ii.)
A hot test as Fig. 7, the head being flattened out to two and a
(//),
half times the diameter of the rivet without

cracking,
(iii.)
The shank is nicked and bent over to show the structure of the steel.
Steel Castings, Tests of. There are many parts of a ship's

made of iron forgings these can
structure which formerly were ;
now be more conveniently made of cast steel, and this material

is always used in steel ships for the stem, sternpost, rudder-frames,
shaft brackets, etc. The following are the tests laid down to
ensure satisfactory castings :
On each casting three projections are cast, to provide, when cut off, test
pieces. One of these is turned and must stand a tensile test of 26 tons per
square inch, with an elongation of 13J per cent, in a length of 4 diameters.

One planed to a section 1 in. square must stand bending cold through an angle
of 45 over an edge having a radius of 1 in. The other test piece is available if
the results of either of the others are disputed. It will be noticed that these
tests are less severe than for mild steel.
Although the material of the casting may satisfactorily stand the above tests,
it is most necessary to find out, if possible, whether any blow holes exist in
the body of the casting. In order to thoroughly shake up the casting, it is
either (a) raised to an angle of about 60, or (6) lifted bodily to a height of about
12 ft. and dropped on hard ground. The latter would be adopted for small
castings. The casting also is suspended in chains and hammered all over with
a heavy sledge to find out if it rings true ;
the surface is also carefully examined
for defects.
All the important castings of the structure, etc., are tested to the above
"
specification, and are described as " quality. A
Less important castings as
bollards, fairleads, hawse pipes, etc., are of "B"
quality, the tests being much
the same except the elongation required is 10 per cent, in a length of 4
diameters. Castings for coaling scuttles, scuppers, and fittings of minor
"
importance are of" C quality. In this case the tensile and bending tests are
not required, but the falling and hammering tests have to be satisfactorily made.
" "
Instead of cast steel quality C special malleable cast iron may be employed.
Castings have to stand dropping on to an iron or steel slab from a height of 9 to
15 feet. One test piece planed to a section of 1 in. x f in. must have a tensile
strength of 18 tons per square inch and an elongation of 4^ per cent, in a length
of 3 in. Another test piece of similar section must stand bending cold parallel
to the shorter side through an angle of 90 over a radius of 1 in.
Phosphor Bronze Castings, Tests of. When a vessel is

sheathed with wood and copper, it is not possible to use cast
steel for the stem, sternpost, etc., because of the galvanic action
that takes place between copper and steel, if metallic connection
is established between them. For such ships, therefore, the stem,
sternpost, etc., are made of the copper alloy, phosphor Ironze.
This is now specified
to consist of copper 85 per cent., and
.6 WAR-SHIPS.
phosphor tin 15 per cent. This material should stand a tensile

test of15 tons per square inch, with an elongation of 10 per cent.
in a length of 4 diameters, the test piece being turned to 1 in.
diameter. This strength is seen to be considerably lower than
for cast steel. A
chemical analysis of some drillings is made to
ascertain the composition of the material, and this analysis must
not show less than 0*45 per cent, of phosphorus.
Phosphor bronze is largely employed where a good
bearing
surface is required, as in rudder bushes, etc.
Sections of Steel. One great advantage attending the use of
iron or steel for shipbuilding purposes is the possibility of obtaining
FIG. 8.
the material in convenient and efficient forms. Thus, to form a

beam in a wooden ship, it would be necessary to take a log of
timber of solid rectangular section. We
have seen in the previous
chapter that in such a section the material is not disposed to the
best advantage to resist bending. A
beam of iron or steel can be
rolled so that it has a broad upper flange and a bulb or flange at
the lower part, both of which are so situated as to be most effective
against bending. Examples are seen in Fig. 8.
The various sections used in Admiralty shipbuilding are shown
in Fig. 8.
The angle bar (a) is used to connect plates together; as a
stiffener to a plate ;
for beams, frames, etc.
The tee lar (b) was formerly used as a stiffener to important

bulkheads as (n). The principal present use is inside masts
(see Fig. 8).
The channel bar (/) is used as a stiffener in certain cases and
for the framing behind armour.

The zed bar (e) is largely used for transverse framing, to avoid
the use of two angles riveted back to back. It is also used as a
bulkhead stiffener.
The I or H
lar (g) is used for the main stiffening of important
bulkheads, and for supports behind armour.
The angle bulb (c) is used for deck beams.
The tee bulb (d) was formerly used for deck beams, where a
steel deck was not fitted.
A flange (k) is frequently put on the edge of a plate to serve
the purpose of an angle bar for stiffness or connection.
The half round (I) is usually hollow, and has been used as a
moulding round the ship.
The segmental lar (ra) is used as a finish round the top of hatch
coamings, etc.
The sections shown in Fig. 8 are drawn to scale, and show the
exact form of sections now used. It is worth noting that the
flanges of the zed, channel, and I bars are more substantial than
the webs. These sections thus illustrate very clearly the principles
touched upon in Chapter I. in connection with beams.
Rivets, Forms of, etc. Fig. 9 shows the usual form of rivets
and riveted connections employed in Admiralty shipbuilding.
A is the most common form of rivet, called a pan head rivet,
from the shape of its head. It will be noticed that the rivet is
formed with a conical neck. All rivets ^-in. diameter and above
are thus formed, because the hole formed in the plate by punching
has a slight taper, and it is most necessary that the rivet should
completely fill the hole (see D, etc., Fig. 9).
D, E, F, Fig 9, show various points associated with the pan
head. D the countersunk point necessary when the surface
is
has to be flush, as for the skin plating. The hole formed by the
punch has to be made conical by the countersinking drill. E is the
point adopted for most of the internal work. No
countersinking
is done, and the point is hammered up full as shown. F is the
point adopted where a finished appearance is desirable this point ;
is called a snap point.
B is a snap head rivet, now seldom used. G is a rivet closed

c
iS WAR-SHIPS.
up by a hydraulic riveter, the pan head of rivet being squeezed

into a snap head.
In some few cases it is necessary to have a flush surface on
both sides, and in countersunk head rivet is used, as C,
this case a
and the point also countersunk, as H.
is
Certain parts of the structure cannot be riveted by the ordinary

means, as, for example, where outer bottom plating is connected to
a stem casting. In such cases tap rivets are used, as L, M, and N
(Fig. 9). In L, the most usual form, the head is countersunk, and
the square projection is chipped off after screwing up. Where it is
necessary to have the work portable, a recess is fitted in the head,
FIG. 9.
as N. Where a flush surface is not necessary, as for some internal

work, the head is made
hexagonal, as M.
In parts of the ship behind armour, where men would be
stationed in action, the rivet heads are likely to fly off and become
dangerous if the armour is struck. In such places the inside of
the framing is closed in with a steel lining (see Fig. 13, on the
main deck). This lining is secured to the framing, etc., by screw
rivets, as K, Fig. 9. Tap rivets, as L, M, or N, would be undesir-
able, because of the likelihood of the rivet breaking and the head
flying out if the armour was struck.
Laps, Butts, etc. When two plates are connected together,
they may either be lapped or "butted.Laps are shown at F and G
in Fig 10. For thin plating a single row of rivets is sufficient, as

F ;
for thicker plates a double row of rivets is necessary, as G.
In the first breadth of the lap, F, is
case, the three and a half times
the diameter of the rivet used in the second case, G, it is six times
;
the diameter of rivet. This gives a clearance between the rows

of 1J diameters, and rather more than a diameter clear of the
edge. The edges of plates are usually lapped, but in some cases,
where a flush surface is necessary, the edges are connected by an
edge strip. A special form of edge connection is seen when the
edge strip takes the form of a tee bar. Here the tee bar forms
FIG. 10.
an edge strip and stiffener at the same time (see n, Fig. 8). When
the end connection of plates has to be flush, as is usually the case,
the connection is called a lutt-strap. Butt-straps are single,
double, treble, or quadruple riveted, according to the importance
of the connection. These are shown by A, B, C, and D, Fig. 10,
the breadths being respectively 6J, 11J, 16J, and 21J- times the
diameter of the rivet used. Quadruple riveted straps are used for
for one instance). In some
specially important joints (see Fig. 51
cases butt-straps are made double, i.e. in two halves on either side,
each being rather more than half the thickness of the plates
connected.
20 WAR-SHIPS.
Angle bars, etc., are connected together by a piece of angle,

fitted as shown in Fig. 14, long enough to take two, three, or more
rivets each side as necessary.
The breadths given above for butt-straps and laps are somewhat
exceeded when
dealing with high tensile steel. In this case a
slightly greater clearance from the edge is considered necessary.
Spacing" of Rivets. The spacing of rivets from centre to
centre along edges, butts, etc., is termed the pitch. This pitch
varies according as it is necessary to have the joint watertight or
not. For the former case the joint has to be caulked, and in order
to do this it is necessary to have the rivets
closely spaced to draw
and keep the work tightly together. The usual pitch for water-
tight work is from 4 to 5 diameters. For oiltight work the spacing
used not to exceed 4 diameters, but experience has shown that for
oil fuel the W. T. spacing is sufficient. For special compartments
containing lighter oils a spacing of 3 to 3J diameters is essential.
For non-watertight work a pitch of 7 to 8 diameters is all that is
necessary.
Caulking". All caulking should be metal to metal, filling
pieces being avoided as far as possible. For laps the caulking edge
must be made square (planed for important parts, as the outer
bottom plating). The edge near the joint is nicked with a sharp
tool and the piece so left is driven against the adjacent plate, as in
Fig. 10. For butts the edges must be planed a split is made
;
either side of the joint and the two edges are forced together with
a hollow tool, giving the shape to the butt, as in Fig. 10. Butt
caulking is not so efficient as lap caulking, because a pull on the
joint, or in-and-out working of the plating, is more liable to open
the caulk in the former than in the latter case.
An interesting and very efficient form of butt-strap, shown at
E, Fig. 10, has been adopted in some destroyers. The strap is
double the inside portion is treble riveted, and the outside portion
;
double riveted. The alternate rivets are omitted in the last row,
so that the plate is not weakened more than at the adjacent frame.
The middle row is closely spaced to allow the edge to be lap
caulked. If fracture of the plate occurred through this line of

rivets itwould be necessary to shear also all the rivets in the last
row. The rivets in the two inner rows are in double shear.
It is interesting to note that this form of butt connection was
adopted in many places on the Cunard liner Lusitania.
CHAPTER III
FRAMING OF VARIOUS TYPES OF SHIPS.
THE British Navy, intended to operate in all parts of the world, is
necessarily made up of many different types of vessel. It would

be obviously impossible in a work of this character to exhaustively
f-4.
FIG. 11. Section of H.M.S. Royal Sovereign.
describe the construction of each type of vessel in the Navy. All

that is possible, or desirable, is to take certain main types, and deal
with the principles of their construction. It will be seen that
21
WAR-SHIPS.
the intended service of a ship has a distinct influence on the

construction.
First Class Battle-ships. These ships are
heavily armoured
and armed, and possess speed which is moderate compared with that
obtained by cruisers.
One distinctive feature l~ u "" Pcc * -
of the construction of
these ships is the pro-
vision of an inner skin
up to the protective
deck. Sections of recent
battle-ships are shown

in Figs. 12, 13, and 13A.
The inner skin is
3^ ft.
in at the middle line,

the depth being some-
what less up the bilge,
and it is continued up
to the protective deck
in theform of a vertical
bulkhead. There is, in
addition, the inner coal-
bunker bulkhead, so FIG. 12. Section of H.M.S. Duncan.
that at the side and
bilge, at the lines AA
and BB, there are three skins to pierce before
a vital portion of the ship is reached at the bottom there are only
;
two skins to pierce. In recent battle-ships the inner bottom is

continued right up to the lower protective deck, forming four skins
at the side through AA to the boiler-room. This arrangement is
important, in view of the possibility of being rammed or receiving

other damage under water.
In the Lord Nelson, Fig. 13 A, it will be noticed the bottom is
quiteflat for docking purposes, and this is now the general practice.
The special shape of the section was necessary for the purpose of
entering a particular dock.
The double-bottom arrangement is adopted in all ships of the
Eoyal Navy above third class cruisers. A double bottom is valuable
as providing an inner skin in the event of damage to the outer
bottom, and it is always subdivided into a number of watertight

compartments, so as to localize any damage that might occur. A
part of the space is conveniently arranged for the stowage of fresh
water forming the boiler reserve feed, and oil fuel. Any of the double-
bottom compartments, including the wings, can be flooded if desired
to correct heel or trim caused by damage (see Chapter IX.).
In the Royal Sovereign the protective deck was level, with a
thick belt, 8J ft. broad, as shown in Fig. 11. In the more recent
ships, the armour belt has been reduced in thickness and increased
in area, and the protective deck has been made level at the middle
line, but sloping down to the lower edge of armour at the side (see
Figs. 12, 13,

and 13AX
Longitudinal Framing-. The main framing of these ships is
arranged on the longitudinal system, the presence of the double

bottom lend-
ing itself ad- "

/-'"
mirablytothis
arrangement.
The fore-and-
aft framing
below protec-
tive deck over
the length of
the double
bottom (about
two-thirds the
length) con-
sists of a ver-
ticalkeel,3ft.
deep, and five
or more longi-
tudinal gir-
ders on each
side, the num-
ber depending
on the breadth
of the ship
(see Fig. 12). Fio. 13. Section of H.M.S. King Edward VII.
Of these the
verticalkeel, second and fourth longitudinals are watertight,
thus dividing the double bottom from side to side into six
watertight compartments. The vertical keel
and longitudinals are
continuous, and the latter are allowed to taper somewhat in depth
towards the ends of the double botton. The vertical keel is
24 WAR-SHIPS.
25 Ibs. (f in.) thick, with two angles along the bottom 4J in. X 4J in.,
and two along the top 4 in. X 4 in. These, together with the
middle plate of the inner bottom, and the inner and outer plates of
FIG. 13A. Section of H M.S. Lord Nelson.
the flat keel, form a substantial backbone

to the ship (see Fig. 3).
The adjacent plates of the vertical keel are connected by 30-lb.
(J in.) treble riveted butt- straps (Fig. 14). Adjacent lengths of
INNER BOTTOM
|HNER
26 WAR-SHIPS.
Of the second type is the most numerous, as it is the type

tliese
adopted beneath the armour from No. 4 longitudinal upwards,

and under the heavy weights of the barbettes and the engine?.
The third class is fitted under the main transverse bulkheads
(the inner bottom being continuous), and at intervals between
(see Fig. 52). The bracket frames are fitted to the remainder.
Fro. 17.
In one ship out of a total of 780, 456 were solid, 180 were water-
tight,and 144 were bracket frames.
The bracket frame built up as shown in Fig. 17.
is Deep
angles, 5 in. x worked to the outer and inner bottoms,
3^- in., are
and to these in each bay are riveted two bracket plates, 15 Ibs.
(| in.), with their inner edges stiffened by a 3 -in. flange. These

FRAMING OF VARIOUS TYPES OF SHIPS. 27
are connected to the vertical keel and longitudinals by pieces

of angle 3 in. x 3 in.
The plate frame
(Fig. 17) consists of a 15-lb. plate connected to
the inner bottom, vertical keel, etc., by angles 3 in. x 3 in., and to
the outer bottom by angles 3.]- in X 3 in. Each plate is stiffened
by an angle bar, and holes large enough to enable a man to pass
through for inspec-

tion purposes are
cut to lighten the _
plate.
The watertight
frames (Fig. 17) _]_
with the vertical !
keel and water- jjj
tight longitudinals _i_

divide the double-
bottom space into
a largenumber of
watertight com-
partments. The
frame is made of
15-lb. (f-in.) plat-
ing under the trans-

verse bulkheads,
andl2^1bs.( TVn.)
in other places.
The space is filled
in solid with the

plate, and staple
angles are worked
round the top and
bottom as shown.
This enables a tight FIG. 18. Support to 9-in. armour.
fit to be made, and
work and caulked.
the whole closely riveted for watertight
is
These frames are specially stiffened, as shown, by three channel

bars in the first bay, one channel and two angles in the other
This extensive stiffening is fitted
bays up to No. 4 longitudinal.
to make the framing strong enough to stand
the severe strains
which exist when such heavy ship

a is in dry dock. Eecent ships
are being fitted with docking keels at the side, as in Fig. 13A. In
WAR-SHIPS.
these cases the stiffening of the watertight frames is not so

extensive as described above.
Extra frames, both longitudinal and transverse, are worked
under the engines in order
DECK to provide a rigid support.
Framing behind and

above Armour. The cha-
racter of the framing behind
armour is governed by the
necessity of providing a rigid
support to the armour. The
framing behind the 9 in.
armour of the Majestic class

is formed of plate frames
15 in. deep, 24 in. apart,
with angles on the edges as
shown in Fig. 18. There
are fore-and-aft girders in
addition, as shown. The
frames are well supported by
bracket plates at the heads
and heels.
In recent ships built up
structures behind armour
have been avoided, and the
framing formed of deep
channel bars, 10 in. to 12 in.
deep, with fore-and-aft stif-
fening girders. 7-in. zed
bars are employed behind
thin armour, 4 in. to 6 in.
thick.
The framing above the
armour consists of 6-in. zed
FIG. ISA Support to 12 in. and 8 in. armour. .
H.M.S. Lord Nelson. bars.
Fig. ISA shows the

method of support behind the 12 in. and 8 in. armour of Lord
Nelson. It will be seen that the frame is composed of two
10-in. angles riveted back to back to the main deck, the outer
one only being continued up through the main deck to the
upper deck.
Framing Ends of Ship. The above description refers

at the
to the framing over the length of double bottom, or about two-
thirds the length. The ends of the ship are framed on a somewhat
different principle. Here the longitudinal strength is of less im-
portance, and the main function of the framing is to stiffen the outei
bottom plating. Accordingly we find that the transverse framing
is continuous either side of the vertical keel to the protective deck,
and this framing is more closely spaced, viz. 3 ft.
The vertical keel is still continuous, but not watertight, and
TRANSVERSE
FRAME
AT ENDS.
FIG. 19.
the frame consists of a vertical floor-plate with outer and inner

angles to the bilge. Above this the frame is formed of a 6 -in.
zed bar. In Fig. 19 the top of the floors is covered
in with a
watertight platform, and the zed bar is passed through with the
inner flange cut away. This makes the watertight work at the
ship's side more readily performed than if the complete zed bar
went through (see Fig. 41).
Asbefore stated, the vertical keel is continuous from end to
end of the ship, but the five longitudinals on each side, which
30 WAR-SHIPS.
extend over the length of double bottom, are altered in character

at the ends. They are either twisted round to connect on to a
fore-and-aft bulkhead or a flat, or tapered down to one of the
forms shown in Fig. 19. The lower one is simply formed by an
intercostal plate between each pair of floors, and the upper ones
are each formed by a 10-in. zed, slotted over each frame, with a
continuous 3 in. x 3 in. angle on the inside. Fig. 20 shows in
detail a longitudinal is tapered down over three frame spaces
how
to the zed bar form. This is necessary to avoid any discontinuity
in the fore-and-aft strength.
Above the protective deck the transverse framing is still 3 ft.
apart, and is formed of 6-in. zed bars. In recent ships armour of
varying thickness is fitted at the forward end, and behind this the
frames are 2 ft, apart to well support the armour. At the extreme
FIG. 20.
forward end, before the collision bulkhead, and below the pro-
tective deck, the frame consists of a solid plate connected by angles
to the outer bottom, etc. This plate is well lightened by holes.
It will be noticed that, although the inner bottom proper stops
at about one-sixth the vessel's length from each end, yet an inner
skin is obtained well towards the ends by the provision of the water-
tight flats forming the floors of the store-rooms, etc. (see Fig. 19).
First Class Cruisers. Vessels of this type are in some cases
of equal or even greater displacement than battle-ships, by displace-
ment being meant the total weight of the ship. The broad dis-
tinction between a battle-ship and a first class cruiser of recent
design, is that the former has thicker armour with possibly a
greater proportion of the side area protected and a heavier arma-
ment, while the latter has lighter protection and armament, but
with high speed. The following comparison will illustrate this
distinction :
32 WAR-SHIPS.
Fig. 23, the inner skin was only carried to the upper part of bilge.
This gave a larger space available for coal. In some cruisers, also,
the upper coal-bunker bulkhead was dispensed with, to increase
the facility of transporting the coal. It is to be observed that in
the more recent practice, the double bottom is carried up to the
protective deck and a fore-and-aft intact bulkhead worked in
the upper bunkers, as in Fig. 2 3 A.
Manyfirst class cruisers, including some of Edgar class and
all the ships of Diadem, Powerful, and Cressy classes, were con-
structed with the bottom sheathed with wood and copper. This
UPPER DECK
FIG. 22. H.M.S. Diadem (sheathed). FIG. 23. Armoured cruiser.
was done because these ships were intended to keep the sea
for long periods without docking. The sheathing of copper is
found to be the best anti-fouler. In building a ship, however,
sheathed instead of unsheathed, certain sacrifices have to be made,
viz. increase of first cost and reduction of measured mile speed.
This subject will again be referred to later when dealing with
methods of preventing fouling.
Framing of a First Class Cruiser. Taking the framing of
the cruiser whose section is shown in Fig. 23, we notice that the
vertical keel is 42 ins. deep and 25 Ibs. (f in.). This depth
is maintained over the whole length of the ship except in the
engine-room, where the depth is about 5 ft. The two top angles
are 3J in. x 3J- in., and the lower angles 4J in. x 41 in. There
are four longitudinals on each side, of which the second and
fourth are watertight. These extend over the length of the
double bottom, i.e. about half length of the ship.
The transverse framing in the double bottom
is similar in
character to that already described for a battle-ship. The framing

behind the 4-in. armour is formed of 8-in. zed bars, with a fore-
and-aft stiffening girder. The
remainder of the framing, both
longitudinal and transverse, is generally similar to that adopted
in battle-ships as described above.
In this type of ship, in consequence of the great power of the
reciprocating machinery, the framing in the engine-room was built
I
F'csle Deck
Upper Deck \^>
Mam Deck
Lower Deck
"[wrT ELZ
FIG. 23A. FIG. 24. Sheathed second class cruiser.
keel was 5 ft. in depth, and

exceedingly strong. The vertical
extra fore-and-aft girders were worked, in addition to the ordinary
to the engines.
longitudinals, to give a rigid support
Second Class Cruisers. Many of the vessels of this type
are sheathed with wood and copper, having been intended for
service on foreign stations, where the ships have to keep the sea
for long periods. A
typical section of a second
class cruiser is
given in Fig. 24. noticeWe

that the protective deck and the
coal above constitute the protection. The advantage of a double
bottom is retained, but it is of less extent than in the previous
cases considered, only extending to the coal-bunker bulkhead.
The construction of this ship is shown in Fig. 25. There is a
7 with double
continuous vertical keel 36 in. x 17J Ibs. ( T^ in.),
34 WAR-SHIPS.
angles along the top 3 x 3 in., and double angles along the
in.
bottom 3-J
in. x 3J in. There are two longitudinals on either
side, 15 Ibs. (f in.), the second one being watertight. The double
bottom extends over the length of the engine and boiler rooms, or
rather less than half length.
For the transverse framing, the outer bar is 7 in. x 3 in., and
is continuous from the keel to the protective deck, about 4 ft.
apart. Within the double bottom, flanged bracket plates are
COAL
WATER LINK
The protection is obtained by a protective deck 1J in. on the

flat, 3 in. on the slope, and the disposition of the coal is arranged
to afford as much protection as possible.

The second class cruisers of the Bristol and Weymouth classes
are constructed similarly but of heavier scantlings generally,
having a much greater proportion of length to depth. They are
not sheathed.
1
The second class cruisers of the Arrogant class were intended
and these vessels are not sheathed. The
for use as Fleet Cruisers,
is obtained by a thick deck and side coal bunkers
protection
as usual. An extra longitudinal bulkhead is worked below at
the side, thus continuing the inner skin up to the protective
deck (see Fig. 26).
-THIRD CLASS CRUISER .
FIG. 26. Section of H.M.S. Arrogant. FlQ. 27.
Second class cruisers are the smallest vessels of the Eoyal

Navy in which double bottoms are fitted, and in the remaining
vessels we have to consider, the valuable element of safety pro-
vided by the inner skin has to be dispensed with.
Third Class Cruisers. In these vessels we have a different
set of conditions to consider. There is no double bottom, and the
plating, being of small thickness, requires to be well stiffened.
For such vessels, therefore, the adoption of the longitudinal system
of framing would be unsuitable, and the best method is the
transverse system.The section of such a ship is shown in Fig. 27,
1
Arrogant, Vindictive, Furious.
36 WAR-SHIPS.
and the arrangement of framing is shown in some detail in

Fig. 28. This framing is 24 in. apart throughout. The floor-
plate, 10 Ibs. (J in.), extends
from bilge to bilge. This plate
is slotted out at the middle line to allow the lower bar of the
middle, line, keelson to pass continuously through. The frame
bar, 4 in. x 3 in., is continuous from the middle line to the pro-
tective deck on either side, and the reverse bar, 3 in. x 3 in., is
FIG. 28. Framing third class cruiser.
continuous from the deck on one side to the deck on the other
side, running along the top of the floor- plate as shown. Some
difficulty isexperienced in making the bunker bulkhead water-
tight where the frame, etc., pass through. In order to make a
satisfactory job, the bulkhead is stopped at the reverse bar,
and a dished plate is fitted below between each pair of frames
as shown. The frame above the protective deck is formed of a
4-in. zed bar.
The longitudinal framing consists of a middle line keelson

and two side keelsons on either side. The middle line keelson is
formed by intercostal plates, 15 Ibs. (f in.), between adjacent
frames, connected to the floor- plates by angles 2^ in. x 2.1 in.
LB
'
Ss
-^~
L STEB.I. VV REVERSE B/\R 3*x3
FRAME CONTINUOUS
(PR/WE c M.LroPROTOK FLOOR 10 LBS.
JFLOOR
FLOOR
.VERTICAL
[VjLRTlO; KEEL CONTIMUOUS FRAME BA* 4.3
FIG. 28A. Section of a third class cruiser.
These plates project above the floors and connect to two con-
tinuous angles, 3 in. x 3 in. The lower edges of the intercostal
plates connect to a continuous angle, 3 in. X 3 in. There are in
addition two continuous rider plates, each 9 in. x 12^ Ibs. ( T5în.),
running along the top of the floors. The side keelsons are
intercostal, with a rider plate on top as shown. Below the
38 WAR-SHIPS.
protective deck, before and abaft the machinery space, the frame
is formed as shown in Fig. 28, with a 4-in. zed bar connecting
on to a 10-lb. (J in.) floor-plate. Above the protective deck the
frame is formed of a 4-in. zed bar.
In a more recent third class cruiser, the section of which
is shown in Fig. 2 8 A, greater longitudinal strength was found
necessary because of the greater proportion of length to depth.
The vertical keel in this case is continuous with double angles,
3 in. X 3 top and bottom, the vertical plate being 21 in. X f in.
in. at
The and reverse bar are not continued through the bunker
floor
bulkhead, which enables the watertightness at the bottom to

be more readily obtained. The outer bottom plating is also of
greater thickness than in the previous instance, especially the
sheer strake and the keel plating.
Sloops. A large number of small vessels, about 1000 tons
displacement, called sloops, were formerly employed on foreign
stations. They carry small sail-power, and
are sheathed with wood and copper for the
reasons given above.
A section of such a ship is given in
Fig. 29. One feature of these vessels is
the absence of any protective deck. In the
side bunkers, however, there is a division,
at about the level of the waterline. It is
FIG. 29. Sheathed sloop, intended that the coal should remain in
the upper part of the bunker as long as
possible, in order to serve the purpose of protection, and to assist
in preserving the stability if the side were pierced.
These ships are framed on the transverse system with the
frames 24 in. apart. Below the watertight division in the
bunkers a frame bar, 4 in. x 3 in. (Fig. 30), is worked from side
to side. Between the coal-bunker bulkheads a floor-plate is
worked, 10 Ibs. (J in.), with a reverse bar, 3 in. X 2J in., on the

upper edge. A 3 in. x 2 in. reverse bar is connected to the
frame bar above the bulkhead. Above the bunker division
the frame consists of a 4-in. zed bar. It will be noticed that
the bunker division severs the transverse frame completely, and
to maintain the continuity of the transverse strength bracket
plates are worked as shown.

The middle line keelson (Fig. 30) is formed by an intercostal
plate, 15 Ibs. (f in.), between each pair of frames, with staple
angles, 3 in. x 3 in., connecting to the flat keel and to the floors.
The intercostal plates project above the floors and connect to two
continuous angles, 3 in. x 3 in. An intercostal keelson is also
worked on either side.
Torpedo-boat Destroyers. The essential feature of this

type of vessel is speed, and every effort is made by careful design,
high quality material and careful workmanship, to keep down
FIG. HO. Midship section sheathed sloop
the weight of the hull structure to the lowest amount possible.

Sections of a typical destroyer are given in Figs. 31 and 32, the
first showing the transverse portions of structure, and the second
those portions which give longitudinal strength.

The ship is framed on the transverse system with a frame,
floor, and reverse frame. Deep frames are fitted at intervals.
There is a middle line intercostal keelson with continuous angles
40 WAR-SHIPS.
at the top and bottom. The engine and boiler bearers are so
arranged as to assist materially in providing longitudinal strength
to the structure. Flanging of plates is largely adopted to save
the weight of connecting angle bars. The deck is of special
importance, as it is severely strained if the ship is subjected to
a sagging moment. In this case the deck between the beams is
liable to buckle, being of thin plating, and in order to enable it to
effectively stand the strains, it is well stiffened by fore-and-aft
girders. Deep beams are fitted at intervals.

CHAPTEE IV.
BEAMS, PILLARS, AND DECKS.
Beams. The transverse framing we have been considering ends

at the upper or forecastle decks. To complete the transverse
structure we have learns connecting the sides of the ship together
at the level of the various decks and platforms. Beams nob only
\
ANGLE BULB
FIG. 33.
tie the sides of the ship together, but they form the support to
the decks and platforms. Beams for decks on to which water
is likely to come, as the upper and main decks, are made with a
round down in order that the water may run to the side scuppers.
The amount of this round is 9 in. in a 75-ft. battle-ship, and
42
BEAMS, PILLARS, AND DECKS. 43
6 in. in a 40-ft. cruiser. Beams to the lower protective decks

are of the same shape as the deck, usually level at the middle
line and sloping down to the sides (see Fig. 12). Beams to the
lower platforms and decks are level (see Fig. 19).
Beams in a large ship are spaced every 4 ft. where the frame
spacing is 4 ft., and every 3 ft. at the
ends of the ship. For
a small cruiser the beams are
placed on alternate frames, i.e.
every 4 ft.
Beams are most commonly

formed of angle hulb (c, Fig. 8).
Decks are now covered with
steel plating, and the angle bulb
is a convenient beam to use.
Angle bars are used as beams

to flats in which the greater
strength of the angle bulb is not

required.
The connection of beams to
the transverse frames is of great
importance, as this, together with

the transverse bulkheads, helps
to prevent the racking of the
ship due to rolling. To ensure
an efficient connection, the beam
isconnected to the frame either
by a beam arm, or a bracket
plate. The beam arm is used
where a neat appearance is desir-
able, as in cabins. To form the
beam arm, the beam is cut at
the middle of the web and the FIG. 34.
lower part is bent down. A

piece of plate is then welded in, giving the form shown in Fig. 33.
The bracket is used in places where a neat appearance is not so
desirable (Figs. 25, 30, and 34). The usual depth of the beam
arm or bracket is two and a half times the depth of the beam, so
as to get a good riveted connection to the frame. In the special
case of the beam to the middle deck, a bracket is not necessary,
as sufficient rivets are obtained through the solid plate frame,
which is worked beneath the armour (Fig. 34). Where the
44 WAR-SHIPS.
inner bottom is worked up to the protective deck as in recent

ships the beam is connected to the inner bottom plating by a
flanged bracket.
Half-beams and Carlings. Some beams come in way of
openings in the deck, as the engine hatch, ventilators, funnel
casings, etc. In these cases it is necessary to cut the beams;
these are then termed half-learns. The inner ends of the half-
beams are connected to a fore-and-aft carling, which extends to
\ OPENING
/ \\
/
'HATCH W AY. ^
1
FIG. 35.
the complete beam at each end of the opening. Fig. 35

first
shows thisarrangement for a small opening for a hatchway.

The carling in this case is formed of an angle bulb of the same
size as the beams. It will be noticed how the half-beam hangs
on the carling, which in its turn hangs on the through beams
at the ends.
At large openings the carling is made exceptionally strong,
because of the large number of half-beams hanging on to it.
In Fig. 39 it is formed of a deep 20-lb. (\ in.) plate, the angle
bulb shown is usually omitted. The deep plate also provides a
convenient attachment for the funnel casing. The specially large

openings in the protective deck caused by the spread of the
funnels as they go down to the boilers cut off such a large number
of beams that some compensation has to be provided. This is
arranged for by running

across deep plate girders
at intervals.
The weather deck needs
to be strengthened in way
of the blast of the heavy
guns. This is arranged for by

working fore-and-aft girders,
well supported by pillars,
under the deck (see Fig. 5 A).
Pillars. It is essential
that the long beams in a
ship should be supported at
other places than at their
ends, as, by thus supporting
them, their capacity for
bearing a load is
greatly
increased. In ships of the
Royal Navy nearly every
beam is supported by pillars
or in some other way. Where
possible, the fore-and-aft

bulkheads are arranged so
Fio
that the stiffening bars to
the bulkhead act also as supports to the beams. One instance
of this be noticed in the cabin bulkheads below the upper
may
deck in a battle-ship. In other places pillars are fitted. These
pillars are made of wrought steel tubes, a usual size for the 'tween-
decks being 5 in. diameter and in. thick.
A considerable saving of weight results from using hollow pillars instead of

solid, as well as the advantage of the greater stiffness of the hollow form of larger
diameter. Thus, suppose a pillar if hollow could be 6 in. diameter and in.
thick,and if solid 4| in. diameter (these two pillars in merchant shipbuilding

would be considered of equal strength), the saving of weight by using the hollo\r
form would be over a ton for every 100 ft. of pillar worked into the ship.
Pillars must be secured at the top and bottom so as to form

a tie, between the decks as well as a strut (Fig. 36). Pillars are
WAR-SHIPS.
so arranged as to form, with the bulkheads, continuous vertical

lines of support right
down to the bottom of
the ship, The heel of
the pillar should be
riveted direct to the
beam, and not bolted to
the wood deck (if fitted).
This latter may be done
when the pillar sup-
ports the skid beams, and
is thus nofc fitted to
assist the structural
strength of the ship.

Pillars have usually been
formed as shown in Fig.
36, with a solid head and
heel on to which the
tube is welded. The
type of construction in-
dicated, however, in Fig.
3 6 A, is now coming into
use, being cheaper in con-
struction and involving
FIG. 36x.
less weight. The figure
indicates a 7-in. pillar.
The head is formed by
pressing the tubular end
on to a piece of 1-in.
plate, and the whole is
then cut off to the shape

indicated, large enough
to take five rivets
through the flange of the

beam. The heel is formed
by simply riveting the
tube on to a circular
forged angle-bar. The
FIG. 37.
connection to the deck
and beam below is by means of five rivets, three through the deck
and two through the beam.
In many cases where pillars are necessary, the obstruction

caused very inconvenient, as in
is way of capstan gear, torpedo-
tubes, etc. In such cases the
pillars are made portable. The
head is fitted with a pin (Fig. 37),
so that the pillar can be hinged
up clear, or, if desired, removed

altogether for the time. Figs.
38 and 38A show two forms of
heel fitting ;
in the first case
there an inconvenient obstruc-
is
tion left, in the second this heel
fitting is also made portable.

PORTABLE^
Deck Plating. Plating is
worked on decks and flats for

FIG. 38. Heel Fio. 38x. Heel
various reasons, (a) for the pur-
fitting to port- fitting to portable
able pillar.
pose of contributing to the pillar.
structural strength, as on the

upper deck (b) for the purpose of protection, as on the main and
;
middle decks in Fig. 12 (c) to divide the ship into watertight

;
compartments, as in the hold forward and aft (d) to distribute ;
the strains on the deck, as under bollards, boat-hoists, etc.

The Upper or Forecastle Deck is a most important part of
the structure, especially in a vessel of large proportion of length to
depth as a cruiser. It forms the upper flange of the girder, and
so contributes materially to the ship's structural strength. Fig. 39
gives the arrangement of the upper deck plating amidships in such

a ship. The side stringer has a total breadth of 10 ft., worked in
two strakes of 20-lb. (^ in.) high tensile steel. There is also a
funnel stringer 5 ft. wide, also of 20-lb. high tensile steel. This
runs fore and aft along the funnel and engine hatches as shown.
The remainder of the plating is 10-lb. (J in.) mild steel. The
plating of this deck is laid direct on to the beams, with single
riveted edge strips on the upper side connecting the edges
together. The butts are all double riveted ;
those of the 20-lb.
plating being double, i.e. on both Holes that
sides of the plate.
are necessary in the deck for coal shoots, hand-ups, etc., are care-
fully compensated for by pieces of plate on either side of the hole
to make up the sectional area cut away. The butts of the various
etrakes of plating are carefully shifted clear of one another, and
clear of the butts of the sheer strake (see Chapter V.).
WAR SHIPS.
The above instance has been taken to show the care taken to
render the upper deck specially strong when such strength is
FUNNEL^
OPENING,
DETAIL or
CARLING
FIG. 39. Upper deck plating of cruiser.
of length to
required, but in many cases, where the proportion
is not so this strength is not necessary.
depth great, exceptional
In such ships, as in Majestic, the plating does not completely

cover the deck, the most important part being the stringer next
the side. We
also find that only such a stringer has been
worked at the ends of
many ships.
In the most recent
ships, however, decks
are nearly always com-
pletely plated ;
when
wood is worked above
as for the weather deck,
there is less likelihood
of the wood catching
fire with the steel under-
neath. For other decks
we have linoleum, and
this necessitates a com-
plete steel deck to re-

ceive it.
In some large ships

armour has been worked
between the main and
upper decks, and this
is covered in by a pro-
tective deck, about 1 in.
thick. This deck is
very advantageously
situated to assist in
EDGE
the ship's structural CONNECTION.
strength.
With the method
adopted of mounting
all the guns above the I
weather deck, large Fi<*. 40. Riveting in protective deck.

holes have to be cut
through the deck plating to pass the armoured barbettes through.
In consequence of this, the plating between the barbettes and
alongside the funnel casings has to be made specially thick in
order to obtain sufficient strength, and the riveting at the butts
and edges has to be of a special character to develop the full
strength of the plating.
5o
WAR-SHIPS.
Protective Decks. In decks fitted for the purpose of protec-

tion the plating is usually worked in two or more thicknesses.
Thus in the Duncan the main deck is 2 in., worked in two thick-
nesses of 40 Ibs. The middle deckis 1 in., and here two 20-lb.
plates are worked. In this

way no edge strips or butt-straps are
necessary, as each thickness acts as security for the edges and butts
of the other thickness. The lower thickness only is riveted to
the beams ;
the general arrangement of the riveting is shown in
Fig. 40.
Where such a deck is used for purposes of structural strength,
butt-straps are necessary to develop the full strength of the
plating, and the edges are separated from each other.
Platforms, etc. The plating
of platforms and flats is usually
of 10-lb. ( in.) plating, with both edges and butts lapped and
single riveted. For compartments where a good foothold is
desirable, and where there is a lot of heavy wear and tear, the
FIG. 41.
flat is formed of 15-lb. ribbed plating. In a large ship the flats of

the submerged topedo-rooms, auxiliary machinery compartment,
etc., are fitted in this way.
Watertightness of Flats, etc. In watertight platforms, etc.,

the connection of the plating with the ship's side must be made
watertight. Fig. 41 shows two methods of doing this, one when
the zed bar frame passes through, and the other when a simple
angle bar passes through. In either case angle bars are smithed
round to cover the holes completely the riveting is closely spaced
;
and the whole carefully caulked and made watertight. The

such a 20 ft. below water, would have to stand if
pressure flat,
the compartment below were bilged amounts to about 60 tons per

100 square ft., so that substantial work is seen to be very necessary.
There is no difficulty in securing a watertight connection with the
side at the main or middle decks of an armoured ship, if the
transverse framing is not continuous through the deck, and a fore-
and-aft angle can be run along and caulked (Fig. 34).
Wood Decks. The use of wood decks has been very much
reduced in recent years in vessels of the Royal Navy, on account
of the probability of the wood catching fire in action. Wood is now
only used for weather decks. Formerly it was fitted in Admiral's

apartments and the ward room, in casemates and ammunition passages,
and for magazine flats. In the Admiral's apartments and ward
room the flat was of Dantzic fir, 2 in. thick. In the other cases
the flat was of teak, on account of the suitability of this timber to
stand heavy wear and tear. For the weather decks the thickness is
Fio. 4'2A.
generally 3 in., with planks called waterways round the edges of

the decks, forward and aft, and round barbettes, etc. Thicker planks
are worked in way of capstans, etc., to obtain a level surface.
In casemates the teak is 2 in., and in ammunition passages 1 in.
When a steel deck is laid, the wood deck is fastened by
through bolts to the plating between the beams. To decks which
have to stand severe compressive strains, the connection of the
wood deck to the steel deck is made a very efficient one, so that
the wood and the steel may act together in resisting the buckling
(see Fig. 42A).

WAR-SHIPS.
When, however, as is sometimes the case, thewood deck has

to be laid direct on to the beams, the bolts must be taken through
the upper flange of the beam. For this reason, when a wood deck
is thus worked, it is desirable to make the beam of tee bulb, so
placed zig-zag, as in Fig. 42B, and not
that the bolts may be in
one continuous line, as would be necessary with an angle bulb

beam. When thus laid direct on the beam the bolts of the butt
of deck plank would be too close if taken through the flange
re o3^
ur
'
Fio. 42n FIG. 43.
of the beam, and at each butt, therefore, a short piece of plate,
having the width of the plank, is fitted to take the bolts (Figs.
42B and 43).

Planking is fastened to the steel deck or the beams by means
of galvanized iron bolts with round heads, as Fig. 44. The heads
are let in well below the surface of the deck, and the cavity is
filled in with a wood plug called a dowel, well steeped in white
lead. A hempen grummet is placed under the head, and the nut
Fio. 44.
underneath has a plate washer and grummet (Fig. 43). All this
is necessary to make the hole in the deck It will
properly tight.
be noticed that the neck of the bolt is square to prevent the bolt
turning when the nut is being screwed up.
The edges and butts of deck planking are caulked to malce
the deck watertight. The edges and butts are left with a slight
opening at the surface, as Fig. 43, and into this oakum is forced,
filling the opening right down to the bottom of the seam. The top
is then payed with pitch. When carefully done this caulking

should make the deck properly tight.
Linoleum. In the living spaces of the ship wood decks are
" "
not fitted, but a thick linoleum called corticine is laid direct
on the surface of the steel deck. In cabins and similar places

" "
cork carpet employed. The steel deck has to be laid flush for
is
this purpose, with edge strips on the under side worked between
the beams. The upper deck of torpedo-boat destroyers is also
laid with linoleum.
The linoleum is secured to the deck by means of a mixture of
three parts resin and one part tallow, and when secured the edges
are stopped with putty. Eound scuttles and exposed edges the
linoleum is secured in addition to the deck by galvanized iron
strips screwed iuto the plating. For thin decks, as in destroyers,
these strips are secured to the plating by bolts with a nut below
having a washer and grummet. Where heavy wear takes place

on a steel deck covered with linoleum, as at the foot of ladders,
the linoleum is covered with plates of ribbed iron. When laid
the corticine on decks is coated with a mixture composed of 1 Ib.
of orange shellac to \ Ib. of dry Venetian red and 1J pints of

methylated spirit.
Watertight Hatches. The hatches to ordinary flats and
decks are fitted with a raised coaming and a steel cover with an
indiarubber joint. This indiarubber should be periodically
examined, and where perished should be renewed, as a defective
cover may detroy the watertightness of a complete deck. Scuttles
and armour gratings to thick decks are dealt with in Chapter XIII.
CHAPTER V.
PLATING OF THE OUTER AND INNER BOTTOMS.
Outer Bottom Plating. The weight of this plating forms a

good proportion of the total weight of the hull structure. It is
a most important portion of the structure, because it not only
contributes largely to the structural strength, but it keeps the
ship watertight. One advantage of forming this plating of mild
steel has already been referred
to, viz. the advantage of the ductility
of steel as compared with iron. Steel ships have frequently
grounded without making water, under circumstances in which an
iron ship would have been in a serious condition owing to the
rupture of the plating.
Shift of Butts. An important point in connection with the
longitudinal structure of any ship is the arrangement of a good
a*. FEET
|-
I
L
Mli- T H-^
'
Tri;no
FIG. 45. Shift of butts.
shift of butts. A butt must be a place of relative weakness, and

the butts of the various portions of the fore-and-aft structure are
arranged well clear of each other. Thus, for the outer bottom
it is laid down that butts are not to be closer
plating, together in
54
PLATING OF THE OUTER AND INNER BOTTOMS. 55
the same frame space than two passing strakes. With plates as
WAR-SHIPS.
now worked, about 32 ft. in length, there is no difficulty in
considerably exceeding this condition. Thus, in Fig. 45, 24-ft.

plates are worked and
five passing strakes are
obtained. The same
principle has to be borne
in mind when arranging
the butts of the vertical
keel and longitudinals,
inner bottom, etc., so as
to avoid any weakness
of the structure in any
one frame space (see
Fig. 46).
Plating of a Battle-
ship. The outer bottom
plating of the older
battle-ships was 25 Ibs.
FIG. 46A. Special form of FIG. 47.
bulkhead liner. (f in.) over the length
of the double bottom,
and 20 Ibs. (J in.) at the ends. In more recent ships, for reasons
of structural strength, the thickness is 40 Ibs. amidships with
treble riveted straps. The fore-and-aft rows of plating, called
strakes, are riveted to one another by double riveted laps. The
plating is worked on the raised and sunken system as shown in
Fig. 47. At every frame in way of each outside strake a filling
piece or liner is necessary to get good riveting between the frame
and the plate. These liners are often lightened, as shown, by oval
holes between the rivets. At the middle line an additional outside
plate, 30 Ibs. (| in.), is worked, called the outer flat keel, the inner
plate being the inner flat keel (see Fig. 3).
Riveting in Outer Bottom. Fig. 46 shows in some detail
the arrangement of the riveting, etc., in the outer bottom plating
near the middle line. The lower angles of the vertical keel are
connected to the flat keels by 1-in. rivets, and as the vertical keel
has to be watertight, these rivets on one side are closely spaced,
viz. 4J to 5 diameters. On the other side a wider spacing, viz.
8 diameters, is all that is necessary. Close spacing is necessary
for the watertight longitudinal, No. 2, and wide spacing for the
non-watertight longitudinal, No. 1. The butt-straps f the inner
and outer fiat keel are treble riveted with 1-in. rivets, and extend
from the keel angles to the edges of the plate concerned. The
edge of the inner keel is connected to the outer keel by a
single
row of rivets, to get a good connection and to allow the edge to be
caulked. The edge riveting of the outer flat keel is double riveted
with The remainder of the riveting is J- in. diameter,
1-in. rivets.
as shown. The butt-straps are double riveted, and for the outside
strakes extend from the longitudinals to the edge of the inside
strakes. For the inside strake the strap extends the whole width
of the plate.
The ordinary frames are connected to the outer bottom by
rivets spaced 7 to 8 diameters. For the watertight frames it is
I -in.
necessary to have the rivets closely spaced, viz. 4J to 5 diameters.

This close riveting cuts away a lot of material from the outer
bottom plating in one transverse section of the ship, and causes
this section to be a distinct line of weakness. In order to
compensate for this, a wide liner called a bulkhead liner is fitted
to each outside strake instead of the ordinary liner. This wide
liner forms a sort of strap over the weak place, and in this
way the strength at the watertight frame can be brought up to

an equality with that at an ordinary frame. In the case of the
second strake from the keel in Fig. 46, it will be noticed that
the watertight frame comes next to a butt-strap, and the strap
is made wide
enough to act as a bulkhead liner and butt-strap
combined.
Fig. 46A shows a form of bulkhead liner adopted in some ships
to economize weight. The outside strake is only pierced at every
other rivet, so that the reduction of strength is not so great at the
section, and the liner can be smaller than would otherwise be
necessary.
The side plating above the protective deck is recessed back
from the side of the ship to make room for the armour and backing
(see Fig. 18). This plating was formerly in two thicknesses, each
of 20 Ibs. ( in.), no butt-straps or edge strips being necessary for
this plating, as each thickness acts as security to the edges and
butts of the other thickness. This plating is now worked flush
in one thickness as Fig. 18 A, and butt-straps and edge strips are
provided so that the plating contributes properly to the structural
strength. Above the armour the plating is 20 Ibs. (J in.) to the
upper deck, except where an armoured battery is worked between
the main and upper decks, as Fig. 13, in which case the plating
behind armour is carried right up as in Fig. ISA.
WAR-SHIPS.
The side plating is doubled in way of any protective plating

to form a flush surface. Thicker or doubling plates are also
worked in way of the stem and sternpost, in way of hawse-pipes,
and where necessary to take the chafe of anchors and cables, and
in other places where special local strength is required, as in wake
of torpedo tubes. Covering plates are worked at the lower and
upper edges of the side armour, as Figs. 18 and 34, connected to
the armour by tap rivets. The edges of these plates are caulked
so as to make the
joint at the armour
watertight.
some battle-
In
ships and cruisers the
topside above the

upper deck has been
worked to the boat
deck, forming a
closed-in super-
structure (see
Fig. 22). This plat-
ing is not intended
to take any structural
strains, and the sides

are severed by the
FIG. 48. Watertight manhole. gangway ports. The
plating of the boat
deck is also severed, and a sliding joint is made as shown
in Fig. 5.
Inner Bottom. The inner bottom of a battleship is gene-
rally 15 Ibs. (f in.), with the middle line strake 20 Ibs. (^ in.),
unless thicker plating is necessary for structural reasons. This
plating extends to the fourth longitudinal, and the inner skin is
continued to the protective deck by means of the wing bulkhead,
which is 15 Ibs. 1 It is most important that convenient access
should be provided to all the watertight compartments of the
double bottom, in order that the condition of the plating, etc.,
may be periodically examined. The best arrangement would be to
have the two manholes at opposite corners of the compartment, so
that the space might be well ventilated before entering. Owing
1
In some ships the inner bottom is carried right up to the protective
deck.
however, to the obstructions caused by the boiler bearers, etc., this

is not generally possible an actual arrangement adopted in one
;
ship is shown in Fig. 49. The manholes are on raised coamings,

with a hinged cover, secured by a number of butterfly nuts as
shown in Fig. 48. In some cases a hinged cover cannot be obtained,
owing to some obstruction, and in these cases the cover is con-
nected to the coaming by a chain. A
plug is fitted to the manhole
cover to enable the compartment to be sounded when necessary.
SECTION AT A B
FIG. 48x.
This ping is hollow, as shown in Fig. 48, with a small vent-hole

at the side. When the plug is slightly unscrewed, this hole allows
the air to escape when flooding, and will also show when the com-
partment is full. The plug can then readily be screwed up tight.
It
to escape so that
is very necessary that the air should be allowed
the tank may be completely filled. At other places, as on watertight
flats or bulkheads, where occasional access through is necessary
for inspection, the edge of the hole is stiffened by a strip, and a
as shown in Fig 50. Manholes
portable plate is secured by studs
are now formed as shown in Figs. 48A and 50A, the cover being
6o WAR-SHIPS.
connected to the coaming with a chain. For an ordinary compart-

ment a red lead joint is used, for F.W. tanks indiarubber, for
FIG. 49. Manholes in inner bottom. Fig. 50. Watertight manhole on

bulkhead.
reserve feed tanks asbestos cloth, and for the oil fuel compart-
ments liquored leather or dexine is used. Air escape plugs are not
necessary in the case of oil fuel compartments, pipes being led into
the open for this purpose. In the case of the manholes to oil fuel
compartments the nuts securing the cover are circular with a slot
requiring a special spanner to open, and with this fitting locks are
not required as would otherwise be the case.
We have already noticed that although the inner bottom proper
ends about one-sixth the length from each end, yet a virtual inner
bottom is obtained well to the ends by the watertight flats, etc.
PLATE COVER
STIFFENING STRIP
FIG. 50A.
These double bottom spaces are carefully tested for watertight-

ness by filling with water, the pipe conveying the water giving
a head of about 5 ft. above the L.W.L. If the test is not satis-
factory, the defects have to be made good and the compartment

retested.
Outer and Inner Bottom Plating of a First Class Cruiser.
This plating is arranged on the same principles as for a battle-ship,
the thickness depending on the structural strength necessary.
Special attention is paid to the structure at the keel and at the
upper and forecastle deck. Fig. 3 shows the structure at the keel
of a cruiser whose ratio of length to depth is 12'4. This is shown
in comparison with the structure at the keel of a battle-ship of
greater displacement whose ratio of length to depth is only 9 '4. It

will be noticed that the vertical keel is deeper, and the outer flat
keel is 45 Ibs. (nearly lîn.), as against 30 Ibs. (f in.) in the battle-
ship. The middle strake of the inner bottom is 25 Ibs. as against
20 Ibs. inthe battle-ship. Fig. 51 shows the sheer strake of this
cruiser, i.e. the plate of side next the upper deck. It is of 25-lb.
high tensile steel, and the butt-strap is 30 Ibs., quadruple riveted.
1
The inner bottom plating of a large cruiser presents no special

features, and the remarks made above apply in this case also. It
TOPSIDE . IS LBSL
rr^r
FIG. 51. Sheer strake of cruiser.
has already been noticed that the wing bulkhead has been
coal capacity.
dispensed with in some ships to give a greater
52 and 53 show the whole of the watertight subdivision of a
Figs.
large cruiser,in which it will be noticed that an inner skin is
obtained well towards the ends by means of the flats, etc., to the
magazines and store-rooms.

Plating of a Second Class Cruiser. The outer bottom
plating of the cruiser shown in Fig.
25 is generally 17 Ibs.
The flat keel is 25 Ibs. (f in.), and the sheer strake

(^ in.).
5
25 Ibs. The inner bottom is 12 Ibs. ( T 5 in.).
1
In some very long ships it has been necessary to form the upper portion of
the side plating in two thicknesses (say 30 Ibs. each), and in such cases it is found
necessary to work double butt-straps, double riveted on the outside and treble
riveted on the inside, as E, Fig. 10.
62 WAR-SHIPS.
Plating of a Third Class Cruiser. The bottom plating of

the cruiser shown in Fig. 28 is generally 12 Ibs. ( T5^ in.). In
way of the engine-room, however, it is 17J Ibs. (^ in.), to stiffen
the ship in way of the fast-running machinery. The flat keel and
sheer strake, next the upper deck, are both 20 Ibs. (J in.). These
are both reduced to 17^ Ibs. beyond the half length.
The cruiser whose section is given in Fig. 28A was con-
siderably longer than that given in Fig. 28, and it will be
observed how much greater was the strength of the shell plating
provided.
Plating of Sloop. The bottom plating of the sloop shown
in Fig. 30 is generally 10 Ibs. (J in.), with flat keel and sheer
strake of 15 Ibs. (|
NO in.).
/
Plating of Destroyer.For this type of vessel high tensile

steel isused for the outer bottom plating in the more recent
ships. The flat keel in Fig. 32 is 36 in. x 10 Ibs. ( in.),
sheer strake 30 in. x 8J Ibs., the remainder of the plating
being 6J Ibs.
For a more recent destroyer than that given in Figs. 31 and
32 the following is a general statement of the scantlings :
Flat keel 42 in. X 17 Ibs. H.H.T., 10 Ibs. forward, 12J Ibs.

aft.
Garboard15 Ibs. H.H.T., 10 Ibs. forward and aft.

Plating to sheer strake 12% Ibs., 10 Ibs., 8 Ibs. H.H.T., 8 Ibs.,
7 Ibs., 7 Ibs. forward and aft.
Sheer strake 4:5 in. x 12J Ibs. H.H.T. Forward and aft
8 Ibs.
Stringer 15 Ibs. H.H.T.

Stringer angle 3J in. X 3J in. X 8 Ibs. H.H.T.
Deck 12% Ibs. H.H.T.
Frames 4 in. x 2 in. x 2 in. x 6 Ibs., Z bars, 21 in. apart.
Floor plates 7 J Ibs.
Special care is taken to incorporate the engine and boiler
bearers into the general structure of the ship, so that they may
contribute to the structural strength.
Fore and aft girders are worked beneath the deck, to give
longitudinal strength and to support the deck plating against
buckling tendencies, and fore and aft stringers and keelsons
are worked to support the outer bottom plating in a similar
manner.
The middle line keel is continuous, and formed of plating
18 in. deep X 15 Ibs. H.H.T., with continuous angles at the

bottom 3 in. X 3 in. 7 Ibs. H.H.T., and at the top 2J in. x
x
2J in. X 5 Ibs. H.H.T., with a rider plate on the top of the
floors.
FIG. 52. Watertight subdivision of a battleship.

From "MoDEBN WABSUIPS" {Cambridge University Press').
CHAPTER VI.
WATERTIGHT BULKHEADS, DOORS, ETC.
THERE are four main methods of watertight subdivision, viz.

by
means of
(i.)
A watertight inner bottom with watertight vertical keel,
longitudinals, and frames ;
(ii.) Watertight decks and flats ;
(iii) Transverse bulkheads ;

and
(iv.) Longitudinal bulkheads.
Wehave already dealt with the first two of these. The
valuable feature of a double bottom has to be dispensed with in
small vessels on account of the space thus occupied. In all ships,
however, we get watertight subdivision from the last three of the
above (see Figs. 52 and 53).
We now deal with the bulkheads. These are not only useful,
in dividing the ship into anumber of watertight compartments,
but they form a most valuable addition to the ship's structural
strength.
Transverse Bulkheads. These are watertight partitions which
go transversely across the ship. Fig. 52 shows the large number
of such bulkheads fitted in a battle-ship. The one near the stem,
extending to the forecastle deck, is the collision bulkhead, and
many have occurred, especially in merchant vessels,
instances
in which, after collision, this bulkhead has remained intact
and saved the ship from possible foundering. On account of
its importance it is well stiffened, and in recent ships no openings
of any kind are allowed in it. Any access required to the

forward side must be by means of scuttles through the decks,
and if the forward
space does not require draining, it must be
pumped out by means of a hose.
In some battle-ships an additional bulkhead has been fitted
"
3 ft. abaft the collision bulkhead. This is termed the " cofferdam
65 F
66 WAR-SHIPS.
bulkhead, and the 3 ft. space thus formed is intended to be

packed, like an ordinary cofferdam, before ramming, to limit the
flow of water aft, supposing the collision bulkhead to be damaged
(see Fig. 65).
A governing feature in the construction of any bulkhead is the
area and depth of unsupported plating likely to be exposed to
water pressure. The transverse bulkheads forward and aft of the
machinery spaces are well supported by the decks and flats, and
so do not require any extensive stiffening, as the unsupported
area is not great. Such bulkheads are usually formed of 104b.
(J in.) plating, stiffened with angles 3 J in. x 2 \ in., worked
vertically, the spacing varying from 2 to 2 J ft. The plating is lap
jointed at edges and butts, and single riveted. Bulkheads forming
the sides of magazines in which teak lining is fitted are stiffened by
fiTT*^
S *. 4J4 TEE BAR
FIG. 54 Stiffening to bulkhead.
zed bars 3 in. deep. This is not now necessary, because of the
omission of the lining to magazines.
The collisionbulkhead is formed of 15-lb. (f in.) plating,
stiffened by zeds and 3J in. x 2J in. angles spaced alter-
5-in.
nately 2 ft. apart. Above the protective deck the angles only
are fitted.
The transverse bulkheads forming the divisions between the
engine and boiler-rooms are specially constructed and stiffened
because of the very large area and depth of unsupported plating.
(In one case 46 ft. wide and 25 ft. deep.) The plating is 15 Ib.
(| in.), worked flush jointed, the horizontal joints being covered
with a tee bar 4J in. x 5 in. forming the edge strip. The vertical
butt- straps on the opposite
joints are covered with single-riveted
side to the tee bars. The main stiffening is worked vertical, and
is formed of 5-in. zeds every 4 ft., with 3J in. x 2J in. angles
between. At intervals of about 8 or 12 ft. the zeds are replaced
by I bars 12 in. X 6 in. X

worked on both sides of the bulk-
6 in.,
head, the tee bars being cut and connected to them (Fig. 54).
WATERTIGHT BULKHEADS, DOORS, ETC. 67
In more recent ships the plating is lapped at both edges and
LAP
-5 ZED 12" H BAR
FIG. 55. Stiffening to a main bulkhead.
and T bars are MI DOLE DECK

butts,
not required. Fig. 55
shows one construction.
The 12-in. H bars are
worked on both sides
of the bulkhead 8 ft.
apart with intermediate

5-in. zeds. The hori-
zontal laps of plating
have 5-in. zeds worked
on them as shown. In
other ships the bars H
on both sides are 12 ft.
apart, and 5-in. verti-
cal zeds 2 ft. apart are
worked with no hori-
zontal stiffening. Both
the zed and I bars are
well supported at the
head and heel by bracket
plates, as shown in Fig.
55A.
This extensive
stiffening has been found
necessary to enable the
bulkheads to withstand SECTION AT A.B.
the great pressure that IBULKHEAD

would exist supposing
1
FLANGE OF
FLANG
one of the adjacent com- H BAR CUT
partments filled with FIG. 55A. Brackets to H stiffeners.
water. It is the practice
in each ship while building to actually fill a boiler-room and an
engine-room with water to a height of 12 ft. above the load

68 WAR- SHIPS.
waterline, in order to test the strength of the bulkhead. The

safety or control of a ship might very conceivably depend on
one of these main bulkheads remaining intact if an engine-room
or boiler-room were flooded.
The bulkheads forming the fore end of the fore boiler-room
and the after end of the engine-room do not need this extensive
stiffening, because of the support received from the decks and
platforms. In these cases the 12-in. I stiffeners are not fitted.
The transverse bulkheads in way of the inner bottom are
bounded thereby, and the watertightness is continued to the outer
bottom by means of the watertight frames already considered.
Between the main transverse bulkheads, divisional bulkheads
are fitted in the side upper and lower bunkers, as seen in Fig. 53.
Where a watertight flat is fitted in the upper bunkers of battle-
ships, as shown in Figs. 12 and 13, a further set of bulkheads is
fitted below the flat, thus giving most minute subdivision to the
side in the neighbourhood of the waterline.
Thefore-and-aft longitudinals, vertical keel, etc., are worked
continuously through all the transverse bulkheads in order to

maintain continuity of the longitudinal
strength. Where the longitudinal takes
the form of a zed and angle, as at the ends
of a ship above the floor plate, the water-
tightness is secured by working angle

collars round, as shown
in Fig. 56.
It will be noticed in Fig. 53 that a
number of the bulkheads are carried right
up to the weather deck. This is impor-
tant in view of the and
sinkage, heel,
change of trim that might ensue after
damage. The bulkheads being carried well above water, there is
more likelihood of confining the water on one side.
Longitudinal Bulkheads. There are a number of small
longitudinal bulkheads forming the boundaries of magazines, etc.
These assist in maintaining the watertight subdivision, but being
only of small area are of 10 Ib. (J in.) only, with 3J in. x 2J in.
angle stiffeners. For the sides of magazines with teak lining 3-in.
zeds were formerly used as the stiffeners.
The principal fore-and-aft bulkheads in a battle-ship are
(i.) Longitudinal engine-room bulkheads,
(ii.) Inner coal-bunker bulkhead.
(iii.) Outer coal-bunker or wing bulkhead.

(iv.) Upper coal-bunker bulkhead.
These are all shown in Fig. 53. We have noticed that in some
cruisers (Fig. 23) the two latter bulkheads have been omitted to in-
crease the coal capacity and to make the transport of coal more easy.
(i.) Engine-room bulkheads. These bulkheads extend for the

whole length of the engine-room, and are taken to the height of
the main deck. In battle-ships this is only necessary in the
ventilators, etc. but in large cruisers, where the protective deck
;
has to be lifted to the main deck, the bulkhead is carried right

up to this deck. The bulkhead is carried to this height in order
to prevent water flowing over the top in case one engine-room
was flooded. It is important that this bulkhead should be amply
strong in view of the severe strains that would come upon it if
one engine-room were flooded. On this account it is of 12J Ib.
(^ in.), and constructed
and stiffened in a similar manner to the
main transverse bulkheads.
(ii.)
Inner coal-bunker bulkhead. This bulkhead is well sup-
ported by the divisional bulkheads in the coal bunkers, and is
not so strongly stiffened as in the previous case. The plating is
lap-butted, lap-jointed, single-riveted; the two lower strakes are
12 \ Ib. (${ in.), and the remainder 10 Ib. (\ in.). The stiffeners are
5 -in. zeds at each beam, connected above and to the inner bottom
by bracket plates. Angles 3J in. X 2J in. are worked between.

(iii.) Wing bulkhead. This bulkhead when forming a virtual
continuation of the inner bottom is formed of 15-lb. (| in.) plating,
stiffened the same as the inner bulkhead.
Upper coal-bunker bulkhead. This bulkhead, only being
(iv.)
from deck to deck and not likely to have to stand any great pressure
of water, is of 10-lb. ( \ in.) plating, stiffened by angles 3| in. x 2| in.
worked 2 ft. apart.
The beams of the ship run continuously through the longi-
tudinal bulkheads, and in order to make the upper portion
smithed fit in the space between the
watertight, angle bars are to
beams, as Fig. 57. The riveting is closely spaced and the whole
carefully caulked. Another method that is now being adopted is
shown in Fig 5?A. The bulkhead between consecutive beams is
formed of a single flanged plate as shown, projecting sufficiently

below the beams to lap on to the upper strake of bulkhead plating.
The bulkheads underneath barbettes are specially stiffened to
efficiently support the barbette, etc.
70 WAR-SHIPS.
Numbering of Bulkheads. Transverse bulkheads are dis-

tinguished by the number station at which they
of the frame
come. Longitudinal bulkheads are distinguished by the numbers
of the frames at which each begins and ends. Thus, bulkhead
116 to 140 would be the bulkhead extending between frames 116
BEAM DECK
BULKHEAD
FIG. 57.
and 140. Formerly bulkheads were distinguished by letters, but

it has been found more convenient to use numbers instead.
SECTION BEAM _ DECK
FLANGED PLATE
BULKHEAD PLATING
FIG. 57A.
Water-testing". All watertight flats and bulkheads, outer

bottom, inner bottom, and decks are tested for watertightness
under water pressure. Wherever practicable, the compartments
are actually filled with water, the head of water now applied being
12 ft. above the load waterline. Otherwise the watertightness is
tested by means of a hose. Usually one of the boiler-rooms and
one of the engine-rooms are filled with water to a height of 12 ft.
above the load waterline, in order to test the strength of the
bulkheads, as well as the watertightness.
A
large number of holes are necessary in a bulkhead for the
purpose of connecting fittings, etc., to the bulkhead. It is
important to note that if any such fittings are removed for any
purpose, the holes left behind should be properly filled up again
by a tap rivet. Otherwise the watertightness of the bulkhead is
destroyed. The glands on bulkheads, where voice-pipes, electric
wires, pass through, should be periodically examined to see
etc.,
that the screws for attachment have not worked out.
WATERTIGHT BULKHEADS, DOORS, ETC.
An Admiralty circular was issued in 1904 directing that

one compartment is to be flooded in each ship at least once a
year, to test the watertightness of the bulkheads, doors, etc.
(see S. 32111/1903, January 29, 1904).
Air-pressure Test. After the water-pressure tests referred
to a number of holes have to be made in the bulkheads for the
attachment of fittings, electric leads, etc. (the latter, however, at
HINGE
FIG. 58. Hinged watertight door.
any rate bulkheads below water, are usually secured to light

for
plating, to obviate piercing bulkheads). In order to ensure that

all such holes are properly filled up, a rigid air-pressure test is
specified. Two selected compartments are tested by air pressure

shortly before the completion of the ship, the pressure of air
applied being about equal to that obtained during the water-
pressure Any leakage is at once made evident by the failure
test.
to keep up the pressure when the supply from the air-pump is

shut off. If any leakage is discovered, the compartments are
72 WAR-SHIPS.
again tested after the defects discovered are

made good, and power
is reserved under the contract to similarly test any or all the
other compartments of the ship.
Watertight Doors. The transverse and longitudinal bulk-
heads being watertight, it is necessary that openings in them
should be capable of being made watertight. These openings
are made as few in number as possible.
The most recent practice is to dispense altogether with doors
in the most important bulkheads. Communication between the
engine and boiler-rooms is facilitated by means of electric lifts
running up to the main deck. There are also scuttles fitted in
the bulkheads for observation purpose.
Watertight doors are of three kinds, viz. hinged, vertical
sliding,and horizontal sliding.

Hinged Doors. This type of door is the most common in
H.M. ships, being fitted to the less important bulkheads below
water, and to nearly all the watertight bulkheads above water.
(In a few cases above water, a hinged door would be inconvenient,
and a sliding door is fitted.) No provision is made for closing
these hinged doors from above, they must be closed at the door itself.
The opening made in the bulkhead is stiffened round by an
angle bar frame (Fig. 58). The door is formed by a piece of plate
of the same thickness as the bulkhead this also has a stiffening
;
"
bar round the edge. In " Median's patent door this stiffening
is obtained by embossing the edge to the shape shown, giving us a
lighter door. India-rubber is worked all round the edge of the

door as shown, which engages with the stiffening bar on the
bulkhead. The door is forced home to the rubber by means of
handles, each of which presses against a wedge-shaped piece of
metal on the door. Spring clips are placed to keep the handles
away from the door when not in use. The hinges are made with
an elongated hole for the pin, so that a certain amount of play is
allowed to force the door home. The handles pass through a
metal collar on the bulkhead, and watertightness is obtained by
a leather washer. The india-rubber on the door, when perished,
can be readily removed by means of the securing strips, which
screw into the door. This india-rubber must never be painted or
greased. It should be chalked to prevent sticking. The leather
washers on the door clips should be periodically examined. No
tread over the sill should be allowed, which will prevent the door
from closing.
Vertical Sliding Doors. These doors are fitted to bulkheads

where it is necessary to close from above, and where sufficient
head room is possible above the opening to take the door when
open. They were formerly of two kinds (1) ordi-
nary, and (2) quick closing, the latter being fitted
on the bulkheads between the engine and boiler-
rooms. Quick-closing doors were formerly fitted
with balance weights, to render the closing easy
and quick. Now, however, a coarse thread of large
diameter is used on the spindle. For ordinary doors V
a smaller thread of less pitch is used. Otherwise
\
the construction of the doors is substantially the
\
same (Fig. 61).
A cast-steel frame, shaped as shown in Figs. 59
FIG. 59.
and 61, is bolted to the bulkhead. The sides are
formed, with a taper of ^
in. to a foot, to correspond with the frame
on the door. The door is formed of a steel plate, with a steel casting
to stiffen it, shaped as shown. The section of the door shows very
clearly the construction, and how the door jams in between the bulk-
head casting and the check plate. The top edge of the casting is pro-
vided with a metal nut, with a thread to correspond with that on the
spindle. This spindle extends from the top of the opening to the main
deck, the thread, of course,
extending only sufficient to open the door.
It hasbeen found necessary to fit a clip at the bottom into which the
door can jam to render the bottom watertight. The door can be closed
by a set of gearing from below, as well
as from above, and, when necessary for \
purposes of escape, it can be worked \

from either side of the bulkhead.
The doors between engine and boiler-
rooms, and to coal-bunkers, have their
sills kept well above the inner bottom. \
The only doors of this nature now fitted
are those on the inner coal-bunker bulk- \
heads opening out of the boiler-rooms.
FIG. 60.
The thread is a double one of 1 J in. pitch.
Horizontal Sliding Doors Doors of this type (Fig. 62) were
below water (and to a few
formerly fitted to important bulkheads
of the
above water), to which vertical doors could not be fitted because
lack of sufficient head room. The opening in the bulkhead is bounded
a steel as
shaped 60. This casting is secured to the
Fig.
by casting,
SPINDLE FOR ORDINARY DOOR.
SPINDLE FOR QUICK -CLOSING DOOR
Fio. 61. Veitical sliding watertight door.

bulkheads by bolts At the bottom and top there is a groove, having

.
a taper of -fg in. to a foot, in which the door slides, and in which
the door jams when closed. The door is formed by a steel plate,
\\
by a steel casting, shaped as shown. The weight of
ell stiffened
the door taken by two wheels. The inside edge of the door has
is
a gunmetal strip all round, which engages with the bulkhead
BOTTC
FIG. 62. Horizontal sliding watertight doors.
casting. The leading edge of the door jams into a number of clips
on the edge of the frame, and the following edge of the door has
some back pieces, which jam against the back of the bulkhead
frame. The door has two horizontal steel racks, and into these
two gunmetal pinions engage, these being rotated by a vertical
spindle. This spindle extends to the main deck, so that the door
76 WAR-SHIPS.
can be worked from above if desired. The door can also be closed
at the door by a spanner, working on a nut on the spindle.
A disadvantage with horizontal doors is the existence of the
bottom groove. The dirt from the traffic soon fills up the groove,
and this gets hard and prevents the door from closing. To obviate
this a sill 'plate is fitted over the sill. When the spindle for closing
is first turned, it plate up out of the way, to allow
turns the sill
the door to slide along and shut. When the door is again opened,
the sill plate automatically drops over the groove again. Hori-
zontal sliding doors are not now being fitted.
The vertical spindles for closing watertight doors are always so
arranged that a right-hand motion closes the door. The deck
plate at the main deck, where the spanner is worked, is provided
with an indicator to show whether the door is shut or not. The
spanner, however, should always be turned as far as possible to
close the door, as frequently after the gearing is worn the indicator
may point to "shut" when the door itself is not completely

shut.
Screens are fitted to the doors of coal bunkers on the inside to
protect the doors from the pressure of the coal, and thus enable
the doors to be opened and closed when the bunkers are full.
A book supplied to the commanding officer of each ship
is
giving a list of the watertight compartments, with their capacity

in cubic feet, the boundaries and means of closing, and the places
at which the doors can be closed. In addition to this the means of
pumping and ventilating each compartment is given. A specimen
page of such a book is given at the end of Chapter IX.
In the Admiralty circular above mentioned attention is drawn
to the defects found in
watertight doors, and the methods to be
adopted to keep them efficient.
Side Scuttles. These are fitted on to a ship's side plating and
elsewhere as necessary for the purpose of giving light and ventila-
tion. They are necessarily watertight, and the present method of
construction is illustrated in
Fig. 63. A cast-steel frame F is
riveted to the plating, and on this frame is a circular rim which
engages with india-rubber let in to the frame D of the glass scuttle.
This scuttle is hinged at the side, and can be secured by means of
the two butterfly nuts shown at the side which engage in tke two
lugs of the scuttle. On practically all the scuttles there are also
deadlights which are made of cast steel. When it is desired to
screen the light from showing outboard, the deadlight is simply
hinged down and secured by means of the two lower butterfly nuts.
It has a sunken strip of indiarubber which engages with a rim on
the inner side of the scuttle. In
action, however, or when guns are
fired in
proximity to the scuttle, it is necessary to hinge back the
glass scuttle, and then the pin of the deadlight is put in the inner
hinge, and the deadlight is secured

by the two lower butterfly nuts
right home to the side, so as to be
watertight. The deadlight is made
S ECTION
FIG. 62 A. Sidelight and Deadlight.
specially strong where exposed direct to the blast of heavy guns.

A light sheet steel light- excluding cover is sometimes fitted over
the glass scuttle instead of using the deadlight for the purpose of
excluding light. The glass is 1 in. thick, and is bedded in india-
rubber in the gunmetal frame D.
Sluice Valves are fitted to the lower parts of certain bulk-
heads to drain water from one compartment to another. These
valves are worked like a small vertical door, with a spindle
extending to the main deck, having an indicator on the deck plate.
In recent ships no sluice valve is fitted on the collision bulkhead.
Automatic Ventilation Valves. In recent ships the artificial
ventilation is arranged so that the main transverse bulkheads
are not pierced. In previous ships, however, the ventilation was
;8 WAR-SHIPS.
provided by a series of fore-and-aft trunks, having louvres at

intervals. These trunks were supplied with air from large steam-
driven fans. In this way the watertightness of the bulkheads
was completely destroyed, and the openings were provided with
automatic valves, arranged to shut automatically if water rose in
the compartment on either side. Provision was also made for
shutting all these valves if desired by pipes led from the main
or upper deck. These valves under the conditions on board
ship, however, were found to be far from efficient, and required
jamming home by hand.
Inspection of Watertight Doors. It is obviously of the
highest importance that all watertight doors should be in perfect
working order, and the crew well exercised in closing them, so
that they are always efficient. The following extracts from the
" "
Steam Manual gives the instructions regarding this :
"
Constant attentionto be paid to all watertight doors, valves, and fittings
is
to ensure them beinga state of thorough efficiency and watertightness, and
in
they should always be kept clear for immediate closing. No fitting of any kind
is to be allowed which will
require to be removed before the doors can be shut.
Certain doors are to be permanently closed, and doors fitted for facilitating the
removal of certain parts of auxiliary machinery should be kept closed and water-
tight until actually required for use.
" All
watertight doors and valves below the protective deck, automatic or
otherwise, are to be closed by hand as soon as the anchor is off' the ground, and
not opened until the anchor is let go again, except in the following cases :
"
(a) For purposes of ventilation for half an hour in the day, or for as long
as considered absolutely necessary by the Captain ;
"
(6) In parts of the ship which it is absolutely necessary to keep constantly
ventilated, such as where men are employed or sleeping.
" In both
the above cases men should be specially stationed to close by hand,
if
required, all openings which are of necessity left open."
Instructions were issued in Admiralty circular, S. 32111/1903,

of January 29, 1904, in regard to watertight doors and hatches.
See also S. 31157/1903, of January 9, 1904, for Admiralty circular
H.M.S. Prince George.
referring to
Automatic Doors. The following extract from Sir William
White's report l on the loss of the Victoria may be quoted with
reference to the suggestion frequently made that automatic or
" "
self-closing doors should be adopted instead of existing arrange-
ments :
"
This suggestion is a revival of one made long ago, then carefully considered
and put aside after certain experimental doors had been tried.
"Automatic arrangements are applied in valves to ventilating trunks and
1
Parliamentary Paper, No. C. 7208/1893.
other small openings in bulkheads and platforms. Even in such cases the feeling
of the Naval Service has led to the automatic fittings being supplemented by the
means of closing the valves when desired. In doors and scuttles the risks of
the automatic appliances failing to act, or of solid materials being carried into
openings by a rush of water, and preventing doors from closing properly, would
be much greater. These considerations have led to the retention of existing
fittings, the design of which provides that, when properly closed and secured,
doors and hatchway covers shall be as strong as the neighbouring partitions, and
watertight under considerable pressure.
"There is no difficulty in making automatic appliances. It is a question of
what plan secures the maximum of safety under the working conditions of the
Royal Navy. With large numbers of disciplined men, familiar with the fittings,
and constantly drilled in their use, it is possible to close and properly secure all
the doors, etc., in a battle-ship in three to four minutes, or possibly a less time
for ships after long periods in commission.
" In the
Victoria, no orders were given to close doors until one minute before
collision. It is established by the evidence that the doors, etc., were in good
order. The failure to close doors, therefore, was due entirely to the insufficiency
of time available, especially in compartments breached by the collision.
"
Under these circumstances no new argument in favour of the use of auto-
matic doors seems to arise out of the loss of the Victoria."
CHAPTER VII.
STEMS, STERNPOSTS, RUDDERS, AND SHAFT BRACKETS.
Stems. The simplest form of stem is that formed by a flat bar,

to which the plating at the forward end is secured. This form of
stem is adopted in merchant vessels and in the smaller classes of
ships in the Koyal Navy, as third class cruisers and destroyers.
For larger war vessels, however, a stronger form of stem has until
recently been considered necessary, so that such vessels should be
able to effectively ram an enemy's ship, without at the same time
sustaining serious damage herself. The most effective form of stem
for this purpose is one having a ram below water, projecting well
forward, so that it shall damage the slight underwater portions

of the structure well in from the side before being brought up by
the strong structure of the armour or protective deck of the other
vessel. In recent designs a ram stem has not been provided.
Stems of steel vessels are made of cast steel, a material pos-
sessing good strength and ductility (see Chapter II.), and capable
of being cast into most efficient forms for the special purpose
required. These castings are a great advance on the iron forgings

formerly in use, and a much more efficient ram has been by this
means rendered possible.
When a vessel is sheathed with wood and copper, cast steel
cannot be used for the stem because of the galvanic action that
would in all probability ensue between the copper and the steel.
In such ships, therefore, the copper alloy, phosphor bronze, is
used, but on account of the low strength of this material, the
casting has to be much more massive than a corresponding one of
cast steel.
In the earlier ships with cast steel stems (Royal Sovereign to
Canopus), the stem casting was carried well down into the body
of the ship (Fig. 64), and on this account it had to be made in two
pieces, because of the difficulties attending the manufacture and

transport of such a large and intricate casting. The two pieces
80
STEMS, STERNPOSTS, RUDDERS, SHAFT BRACKETS. 81
were connected together, as shown, by a scarph (Fig. 64), a

tapered key being driven in to draw the parts together, and the
whole well secured by screw bolts. The scarph was necessarily
UPPER DECK.
FIG. 64. Stem of battle-ship.
a place of weakness, and this has been avoided in more recent

ships by making the casting all in one piece. In order to do this
the lower portion is made much shorter (see Fig. 65). This sketch
shows the general shape of the casting, the ship being swelled out
in way of the ram. Inside the ram projection two webs are cast,
one vertical and one horizontal.
When a ship is intended for ramming, it is most important to
support the stem effectively by the adjacent structure. The follow-
ing arrangements are made with this object in view, viz.
(i.) The outer bottom plating is doubled in thickness and
recessed into the casting.
(ii.) The bow protection, where fitted in this case 2 in. is
recessed into the casting for one half its thickness.

(iii.) The lower deck, which is a thick deck, is well connected
to a large projection on the stem casting.
G
82 WAR-SHIPS.
(iv.) At the level of the platform

a 2-in. plate is worked, well
connected to the horizontal web inside the ram. This 2-in. plate
extends back to within 3 ft. of the collision bulkhead.
(v.) The vertical keel runs up
to the stem, and is well con-
nected to the vertical web inside the ram.
It is thus seen that every precaution was taken to support the
stem to make it efficient for ramming purposes, and (iii.) and (iv.)
SECTION AT RAM -
Fio. 65, Stem of battle-ship.
above not only provide a direct support, but they would resist the
side bending action that would ensue when the ships swung together
after the blow was struck.
"
The sketches each show a " cofferdam bulkhead 3 ft. abaft
the collision bulkhead. It would be advisable to pack the space
between these bulkheads, like an ordinary cofferdam, before ram-
ming, to limit the flow of water aft, in case the collision bulkhead
was damaged; access is obtained through the various decks for
this purpose. This bulkhead has not been fitted in recent vessels.
The phosphor bronze stem of a sheathed second class cruiser is
shown in Fig. 66. In this case the stem is cast in two pieces, and
the plating and planking have both to be recessed into the casting.
The wood keel also has to be recessed as shown.
In the stems of recent ships the casting has been stopped at
FORECASTLE DECK
FIG. 66. Stem of sheathed cruiser.
the main deck, the strength to the upper deck or forecastle being
provided for by a bent plate.
Sternposts. The remarks already made as to
the necessity of
in this case also.
phosphor bronze stems in sheathed ships apply
The of a single screw ship has to be formed to receive
sternpost
84 WAR-SHIPS.
the propeller as well as to form a support on which to hang the

rudder. Very nearly all the vessels now in the Royal Navy
have two or more screws, so that the main function of the stern-
post is to receive the rudder.

The shape of the sterns of ships varies considerably in different
classes, all war-ships (except the smallest) an essential

but in
condition to be fulfilled is, that the stern shall be so formed
that the rudder and steering gear are well below water and under
protection. In order to do this the stern is carried well abaft the
rudder-head, as seen in Figs. 69-74, to house the rudder-head
STUFFING GLAND
AT RUDDER HEAD.
FIG. 67. Sternpost of battle-ship.
and steering gear. See also Figs. 75, 76, 7 6 A, for the sterns of
Lord Nelson and recent battle-ships and battle-cruisers.
Taking first the case of battle-ships, the stern in vessels up to
the Canopus class was of simple construction. The sternpost was
a casting shaped as shown in Fig. 67, with projections on which
the rudder could be supported. In more recent ships the flat
"
portion of the stern, called the deadwood," has been cut away,
as in Fig. 68, and this makes the shape of the sternpost casting
" "
rather more complicated. The
object of this cut away was to
maintain good turning qualities ; this will be referred to again in
Chapter XX.
Fig. 68 shows in some

detail the sternpost of a battle-ship of
the Duncan the lower figure giving on a smaller scale the
class,
shape of the stern. The keel is cut up about 50 ft. from the stern,
but is brought down again to allow about 7 ft. to take the blocks
when docking. A 2-in. plate is worked horizontally from the stern-
post, extending well forward and securely fastened to the ship's
structure. This thick plate forms a substantial bed, to which the
lower palms of the shaft brackets and the forward end of the stern-
SECTION AT A. A. I
SECTION
.
FIG. 68.
post casting can be secured. It also forms an excellent stiffening

to the ship, to take the side bending due to putting the rudder
over. The sternpost has projections to take the weight of the
rudder, and is swelled out at the upper part to receive the rudder-
head. This has to be made watertight by means of a stuffing gland,
as in Fig. 67. The casting ends a short distance from the rudder-
head, and the shape of the stern is maintained to the upper deck
by means of a thick steel plate. The stern of the Lord Nelson
class was cut up as indicated in Fig. 75, and in the Dread-
nought and subsequent ships the stern has been formed similar to
Fig. 76, the stern being cut right up and the two rudders com-
pletely underhung. The sternpost of Lord Nelson is constructed
as in Fig. 78A, the weight of the rudder being taken
by the pintle.
Coming now to the sterns of cruisers, we notice that the stern
Fia. 75.
Fia. 7G.
BATTLESHIP
W.L. FIG. 76A.
BATTLE -CRUISER
"
has been shaped in two ways, in both of which a " balanced
rudder is obtained
1. As Fig. 69, as adopted in large cruisers up to and including
the Diadem, and in second and third class cruisers up to recently.

2. As Fig. 71, as adopted in large cruisers since the Diadem.
In type of stern the deadwood is cut right away to facilitate
this
turning, and the rudder is underhung with a portion of the area

before the axis.
In the former case the weight of the rudder is taken at the top
of the sternpost casting, as in Fig. 77, which has therefore to be
z
made specially strong on this account, and a steadying pintle is

provided at the lower part. Fig. 79 shows in some detail the
construction of the sternpost of a large cruiser with underhung
rudder, the weight of rudder being taken above. In large cruisers
previously to Invincible class the sternpost was constructed on
similar lines to Fig.78A and smaller cruisers are similarly fitted.
Fig. 76A shows the shape of the stern in battle cruisers with
double rudders.
Rudders. The shape of rudders in battle-ships up to Duncan
* SECTION
SECTION AT C.D.
FIG. 78. Rudder of battle-ship.
class has been, as in Fig. 73, of nearly rectangular shape, hinged

at the fore side. Fig. 78 shows in some detail the construction of
such a rudder. The weight
is taken on the sternpost projections,
and in order to make

the friction as small as possible the bearings
are, as shown, of hard steel. The frame of the rudder is a steel
1
casting, lightened out as much as possible, but necessarily of
massive construction at the forward end to stand the large twisting
moment. The sides are covered with 15-lb. (f in.) steel plating, 1
1
Phosphor brouze in a sheathed ship.
90 WAR-SHIPS.
and the space inside with fir. The rudders of the battle-ships
is filled
of King Edward VII. have a portion of the area before the

class
axis (Fig. 74). This renders the steering of the ship easier, because
the centre of pressure on the rudder is brought nearer the axis.
In the Lord Nelson the rudder is shaped as Fig. 78A, the stern
being cut away, as shown in Fig. 75, to improve turning. The
weight of the rudder is taken on the bottom of the sternpost and not
inside the ship at the rudder-head, as has formerly been the practice.
The casting is not continued at the fore and after edges, flat bars only
being necessary there. The wood filling shown at Ais to prevent
the rudder lifting, and that at the upper part is necessary to enable
the rudder to be lifted for getting out. These portions are portable.
In the Dreadnought there are two rudders, completely under-

the ship. See
hung, and the weight is necessarily taken inside
stem and rudder in a recent
Fig. 76 for the shape of battle-ship.
92 WAR-SHIPS.
"
For cruisers rudders are always balanced," 'i.e. a portion of
the area is before the axis. If we
deal with a rectangular plate
towed through the water at an angle of 30 to 40, it is found that
the centre of pressure is about one-third the breadth from the
leading edge. So to balance a rudder, i.e. to get the centre of
pressure close to the axis, we need to make the area before the
axis considerably less than one-half the total (see Figs. 69 and 71).
In such a rudder the twisting
moment even at high speeds
is small, and much smaller
power is needed in the steer-

ing arrangements than with
an unbalanced rudder. This
is specially desirable in
cruisers, because of their
high speed and the limited

room available aft to house
the gear owing to tho fineness
of these ships (see Fig. 81 A).
The pressure, per square foot
of rudder area, increases as
the square of the speed, so
that, comparing 28 knots
with 21 knots, the proportion
of pressure for equal area is
sy
) = 1'8 or an increase of
(23x2
80 per cent. This represents
the greater pressure, to be
dealt with per unit area in
the fastest cruisers as com-
FIG. 81, pared with ordinary battle-
ships. If we fitted a rudder
hinged on the forward edge in a fast ship, the steering engine and
gear would have to be of great power, and steering by hand would
be difficult. On these accounts the rudders of all recent ships
have been balanced, so that the twisting moment of the water
pressure about the rudder-head is small even at high speeds.
The weight of the rudder was formerly taken by the top of the
sternpost, as shown in Fig. 77. At the top of the rudder-head a
recess is formed, into which a bearing is placed in two halves.
This bearing rests on the rudder cross-head, which has three or

four legs. These are connected to a circular bearing ring, which
slides in the metal path on the top of the sternpost.
Fig. 79 shows in some detail the construction of the sternpost

and rudder of a large cruiser.
More recent cruisers with
single rudders have the rudders formed
like the Lord Nelson, with the weight taken at the bottom of the
sternpost. The Invincible and later classes of cruisers have twin
rudders similar to the Dreadnought, the weight being taken inside
the ship (see Figs. 76 A and 81 A).
In determining the diameter of the rudder-head of an unbalanced rudder,
we have only the twisting to deal with, but in balanced rudders the
practically
bending is of large amount, while the twisting going ahead is only small. Figs. 80
and 81 show how the two forms of balanced rudder would bend supposing each
is held rigidly in the sternpost. In the first case it is like a beam held at one end
and simply supported at the other. In the second case it is held at one end and
supported near the middle, with the other end free. In either case a consider-
able force has to be taken by the lower pintle, and a bending moment at the
rudder-head. The condition going astern has to be investigated, as it may
happen that worst case. The maximum speed astern is assumed to
this is the
be that due one half the revolutions for full speed ahead. The centre of
to
pressure then is nearer the after edge, and the twisting moment about the axis
is of considerable amount. This twisting moment, combined with the bending
moment, will determine the necessary size for the rudder-head, unless the ahead
conditions require a larger diameter.
In the Dreadnought a distinct departure was taken in regard

to the steering arrangements. This ship was considerably longer
than preceding battle-ships, and it will be seen in Chapter XX.
that length is a feature of a ship leading to a large turning circle.
Accordingly two rudders were fitted with their heads in the same
transverse section. The rudders were placed immediately abaft
the inner pair of propellers, which helps in the steering directly
the engines turn and before the ship gathers way. The rudders
are balanced and completely underhung. The weight is neces-
sarily taken inside the ship. The diameter of the rudder-head is
considerable, because of the great bending moment which comes
on the rudder-head where it enters the ship. In this case the
twisting moment is small and the diameter of the rudder-stock is
determined by the conditions due to bending when going full
speed ahead. Similar steering arrangements have been fitted in
battle-cruisers.
In any rudder, the head being under water, it is necessary that
the hole in the sternpost should bo made watertight. The hole
94 WAR-SHIPS.
is lined with gunmetal or phosphor bronze, and the rudder-head is

cased with a gunmetal or phosphor bronze sleeve, as shown in
Figs. 67, 88, and 8lA. A
stuffing gland is fitted at the top to
make the hole watertight.
Projections were formerly cast on the rudder on each side to
bring up against the sternpost when the rudder is hard over ; these,
however, have been omitted in recent ships.
It was formerly the practice to supply each ship with a mould
giving the actual shape of the rudder. It is the present practice

to supply a sketch of the rudder on a large scale, giving complete
figured dimensions.
Rudders, design of. It will be of interest to consider briefly
some points of importance in connection with the design of the
rudders of a large, fast war-vessel. We have already discussed
the reasons for adopting double rudders. Fig. 81 A, showing a
transverse section through the rudders, has been drawn to illus-
trate these remarks. The rudders must be completely under
water and under the protection of the protective deck, and no
portion of the rudders must be in contact with that deck, because
of the risk of injury if that deck were struck. With an under-
hung rudder the bending moment on the stock when put hard
over at full speed is very considerable, and the diameter of the
stock must be made of sufficient size to withstand this bending
moment. The bottom of the ship in way of the rudders is formed
of a steel casting, A, which is well connected to the bottom plating
of the ship which surrounds it. A metal (phosphor bronze) cast-
ing, C, is fitted in the casting at each rudder, and the rudder-stock
is enclosed by a metal sleeve. Thus the bearing of the rudder in
the stern post is metal to metal. The hole in which the rudder
works must be watertight, and this is obtained by a stuffing-box,
SB, at the upper end. This stuffing-box is screwed down and
compresses packing placed in the space between the metal casting
C and the sleeve on the rudder-stock. The rudder is turned by
means of a crosshead (CH) keyed to it and connected to the screw
steering gear by means of connecting-rods (CE). The weight of
the rudders is taken at the top. A steel casting, B, is fitted and
supported by a massive structure going right across the ship. In
the casting B are fitted metal castings, D. The head of the
rudder is slotted out and a metal casting, E (fitted in two halves),
is fitted and keyed to the rudder, and this bears on the metal
casting D, proper provision being made for lubricating the bearing
surfaces. A ring in two halves is fitted immediately under the

casting B to prevent the rudder from lifting. Careful attention
is necessary in the design of the rudder and stern post to make
sure that the rudder can be taken out when the vessel is in
dry
W.L.I
FIG. 8lA Twin rudders.
dock. In the case given in Fig. 81 A the rudders are inclined

inwards at their upper end, this being necessary in order to obtain
room for the crossheads. In ships not so fine at the stern as in
this case, the rudders are fitted vertical. Zinc protecting plates
(ZP) are secured to the rudder-stock just below the sleeve. When
96 WAR-SHIPS.
steel and metal are in contact in sea-water a galvanic action is set

up, which corrodes the steel ;
but if zinc is fitted this corrodes
instead of the steel, and can be periodically examined and renewed
whenever the vessel is put into dry dock.
Shaft Brackets. In twin screw vessels a considerable length
of the propeller shafting is outside the ship, and brackets are
fitted on either side, just forward of the
propellers, to take the
weight of the after end of the shafting, etc. For steel ships the
brackets are of cast steel, for sheathed ships of phosphor bronze.
These brackets do not have to take any fore-and-aft thrust (this
FIG. 82.
being taken in the engine-room by the thrust block), but they

have to bear the very considerable weight of the propeller, etc.
Because of this the attachment to the structure of the ship has
to be very secure.
Fig. 82 shows the arrangement in a battle-ship of the Duncan
class. The arms are flattened out at the top and bottom. The
upper palm passes inside the ship, where it is riveted to a
thick fore-and-aft plate. The lower palm is shaped with a
scarph to fit a
corresponding scarph on the other bracket. The
two brackets are then securely fastened to the 2 -in. fore-and-aft
plate, and to the forward end of the sternpost casting (see also
Fig. 68). The arms of the brackets are of pear-shaped section,

with the blunt end foremost. This is in order to diminish the
resistance of these brackets, it having been proved by experiment
that this form offers the least resistance when fully submerged.
(It will be remembered that a torpedo has a blunt nose and
a fine run.) Similar arrangements are necessary in ships with
four screws.
CHAPTER VIII.
STEERING GEARS.
THE steering gear, fitted at the stern
adiac
most important
/i An ~~_ - -,
fitting of a war-shiu
1 P> and, excent
* in tho ornn ^*>
always
"
* * t0
steerig gear.pends on -
(i) The area of the rudder
moment small, even at high

speeds
is iRi tu"S A
e
e:
rd
TH Iimit
?
the
an aCC m Panied
Blower motion at
an^TZT ^
Zge *
by a
steering gear is
,
iJS-^l
,hieh have been largel
gear, and takes up a lot of room.

STEERING GEARS. 99
3d in battle-ships. A
cross-head on the rudder-head is connected
r
parallel rods to a second cross-head, on to which a long tiller is
ached. The forward end
ot this tiller is of parallel
section, and passes through

:
block, which can swivel
side another block. This
cond block is made to
ivel across the ship on a
jwartship path, by means
f a sprocket chain. This
lain passes to the sides of
e ship, and down to the
hntre line, where passes it
ader a sprocket wheel,

Vhich fits into the chain.
4? his
sprocket wheel is made
o revolve either by the hand
or steam gear, and so the
rudder moves as required.
This gear takes up a lot of
room, because of the travel
of the tiller from side to SCREW GEAR.
side, and for this reason its ~~J^ 83
use has been confined to
battle-ships, which are full at the stern. It has the advantage of
compensation, but has the disadvantage of being reversible. If
the chain or sprocket wheel broke, the tiller would swing from
side to side with the movement of the rudder. On this account a
friction brake is fitted, to hold the tiller ifnecessary. This brake
is tightened up to hold the tiller when the gear has to be changed
from hand to steam, or vice versa.
Theory of Rapsori's slide steering gear. In Fig. 85 a tiller is shown,

passing through a ball, which is made to move along the thwartship path.
This is similar to the state of things that obtains in the actual gear. The
constraint of the slide brings into action a side force on the ball, as shown, Q.
The pull of the chain P, combined with the force Q, gives, by the parallelogram
T>
of forces, a resultant force acting square to the tiller. This acts at a
of so that the moment at the angle is as compared

leverage ,
cQg 2fl
,
100 WAR-SHIPS.
1
with the moment PA at the middle line. At 35, = 1*5, so that at the
cos
extreme angle the moment is 50 per cent, greater than at for the same
DETAIL OF SLIDE
MKI KJL.
SPROCKET CHAIN SPROCKET WHEEL

L J
FIG. 84. Rapson's slide steering gear.
pull on the chain. This increase in moment is accompanied by a slower

motion thus, if a certain number of
;
revolutions moves the tiller through

10 at the start, the same number
will move it through about 7 at the
end of the travel. The detail of the
slide shown in Fig. 84 shows how
the side thrust is provided for.
2. Harfield's Steering
Gear. This patent gear has
been fitted in a number of
ships, both battle- ships and

cruisers. It has the advantage
FIG. 85.
of compensation, which is
obtained as shown in Figs. 83
and 86. The forward cross-head is fitted with a curved rack,
STEERING GEARS. ldf
which engages with an eccentric pinion. This pinion is made to

revolve as required. Assuming a constant moment acting on
FIG. 8G. Harfield's steering gear.
this pinion, theforce on the teeth is greatest at large angles,

because the leverage from the turning centre is then smallest.
FIG. 87. Ollis's steering gear.
This large force acts on the cross-head, and at the large
angles acts at a larger leverage than at small angles. Thus,
FIG. 88. Screw steering gear.
at the extreme angles we have a greater moment on the rudder

obtained in this double way, a larger force on the teeth, and a
102- . WAR-SHIPS.
larger leverage. Thus the compensation desired is obtained.

With a hydraulic brake is fitted, which can be made
this gear
to hold the rudder in any desired position. The safety of the
gear depends absolutely on the teeth of the rack and pinion in
contact, and on this account these are made of forged steel of
ample strength. The gear is reversible, like the Eapson's slide,

unless a worm and worm-wheel is introduced, which has been
done in some ships.
3. Ollis's Steering Gear (Figs. 83 and 87). This is a gear
1
designed at the Admiralty by Mr. F. B. Ollis. It answers the
same purpose as Harfield's gear, with simpler construction. The
pinion and rack are both circular, but the connecting rods to the
rudder cross-head are brought close together at the forward end.
Assuming a constant moment on the pinion, it is seen that as the
angle increases the connecting rods come closely together, and so
the force along these rods increases, giving a larger moment at
large angles than at small. This is the compensation desired.
The gear is reversible, as in the previous cases, and it is necessary
to provide a friction brake, to hold the gear in case it becomes
disabled, and when changing from hand to steam, or vice versa.
4. Screw Steering Gear (Figs. 83 and 88). This gear is
quite different to the previous gears, being the reverse of com-
pensating. It has been very largely adopted. It consists of a
right- and left-handed screw, which, on rotating, causes two nuts

to slide along in opposite directions; these nuts are prevented
from rotating by being made to slide along parallel bars. Con-
necting rods are led from these sliding nuts to the rudder
cross-head.
The gear takes up very room. If the pitch of the screw
little
is can be made non-reversible.

sufficiently fine the gear Cases,
however, have occurred in which the thread was coarse enough
to allow the gear to reverse, and such gear is not self-holding.
This self-holding is one of the features of this type of gear, which
is of extreme value, and recent gears have been made with the
ratio pitch -f- circumference about 0'075 in order to secure this.
The self-holding of the gear, when

subject to shocks from the
rudder, must vary with the degree of lubrication of the screw.
Screw gear is being now fitted to all the new ships of large size
in the Royal Navy. It is, however, the reverse of compensating,
and a rudder not balanced would be difficult to move by the hand
Now Constructive Manager at Chatham
1
Dockyard.
STEERING GEARS. I03
wheels at large angles when a high speed is reached. All the

large ships at present building are, however, being fitted with
rudders more or less balanced, and in these ships the screw
steering gear proves a satisfactory fitting. A friction brake has
sometimes been on the rudder cross-head, to hold the rudder
fitted
in case anything has to be done to the gear.

The projection on the rudder cross-head to take locking bolts,
shown in Figs. 86, 87, 88, and the corresponding projection on the
sternpost, shown in Figs. 67, 68, 77, are not now fitted.
Steam Steering Gear. Whatever is adopted

steering gear
at the stern, the motive power for working it under ordinary
circumstances is a steam steering engine. Formerly the engine
was placed aft, just forward of the steering gear, but this is
objectionable for the following reasons :
(i.)
The steam-engine heats the compartment, making special
ventilation necessary.
The steam and exhaust pipes led from the boilers heat the
(ii.)
compartments at the after end of the ship, through which they pass.
(iii.)
Attention of the engine-room staff is required at the
engine.
In recent ships the steering engines, of which there are two,
are placed on the after bulkhead of the engine-room. In some
cases duplicate shafts are taken aft, one from each engine, to the
gear; in recent vessels, however, only one shaft is taken aft, the
connection being made in the engine-room to enable either engine
to work the steering gear.
The steam steering enginesare of ample power, and in a large
ship the time specified to take the rudder from hard-over to hard-
over is only 30 seconds when the vessel is going full speed, A
worm and worm-wheel between the shaft and the steering
is fitted
engine, so that shocks from the rudder are kept from the engine,
i.e. supposing the steering gear is reversible.
The steam-engine is operated by means of steering wheels

placed as follows in recent ships forward bridge, forward con-
:
ning tower, lower deck forward in the lower conning tower. The
engine worked from these positions by means of controlling
is
shafting or telemotor gear. The shafting is led down inside the

armoured tube from the conning tower to below the protective
deck, whence it is taken aft to the steering engine. Care is taken
to avoid any rigid connection of the shafting to the protective
deck or any portion of the vessel exposed to damage in action.
104
WAR-SHIPS.
Some ships have been fitted with telemotor controlling gear

instead of the shafting. In this gear the motion is conveyed
to the steering engine from the steering wheel by means of small
pipes. A
small piston is worked at the wheel; this causes
motion of water throughout the system, and exactly similar
motion at the steering engine.
Steering by Manual Power. Most vessels have one position
below, aft, for steering by manual power, in case the steam gear
is not available. The hand wheels are three or four in number
in the largest vessels, three and two being fitted in smaller ships
FIG. 89.
the number depending on the twisting moment on the rudder-

head. In a gear not self-holding, any shock on the rudder is
transmitted to the hand wheels, and considerable power is re-
quired to hold the gear at any required angle. To hold the hand
wheels in such a gear a "Fayrer's brake" is fitted. This is
a band brake bearing on a wheel on the hand-steering shaft.
It can be tightened by standing on a foot-plate either side
(Fig. 89).
In Harfield's gear an interesting clutch is fitted just abaft the
hand wheels (Fig. 90). The wheels A and B are loose on the
shaft, and are provided with pawls at A and B, so that motion of
A in a right-handed direction and motion of B in a left-handed
direction (looking from the bulkhead) is impossible. The pinion,
which gears with the spur wheel on the steering gear, moves
STEERING GEARS. 105
endwise on the shaft by means of a thread, the shaft acting as a

bolt and the pinion as the nut. If the hand wheels are rotated in
either direction it will be noticed that the pinion is drawn along
to the wheel A or B, which is free to rotate. If, however, the
motion is reversed, i.e. the rudder becomes the motor in either
drawn along to, and jams against, the wheel
direction, the pinion is
A or B, which is prevented from rotating, so that the rudder
cannot react on the hand wheels.
Such a clutch or a Fayrer's brake is unnecessary when a worm
of sufficiently fine pitch and worm wheel is introduced into any
HAND WHEEL
SHAFT.
SPUR WHEEL TO
STEERING GEAR
Fio. 90.
portion of the gear, as this renders the gear non-reversible. It is

also unnecessary in a screw gear which is non-reversible.
Spare Gear. With such an important fitting as the steering

gear, a large margin of strength is provided in all the parts, and
spare parts are provided for certain portions of the gear, so that
these may be replaced in case of disablement. The following is
the list of spare gear provided in a recent vessel fitted with screw
gear.
One connecting-rod pin for crosshead, one ditto for sleeve, one length of the
weakest shafting 7 in. diameter, one spur wheel 34 teeth and one with 64 teeth
being the wheels from the steam gear, one complete set of brasses, one bush and
one gland for each size fitted. All of these are tried in place and stowed.
In some gears a length of the shafting leading to the steam

gear is made the weakest part of the gear, so that this would
106 WAR-SHIPS.
be the first to go, and so save the steering gear itself. spare A
length of this shafting is carried to replace.
Clutches, etc. All steering wheels are so arranged that the
upper part of the wheel moves in the same direction as that of
the ship's head. In a large ship the steam steering wheels take
the rudder from hard-over to hard-over in eight complete turns, the
hand wheels requiring twenty-four complete turns.
Clutches are fitted to enable the gear to be changed quickly
from connection with one engine to the other, or from either
engine to the hand wheels. In doing this it is necessary to hold
the gear (unless this is self-holding), by means of a brake, usually
a friction brake. Indicators are provided in a prominent position,
showing the angle at which each engine, hand wheels, and the
steering gear itself are situated. The angle shown on the indi-
cating dials for the rudder, steam steering shaft, and hand steering
shaft all now show the position of the helm and not the position
of the rudder. It is of the greatest importance that connec-
tion should only be made when the engine or hand wheels,
as the case may be, indicate the same angle as the steering
gear itself. For, suppose the engine at 5 port is coupled with
the steering gear at 0, then when the steam-engine is taken over
to hard-a-starboard the gear will bring up against the stops with
the engine at full speed, and some part of the gear will be fractured.
In recent ships the steam gear brings up at 35, and the gear
itself brings up with the rudder at 38. The clutches to the
steering gear should be so arranged that it is not possible for the
steam steering engines to become geared up together, or for either
to be geared up to the hand wheels under any circumstances.
Steering Gear for Twin Rudders. Fig. 90A shows in out-

line the arrangement of steering gear for a recent battle- ship
having twin rudders. Each rudder has its own set of screw
steering gear, either of which can be disconnected by a clutch
CL 1. The screw in this case was 6f in. under thread, 8J in.
over thread, and the pitch was If in. single thread. 1 It is seen,
by putting in the clutch CL 2, which throws the hand-gear clutch
CL 3 out, the shaft from the steering engine works the port set of
gear direct, and by means of the spur-wheels A, B, C, and D, the
starboard set of gear is given an identical motion. The shaft
takes 16 revolutions to take the rudder from hard over to hard
1
It will be noticed that the ratio pitch -4- circumference is below the value 0-075
mentioned on p. 102.
STEERING GEARS. 107
over 35 each way. To put the hand gear in, the clutch CL 3 is
put in and keys the wheel

E on to the shaft, this motion at the
* te
fcD
R
1
_0
00
same time taking the steam-gear clutch CL 2 out. The hand-

wheels then drive the port steering gear through the spur-wheels
io8 WAR-SHIPS.
G, F, E, and the starboard set obtains an identical motion through

the wheels A, B, C, and D, as before. The shafts to each set of
screw gear is 6 in. diameter, and the shaft to the steering engine
which operates both sets is 7J in. diameter. A portion of this
shaft 5J ft. in length is made 7 in. diameter, and a 5J ft. length
of shaft of this diameter is carried as spare. The hand- wheels are
available in case of emergency, and the helm angle that can be
employed at various speeds can be ascertained by experiment.
Thus, in this case, 25 of helm could be obtained by hand-power

with the ship steaming at 12 knots. There is telephonic com-
munication from the steering flat to the bridge for this purpose.
Auxiliary Steering Gear. It was formerly the practice to
provide auxiliary means of steering, for use supposing the steam-
engines and hand wheels were disabled. This usually consisted of
a set of blocks and tackles, which could be fastened to the gear to
work it by hand. The gear provided entailed considerable weight,
and was difficult to rig up in place. Ships which had the auxiliary
provided never used it, and it has been removed from a number of
ships. The more recent practice was to rely on the hand wheels
as the alternative in case the steam-engines are not available.
From this point of view, the use of twin or quadruple screws is an
alternative method of steering, and cases have occurred in which
ships have had to go long and stormy voyages without a rudder at
all, the steering all having been performed by the twin screws.
In the most recent ships it has been found impossible to steer
satisfactorily by manual power, and hand wheels and gear are not
fitted. An auxiliary gear is provided in some ships, worked by
" "
means of the Williams-Janney variable speed gear, of sufficient
power to steer the vessel at a speed of about 15 knots, say to 15

of helm, port and starboard. This gear is worked by an electric
motor rotating continuously in one direction, and by manipulation

of gearing on the variable speed gear worked from a small steering
wheel placed in the steering compartment, the steering gear can
be worked in either direction. In some ships also the gear can be
connected to the main telemotor system so that the ship can be
steered from the conning tower by its means. This variable
likely to be largely
is employed in the future for boat
speed gear
hoists, electrical capstans, etc.
CHAPTER IX.
PUMPING, FLOODING, AND DRAINAGE.
THE present chapter deals with the methods adopted in ships of

the Eoyal Navy for clearing water out of the ship, which has
entered through damage or has accumulated in the bilges under the
ordinary conditions of working. Also the means for voluntarily
admitting water into certain portions of the ship to preserve
stability, or to keep the ship upright or on an even keel after

damage, or to flood the magazines, etc., in case of fire.
By Drainage is meant the means of allowing water to pass
from one compartment to another until it reaches a pump- suction
by which it can be removed.
By Flooding is meant the deliberate admission of water into
the ship through a sea-cock. This may be necessary to correct
heel or trim after damage, or in the event of fire.
By Pumping is meant the general arrangements
adopted to
remove water from the ship by means of the fire and bilge pumps
in the engine-rooms, or by the Downton or electrical pumps.
Before dealing generally with the subject, it will be advisable
to see how far the pumps fitted to a ship are able to deal with the
inflow of water.
If A be the area of a hole in a ship's bottom in square feet, and d the distance
of the centre of the hole below water, then the initial velocity of the water
through the hole is given by 8*/d ft. per second, so that jjvery second there
would be SAy'd cubic ft. of water entering, or about lK*/d tons per minute.
If the hole is 16 ft. below the surface, about 56 tons will enter in a minute for
every square foot of area.
56 tons a minute is 3360 tons an hour.
It is quite impossible to expect that any ordinary provision of
pumping capacity could cope with the inflow of water through

even a moderately sized hole. All that can be provided is first,
109
i io WAR-SHIPS.
extensive W.T. subdivision to localise the damage, and second,

such provision of suction pipes and pumps as will keep under the
inflow of water through bulkheads or flats strained and leaking
owing to the damage. To pump out water entering a com-
partment direct from the sea is simply "trying to pump the
ocean dry."
In recent ships the arrangements for pumping and drainage are
considerably simpler than formerly adopted it is proposed only
;
to deal with the arrangements in ships of the Duncan and recent

classes.
PUMPING ARRANGEMENTS IN "DUNCAN" CLASS.
Main Drain.Special arrangements are made for clearing the

ship of large quantities of water for which the ordinary steam and
hand pumps would be inadequate. For this purpose a main drain
(Fig. 91) is worked above the inner bottom from the forward boiler-
room, branching to either engine-room. This drain is 15 in.
diameter in the middle boiler-room, and 20 in. x 15 in. in the
after boiler-room. By means of this drain, water can be passed
from either boiler-room into one or both engine-rooms. Sluice
valves, with non-returns, are fitted as shown to control the flow of
water to the engine-rooms the non-return valves automatically
;
prevent the passage of water from the engine-rooms or from one

boiler-room to another. In each engine-room, suctions are taken
down from the circulating pumps of the condensers, so that these
pumps can be made to draw from the engine-room bilge if desired
instead of from the sea. These circulating pumps work indepen-
dently from the main engines, and their engines are placed as high
in the engine-rooms as possible, in order that they may continue
working even if the engine-room is flooded to a considerable depth.
The sluice valves on the main drain are worked from below and
also from the main deck.
In addition to the main drain, the forward compartments likely
to have large quantities of water are drained into the forward
boiler-room (Fig. 91). Similar drains are fitted aft, draining into
the engine-rooms. At the forward end we notice that two pipes,
8 in. we have pipes leading from
diameter, are used, and into these
1
barbette, submerged torpedo -rooms, torpedo-tubes, chain lockers,
1
In recent ships other means are adopted for clearing cable lockers of water
and mud.
PUMPING, FLOODING, AND DRAINAGE. 113
etc. The forward and after drains, and also the main drain, each
have connection to a sea valve for flushing purposes. The
wing
compartments and inner coal-bunkers are drained on to the of top
the inner bottom, as shown in Figs. 91 and 93.
When a flat requires draining, a valve, as Fig. 99, is fitted.
FIG. 93. Pumping section.
Above the flat there is a strainer with screw-down valve, and

below the flata flap valve to act as non-return.
Pumping. For freeing the ship of ordinary quantities of
water, a main suction is fitted above the mn6r bottom, extending
nearly the whole of the ship's length. Screw-down valves, marked
S in Fig. 92, are fitted on this main suction at each main transverse
I
114 WAR-SHIPS.
bulkhead in order to preserve the watertightness of the bulkhead in

the event of the pipe being damaged on either side. Fig. 92 shows
a portion of this main suction. Pipes are taken from it to the double-
bottom compartments, and on each of these branches a valve, F, is
fitted, which under ordinary circumstances acts as a screw-down
non-return valve. The valve, however, can be unlocked and lifted
off its seat, when it will allow the compartment to be flooded (see
Fig. 97 for construction of valve). The double bottom between
Nos. 2 and 4 longitudinals is drained through S.V. on No. 2
longitudinal, as shown in Fig. 93, to the suction at the middle

line (except in way of the portion of the double bottom used for
the reserve feed water). Suctions are also led from the main
suction to shallow pockets in the engine and boiler-room bilges ;
these suctions have a S.D.N.R. valve, without the flooding arrange-

ment shown in Fig. 97. A
suction is also taken from these
pockets, independent
main suction, direct to the steam
of the
pumps. Mud
pockets, are placed at intervals on the main
M,
suction as shown. This main suction also has branches to the
bottom of the ship, forward and aft of the double bottom,
as shown in Fig. 92. The main suction is connected with the
steam pumps in the engine-room, so that any desired compart-
ment may be pumped out by the use of the steam pumps of
the ship.
Downton Pumps. A connection is also made with each of
the four 9-in. Downton hand pumps. These

ammunition pas-
are placed in recesses in the
sages (Fig. 93). Each Downton is geared to
work from either the main or middle decks, as
most convenient, by means of long cranks, on
towhich a number of men can be placed.
The Downton is a double-acting pump, shown in out-
line in Fig. 94. valves, A, B, C, and D,
There are four
allopening upwards. A piston works up and down as
shown. Suppose water is in the tail pipe and the piston
comes down. This causes B to close and A
to open, so
that air forced up the left-hand side of the pump.
is
Fio. 94. Downton This action creates a partial vacuum above the piston,
pump. and water will rise and fill the upper part of the piston
chamber. During the upward movement this water is
forced up through the valve C, the valve D closing. The water which then
collects below the piston is forced away during the downward stroke. In this
way a continuous flow of water is delivered either overboard through the
discharge or into the rising main (Fig. 93).
Fig. 95 shows in diagram the various leads from a Downton

pump. There are three valves, viz.
1. A
screw-down (S.D.) valve connecting to the Kingston
valve (K) (see Fig. 98 for construction of a Kingston).
2. Ascrew-down non-return and flood (S.D.N.R. and F.)
valve (Fig. 97) connecting to the main suction (M).
3. A S.D. valve connected to the tail pipe of the pump.
:ATOR PLATES.
A. o. -
MAIN DFCIT
d>
FIG. 95.
The pump has

(i.) A discharge overboard through S.D. valve (D).
(ii.) A discharge into a rising main with a branch into the fire
main.
The branch to the fire main has a N.E. valve, E, so arranged
that when the fire-engines are working the pressure from the fire
main is prevented from affecting the hand pumps or their system
of pipes. The rising main from the Downton is shown in Fig. 93,
and this can be used for wash deck purposes, etc., hose connections
being fitted as shown. A connection is made across the ship to
WAR-SHIPS.
the pump on the opposite side. Fig. 95 shows how the valves at
the ship's side are geared to be worked from the level of the pump,
and also from the
main deck if desired.
To pump any com-
partment out, all the
valves between the
fire-engine or Down-
DFLIVFRV ton and the compart-
ment must be opened.
Thus for a forward
compartment all the
bulkhead stop valves
FIG. 96. Screw-down FIG. 97. Screw-down must be opened be-
valve. non-return and flood- i
va lve. tween the pump and

the compartment, be-
sides the valve on the particular branch.
" "
In protected cruisers suctions are taken down to the
FIG. 98. Kingston valve to a sheathed ship.
bottoms of the upper bunkers to clear water away. These ships

would get water above the protective deck if the thin side were
riddled.
In working a valve it should be noted that the valve is not

"
necessarily closed when the indicator points shut," as after some
wear of the gearing it is necessay to work beyond this point to
get the valve well down on its seat.
Flooding. In order to permit flooding of the compartments
of the double bottoms, the valve 2 in Fig. 95, and the valve to
the particular compartment, both of which under ordinary circum-
stances of pumping act as non-return valves, can be lifted up off
the seats so that flooding can take place. This, however, can
only be done deliberately, as the valve cannot be lifted right up
unless a locked pin is withdrawn (see Fig. 97). In order,
therefore, to flood a double-bottom compartment we have to open
FIG. 99. Non-return valve. FIG. 100.
a numberof valves (Fig. 95), viz. Kingston (K) valves A

and 1,
unlock and lift up valve 2, and also unlock and lift up the valve
leading to the particular compartment.

The wing compartments are flooded direct from sea valves
It will be noticed that each pair of
placed as shown in Fig. 91.
wings has a flooding valve. The trimming tanks forward and aft
can also be flooded direct from the sea.
are made for flooding magazines (and in
Special arrangements
recent ships the shell rooms) direct from the sea. Usually one
sea valve supplies a number of magazines, etc. Besides the sea
valve itself a valve is placed at the magazine, both of these being
locked valves, which must be deliberately released to allow flooding
deck or below
to take place. They can be worked from the main
as may be desired. For trial purposes the outer valve should be
ii8 WAR-SHIPS.
firstopened so that the connecting pipe would be filled. This valve

is then closed and the magazine valve opened. The water which
then enters can be caught in a bucket. When a sea valve is used
only for magazine flooding, it is only necessary to fit one locked
valve close to the magazine in addition to the sea valve. Where
a sea valve serves other purposes a valve must be fitted at the
sea valve, so that it may be open without flooding the magazine
flooding-pipe (see Fig. 100). No means are provided for draining
magazines.
Fire Main (Figs. 101 and 102). The fire main and its system
of pipes obviously a most important set of fittings in a war-ship.
is
The fire main itself in the Duncan is a copper pipe 5 in. diameter,
running all fore and aft under the protective deck. It is con-
nected in the engine-room with the fire and bilge pumps, stop
valves being fitted to shut off the fore-and-after portions of the
system as desired. The fire main system can be charged from
the Downton pumps if required, but the connection is fitted
with a non-return valve, so that the pressure in the fire main
may not affect the Downton pumps. The fire main is not
rigidly connected to the protective deck, but is either supported
from bulkheads or from slings connected to the deck. Branches
are taken from the fire main to each stokehold, to submerged
torpedo-rooms, capstan engine-room, ammunition passages, etc.,
with connections at each place for attaching a couple of hoses.
Rising mains are led at intervals to the upper deck, two of
these are led up under protection inside the barbettes. Each
rising main has a stop valve beneath the
protective deck at its
junction with the main, and these valves are placed in
fire
accessible positions so that they can be worked from below the

deck if desired. They are geared to work also from above the
deck. To most of the rising mains there is also a stop valve
beneath the main deck, to shut off the pipe above that deck in the
event of the pipe being damaged in action. Branches are taken
from the rising mains for flushing w.c.'s, washing out ash-shoots,
washing out barbette guns, etc. The rising mains are fitted with
branches with hose connections in convenient positions on the
various decks for fire and wash deck purposes. One-inch bib
valves are placed on rising mains between middle and main decks,
main and upper decks, and in ammunition passages for drawing
off small quanties of water if desired.
Dry Dock Flood. Fittings are also provided for flooding

Ys ^d
INDWWy
.3. o
JS1
120 WAR-SHIPS.
magazines when the ship is in dry dock. A pipe is taken, between

the Kingston and the magazine (Fig. 100), to the upper deck, and
fittings, to take hose connections, are placed on this pipe above
the main and upper decks. In the event of fire, hoses could be
taken from the dockyard and the magazines flooded. In recent
ships the hoses from dockyard are arranged to connect on to the
Kingston direct, by means of a fitting which can be secured to the
Kingston opening when the ship is in dry dock.
Air Escapes. When flooding any closed compartment it is
necessary to provide an escape for the air. Pipes are led from the
crown of the compartment, as in Fig.
93, with plugs on the upper ends.
These plugs, when unscrewed slightly,
allow the air to escape, and when the
compartment is full they can be readily
screwed up again. Plugs are also pro-
vided on the manhole covers (Fig. 48).
The air escapes from the maga-
zines are constructed with a simple
liftvalve, as Fig. 103 ; the pipe is
led well up above the crown of the
magazine and bent round, with the
end perforated. In ships in which
FIG. 103. Air escape to
magazine.
the crown of the magazine is near the
waterline, these air escape valves are
provided with a spring to cause them to lift easily under the small
pressure. The recent practice in fitting air escapes to magazines

is simply to take a pipe from the crown of the magazine to a
position beneath the upper deck the top of the pipe is turned
;
round and perforated, a cock being fitted above the magazine

crown to prevent escape of air when the magazine is being
ventilated.
EECENT PRACTICE FOR PUMPING, ETC.
Pumping, etc., Arrangements. Fig.103A shows in diagram

the arrangements adopted for pumping, flooding, and firemain
purposes in a recent large ship. Each section of the ship, divided
by main transverse bulkheads, is dealt with quite independently,
and these bulkheads are not pierced at all. There is no main
drain or main suction as formerly fitted. Each section has a
motor-driven centrifugal pump capable of pumping 50 tons of
water an hour at a pressure of 60 Ibs. per square inch. The

suction from the seais 4 in. and is taken from the sea side of a
magazine flood where one is in the vicinity, otherwise direct from

the sea. The main valve of this flood, marked A, and the valve
at the magazine marked B, are worked from the main deck and
UPPER DECK
FIG. 103A. Pumping and flooding arrangements.
adjacent to the valve, the detail of the

locked fitting below being
shown in Fig. 103B. A
valve, such as A, usually supplies the
flooding pipes for several magazines
and shell rooms.
Suppose it is desired to use the pump for fire or wash-deck
purposes. The valves B, D, and C are opened and E is closed.
The valve M, on the rising main M, is opened and the discharge
122 WAR-SHIPS.
overboard L is closed. The valves N or P, or both, on the fore

and aft salt-water
main, are opened as required. Hose connections
with valves are taken off the rising main and the salt-water main
in convenient positions with the hoses stowed in
adjacent places.
The valve M
is fitted to shut off the rising main when
discharging
overboard through L, and also to shut off the rising main in the
event of its being damaged in action. If the pump has broken
down, a portable pump is provided, and its suction can be attached
to a hose connection Qand its delivery to a hose connection E.

The salt water main shown in Fig. 103A is carried all fore and
aft with branches as required for hose connections. Connections
are taken from this pipe for supplying washplaces, water service
to auxiliary machinery, sanitary tanks, heads, cables, etc.
An alternative salt water supply under pressure is provided
independently of the above by rising mains from the fire and bilge
pumps in the engine and boiler-rooms, hose connections being
provided as necessary.
Suppose now it is desired to pump out a double bottom com-
partment or the bilge. The system has first to be flooded from
the sea, as the centrifugal pump will not draw otherwise. The
valves B, D, E are opened, as also that leading to the compart-
ment, say F. The valve I) is then closed, C, E, and F are opened,
and the pump can draw and discharge through L. The valve K
is formed as shown in Fig. 103D, the flap being very light. It has
been found that the type of N.K.F. valve, shown in Fig. 97, is not
suitable for use with a centrifugal pump. To flood a compart-
ment, C is closed, B, D, E, and F are opened, and K is lifted off
its seat (for detail of this valve, see
Fig. 103D). The valve C is
fitted to isolate the pump in the event of repairs being necessary,
for using the portable pump, or when
flooding. E is fitted to shut
off all the suctions when using the
pump for charging the rising
main. D is fitted as an auxiliary to B in case the latter could
not be completely closed. It will be noticed that the valves C, E,
F, G, and H
are of the double-faced straight through type given
in detail in Fig. 103c this type offers no obstruction to the flow of
;
water. In some ships a mud box with perforated grid inside has
been fitted in the position shown, and the cover can be readily
removed for clearing out (see Fig. 103E). The hand- wheels and
deck-plates for flooding are marked L, which indicates that they
" "
are locked and must only be opened under proper instructions.
Flooding Magazines. Magazines and shell rooms can be

FIG. 103o.
FIG. 103B.
Fio. 103R
FIG. 103D.
124 WAR-SHIPS.
flooded by valves which can be operated either from the main

deck or below, the alternative position being provided in case one
was untenable owing to fire (see Fig. 10 3 A). Both positions are
"locked." The arrangement above is an ordinary deck-plate
worked by a spanner. The fitting shown in Fig. 10 SB is adopted
below. As shown, the wheel loose on the spindle.
is If it is
desired to use this wheel for opening the valve, the padlock to the
cotter pin is unlocked, the sleeve is lifted up, and the cotter pin
is inserted in the dotted position. The upper portion of the
spindle then free and the wheel can operate the valve.
is
A book
is issued to each ship, and among other things it is
stated how each compartment may be drained and pumped, or

flooded, and where the valves are worked from. A specimen page
(leaving out ventilation) is appended.
CHAPTER X.
VENTILATION.
THE problem of the effective ventilation of a ship is a more

difficultone than the ventilation of a building. In the case of a
building, ventilation is assisted because the porous nature of the
walls allows air to diffuse through, this diffusion being all the
more effective as the difference of temperature within and without
is greater. In the case of a ship, however, the conditions are
quite different, and the impervious nature of the skin renders
diffusion through impossible; ventilation must be obtained by
the actual introduction of fresh air and the withdrawal of the
foul air.
Ventilation can be either natural or artificial. By natural
ventilation is meant supply and exhaust without the aid of fans.
By artificial ventilation is meant that fans are used to draw pure
air down to, or draw foul air away from, the space to be ventilated,
or both. Natural ventilation is used for the special case of the
coal- bunkers. Artificial ventilation is used for the other spaces
below, as mess spaces, spirit-rooms, magazines, shell-rooms, store-
rooms, etc.
The side-lights are useful for ventilation purposes. For the
spirit-room it is necessary to have an independent supply and fan
exhaust, with mushrooms on the weather deck, as it is undesirable
to have any connection between the spirit-room and any other
system of ventilation which might communicate with a magazine.
Coal-bunker Ventilation. The efficient ventilation of coal-
bunkers is of extreme importance, because of the gas that comes
from the coal. This gas when mixed with air forms an explosive
mixture, and if it is allowed to accumulate may cause serious
explosions. The gases are especially liberated from the coal,
if a sudden fall of the barometer occurs, or if the temperature
rises.
126
VENTILATION. 127
It is worth noting in this connection that it is undesirable to take wet coal
on board, because the moisture causes a rapid and dangerous generation of heat
and gas. The coal should also be kept as dry as possible (thus when cleaning
the main deck the coaling scuttles should be kept closed.) Temperature tubes
are provided in all coal-bunkers, so that the temperature in the body of the coal
may be ascertained at frequent intervals. No light except a safety must be

lamp
used inside a coal-bunker until it is ascertained that the bunker does not contain
explosive gas. Special precautions in this respect are necessary for a few days
after coaling.
In the ventilation of a coal-bunker two pipes are employed, one

for the introduction of fresh airand one for the withdrawal of the
foul air. The latter is led where possible up a funnel casing.
These casings under ordinary circumstances will be hot, and this
causes the air in the pipe to rise. This induces a current of air
over the top of the bunker, and fresh air is then drawn down the
ejupply pipe leading from the upper deck.
There are slight differences in the fittings of different ships,
but the general principles will be understood by Fig. 104, which
shows the ventilation of the coal-bunkers of a battle-ship of the
Canopus class. In this type of ship there are three separate series
of bunkers, which are each independently ventilated.
1. The upper bunkers, behind the armour, which extend over the
length of the double bottom.
2. The lower bunkers, abreast the engine- and boiler-rooms.
3. The wing bunkers, abreast the engine- and boiler-rooms (can
be used for coal if desired, and in the deep load condition these
are assumed to be filled).
1. Upper bunkers. These bunkers are formed into groups,
Fig. 102 shows the arrangements for the forward group of four. A
trunk is taken into the funnel casing into which all the four bunkers
exhaust. Each bunker has a separate supply by means of a pipe
leading from the upper deck. The coaling scuttles in the main
deck can also be removed and the gratings put in if desired.
2. Lower bunkers. The supply pipes for these bunkers are
taken down the ventilators in order to avoid piercing the thick
decks, and the exhausts are taken up the funnel casings. A
separate supply and exhaust is taken from each bunker, the
supply pipe being taken to the far corner so as to get the current
of air well over the top of the bunker.
3. Wing bunkers. These bunkers are ventilated similarly to
the lower bunkers.
128 WAR-SHIPS.
The upper end of each supply pipe is fitted with a louvre with
"
fittings for locking, and the louvre has engraved upon it, Not to be
closed without special order." The top ends of the exhausts are led
above the upper deck, each with a mushroom top and a throttle
_BO*T
T THROTTLE VALVE.
L. LOUVRE.
I
V. SLIDE VALVE.
SECTION.
PLAN.
UPPER BUNKERS
~ LOWER BUNKERS.
Fio. 104. Ventilation of coal-bunkers.
valve for closing. Where these pipes have a bend where water
is likely to accumulate, a plug is fitted so that the water can be
drained away. The flanges of the deck beams have holes as

necessary to allow air to circulate freely over the top of the coal
when the bunker is full. The slide valves and throttle valves to
the supply and exhaust pipes have been omitted in recent
ships.
All the exhausts end in a louvre on the side of the
casing.
VENTILATION. 129
It is to be observed that more recently both supply and exhaust

pipes end in louvres at the upper end. These pipes are made
watertight, and the sluice valves fitted in the above case at the
watertight bulkheads are not therefore necessary.
Bunker Ventilation. The type of louvre now used for
bunker ventilation is of the watertight pattern shown in Fig. 105.
As shown, the shutter is up, and secured in position by a key,
which prevents unauthorized interference. When it is desired to
close the opening as, e.g. when the boiler-rooms are under forced
draught, the shutter is lowered and secured in the lower position.

130 WAR- SHIPS,
are in pairs (see Fig. 49), fitted as far apart as possible, and hand
fans can be used to ventilate the spaces before entering.
Ordinary Ship Ventilation. 1. With large steam-driven fans-
For the artificial ventilation of spaces below, the system adopted,
up to and including Canopus class, was by means of a number
of large steam-driven fans. Trunks were led from these fans, and
these trunks were pierced by louvres to allow air to pass into the
compartments through which the trunks pass. Fig. 106 shows
some specimen leads in a ship of the Majestic class. In these
ships there were ten fans, eight being 6 ft. in diameter, two being
4 ft. 6 in. in diameter. Some of the most important of these are
connected together, so that if a fan breaks down the other can do
its work. With this system the watertight bulkheads are con-
stantly pierced by the non-watertight trunks, and it is necessary
at such places to fit valves which will automatically close the
opening. These valves are designed so that if water rises on

either side of the bulkhead or floods the flat, a float is lifted which
releases the valve closing the opening. Two forms of these auto-
matic valves have been largely adopted, viz. Beck's and Broad-
foot's. In Beck's valve an ordinary slide valve is used, which
is connected to a balance weight. When the float rises it
releases this balance weight, and the valve then shuts. In
Broadfoot's valve, the valve turns round and closes the opening.
In either case a small hole is made in the bulkhead with a pipe
leading to the float, so that if water rises on the opposite side to

the float, the valve will be shut. This hole is closed by the
movement of the lever carrying the balance weight. Small
pipes are led from wells in the main or upper deck, so that in
case of emergency the valves can be closed by pouring water into
these wells.
There are objections to this system, viz.
(i.)
The large fans require steam-engines, which heat the com-
partments, and which require long lengths of steam and exhaust
pipes.
(ii.) The use
of the automatic valves to keep the flats and
bulkheads intact. These are frequently found either to work too
stiffly, so that they do not act, or to be too sensitive, so that the
motion or vibration of the ship continually closes them. In the
former case the watertightness of the bulkheads is completely
destroyed, and in the latter great inconvenience is caused by the
shutting off of the ventilation.
I
32 WAR-SHIPS.
In the above system of ventilation the exhaust is usually

obtained through hatchways and doors, but for some compartments
this is insufficient, and special exhaust trunks are led to the upper
deck with cowls. Such spaces are compartments for auxiliary
machinery, capstan engine, and steam steering engine. (These
trunks can be used for escape purposes.)
The special arrangements for magazine ventilation will be
referred to later.
2. With fans driven ly
electric motors. The system now
adopted have
is toa large number of fans driven by electric
motors taking the current from the electric power circuit of the
ship. The principle of the system is that the vessel is divided
into a number of sections, so that the main watertight bulkheads
of the ship below water are not pierced, and the compartments in
each of these sections are supplied with air by one or more motor
fans. In this system automatic valves are unnecessary, and the
inconvenience of heating by steam-engines and steam-pipes is
obviated.
Compartments below the main deck are ventilated by means
of a fan which delivers air into a watertight valve chamber
from which watertight pipes, usually six in number, are led
to distribute the air to the several compartments, a separate
pipe to each compartment. A
slide valve is provided to cover
all the pipes leading from the chamber, and a watertight valve
is fitted to each of the pipes at the lowest deck or bulkhead
pierced.
Living and sleeping spaces have artificial supply and natural
exhaust, and the supply after leaving the fan is passed through a
steam heater for warming the air. The great difficulty is to avoid
draughts which lead to the pipes being stuffed up, and as far as
possible the outlets are taken so that the air issuing will not cause
inconvenience either to men sleeping in hammocks or sitting at
the mess tables.
Compartments containing auxiliary machinery, as dynamos
and hydraulic engines, are now always ventilated by the exhaust
system. That is, one or two fans are provided for exhausting
the hot air, discharging it above the weather deck, and the supply
" "
is natural and comes down the large escape trunk with opening
at or above the weather deck. The great advantage of this system
is that the hot air is drawn directly out of the compartment and
the fresh air is supplied without draughts, this latter is almost
VENTILATION. 133
impossible when supply fans are employed, as they deliver the air
necessarily at a high velocity.

The provision of suitable openings at or above the weather
deck to provide suitable supply or exhaust in all weathers is a
difficult matter. If such openings are provided with
closing
arrangements (as a screwed mushroom top) there is at once the
disadvantage that when ventilation is most necessary the openings
have to be closed, and the spaces below become untenable.
Water - excluding mushroom
ventilators of various types
have been employed, and the
type of mushroom top shown
in Fig. 107 has been found
to give good results in bad
weather. Some water finds
itsway down in very rough
weather where such tops are
in exposed positions, and in
such cases it is necessary to
fit water-tight slide valves
beneath the deck. Wherever
possible, openings for supply
and exhaust, for important
places occupied in action, are
carried well above the weather
deck.
In either of the above
systems of ventilation, spaces
like provision-rooms, etc., are
not supplied with separate
FIG. 107.
ventilating pipes, but they
can be ventilated when re-
a nozzle on an adjacent
quired by means of a hose taken to
ventilation pipe.
forventi-
Magazines. Magazines require special arrangements
because of the of the ammunition. It is necessary
lation, presence
to avoid any undesirable rise of temperature, and also any excessive
amount of moisture, because the ammunition is very sensitive in
these respects.
which
Magazines are now fitted with cooling arrangements
are incorporated with the ventilation. A closed circuit of air
g
I S a
|JI5
D
-J55 a .3
S s
rf
II
111 _l-g
*JS 78
_-c s
',
cc - |11.^
CI.-M ci
S T3
Ha
.52 S
1
S 2
OD 2-
STS J
II
^ îj
55 ^q P
Cu o
= o o-o
VENTILATION. 135
in which there is a turbine-room and a condenser-room on either

side of a middle-line bulkhead.
Each turbine-room is ventilated by two 40-inch supply fans
and two 50-inch exhaust fans, and each condenser-room by two
30-inch supply fans and two 35-inch exhaust fans, all
electrically
driven. Natural supply and exhaust is arranged for in each case,
hinged shutters being provided to be placed over the fan inlets.
The diagram has been drawn showing for the turbine-rooms
natural supply and fan exhaust, and for the condenser-rooms fan
supply and natural exhaust.
Natural supply to turbine-room. The air comes down the
trunk and then through a door and down the engine room hatch,
the shutters being placed over the eyes of fans.
Fan exhaust to turbine-room. The shutters shown on upper
deck level are closed, and the air goes into the eye of the fan to be
discharged upwards.
Natural exhaust to condenser-room. The shutters are placed
over the eyes of the fan and the cover at upper deck is opened.
Fan supply to condenser -room. Air is drawn down into the
eye of the fan and is discharged into the condenser-room.
Of course, both fan supply and exhaust might be employed
simultaneously, and this would doubtless be done in action, with
all the armoured scuttles closed.
The scuttles to engine-rooms are shown by H, these are fitted
with a balance weight, and usually also with a small scuttle in the
cover capable of being lifted from below in case the wire of the
balance weight is shot away. Ladders are indicated by L. These
scuttles are not made watertight as the spaces are bounded by
watertight bulkheads with watertight doors at W.
The openings in the armoured decks are provided with armour
gratings.
The supply openings are provided with trunks to lead the air
to various portions of the engine-rooms and below the platforms.
GENERAL SCHEME OF ARTIFICIAL VENTILATION.

Onp. 136 is shown in tabular form the general
scheme
"
of ventilation as between " natural and " artificial."
The book referred to in Chapter IX. issued to each ship, con-
tains also information as to the ventilation of each compartment,
and the type of information, so far as ventilation is concerned, is
given in the table on p. 137.
-
-a d
!!'
g
53
ci'd
9cc
O
S.S
*-cS
08 08
ct ft. O J PM
VI *
isl
-.9 gog-o
2 fiP.1^
all
"si
a^
-
* af
o a
fie
7?.^l|lll
- *3
CHAPTER XI.
CORROSION AND FOULING.
Rust. If bare iron or steel is allowed to remain in contact with

moist air containing carbon dioxide (CO^), a chemical action goes
on, by which the oxygen in the gas unites with the iron, and
forms certain oxides of iron. We
term the resulting compound
rust. A
similar action goes on if the iron or steel is immersed in
fresh or salt water, owing to the carbon dioxide contained in the
water. Eusting action is much hastened by heat, and where heat
and moisture exist together, rusting goes on very rapidly indeed.
Eust is about six times as bulky as the iron from which it is
formed.
Corrosion. is very much accelerated by galvanic
Corrosion
action. and copper are immersed in dilute acid and
If iron
metallic connection is made between them, an electric current is
set up, and the energy of the current is provided at the expense
of the iron. If iron and zinc are similarly immersed, it is found
that the zinc wastes away. Not only do different metals act in
this way, but different parts of the same plate may be sufficiently
apart in the electro-motive series, owing to differences of density,

etc., to give rise to an electric current if immersed in dilute acid.
Iron and its rust are sufficiently different as to give rise to a
current, and the iron which becomes corroded, so that when
it is
rust is once formed it does not cover up and protect the material,
but itself is a cause of further corrosion. Eust is hydroscopic, i.e.
it up moisture, and so dampness gets between

takes it and the iron,
and hastens the rusting and corrosion.
A black oxide is formed on
steel during the process of manu-
facture, called mill scale.This stands in the same relation to the

steel as ordinary rust, in that it is electro-negative to the steel,
giving rise to an electric current, by which the steel is further

corroded. This mill scale clings very tenaciously to the steel, and
138
CORROSION AND FOULING. 139
does not come offby ordinary scraping. It is, however, most
necessary that it should be completely removed before any paint

is applied, and this is done in Admiralty practice by the process
termed pickling. The plates before being worked are immersed for
a few hours on edge in a bath of dilute hydrochloric acid (1 part
acid to 19 parts water). This loosens the scale, and the plates
when removed are brushed with hard wire brushes, and washed
with a hose to remove all traces of the scale and acid. The
portions of the structure thus pickled are those liable to come
into contact with sea or bilge water, viz. outer and inner bottom,
lower plates of bulkheads, plates of fresh- water tanks, and plates
of frames.
Steel is more liable to suffer from corrosion than iron, and
as the steel thinner than corresponding iron of the same
is
strength, a given amount of corrosion is relatively more serious

in a steel structure than in iron. Small vessels are usually
built relatively much stronger than large vessels, because of the
necessity of providing a margin against corrosion. In the case,

however, of vessels of special construction like torpedo-boat
destroyers, the conditions are such that the weight of the hull
structure must be the least possible consistent with sufficient
structural strength, and the thickness of the plating, etc., is in
consequence not enough to leave much margin against corrosion.
Thus in these ships the greatest care and frequent examination are
necessary, in order to guard against corrosion in its early stages.
In these vessels the plating, etc., is galvanized before painting, i.e.
after being thoroughly cleaned the material is immersed in a bath
of molten zinc. The layer of zinc thus deposited is a valuable
preventive against corrosion.
It is clear from the above remarks that no part of the structure
of a ship must be allowed to remain bare, but all must be securely
protected against rust and corrosion. A

point of considerable
importance is the necessity for the thorough cleaning and drying
of the iron or steel before applying any paint, as if any rust or
scale remains the corrosion will go on underneath the paint, and
if the surface is not dry the paint will not adhere. In building
H.M. ships it is laid down that each portion as finished must be
thoroughly cleaned and painted with linseed oil or thin red-lead,
to prevent oxidation while building. Subsequently three coats
of red-lead paint are applied, except in confined spaces, where
oxide of iron is used. The object of these successive coats is to
140 WAR-SHIPS.
hermetically seal the surface of the steel. In store-rooms, etc., the

first coat is of red-lead, and the two succeeding coats white.
The outer bottom is coated with three coats of special protective

composition, and the portion in the vicinity of the waterline is
coated with a special boat-topping composition. Above this grey
oil paint is employed.
Wherever any fitting, made of a copper alloy, like gunmetal, is
1
attached or adjacent to the steel structure, an electric current is set

up, as already stated, if the paint comes off the steel. The necessary
condition for this is that there shall be metallic connection between
the copper and the steel, both being immersed in dilute acid (sea
FOR ST&EL SHIP

WITH
DOUBLE BOTTOM.
FOR SHEATHED
SHIP
<XB. PLATING.
FOR STEEL SHIP

WITH FIG. 108. Zinc protectors, etc.
SINGLE SKIN.
water or bilge water is acid enough for this purpose). Under

these circumstances very rapid pitting would take place in the
plating, etc., in way of gunmetal valves, rudder-head, propellers,
1
Gunmetal has the composition copper 88, zinc 2, tin 10.
etc., and in all such places a strip of zinc is secured to the plating,
called a zinc protector (see Figs. 77 and 108 for examples). (For a
sheathed ship it will be noticed that these zinc protectors are not
necessary (Fig. 108)). The current then goes on between the gun-
metal and the zinc, rather than between the steel and zinc or the
steel and copper, and the zinc becomes corroded. The zinc thus
protects the steel structure by its own decay, and it can be readily
examined and renewed as opportunity offers. It is evident that
these zinc protectors should le left unpainted.
Gunmetal and brass propellers of steel ships are now kept
bright, not being painted at all, in order to reduce the friction of
water on the blades, which is an important factor in
propeller
efficiency. This, however, results in a tendency to corrosion
between the metal propeller and the adjacent hull due to galvanic
action, and special attention has to be paid to the coating of the
hull and to the cleaning of the zinc protectors in the vicinity.
An extra coat of composition is applied to the portions likely to
be liable to corrosion.
Fouling". In dealing with the outer surface of a ship below
water, we have notonly to seal up the plating to prevent corrosion,
but also to avoid, as far as possible, the attachment of marine
gro^hs and animals. This is termed fouling, and some means
of preventing fouling must be adopted in order to keep the
resistance as low as possible. The increase of resistance due to
fouling very considerable, and if a certain speed is desired it
is
means a larger expenditure of I.H.P., with consequent increase in

coal consumption, or with a given I.H.P. a serious loss of speed.
Copper is the most efficient anti-fouler, and in vessels which

have to serve on distant stations with the likelihood of remaining
afloat for long periods without docking, it is necessary to use
copper sheets to prevent fouling. The action of copper in this

regard is held to be as follows The
: action of sea water on copper
is to form certain copper salts. These form a deposit which dissolves
off the surface, and any marine growths, etc., which have attached
themselves to the ship are thus washed off as the ship moves along.
This action of copper is termed exfoliation. Although the copper
ispoisonous, it is not by poisoning that the growths, etc., become
detached, since these only use the ship to attach themselves to, the
nourishment being obtained from the water. There is, however,
said to besome poisonous effect on germs that are deposited.
With wooden ships, therefore, for many years copper was the
142 WAR-SHIPS.
FIG. 110.
standard material to avoid the attacks of marine animals, etc., and to

prevent fouling. When, however, iron came into use for ship-
building, it was at once found that copper was impossible because
of the galvanic action between the copper and the iron skin
unless they were insulated from each other. For ships on isolated
stations and in tropical waters, the composite system of construction
was largely employed. In this system the framing was of iron
or steel, and the skin was formed of two thicknesses of wood.
On to this the copper sheets could be nailed, and insulation
Fi. in.
between the copper and the structure was obtained. Fig. 109
shows the method of securing the wood sheathing. The inner
l
thickness was connected to the frame bars by naval brass bolts
screwed through, with a nut on the inside. The outer thickness
was connected to the inner by copper through bolts. The com-
for
posite system is not, however, structurally strong enough
vessels of any size and power, and no ships of the Eoyal Navy
have been built on this system for some time. The system
like
adopted instead has been to build the ship completely
an iron or steel ship, with skin plating as usual but on to this ;
skin woodsheathing, usually teak, is fastened.

thus get a We
ship structurally as strong as an iron or steel ship,
with the
of the
advantage of the copper sheathing. It is obviously highest
1
Naval brass is composed of copper 62, zinc 37, tin 1.
144 WAR-SHIPS.
importance to have complete insulation between the copper and

the iron or steel skin, and in the most common method
adopted
up to 1887, two thicknesses of planking were used, fastened as
shown in Fig.The inner thickness was secured to the
110.
plating by galvanized iron bolts, and the outer to the inner by

naval brass screws, care being taken to keep the points away
from the skin. The system was expensive, and in practice it was
found that water found its way between the planking and the
plating, and the iron bolts became corroded. The Calliope, was
sheathed on this system, and when she was taken in hand in
1891 for refit and repair, the outer thickness of planking was
removed, the inner thickness refastened with naval brass bolts,
and the outer thickness refastened.
The system adopted since 1887, for sheathed vessels in the
1
Eoyal Navy, is by a single thickness of leak. The principal
features of this system are as follows :
1. The adoption of such a thickness of wood sheathing as will
admit of thorough caulking. The mean finished thickness of

teak accepted is 4 in. for large ships, and 3J in. for the smaller
classes.
2. The use of naval brass bolts and nuts with their points
screwed through the bottom plating, and with plate washers under-
neath the nuts.
3. The thorough water-testing of the skin plating before the
planking is worked.
4. careful fitting, fairing, and fastening of the planks
The most ;
the coating of all faying surfaces with suitable compositions before

the planks are fitted in place, and the subsequent injection of
composition in order to fill any spaces left between the planking
and the plating.
The use of hempen grummets steeped in red-lead, under
5.
the bolt heads and under the plate washers to secure watertightness
in the bolt holes. The use of a plug of cement over the bolt head
to prevent the sheathing nails coming into contact with the bolts
and destroying the insulation.
Ill shows the method of fastening above described.
Fig.
For any vessel sheathed with copper it is impossible to use
iron or steel for stem, sternpost, etc., in consequence of the
galvanic action referred to above. For these vessels these portions

are made of the copper alloy phosphor bronze, a material which
See paper by Sir W. H. White (I.N.A., 1896).

can be made to give good castings, and which also possesses good
strength.
Copper sheathing. The surface of the vessel is payed over with
pitch, and tarred paper is placed on, and the copper sheets are then
fixed on with brass nails.
Ships of large size, including vessels of Powerful, Diadem and

Or'ess y classes, have been built on this system, and for vessels on
foreign stations likely to remain undocked for long periods, the
prevention of fouling obtained is worth the extra expense involved.
The following comparisons show the cost in money and measured
mile speed in the case of ships of Edgar and Apollo classes. In
these classes a direct comparison is possible, as the vessels were
similar in all respects except in the matter of sheathing. No
vessels have been so constructed since the Cressy class.
1 46 WAR-SHIPS.
that are deposited. It is clear, therefore, than an iron or steel
ship requires frequent docking in order to renew the anti-fouling

paint. On these occasions the bottom should be carefully
examined to see if the surface has become corroded.
Prevention of Corrosion Inside. We have seen above that
no part of a ship's structure must be left bare, or else rusting and
corrosion will certainly ensue. Examination is continually
necessary to ascertain how far the paint is protecting the steel
structure, and it is necessary to provide access to all parts for this
purpose. For places of which no use is made, access is usually
obtained by manholes, Figs. 48, 48A, 50, and 50A, the latter being
the type fitted on bulkheads, etc., and the former to the double-
bottom compartments. The means of access to every compartment
of a ship is given in the book of watertight compartments supplied
of which a specimen page is given at the end of Chapter IX.
It is laid down that every accessible part of the outer and inner
bottom and framing is to be inspected once a quarter by the engineer
officer and the carpenter, and any defects discovered are to be
made good. In the event of dampness, the steel must be
thoroughly dried and all traces of rust removed before applying
the paint. Well-slacked lime is to be used in places from which
water cannot be removed.
The survey of the hulls of steel vessels is made by the dock-
yard officers as follows :
Battleships, first and second class
cruisers, every four years. Third class cruisers, scouts, and small
vessels with plating over 10 Ibs. (J in.), every two years. Torpedo-
boat destroyers, torpedo boats, and other small vessels with plating
10 Ibs. and under, every year.
Pipes at the lower parts of a ship should preferably be of
galvanized iron, and not copper or lead. If, however, copper or
lead pipes have to be used, it is necessary that they be well
painted, covered with canvas, painted to make quite waterproof.

In way of metal valves zinc protectors are fitted to assist in
preventing corrosion.
The inner bottom plating under engines and boilers is specially
liable to corrosion, especially the upper surface, due apparently to
the fretting action of the ashes and hot water. These parts should
be frequently examined, and where rust is found to be forming, or
where the paint is abraded, the surface should be thoroughly scaled,
cleaned, and dried, and coated with three coats of red-lead paint.
Cement. The cementing carried out in recent ships is of far
less extent than that formerly adopted. Cement may even be

detrimental, as if it gets cracked through any cause, water will
get down to the plating, and corrosion may go on unnoticed.
The insides of fresh- water tanks are coated with " Eosbonite."
In living spaces, etc., corrosion and discomfort is caused by the
sweating of steel work, owing to the condensation of moisture from
the air. In such spaces the underside of decks, bulkheads, etc.,
are painted with one coat of red-lead and then covered with fine
cork and painted white. The cork does not cool so quickly as the
steel,and so condensation does not take place so readily. This
application is termed cork cementing.
(A very complete discussion of rusting, corrosion, and fouling
" See also a
is given in Professor Lewes's Service Chemistry."
paper by Mr. Holzapfel, I.N.A., 1904.)

Enamel. Enamel is now widely used instead of paint for
portions of the ship both inboard and outboard not subject
to
rough wear. This enamel can be washed and kept clean.

CHAPTER XII.
ARMOUR AND DECK PROTECTION.
THE three methods of attack which a war-ship may have to with-

stand are ramming, torpedo- or mine-attack, and gun-fire. The two
former would cause damage principally at and below the waterline,
and the only protection that can be afforded is the extensive
watertight subdivision. For protection against gun-fire, armour
plates are provided over as large a portion of the ship as possible.
These armour plates are, in most cases, backed up by coal and
thick decks. The parts not protected by armour are minutely
subdivided in the neighbourhood of the waterline in order to
localize damage as far as possible. Thick decks, both above and
below water, are also largely employed for purposes of protection.
A large proportion of the weight set aside for protection in
modern ships is necessary for the protection of the armament.
Thus in a recent battle- ship the total weight provided for pro-
tection was 4335 tons, and this is divided as follows :
Armour and deck protection to hull for the preservation of

buoyancy and stability, 2875 tons.
Protection to armament, as barbettes and casemates (not
including gun shields), 1460 tons.
This shows that, in this case, about one-third the available
weight was devoted to the protection of the armament and two-
thirds to the ship.
It is proposed to trace briefly the history of the subject from
the Warrior until the present time, taking typical ships. The
dates given refer to the year in which the ships were laid down.
"Warrior." The first large vessel provided with armour
protection in this country was the Warrior (1859). This ship
was built of iron, and was 380 ft. long, and 9200 tons displace-
ment. The armour was 4^ in. thick, and extended, as shown in
Fig 112, for a length of 218 ft. and a depth of 22 ft. This 4.J-in.
148
\
WAR-SHIPS.
armour was the thickest that could then be produced, and experi-
ments carried out at the time showed that it was sufficient to
withstand the guns then in use. The armour was of wrought iron,
and this material was used for protection right up to the Inflexible,
(1874). Owing, however, to the continuous
development of gun power the thickness and
disposition of the armour in subsequent ships
underwent considerable modifications. One
serious disadvantage in the Warrior was the
unprotected state of the ends and the rudder-

head. This was remedied in subsequent
ships.
" Minotaur." This ship (1861) was larger
than the Warrior, being 400 ft. long and
10,690 tons, the extra size being necessitated
by the increase of the thickness of armour
to 5J in. amidships (Fig. 113); and the
provision of 3-in. armour at the ends.
" "
The next step
Bellerophon (1864).
was the production of a shorter and handier
ship, the Bellerophon, which was 300 ft. long
and 7550 tons. There was a complete belt
at the waterline (Fig. 114), 6 in. thick amid-
ships, tapering at the ends as shown. Above

the belt for a length of 94 ft. amidships there
was an armoured battery of 6-in. armour.
"
Temeraire." The above type of ship
with a complete belt and a battery amidships
continued until the Temeraire (1873), in
which ship the belt was 11 in. maximum
thickness, with a battery 8 in. thick (Fig.
115). In this ship we also find 8-in. and
10-in. redoubts towards the ends for mounting
the heavy guns. The Temeraire was 285 ft.
long and 8540 tons.
"Devastation" (1869). This was a
special type of ship inwhich sails were abandoned and twin screws
adopted. She had low freeboard (Figs. 116 and 117), and was
285 ft. long and 9320 tons. The belt, 12 in. thick amidships, was
continued to the ends 8 in. thick. Above this belt there was a
central breastwork 153 ft. long, 10 in. thick along the sides, and
12 in. round the turrets. These turrets were 14 in. maximum
thickness.
During the above period (1859 to 1873) we have noticed that,

owing to successive improvements in guns, continuous increase
had to be given to the thick-
ness of armour protecting the .
ships, and a stage was at length -BREASTWORK

reached when it was found
impossible to cover any large IE.
area of a ship's side with

armour thick enough to resist
the fire that could be brought
to bear against it. The next
development accordingly con-
sisted in only attempting to
protect a portion of the ship

with very thick armour, and
to depend on deck and other
protection for the remainder.
The principle acted upon was FIG. 117. H.M.S. Devastation.
that it would be better to
efficiently the midship portion of the ship in way of

protect
machinery and heavy guns, than to cover a large area with thinner
armour that could not keep out the enemy's fire. This principle
was carried to its extremest limit in the Inflexible.
"Inflexible" (1874). This ship was 320 ft. long and 11,880 tons.
The armoured citadel (Figs. 118 and 119) was 110 ft. long, with a
maximum thickness of 24 in., the armour being in two thicknesses.
Forward and aft of this citadel the magazines, etc., were protected
by under-water decks, 3 in. thick, and some protection was

afforded by the stowage of coal and cables on these decks, with
cork packing at the sides and cofferdams, as shown.
In this ship the hull armour was still of iron, but the outer
thickness of the turret armour, 9 in., was "compound" or steel
faced.
Figure of Merit of Armour. The resistance of armour to
perforation is compared with the perforation of wrought iron as
found by some empirical formula. There are a number of these
formulae, the one used in England being that obtained by Captain
"
Tresidder (see Gunnery Manual "). He found that perforation
of wrought iron could be represented by the formula
W.L.
SECTION OF
SIDE.
Fro. 119.
o
ARMOUR AND DECK PROTECTION. 153
wv 3
=
* (log 693,500,000 8-841)
I) 693,500,000
where t = thickness of wrought iron in inches
V= striking velocity in feet per second
D= calibre of shot in inches
W = weight of shot in pounds (uncapped).
The figure of merit of any particular plate is the ratio between
the thickness of wrought iron and of that plate which is just
perforable by a given projectile and the same striking velocity.
EXAMPLE. A 6-in. armour plate is attacked by a 6-in. 100-lb. Holtzer
armour piercing shot with 2200 per second striking velocity.
ft. The shot
just perforates the plate, find the figure of
merit
100 x (22QO) 3
we have t z =_ for wrought iron.
6 693,500,000
By using logarithms t is found to be IG'O in., so that the figure of merit
is 2-66.
The factor of perforation of a projectile is the ratio between

the thickness of wrought iron perforable by a given projectile at
a given striking velocity and the thickness of the plate against
which it is fired.
If these two are equal the plate is just perforated.

Compound Armour. A compound armour
plate consists of
a wrought-iron plate artificially attached to a steel face of about
half its own thickness, the result ,
being a plate with hardness of

steelon the face by which projec-
tiles arebroken up, and toughness
of wrought iron at the back which
prevents cracking taking place.
As at first made this armour had
a figure of merit of about 1J, but
improvements in manufacture
brought up the ratio to 1*7. In
France all steel plates were used,
but experiments showed that this
material was lacking in toughness,
and trouble was caused by the
spontaneous cracking of these FIG. 121. Section Admiral class.
plates. Compound plates were
used in ships after the Inflexible up to and including the
Royal Sovereign (1889).
154 WAR-SHIPS.
The next distinctive type of ship was that

Admiral Class.
known Admiral class (1880). The various ships of the
as the
class differ somewhat in detail, but their main features are the same.
The Collingwood, 325 ft. long and 9500 tons (Figs. 120, 121), may
be taken as a typical ship of the class. The armour belt was 18 in.
thick amidships, 7 ft. 6 in. broad, and it extended for a length of
140 ft., or about seven-sixteenths the length of the ship with
athwartship bulkheads. No protection save coal was provided
above this armour belt. Forward and aft were two sloping bar-
bettes of 11 J in., with trunks coming down within the limits of the
armour belt as shown. The floor of these barbettes was 3 in. The
ends of the ship were unprotected by vertical armour, but under
water decks were worked 2 J in. in thickness. The top of the belt
was covered in with a level deck 2J in. thick.
While this class of ship was building, quick-firing guns and high
explosive shell were being developed, and although only 6-pounder
shell were projected, yet it was felt that the large areas of unpro-
tected side in these ships was an element of serious danger. The
influence of this factor is seen in the Nile and Trafalgar (1886),
and in the next main type of ship, the Royal Sovereign class.
" "
Royal Sovereign (1889). This ship was 380 ft. long and
14,150 tons displacement. There were eight of the class built,
Fm. 122. Armour, etc., H.M.S. Royal Sovereign.

(Note six 6-in. casemates have recently een placed on the upper deck.)
seven with barbette mountings for the heavy guns (Royal Sovereign,
Ramillies, Repulse, Royal Oak, Resolution, Revenge, Empress of India),
and one with turrets (Hood). The former ships were able to obtain
a greater height of big guns above water than the Hood on account
of the weight involved in the turrets.
In these ships the waterline belt was 8J ft. broad, with a
maximum thickness of 18 in., covered in with a horizontal deck
3 in. thick (see Figs. 11 and 122). The length of the belt was
250 ft., or two-thirds the length of ship. Bulkheads, 16 in. and

14 in., closed in the belt at the forward and after ends respectively.
The one-sixth of the length at the ends had underwater decks
2J in. thick. Above the 18-in. belt, from middle to main decks.
4-in. armour is worked, with an upper coal-bunker behind. This
armour was of nickel steel, and was fitted in order to determine the
explosion of shell filled with large bursting charges of high explosive
outside the ship, and to prevent the free perforation of the side
above the belt by the smaller nature of quick-firing guns.
The barbettes mounting the 13J-in. guns were pear-shaped and
of very substantial construction, the maximum thickness being 17 in.,
and the armour extending right down to the middle deck. The
armour for this class was compound, except the 4-in. side, which
was of nickel steel, and the main deck casemates of steel.
Harveyed Armour. The next large group of ships were those
of Majestic class, nine in number. In these ships armour made
by the Harvey process was adopted. This armour is estimated to
have a figure of merit of from 1/8 to 2 '2. The development of
armour-piercing projectiles of forged steel was the cause of this
improvement in armour manufacture. In the Harvey process an
all steel plate is used, and the face is cemented, i.e. animal charcoal
is placed next the face of the

plate (two plates being usually
dealt with together, face to face), and the whole is covered in with
bricks and run into a gas furnace, where it remains two or three
weeks, seven days or so being allowed for cooling. In this way
the proportion of carbon on the face is increased, and the front is
then capable of being hardened. The plate is first cemented as
above, and then bent to the required shape and all necessary
holes made in the surface. It is then heated and the face douched
with cold water, which makes the front of the plate exceedingly
hard. We thus have a compound plate, but the junction between
the hard face and the tough back is much more perfect than in
the compound plates. The object to be attained was a steel plate,
without welds, having such a proportion of carbon in the surface
that water cooling would produce a very hard face. As the
thickness of the hard steel is practically constant for all thick-
nesses of plate, it follows that thin plates obtain relatively higher
values of the figure of merit than thicker plates. That is, a 12-in.
plate not twice as good as a 6 -in. plate.
is For Harveyed
armour a figure of merit of 2-2 is obtained when 6 in. thick, but
lower values are obtained for thicker plates.
i
56 WAR-SHIPS.
"
"Majestic There are nine ships in this class (Ma-
(1894).
jestic, Magnificent, Mars, Jupiter, Hannibal, Illustrious, Victorious,
Prince George, Csesar). They are vessels of high freeboard, and
(Figs. 123, 124). A new departure
are 390 ft. long and 14,900 tons
Fio. 123. Armour, etc., H.M.S. Majestic.
was taken in these ships in the disposition of the armour. The

belt was 9 in. only, but it was carried up to the main deck for a
total depth of 15 ft. The length of the citadel was 250 ft., or
about two-thirds the length. The protective deck between bar-
bettes was run level across at the middle line, but sloped down to
FIG. 124.
the bottom of the armour at the side, as shown in Fig. 124. This
deck was 3 in. on the flat and 4 in. on the slope. In this arrange-
ment of armour the influence of the larger nature of quick-firing
guns is seen. The protection at the waterline is not sufficient
to keep out the heaviest projectiles. It is, however, backed
up
by the 4-in. sloping deck (which is about equivalent to 5 in.
horizontally), so that before

the vitals of the ship are
reached a penetration equi-
valent to about 23 in. of
wrought iron is necessary.

Although this does not
compare well with the 30
in., say, in the Royal Sove-
reign, yet it is certain that

a ship will be most difficult
to hit at the waterline, and
it was considered better to
give up absolute protection
at the waterline in order to
obtain a larger area of good
protection, because of the
development of the large
quick-firing guns. The
ends of the ship are not
protected by vertical ar-

mour, but have underwater
decks 2 \ in. thick. In some
of the later ships of this
class the barbettes are cir-
cular,and not pear shaped

as Fig. 123, owing to a
change of the type of gun
mounting.
Apoint worth noting in con-
nection with a narrow belt (as in
Fig. 131 A), is the fact that when
deeply loaded, or damaged so
that sinkage takes place, the top
of the belt may possibly be at or
below the waterline, and the
advantage of a thick belt is then

lost. Another point is that ships
increase their draught as time
goes on (about an inch yearly),

owing to alterations and additions,
BO that the value of the belt gets
less and less during the course of
years, unless steps are taken to

reduce the draught.
I 5 8 WAR-SHIPS.
Krupp Armour. The Harvey process of manufacturing armour

was soon superseded by the Krupp process. The steel for this
process has a high tensile strength, approaching 50 tons per square
inch, and contains small proportions of carbon, nickel, chromium,
and manganese. Plates 3 in. and above are cemented, and are
termed K.C., or Krupp cemented. Plates below 3 in. are not
FIG. 131 A. Section of French ship Charlemagne.
cemented, these are termed K.N.C., or Krupp non-cemented. For

K.C. plates the cementation is carried out in a similar manner to
that of the Harvey process, but in the final face hardening the
plate not heated bodily as in the Harvey process, but the heat
is
is graduated from the face to the back. After heating the face is
placed under the cold water douche.
For K.N.C. plates the composition of the steel is similar, but

the plate is water cooled without previous cementation. These
plates are about equal in resisting power to Harvey plates. K.C.
plates have a figure of merit about 2-3 to 27. In some few cases
a higher figure of merit, approaching 3, has been obtained.
Some of the processes do not use the Krupp patent, and armour
is specified to be K.C. or equivalent quality, or K.N.C. or
equivalent quality, as the case may be.
(An exhaustivesummary of the armour question year by year is given in

Lord Brassey's Naval Annual. See also the supplements to Captain Orde-
Browne's "Armour and its Attack by Artillery.")
i6o WAR-SHIPS.
w-v 2 =
% (log 2,255,000 6-3532)
d 2,255,000
the termsmeaning the same as before.
This formula can be used to construct a diagram as Fig. 132,
in which the thickness of plate can be picked up for various
projectiles at different striking velocities.

A rough rule which obtainable from the diagram is that at
is
2200 f.s. striking velocity an uncapped A.P. shot can perforate a

K.C. plate of its own diameter.
The scales at the right-hand side show the reduced perforation
obtainable when the plate is inclined at 20 and 30 to the direction
of the projectile.
Armour Bulkhead. In the case of a thwartship armour bulk-
head to a ship with a sloping protective deck as Majestic and
following ships, it is necessary to pay attention to the thickness of
this bulkhead where the middle deck does not come behind.
Fig. 133 shows the after armour bulkhead of a battle-ship with

7-in. side armour. The armour protecting the barbette gun
forward and aft, are 2 in. A section of the side amidships of this
class is given in Fig. 127.
" "
Formidable Class. The next type of ship laid down
(1898) was the Formidable class (Formidable, Implacable, Irresis-
tible). These ships are 400 ft. long and 15,000 tons. They are
protected similarly to the Majestic, with 9-in. armour, but the
main deck is made 1 in., and the middle deck 2 in. on the flat and
3 on the slopes (see Fig. 132 for section at side). The lower
in.
decks at ends are 2 in. forward, 2J in. aft, and the side forward
has 2 in., worked on a double thickness of plating. The side
plating aft is increased to a total thickness of 1^ in. The barbettes
have a maximum thickness of 12 in.
" The Formidable class (Bulwark,
Bulwark." five later ships of
London, Venerable, Queen, Prince of Wales) (1899) are of the same

dimensions and displacement as the Formidable, but the armour is
arranged differently at the forward end. The forward armour bulk-
head dispensed with, and the side armour is carried forward from
is
the 9 in steps of 7 in., 5 in., and 3 in., and right forward

in. side
we have 2 in. worked on a double thickness of skin plating. The

main deck is 2 in. is worked to the stem,
over the citadel, and
where it is 1 in. The middle deck in way of citadel is 1 in.
thick.
on the flat, 2 in. on the slopes. The lower deck forward is 1 in.,
and where it slopes to meet the middle deck before the forward
barbette it is 2 in. The lower deck aft is 2J in. The side plating
aft is increased to a total thickness of 1^ in. The whole arrange-
ment is similar to that of the Duncan (Fig. 134).
" Duncan " Class The six vessels of this class (Duncan,
(1899).
Cornwallis, Exmouth, Russell, Albemarle, Montagu) are 405 ft. long
and 14,000 tons. They were designed for a higher speed than
their predecessors, viz.
19 knots on 8 hours' trial, 18,000 I.H.P.
18 30 13,500
In consequence of this, these ships had less displacement and a
finer form than Formidable, in order to avoid excessive I.H.P., and
the armament being the same as in that ship, the weight available
for protection would only allow a 7-in. belt to be worked. This
belt is reduced in thickness to the forward end, as shown in Fig. 134.
Aft the plating is doubled, and on the flush surface thus obtained
20 -Ib. ( J in.) nickel steel is worked. The decks are as follows :
Main deck 2 in. between barbettes to 1 in. right forward.

:
Middle deck 1 in. uniform between barbettes.

:
M
Lower deck : 1 in. forward, 2 in. aft ; slope to middle deck

forward, 2 in.
In above ships back to the Majestic, the secondary

all the
armament of 6-in. guns are carried in isolated casemates having
6-in. fronts and 2-in. backs.
"King Edward VII." (1902). Of this class there are eight,

viz. King Edward VII., Commonwealth, Hindustan, Dominion^
Hibernia, Britannia, Africa, New Zealand.
The particulars are :
length, 425 ft. ; breadth, 78 ft. ; draught,
26 ft. 9 in.; displacement, 16,350 tons; I.H.P., 18,000; speed,
18^ knots ; armament, 4 12-in., 4 9'2-in., 10 6-in. In these ships
the casemate system of protecting the secondary armament has
been adandoned, and a battery, 7 in. thick, has been worked
between the main and upper decks to take the 6-in. guns (see
Figs. 13 and 135). The battery is covered in with 1-in. plating at
the top, and this battery also performs the function of protecting
the funnel casings, etc., to the upper deck.
The armour belt is 9 in. at the waterline, 8 in. above to main
deck. carried in reduced thicknesses to the bow, as
This belt is
shown in Fig. 135. A

bulkhead is worked at the after end of
citadel as usual. In this ship also the after end is protected by
2 in. worked on doubled plating. The decks are as follows :
Upper deck over battery, 1 in.

Main deck, forward of battery, 2 to 1 J in.
Middle deck, 1 in. flat, 2 in. slope.
Lower deck, forward, 1 in., aft 2J in.
The barbettes for the 12-in. guns have a maximum thickness

of 12 reductions being possible where the battery, etc., would
in.,
also have to be pierced before reaching the barbette. The shallow
barbettes for the 9'2-in guns are 4 in. thick.
Figs. 125 to 131 give an interesting comparison of the

armoured sides of British battle-ships from Eoyal Sovereign to
King Edward VII.
"
Triumph
"
and " Swiftsure." These two ships were
designed for the Chilian Navy, and bought by the British Govern-
ment in 1904. Their dimensions are 436 ft. by 71 ft, and 11,800
tons. The guns are two pairs of 10-in. in barbettes forward and
aft, ten 7i-in. guns in main deck battery, and four 7J-in. guns
in upper deck casemates. The distribution of the armour
protection is shown in Fig. 135A.
"Lord Nelson" (1905). This class consists only of two

ships, the Lord Nelson and the Agamemnon (Fig. 135B). The
particulars are :
length, 410 ft. ; breadth, 79 ft. 6 in.; draught,
27 displacement, 16,500 tons; I.H.P., 16,750; speed, 18
ft.;
knots. There are no guns on the main deck, and the 6-in. gun
is dispensed with. The guns in the upper deck are four 12-in.
guns in pairs, eight 9 -2 -in. guns in pairs, two 9'2-in. guns in
single mountings amidships. The bases of the 9'2-in. mountings
are protected by a citadel between the main and upper decks
of 8-in. armour. This citadel is absolutely unpierced, and extends
between the barbettes of the 12-in. guns. The armour belt is
12 in. thick at the waterline and 8 in. to the main deck. Forward
it is 6 in. and 4 in., and aft 4 in. There is an armoured bulkhead
aft 8 in. thick, but the middle deck underruns the bulkhead and
then slopes down to meet the lower deck aft. There is a similar
slope forward. These slopes are 4 in. thick. The barbettes
for the 12-in. guns have a maximum thickness of 12 in. The
middle deck is 1 in. on the flat, 2 in. on the slopes. Lower
deck, forward 1 in., aft 2 in. and 3 in. The main deck forward
is 1J in.
"
Dreadnought" (1905). This ship inaugurated a new depar-
ture in the design and building of war-ships. The reasons are
summarized in the Admiralty memorandum reproduced in the Ap-
pendix. In the Lord Nelson
class, as seen above, the 6-in. gun was
dispensed with, the armament consisting of four 12-in. and ten

9'2-in. guns, all on the upper deck. These ships were limited in
length to 410 ft., and the highest speed that could be arranged for
in the design without an abnormal horse-power and without the
sacrifice of other important qualities was 18 knots (18J knots being
obtained on trial with the specified horse-power). These ships
had a large coefficient of fineness, viz. 0*66, and this causes a very
great increase of resistance near the top speed for a small in-
" "
crement of speed owing to the rapid growth of the wave making
resistance. The step to the Dreadnought was governed by the
adoption of a heavy armament of uniform calibre, viz. ten 12-in.
guns in pairs, and to place these guns so that they were free to
fireover very large arcs of training meant a much longer ship }
the length finally fixed upon being 490 ft. This length in con-
junction with the breadth (82 ft.), the draught (26 ft. 6 in.), and
the displacement (17,900 tons) leads to a coefficient of fineness of
0'59. With a form of this length and fineness a higher speed than
ARMOUR AND DECK PROTECTION 167
formerly was obtainable without great increase of power, and the

speed decided upon, viz. 21 knots, was obtained with 23,000 horse-
power, or 12 per cent, more speed than the Lord, Nelson with
about 14 per cent, more power. The adoption of turbines led to
considerable economy, both as regards weight and space, because
the turbines for a given power require fewer boilers, using the
steam more economically than reciprocating machinery. The
21 knots of the Dreadnought would have been quite impossible
in a vessel of the Lord Nelsons form even with turbines, because
of the enormous horse-power necessary.
The ten guns of Dreadnought are disposed in shields in five
barbettes, arranged as shown in Fig. 135c, eight guns fire on one
broad-side and six can fire simultaneously either ahead or astern.
One pair of guns is mounted on the forecastle and the other four
pairs are on the upper deck. The anti-torpedo boat armament
consists of twenty-four 12-pounder guns.
The armour protection to the ship extends only to the main
deck, the belt at the waterline being 11 in. and 8 in. above. The
belt is continued to the ends of the ship, being 6 in. forward and
4 in. The barbette armour varies from 11 in. to 8 in. in
aft.
thickness, and the shields to the 12-in. guns are 11 in. in thick-
ness. The forward conning tower is 11 in. and the after conning
tower is 8 in. The main deck is a protective deck J in. thick,
and the lower protective deck is 1J in. on the flat and 2| in. on
the slope. There are five submerged torpedo tubes, one being right
at the stern.
" "
Bellerophon (1907). The ships immediately following the
Dreadnought, viz. Bellerophon, Superb, and Temeraire, were of
similar type, but the draught was increased 6 in. and the dis-
placement by 700 tons. The dimensions are 490 ft. x 82 ft.

X 27 ft. X 18,600 tons; 23,000 horse-power for 21 knots. The
same number and disposition of 12-in. guns was adopted, but
sixteen 4-in. guns were provided for the anti-torpedo boat arma-
ment instead of 12-pounder guns.
"St. Vincent" (1908). The ships of this class, St. Vincent,
Vanguard, and Collingwood are of similar type, but a new 12-in.
gun of 50 calibres in length was adopted with 4-in. guns for the
minor armament. The displacement and dimensions are greater
than in the former designs, being 19,250 tons and 500 ft. length,
84 ft. beam, and 27 ft. draught, the speed still being maintained
at 21 knots, this requiring 24,500 horse-power. There are three
submerged torpedo tubes, two on the broadsides forward and one

at the stern.
" This ship, which
is the only one of her
Neptune" (1909).
class, is 510 ft. beam, 27 ft. draught, and 19,900 tons
long, 85 ft.
displacement, a designed speed of 21 knots being obtained with

25,000 horse-power. The disposition of the armour protection
was generally similar to the previous ships, but a different dis-
position of the ten 12-in. guns was adopted, enabling all of them
to be fought on each broadside. The midship guns are arranged
en echelon, and the two after turrets on the middle line are
arranged so that the forward pair can fire over the after
pair.
The boats are carried on girders between the superstructure.
The ship has two tripod masts with control tops, and carries
three 18 -in. submerged torpedo tubes. The an ti- torpedo boat
armament consists of sixteen 4-in. guns.
" Hercules " This vessel and the Colossus form a class
(1909).
by themselves. Their design was a slight variation from the
Neptune, the arrangement of guns being similar, but three 21-in.

tubes were arranged for, and a single foremast (with control top)
only is fitted. The dimensions are the same as the Neptune, but
the displacement is 20,000 tons, and the horse-power and speed
are the same.
" Orion "
(1909 and 1910). This class, consisting of the Orion,
Monarch, Conqueror, and Thunderer, has the following dimensions,
545 ft. x 881 ft. x 27J- ft. X 22,500 tons, 2*7,000 S.H.P., 21
knots. There were two special features in this design, viz., size
of guns and their disposition. The 13J-in. gun was adopted,
ten in number, and they are all disposed on the centre line. Two
pairs are forward, and two pairs aft,
in each case the pair nearer
midships firing over the adjacent pair. One pair is placed

amidships. The armour in these ships is carried to the upper
deck.
"King George V." (1910 and 1911). This class, consisting

of King George Ajax, and Audacious, has the follow-
V., Centurion,
ing dimensions length,:555 ft. breadth, 90 ft. draught, 28 ft.

; ; ;
displacement, 23,000 tons, and with 31,000 horse-power, are de-

signed for 21 knots. Ten 13-in. guns, sixteen 4-in. guns, and
three 21-in. torpedo tubes are provided for as in the Orion class.
The cost of these ships is given as
King George V. (Portsmouth), 1,961,096,
Centurion (Devonport), 1,950,671,
1
70 WAR-SHIPS.
Ajax (Scott, Greenock), 1,889,387,

Audacious (Cammell Laird), 1,918,813,
including guns, but excluding stores.
" Iron Duke" This class, consisting of Iron Duke,
(1912).
MarlborougJi, Benlow, and Emperor of India, has the following
dimensions length, 580 ft.
:
breadth, 90 ft. draught, 28 ft.
; ; ;
displacement, 25,000 tons, and with a 29,000 horse-power have

a designed speed of 21 knots. Ten 13|~in. guns, twelve 6-in.
guns, and four torpedo boats are carried. The armour belt is 12 in.
thick, and is carried to the upper deck with a 6-in. battery for
the 6-in. guns.
" "
Queen Elizabeth This class, consisting of Queen
(1912).
Elizabeth, Valiant, and Malaga, has the fol-
Warspite, BarTiam,
lowing dimensions length, 600 ft. ; breadth, 90 ft. 6 in. ; draught,
:
28 ft. 9 in. ; displacement, 27,500 tons, and with 75,000 horse-

power, have a designed speed of 25 knots. Eight 15-in. guns,
sixteen 6-in. guns, and four torpedo tubes are carried. The
armour belt is 13 in. thick, and is carried to the upper deck with
a 6-in. battery for the 6-in. guns. The boilers are designed to
burn oil fuel only.
Protection of Cruisers. The essential quality to be obtained
in cruisers is high speed and large coal supply, and as much
protection is given as the weight available will allow. In the
belted cruisers of the Orlando class, protection was obtained by
a narrow belt, 5^ ft. wide, 10 in. thick, extending over about two-
thirds the length. There was a level deck 2 in. over the belt, and
decks at the end 3 in. thick. Some disadvantages of a narrow
Fifl. 13G. Armour, etc., H.M.S. Cressy.
belt have already been noticed, and since the Orlando until the
time of Cressy class (1897) large cruisers were "protected," 1 i.e.
they had the vitals covered in with a thick deck, and this thick
deck, in conjunction with the coal stowed above it, was depended
upon to give protection. We shall see in Chapter XVI. how
important this coal is, not only in offering direct resistance to

1
Edgar class ; Blake, Blenheim, Powerful, Terrible, Diadem class.
penetration, but in preserving stability when the side is riddled.
Examples of large protected cruisers are given in Figs. 21 and 22,
and the smaller cruisers of the second and third classes are still
FORCC.TLE Dg
[_
* AMIDSHIPS
FIG. 137. Sections of armoured cruiser Cressy.
constructed on this system (Figs. 24, 26, 27). The side bunkers
are also well subdivided by watertight bulkheads in order to
localizeany damage at the waterline (see Fig. 54).
" "
Cressy (1897). The ships of Cressy class represent a distinct
departure in the design of large cruisers. There were six ships of
this class, all being sheathed with teak and copper (Cressy,
Aboukir, Hogue, Sutlej, Baccante, Enryalus), 440 ft. long, 12,000

tons. Advantage was taken of the improved quality of armour
obtained by the Krupp process to armour the side for about half
length and a depth of 11^ ft. with 6-in. armour, closing in the ends
with bulkheads 5 in. thick (Figs. 136, 137). Two-inch protective
plating was worked to the bow directly on the ship's plating. The
middle deck is 1 J in., and main deck 1 in. in way of armoured side.
The lower deck forward is 1^ in., and aft 2J in. In a cruiser,
armoured in this way, the height of the middle deck above water
can be made than in a protected cruiser, because of the
less
presence of the armoured side (compare Figs. 22 and 23).

" Drake" The four cruisers of Drake class (Drake,
(1899).
King Alfred, Leviathan, Good Hope) are unsheathed, and are
500 ft. long, 14,100 tons, and with 30,000 I.H.P., were designed
for the speed of 23 knots. The protection is similar to the Cressy,
with a 6-in. armour belt and 2-in. protection forward.
II
c 3
c
" Monmouth" The first cruisers of this class

(1899), (Mon-
mouth, Bedford, Kent, Essex, Cumberland,
Suffolk, Cornwall,
Berwick, Lancaster, Donegal) are 440 ft. and 9800 tons, 22,000
I.H.P., 23 knots. They are armoured with a 4-in. belt, with
middle deck f in., and main deck 1 in., as Fig. 23. The later
cruisers have a 6-in. belt, and are somewhat larger, viz. 450 ft.
long, and 10,700 tons, 21,000 I.H.P., 22J knots (Devonshire,
Antrim, Roxburgh, Carnarvon, Hampshire, Argyll).
" Duke of "
Edinburgh (1903). The first-class cruisers (Duke
of Edinburgh, Black Prince) are 480 ft. long and 13,550 tons,
23,500 I.H.P., 22JL knots. They are similar to King Edward VII.
in having an armoured battery (Fig. 138). This battery contains
the ten 6-in. guns of the secondary armament, and is 6 in. thick.
The main armament consists of six 9'2-in. guns on the upper deck
and forecastle in barbettes. The belt is worked over the whole
length, 6 in. thick amidships, 4 in. forward, and 3 in. aft. The
deck over battery is 1 in., the main deck forward and aft is
upper
1 in., the lower protective deck is f in., being thickened up to 3 in.
over the steering gear. The later ships of the class Warrior,
Achilles, Natal, Cochrane have had the battery of 6-in. guns on
the main deck dispensed with, and the armament consists of
six 9*2 in. guns mounted singly (five on upper deck and one
on the forecastle), and four 7* 5 -in. guns mounted singly on the
upper deck.
" Minotaur " The vessels of this
(1905). class, Minotaur,
Defence, Shannon, are 490 ft. long, 14,600 tons, 27,000 I.H.P.,
23 knots. There are four 9'2-in. guns, one pair on the forecastle
and one pair on the upper deck aft, and ten 7'5-in. guns on single
mountings on the broadsides. The disposition of armour is
shown in Fig. 138 A, and is similar to that of the Duke of Edin-
burgh, but there is no armour between the main and upper decks
except that protecting the mountings of the 9*2-in. and 7'5-in.
guns. The lower deck is of J-in. mild steel with 1-in. K.N".C. on
the flat and IJ-in. K.N.C. on the slopes, making a total of 1J- in.
on the flat and 2 in. on the slopes. The main deck is not fitted
with thick plating except over the engines; the lower deck
forward and aft is 2 in. thick.
"Invincible" (1906). There are three vessels of this class,
viz. Invincible, Inflexible, Indomitable. The dimensions are :
length, 530 ft. ; beam, 78 ft. 6 in. ; draught, 26 ft. ; displacement,

17,250 tons. The designed speed is 25 knots with 41,000 H.P.
174 WAR-SHIPS.
ro*.
.?/ I *
l\l I
The ship carries eight 12 -in. guns in pairs, arranged as shown in

Fig. 138B, the thicknesses of armour being as given in the
"
Naval Annual." The midship guns can fire on either broadside.
The six forward guns are on the forecastle level, and the two
after guns on the upper deck level.
"Indefatigable" (1909). This ship, of which type only one

was built, was an improved Invincible. The guns amidship are
arranged more en echelon in order that they may fire simultaneously
on the beam. The dimensions are length, 555 f t. breadth,
:
;
80 ft. draught, 26 ft. 6 in.

; ; displacement, 18,750 tons ; speed,
25 knots with 43,000 H.P.
"Lion" (1909), "Princess Royal" (1910). These vessels
are designed for the great speed of 28 knots, and their dimensions
are:
length, 660 ft. ; breadth, 88 ft.
draught, 28 ft. ; displace-
;
ment, 26,350 tons horse-power (turbines), 70,000. The arma-

;
ment consists of eight 13J-in. guns all arranged on the centre

line. The two forward pairs are in barbettes adjacent, the after
pair firing over the forward pair. The pair amidships are
above the forecastle deck level, and the after pair on the upper
deck level. The armour is carried to the upper deck level,
and the upper deck and lower deck are "protective" decks.
There are two 21-in. broadside torpedo tubes forward. The anti-
torpedo boat armament consists of sixteen 4-in. guns. These

ships were the first in the British Navy costing more than
2,000,000.
" This vessel
Queen Mary" (1911). is slightly larger than
the Lion, being 660 ft. length; 89
beam; 28 ft. draught; ft.
27,000 tons, 75,000 horse-power (turbines), and 28 knots. The

arrangement of main armament is the same as in the Lion.
" This vessel of the following dimensions
Tiger" (1912). is :
length, 660 ft. ; breadth, 90 ft, 6 in. ; draught, 28 ft. 6 in. ; dis-
placement, 28,500 tons, and with 85,000 horse-power has a
designed speed of 28 knots. Eight 13J-iu. and twelve 6-in. guns
are carried with four torpedo tubes. The armour belt is 9 in.
thick, and is carried to the upper deck with a 6-in. battery above
for the 6-in. guns.
Second and Third Class Cruisers. As mentioned above,
these cruisers are built on the deck-protected system, Fig. 24
showing a typical sheathed second class cruiser, Fig. 27 showing

a third class cruiser. The Scouts are also protected in the same
way.
i
76 WAR-SHIPS.
The protected cruisers named after British towns have the

following particulars :
Dimensions. Power and Speed. Armament.

Bristol . . .
Glasgow . .
1909. Gloucester .
Liverpool . .
Newcastle
FIG. 139. Armoured hatch.
FIG. 140 Armour gratings.

on angles as shown in the sketch. For a 1-in. deck the gratings

are 5 in. deep, in. thick, and the clear space between the bars is
2 in. The machinery and boilers down below are thus protected
in some measure. Most of these
gratings are prevented from lifting
as shown, but some have to be
hinged for access below, and these

were formerly fitted with balance
weights to make the lifting easy.
It is now
the practice to construct
the hinged gratings in sections
which can be readily lifted or
lowered, as in action the rope to
the balance weight might be de-
stroyed. Escape trunks are also

provided with sliding watertight
shutters below the lower deck,
worked from above and below.
Armour gratings are now con-
structed as castings shown in Fig.
FIG. 141.
141, as it has been found that
this much stronger against bursting shell than when built
form is
up. Fig. 141 shows the gratings in a boiler-room ventilator with

some of the sections in narrow portions capable of being lifted
from below. The fixed portions of the gratings are fitted with
holding down clips of shape to suit the local conditions.
Splinter nettings are provided below the
armour bars over the
engine-room and auxiliary machinery rooms, so as to protect the
machinery from debris that might get through the armour gratings.
The nettings are usually about 12 in. below the gratings.
Cofferdams. Around all openings in the middle deck which
are necessarily open in action a dwarf bulkhead is carried some
feet above the L.W.L. A similar bulkhead is run close to the
1
upper coal-bunker bulkhead (see Figs. 12 and 22). The space
thus formed, about 12 in., is termed a cofferdam, and into these
cofferdams canvas, oakum, or other such material can be jammed
down to act as a leak stopper, and so limit the flow of water
across the deck, supposing the sides and bulkheads pierced. This
is the primary function of these cofferdams, but in many ships
they are made rather wider, about 18 in., in order to allow bags
1
Omitted in recent ships (see Fig. 13).
MAIN DECK. MAIN-
_L
. 142.
FIG. 143.
FIG. 144. Support to 6-in. armour.
and hammocks to be stowed.

ELEVATION
Their use now is not so common AT A.B.
as formerly.
Armour Backing and
Supports. In the early days
of armour protection with plates
of wrought iron, the wood
the
backing was very thick,
a sup-
object being to provide
port which should be somewhat
elastic. This thick backing
was continued to the Admiral
class, in which the 18-in.
armour had 15-in. backing be-

hind. With modern plates a
small thickness of teak backing
is fitted, of minimum thickness,
2J in., to form a bed for the
back of the armour. This back-
ing is connected to the plating
behind armour by bolts screwed
through the plating with washer
and grommet inside. Formerly
the skin plating behind armour
was in two thicknesses, but the
present practice is to work
this plating in one thickness
of f in. or f in. Tigs. 142 and
143 show the framing behind
the 18-in. armour of the Royal
Sovereign. The 4-in. steel above
is supported by 6-in. zed bars.
In Majestic the 9 -in. armour
has 15-in. plate frames worked
every 24 in., with horizontal
stiffening girders as Fig. 18.
In Cressy the 6-in. armour is
supported by 10 -in. zed bars
worked every 24 in., with a FIG. 14 U. Sup-
horizontal girder as shown in port to barbette
Fig. 144. The framing behind

182 WAR-SHIPS.
armour in Lord Nelson is formed of two 10 in. by 3J in. angle

bars worked back to back and connected by a single riveted lap,
shown in Fig. ISA. The battery armour is backed and supported
by framing. Barbettes, although well fitted to withstand blows
by their shape, are supported inside the plating by closely spaced
vertical girders.
No is now employed behind the armour plating of
backing
barbettes, and in order to further diminish the diameter of the
barbette the framing and plating is stopped some distance from
the top (see Fig. 144A). The reduction of diameter thus obtained
has resulted in a considerable economy in the weight of armour.
Adjacent plates in the upper tier are connected together at the
upper part by keys.
Behind armour, where men are likely to be employed in
FIG. 145. Armour bolt.
action, the inside of theframing is covered in with thin plating

(see main deck in Fig. 13). When armour is struck, rivets to the
framing are likely to break and the heads fly off, so that this
lining forms some protection to men inside. This plating should
be secured by screw rivets to the framing (see K, Fig. 9).
Armour Bolts. With wrought-iron armour, the bolts for
securing the armour to the ship's structure were carried through
with a large conical head flush with the surface of the armour.
With hard-faced armour, however, the surface must not be pierced
because the surface would then be liable to crack badly
for bolts
from hole to hole when struck. Armour bolts are now screwed
into the back of the plate (Figs. 145 and 145 A), and about one
bolt to every 7 square feet is allowed. Fig. 146 shows the holes
in a specimen plate ; important to
it is have good security in order
to keep fragments together, even if the plate is badly cracked.
Experiments have shown

that the pieces are still very efficient,
are held up to the backing.

provided they
FIG. Hfix. Armour bolt.
In order to dimmish the liability of bolts breaking under the

the shank of the
impact of projectiles,
bolt is made slightly less
in diameter
than at the bottom of the thread. The

bolt will then stretch at this
weakest
rather than break under the thread.
part FIG. 146.
A sleeve was formerly fitted to provide
length for this
weakest portion; also, to provide some elasticity
FIG. 147 A. Armour bolt to barbette

armour with no backing.
FIG. 147. Armour bolt.
184 WAR-SHIPS.
to take the shock, the nut securing the bolt to the ship was fitted
with an elastic washer of india-rubber. This was placed inside a
" "
cup washer to keep the rubber washer in place. In some
places, as barbettes, it is not possible to get sufficient room for
the type of armour bolt shown in Fig. 146 ; in such places, the
bolt, as in Figs. 147 or 147A was used, the latter being fitted to
barbettes with no backing.
Armour bolts are now fitted as shown in Fig. 145A. Consider-
able weight, expense, and space are saved, and it has been shown
by experiment that these bolts are quite efficient in securing
armour to a ship's structure.
CHAPTER XIII.
RULES OF MENSURATION FOR THE CALCULATION OF AREAS

AND VOLUMES.
]. AREA of rectangle, as ABCD (Fig. 148). The area is given

by AB x BC. The length and breadth must be of the same
denomination, i.e. if the length is in feet the breadth must be in
feet, the area then being square feet.
2. Area of triangle, as ABC

(Fig. 149). draw per- We CD
pendicular to the base AB, meeting it, or it produced, in D. Then
the area J X = AB
X CD, or one-half the base into the height.
3. Area of trapezoid. This is a four-sided figure, in which
bwo sides only are parallel, as ABCD (Fig. 150). Calling the
parallel sides a and b respectively, and
h the distance between
them, the area = J (a -f b) h, or one-half the sum of the parallel
sides multiplied by the distance between them.
4. Circle. (a) Length of circumference is TT times the diameter,
where TT = 2
31416, or -^ nearly.
72
(b)
v /
Area of circle of diameter d = ?
4
5. Curvilinear figure, as ABCD (Fig. 151). The area of figures
of this character are continually required in ship calculations.
(a) Trapezoidal rule. This rule has found considerable favour,
especially in France and the United States, on account of its great
185
i86 WAR-SHIPS.
simplicity. We divide the base into a number of equal spaces

and erect ordinatesmeet the curve as EF, GH,
to etc. Calling the
common interval h and the length of ordinates y\ t y2 , . . .
y^ we
have, regarding the figures ADFE, EFHG-, etc., as trapezoids
Area ADFE = \ (yL + ?/ 2 ) A

Area EFHG = \ (y2 -f- 3/3) h, and so on to
Area NOCB = J (y + 6 y7) h.
Adding all together-
Area ABCD = \h( yi + 2y a + 2y3 + 2y4 + 2y5 + + 2/7)
+ 2/2 + 2/3 + 2/4 + y5 +
i.. the first and last ordinates are added together and divided by
two, then all the remaining ordinates are added and the total sum
is multiplied by the common interval.
The following example will illustrate the use of the rule and
how near the result obtained is to the real area required.
Thecurve, whose ordinates 2 ft. apart are 0, 2*2, 4*0, 5'4, 6'4, 7*0, and
7'2 respectively, is a portion of a common parabola, and the exact area
ft.
enclosed is 57'6 square ft. Find the area by using the trapezoidal rule.
= + '
7 2
Area 2
( 2
+ 2-2 + 4'0 + 5'4 + 6'4 + 7-o)
= 57-2 square ft.
There is thus an error of nearly 1 per cent. The error involved

in using this rule is lessened by spacing the ordinates
closely,
but it is not used in Admiralty calculations on account of the
approximate nature of the results obtained. The rule that is
employed is
(&) Simpson's first rule. Take a figure as ABCD (Fig. 152),
RULES OF MENSURATION. 187
aud divide the base into two equal parts in the point E. Then,
assuming the curve is a common parabola, the area ABCD is
given by (yi 4- 4y2 + y3). To apply this rule to a longer figure,
as ABCD (Fig. 151), we divide the base into an even number of

intervals, so that the above rule may be applied to each portion
containing a pair of intervals,
thus
Area ADHG = ^ (yi 4- 4y2 4- 2te)
Area GHML = |(ft

+ 4y4 4- ft)
Area LMCB = -
(?/ 5 -f 4ye 4- yi)
6
Combining together we have

Area ABCD = -
( yi + 4y2 4- 2y3 + 4y 4 4- 2^5 4 4y6
The multipliers thus are 1, 4, 2, 4, 2, 4, 1, and there must be

an even number of intervals or an odd number of ordinates for the
rule to be applicable. In working by this rule it is advisable to
use a table as follows, which is the calculation for the area of the
figure considered above, the exact area of which is known to be
57'6 square ft.
Number of
ordinate.
1 88 WAR-SHIPS.
EXAMPLE. The following ordinates, 1-3 ft. apart, give a curve which is an
arc of a circle of 6 ft. radius, viz. 0, 1'56, 2'41, 2-86,
3-00, 2-86, 2'41, 1/56, 0. The exact area of the circular
segment thus obtained is 22' 1 square ft. If the area be
calculated by the two methods considered above we
have
(a) Area by trapezoidal rule, 21*64 square ft.
(6) Area by Simpson's rule, 22'1
It is thus seen that the latter gives a correct result

even though the curve is not a parabola. Also the
trapezoidal rule is in error to the extent of over 2
per
cent.
Volumes. To find the volume of a solid

bounded by a curved surface. The volumes of
such bodies as this are continually required
CURVE OF SECTIONAL AREAS.
\
PIG. 153.
in ship calculations, the most important case

being the underwater volume of a ship.
The method adopted is to divide the
volume by a series of equidistant planes; we
then find the area of each of the figures traced
out on these planes by the surface, as in
Fig.
154, and treat these areas as the ordinates of
a curve having the same length as the
body.
The area enclosed by this " curve of areas " will
give the volume required.
In finding the underwater volume of a
ship
we may divide the volume in two ways, viz.
1. By means of equidistant planes in
transverse sections. The shape of these
sections " "
is given in the body plan in Fig.
155, and in perspective in Fig. 154.
190 WAR-SHIPS.
2. By means of equidistant planes in horizontal sections

parallel to the load water-plane. (In the ship, Fig. 155, it has
been found necessary to introduce a "half" waterline at the
bottom.)
In calculating out the underwater volume of a ship both these
methods are adopted. An excellent check is thus obtained on the
accuracy of the work, because the volume as found by either
method should be the same. The following example will illustrate
this
The underwater body of a yacht is divided by transverse planes 10 ft.

apart,
and the following are the areas, viz.
03, 22-7, 48-8, 73-2, 88-4, 82-8, 58-7, 26-2, 3-9 square ft.
The same body is divided by horizontal planes, 1 ft. 6 in. apart, having the
following areas, viz.
944, 795, 605, 396, 231, 120, 68, 25, 8 square ft.
To find the volume
The following is the calculation using the areas of the transverse sections, the
curve of areas being given in Fig. 153
Nnmber of
section.
RULES OF MENSURATION. 191
Number of
plane.
CHAPTER XIV.
NAVY LIST DISPLACEMENT, TONNAGE, ETC.
Length, The length of ships of the Koyal Navy stated in all

official documents is always the length between perpendiculars.
The forward perpendicular is a vertical through the intersection

of the fore side of stem with the normal load waterline, the ship
being supposed to be floating at this line (see Figs. 67 and 68).

The after perpendicular (a) in ships with the rudder hinged at the
fore side, is taken as the after side of the sternpost, as in Figs. 69
and 75 in ships with balanced rudders, it is taken as the centre
; (&)
line of the rudder-head (see Figs. 71, 72, and 79).
The length on the load wafcerline includes the overhang of the
stern at this line. In the United States Navy, for instance, this
is the length used in stating a ship's dimensions. Comparisons
between ships are apt to be misleading if the lengths are not taken
on the same basis. It is believed that the usual French practice
issimilar to that in the Eoyal Navy.
The length of ship over all includes the overhang of the stern
and the projection of the ram ; the length for docking purposes
includes also the overhang of the stern walk, if any.
Breadth. The breadth stated is the breadth of the hull at
the broadest part as designed. It sometimes happens that the
actual breadth as built is slightly greater than this. For
docking purposes the projection of casemates, bilge keels, etc.,
must be considered, as also the shape of the dock and dock
entrance.
Navy List Displacement, This is always used in official
documents, and is a figure which attaches to the ship so long as
she remains in the Navy. It is the total designed weight, in-
cluding the estimated weight of hull, machinery, armour, and

armament, legend weights of water, stores, and coal, and a weight
appropriated to a Board Margin.
192
NAVY LIST DISPLACEMENT, TONNAGE, ETC. 193
Draught. The draught of water corresponding to this

Navy
List displacement is the normal load draught.
It does not follow that, when the ship is finished, she will,
with legend coal, etc., exactly float at the designed load water-
line. For instance, the Board Margin may not be appropriated,
or only a portion of it. The weights of hull, machinery, armour,
or armament may turn out greater or less than estimated at the
time of the design. Ships of the same class, built at different
yards, from the same drawings and specification, sometimes differ
among themselves by considerable amounts.
The draught marks are not usually set up on the ship at the perpendiculars,
but at the points where the keel cuts up at the bow and stern. In the special
case of destroyers and vessels with propellers below the line of keel, a set of
draught marks are set up on the shaft brackets, showing the draught at the
bottom of propeller sweep. The draught marks at the after cut up of the keel
are used for docking purposes only.
The draughts of water of each ship in the Navy, as completed,

in three conditions of the ship, are given in the stability statement
furnished to the ship's book.
(a) The normal load draught, the ship being fully equipped,
with reserve feed-tanks empty, and with the legend weights of
coal, etc., on board.
1
(&) The deep load draught, the ship being fully equipped,
with
fresh- water and reserve feed-tanks full and bunkers full.
The light draught, being an extreme light condition
of the
(c)
ship. All coals, water (including reserve feed), provisions, officers'
stores and slops, and one-half the carpenter's, boatswain's, and
engineer's stores are assumed to be consumed.

No expenditure of
ammunition or shell is assumed for this condition.
For vessels with considerable sail, like the sloops, which are
likely to proceed under sail alone, the light
condition is taken as
above, but with the boilers quite full, and the engine condensers
and feed-tanks empty.
A specimen stability statement is given at the end of Chapter
XVIII.
It should be stated that ships generally increase in draught
somewhat as time goes on, owing to the alterations and additions
carried out. Large weights of paint are often worked into a ship,
1
In ships carrying oil fuel there is an extreme deep load draught which sup-
poses, in addition to (&), that the oil fuel compartments are filled to 95 per cent, of
their capacity.
P
194 WAR-SHIPS.
one coat succeeding another, until the weight of the whole is

1
very considerable. The question of the draught is looked into
occasionally as required, and when necessary a new stability
statement is issued. This is specially done when a ship has
undergone an extensive refit.
The position of the extreme deep load line is indicated inside
the ship by label plates and a broken paint line for the information
of the ship's officers.
Trials. The steam trials of a ship, carried out to ascertain
how far the speed estimated at the time of the design has been
realized, are run at the normal load draught corresponding to the

Navy List displacement. Ships are sometimes tried for this
purpose when in an incomplete condition, and they are then
ballasted to give the required draught of water. If the trial is,
however, simply to determine the power developed, and for the

acceptance of the machinery, it is not necessary to bring the ship
to the load line, provided that sufficient immersion is obtained for
the propellers. Except in special ships, as in destroyers, etc.,
contractors are not responsible for speed, but only for the power
developed under the given conditions. It has been found a great
convenience, in some cases, to finish the contractor's trials as early
as possible, so that the opening up of the machinery, to determine
the final acceptance, may proceed while the ship is being completed
in other respects.
Tonnage. We have seen that war-ships are known by their
displacement tonnage, this being the total weight as designed with
legend weights on board. This tonnage is specially suitable for
war-ships, as these ships have to carry a fixed load of armour, guns,
etc. For merchant ships, however, a different system is adopted ;
here the tonnage is a measure of the internal capacity of the ship.

Gross tonnage is the total closed in capacity of the ship,
excluding double bottoms, reckoned in tons of 100 cubic ft. The
makes certain deductions from the gross
nett or register tonnage
for the space occupied by the crew, etc., machinery and coal.
The intention is that the nett or register tonnage shall give a
measure of the earning capacity of the ship for carrying cargo

and passengers.
It is necessary that ships of the Eoyal Navy should be
measured for their tonnage, in order to form a basis for the
1
In one case the paint removed from the crew space of a destroyer weighed
over 2 Ibs. per square foot
NAVY LIST DISPLACEMENT, TONNAGE, ETC. 195
payment of dues and other charges at foreign parts. A tonnage

certificate is issued to all the ships of the Navy from measure-
ments made by the officers of the Board of Trade. There are two
systems, viz.the British system and the Danube system. They
differ in the allowances made for deck erections and machinery
spaces. The following shows how these compare with one

another, and howthe gross tonnage compares with the displace-
ment tonnage in several cases. The Danube rule is the one used
when passing through the Suez Canal.
CHAPTER XV.
BUOYANCY, DISPLACEMENT, TONS PER INCH, ETC.
Buoyancy. At every point of the surface of a body im-

mersed in water there is a pressure which acts normally to the
surface. The amount of this pressure will depend on the depth
of the point below the surface. If d be the depth of the point
below the surface in feet, and w the weight of a cubic foot of water,
then the pressure per square foot is w X d Ibs. Thus if a hole of
1 square ft. is made in a ship's bottom 17J ft. below the surface
of the water, a piece of wood would have to be held against the
hole with a force of - = 0'5 ton to keep the water out

It is because of this pressure that diving operations beyond a
certain depth are rendered impossible.
In the case of a floating body like a ship, these normal pressures
all over the surface act in many different directions. In each case,
however, the normal pressure may be resolved into its three com-
ponents at right angles, viz. (i.) horizontal ina fore-and-aft
direction, (ii.) horizontal in a transverse direction, and (iii.) vertical.
If the ship is floating at rest, all these horizontal components must
balance between themselves, since there is no bodily movement
of the ship in any direction. It is the combined effect of all the
verticalcomponents which exactly balances the weight of the ship.
The single vertical force, which is the resultant of an infinite
number of small vertical forces acting on the ship, is termed the
buoyancy. In the same way the forces due to the weights com-
posing the ship have a single resultant, which we term the weight
of the ship. The vertical forces acting on the ship at rest are
therefore
(a) The weight of the body acting vertically down.

(b) The buoyancy acting vertically up.
196
BUOYANCY, DISPLACEMENT, TONS PER INCH, ETC. 197
Since the body has no motion up or down, it follows that the
buoyancy exactly equals the weight.

If we take a block of wood, say pitch-pine or teak, and suspend by a spring
balance, the dial would show the weight of the wood, say 10 Ibs. If we place the
block into water we should notice, as the wood descends, that the dial registers
less and showing that the water is taking some of the weight. A point will
less,
at last be reached when the dial will register zero, and then the wood is floating,
and instead of the 10 Ibs. being borne partly by the balance and partly by the
water, the whole weight is taken by the water. That is, the buoyancy is equal
to the weight.
Displacement. We now come to a most important proposition

in connection with floating bodies. A body floating freely and
at rest displaces or puts aside a volume of water having a
weight exactly equal to the weight of the body. The water
displaced is termed the displacement, and can be either reckoned
as a volume, when it is expressed in cubic feet, or as a weight,
WATER SURFACE,
FIG. 156.
when it isexpressed in tons. That the above proposition must

be true may be seen by the following :
Consider a vessel floating freely and at rest in still water, and

imagine if it were possible that the water is solidified, maintaining
the same level, and therefore the same density. If now we lift
the vessel outwe shall have a cavity left behind, which is exactly
of the form of the underwater volume of the ship, as Fig. 156.
Now suppose the cavity is filled with water. This amount of
water is evidently the displacement of the vessel. Suppose now
that the solidified water outside again becomes water. The water
we have poured in will be supported by the water surrounding it.
The support given, first to the vessel and now to the water we have
poured in, by the surrounding water must be the same, and con-
sequently it follows that the weight of the vessel exactly equals
the weight of the water poured in to fill the cavity, or, in other
words, the weight of the vessel is equal to the weight of the water
displaced.
ig8 WAR-SHIPS.
This fact is of immense assistance in dealing with the
weights of ships. We do not need to estimate the weights, which

would be an almost impossible task, but, knowing the line at
which the ship is floating, and having the drawings giving the
shape of the ship, we can calculate the volume of displacement up
to that line. Then, knowing the density of the water, we at once
have the weight of the ship with everything she has on board.
When a man is floating in water, it is manifestly desirable to keep the arms
below the surface because of the buoyancy due to their displacement. It will
be noticed that, when thus floating, if the arms are held up out of the water, a
certain amount of sinkage occurs. The weight of the body remains the same,
but the buoyancy is reduced and must be made up by the sinkage before the
balance is obtained, viz. that the weight equals the buoyancy.
The same principle has to be borne in mind when constructing a raft. All
the human beings have to be placed on it, but a great quantity of provisions,
etc., may be safely carried under it. For instance, a cask of beef weighs 300 Ibs.,
and itsvolume displaces 184 Ibs. of water, so that if carried beneath the raft we
get 184 Ibs. of buoyancy from it, the net weight of the cask is therefore
only 116 Ibs.
The density of water varies at different places, and often at

the same place at different states of the tide. Thus in places on
the coast and at sea it is practically constant, viz. 64 Ibs. per cubic
foot, giving 35 cubic ft. to the ton. At Glasgow, well up the
Clyde, the water weighs 62J Ibs. per cubic foot, giving 35*84 cubic
ft. to the ton. At Gravesend the water is 63' 7 Ibs. per cubic
foot at high tide, and 63 '4 Ibs. per cubic foot at low tide. It is
because of the difference of density that a vessel decreases her
draught in going from fresh to salt water. This is of importance
in merchant vessels, which are allowed to be loaded below the load
line disc, when floating in water less dense than salt. Thus a
vessel of 20 ft. depth, if in fresh water (1000 ozs. to the cubic foot),
isallowed by the Board of Trade officers to be loaded 4 in.deeper
than the load line disc, because it is known that when she has
got to sea she will rise this amount. If the same ship were being
loaded in Aberdeen Harbour, for instance, where the water weighs
1015 ozs. to in. only would be allowed.
the cubic foot, 1J
The whose displacement has been calculated in the
vessel
previous chapter as 4052 cubic ft., will weigh if floating at the top
4052
waterplane in salt water QK = 115 '8 tons. If floating at the
OO
same waterplane in river water, of which 35'6 cubic ft. go to the
4052
ton, the weight would be ^^ = 114 tons.
OO'O
Curve of Displacement, We have seen how to determine

the volume of displacement of a ship when floating at a given
waterplane. The draught of a ship, however, continually varies
owing to having different weights of coal, stores, etc., on hoard,
and so it is desirable to have the means of determining quickly
the displacement of a ship at any other given draught. The body
having been cut by a series of equidistant planes, we calculate the
displacement to each of these planes in succession. Thus in a
battle-ship the following figures were obtained at planes 4 ft. 3 in.
apart, the top one being at a draught of 31 ft., viz. 18,300, 15,016,
12,121, 9,293, 6,510, 3,923, 1,901, 268 tons respectively. can We

then set up a scale of draughts, as in Fig. 157, and at each line set
out the corresponding displacement. Through the spots thus given
we then draw a curve, which is termed the curve of displacement.
In this case, suppose the ship is floating at a draught of 19 ft. 6 in.
forward and 20 ft. 10 in. aft, i.e. 20 ft. 2 in. mean. We set up this
draught at AB
and measure to the curve, and find the displacement
to be 10,550 tons. It is usual to continue the curve right down
to zero draught, although the ship could never float at a less
draught than would be due to her structure alone. The smallest

displacement obtained in the history of any ship would be when
she was launched. i
Tons-per-inch Immersion. It is frequently necessary to

know how much a vessel, when floating at a given draught (a) will
sink, if certain known weights are put on board, or (5) will rise if
certain known weights are removed. Since the displacement of a
vessel equals the weight, any extra displacement caused by adding
a weight must equal the added weight. If A is
the area in square
feet of the waterplane at which a ship is
floating, then the volume
^
of a layer 1 in. thick is ^
la
cubic ft., and the displacement of this
_:_
A- This must be the weight neces-
l
35 s- tons.
(A\
sary to add to sink the ship 1 in., or to take out to lighten the ship
1 in. This is termed the tons-per-inck immersion.
EXAMPLE. The area of the waterplane at which a vessel is floating is 7854

square ft. Find the rise due to burning 56 tons of coal.
fJQ C A
=
Tons per inch
-^ = 18'7 tons
56
Rise = fQ-^.
= 3 in.
200 WAR-SHIPS.
Approximation to the Value of the Tons-per-inch Im-

mersion, If L is the length between perpendiculars, and B is the
breadth of the hull
Then (1) for ships with fine L.W.P. as cruisers
Tons per inch = approximately.
(2) for ships of fuller form as battle-ships

/T
y TA
Tons per inch = ~or\ approximately.
These approximations will be found useful in the absence of
the correct figure for a particular ship. The correct tons per inch
is usually given in the Ship's Book. The length between perpen-
and breadth of hull are used above instead of the length
diculars
and breadth on L.W.L., as the latter are not usually known with-
out reference to the drawings.
The following show how far the approximation holds in different ships :
Third class cruiser actual 18'7, approximation 18-25

First class cruiser 58 59-2
Battle-ship of fine form 54-2 57*7
Battle-ship of full form 56-7 55-2
Curve of Tons-per-inch Immersion. The above figures refer

to the tons per inch at the L.W.L. In order to determine the
tons per inch at any other line at which a ship happens to float
we construct a tons-per-inch curve. We
calculate the tons per
inch at the several level lines parallel to the L.W.L., and set
these out at the corresponding draughts on a convenient scale.
Thus, for a battle-ship, the tons per inch at level lines 4 ft. 3 in.
apart, commencing at 31 ft. draught, were found to be respec-
tively 57'5, 57-2, 56-5, 537, 51-0, 46 3, 39'0, 24'0. The curve
drawn through the spots thus obtained, as in Fig. 157, is the
curve of tons-per-inch immersion. If the vessel floats at a draught
of 19 ft. 6 in. forward and 20 ft. 10 in. aft, i.e. 20 ft. 2 in. mean,
this draught is set up and the ordinate of the curve, 54*8, is the
tons per inch required.
In ordinary ships the value of the tons per inch varies very
little for considerable changes of draught in the neighbourhood of
the L.W.L.
Coefficient of Fineness. This is the ratio which the actual
volume of displacement bears to the volume of a
rectangular block
having the same length as that of the ship between perpendiculars,
the same breadth and the depth equal to the mean draught of the
ship (the draught should be excluding any keel projection, if any,

as in a sheathed ship).
The value of this coefficient gives us a good idea of the degree
of fineness of a ship, as the following examples show ;
110.000 ,15,000 I
20,000
I I I I l
I I I I
FIG. 157.
"Formidable." 400 ft. x 75 ft. x 26| ft. x 15,000 tons; 15,000 I.H.P.,
18 knots.
Volume of displacement = 15,000 x 35 cubic ft.
Volume of block = 400 x 75 x 26-75 cubic ft.
Coefficient of fineness
= = '
65
400^75 x 2 6'75
'
202 WAR-SHIPS.
"Duncan." 405 ft. x 75| ft. x 2CJ ft. x 14,000 tons; 18,000 I.H.P.. 19
knots.
14,000 x 35 =
_
of fineness
= fl
Coefficient 40 5 x 75'5 x 26'5

"
Dreadnought." 490 ft. x 82 ft. x 26 ft. x 17,900 tons; 23,000 H.P.,
21 knots.
of fineness = 0-59.
"Amethyst." 360 ft. x 40 ft. x 14| ft. x 3000 tons; 9800 I.H.P., 212
knots.
3000 x 35
Coefficient of fineness = =
360 x 40 x 14'5
"Drake. 500 ft. x 71 ft. x 26 ft. x 14,100 tons; 30,000 I.H.P., 23
knots.
Coefficient of fineness = 14,100x35 = nro

500 x 71 x 26
of fine-
The following are average values of this coefficient
ness, viz.
Battle-ships 0'6 to 0'65
Cruisers 0'5 0'55
Destroyers 0*4 045

As the importance of the fineness of a ship in
illustrating
connection with the attainment of speed, reference may be made
to the Formidable and
Duncan. In the former
ship a coefficient of fine-
ness of 0*65 was adopted,
15,000 I.H.P. being in-
tended to realize 18 knots.
In the latter ship, in order
SEA WATER.- to attain 19 knots without
FIG. 158. an excessive expenditure
of I.H.P., a finer form
had to be adopted with a coefficient of fineness of 0'6.
Difference of Draught of Water when Floating in Salt
Water and River Water. (Salt water 64 Ibs., river water 63 Ibs.
to the cubic foot.) The weight of the ship going from the one to
the other remains the same = tons = W
X 2240 Ibs. W
Volume of displacement in salt water = cubic ft.
river
W x 2240
/. volume of the layer between the)

"
W x 2240 W x 2240
two lines (Fig. 158) J 63 64
W X 2240
- ---
63 x -FT
64
If T be the tons per inch in salt water and t the difference of
draught in inches, the weight of the layer
and the volume of, the layer
We have
., ,
--
= T x t
thus found the volume of the layer in two ways, and

^j
X 2240
=T x
,
cubic f
ft.
.
t
.
x 2240 Ibs.,
we can then equate, viz.
W x 2240 T x t X 2240
Thus for a battle-ship 57*2 tons per inch, 15,000 tons displace-
= 15,000 =4
ment, the difference of draught ^ 7
in.
If for W and T we put approximate values, we have

= (0 62 x L X B x D) J*
t
(a) Battle-ships ^ 7~T^ T^ ^\
= jrjj
X draught.
L x B x D) J*
(0 52 x
(I) C^sers . <^ x L x ^f ,
63x(
= =r x draught.
We can therefore say, roughly speaking, that the difference of

draught in inches is one-seventh the draught in feet.
In the general case of a ship passing from water of density d'
f
to water of density d (d being greater than d), the difference of
draught
,, .
is
W .
d'-d
.
-^ ^
Reserve of Buoyancy and Freeboard. We have seen that
buoyancy is the upward support given by the water to the ship,
and this upward force exactly equals the weight of the ship.
Freeboard is the height of the upper deck at side (to top of
deck plank if fitted) from the water surface. Reserve of "buoyancy
204 WAR-SHIPS. ,
is the volume of the ship above the waterplane which can be

made watertight. In many ships this will be to the upper deck,
but in some ships there are erections, as poop and forecastle, which
can be made watertight. The sum of the buoyancy and the
reserve of buoyancy is the total floating power of the vessel.
Reserve of buoyancy is expressed as a percentage of the buoyancy,
and this varies considerably in different types of ship, e.g. in the
Devastation, a low freeboard ship, this percentage was about 50

per cent. For modern battle-ships with good freeboard it amounts
to about 90 per cent., and for cruisers and destroyers higher
values than this are usual.
In merchant vessels sufficient reserve of buoyancy is obtained
by specifying the minimum freeboard, this being obtained from
tables drawn up by the Board of Trade. All British war-ships
have freeboard and reserve of buoyancy considerably in excess of
what the Board of Trade would require for merchant vessels of
corresponding dimensions. Ships like the torpedo gunboats
appear to disadvantage in comparison with merchant vessels of
similar size, because of the absence of bulwarks which extend
several feet above the upper deck of the latter ships. The
presence of the bulwarks, although affording protection from the
sea, does not, of course, increase the reserve of buoyancy. Indeed,
bulwarks are likely to become a source of danger, if provision is
not made
for a sufficient number of large clearing ports for the
purpose of speedily clearing the deck of water.

The reserve of buoyancy possessed by a ship is important,
because this has to be drawn upon if a ship is damaged, and a ship
with small reserve could only stand a small amount of damage
before being entirely submerged. Freeboard and reserve of
buoyancy are also important, however, because of the necessity
of providing sufficient stability at large angles of inclination.
This will be dealt with in a later chapter.
Sinkage caused by a Central Compartment being open to
the Sea. The principles involved will be well illustrated if we
take a box-shaped vessel. Such a vessel 100 ft. long, 20 ft. broad,
20 ft.
deep floats at a draught of 10 ft. What will be the
draught if a central compartment, 20 ft.
long, is laid open to the
sea (Fig. 159).
The weight of the vessel is the same before and after the
bilging, but the buoyancy has been diminished, and the vessel is in
a similar condition to one having the watertight side and bottom
between the bulkheads replaced by a lattice- work. In conse-
quence of the loss of buoyancy the vessel must draw on the reserve
of buoyancy by sinking down to the waterline WL' to the draught
(I feet, say
Original volume of displacement

= 100 X 20 X 10,
New = 80 x 20 X d\
and seeing that the weight of ship is the same, these two volumes
must be the same, so that d = 12'5 ft.
It is important to note that if a compartment is filled with
coals, stores, etc., the space thus occupied cannot be taken up by
the water, and thus the lost buoyancy is much less. For instance,
in the above vessel, if the central compartment were filled with
coal from the ends, the vessel floating at the same draught of 10 ft.
before bilging, the sinking after bilging would only be 9^ in.

If the watertightness of either of the bulkheads in the former
example ceased below a height of 12 ft. 6 in., the water would
w'
CHAPTER XVI.
INITIAL STABILITY, METACENTRIC HEIGHT, ETC.
The weight of a body is the sum or resultant

Centre of Gravity.
of all the particles composing it, and this resultant
of the weights
acts through a definite point
however the body is placed.
This point is termed the centre,
of gravity (C.G.), and the body
being at rest, we can regard
the whole weight as being
concentrated at the centre of
gravity.
if two weights each of 5 Ibs. are

MljN placed, as in Fig. 160, 24 in. apart,
the C.Gr. of the system must be
midway between them at G, and if
FIG. 160. the weights were held by a string, it
would be immaterial, so far as the
string was concerned, whether the weights are as shown, or the whole 10 Ibs.
concentrated at G. Again, if the weights are 10 and 5 Ibs. respectively, 24 in.
apart, we should need to support at the point G, 8 in. from the larger weight,
i.e. at the centre of gravity.
The centre of gravity of a ship is the point at which we may

regard the whole of the weight to be concentrated.
Centre of Buoyancy. The resultant of the upward buoyancy
must have its line of action through the centre of gravity of the
displacement. When the water filled the space, before the ship
was there, the weight of the water acted through this point, and
so the support of the surrounding water, or, as we term it, the
"buoyancy,must also act through the same point. This point is

termed the centre of buoyancy, being the C.G-. of the displacement.
The position of the centre of buoyancy (O.B,), relative to the
206
INITIAL STABILITY, METACENTRIC HEIGHT, ETC. 207
1
waterplane, can be accurately calculated by simple rules (which
it is not proposed to discuss in this work), and this position has
an important influence on the transverse stability of the ship.

The following formula gives the approximate distance of tho
C.B. below the L.W.L., viz.
V5 j_ "*"
^ "\ where D= mean draught (excluding keel pro*
3\ 2 12 T/ jection, if any)
W= displacement in tons
T = tons per inch
This formula is found to give results very close to the figures

obtained from detailed calculation, as the following examples
show :
1. Vessel 2135 tons, 18'7 tons per inch, 13 ft. 6 in. mean draught.
C.B. below L.W.L. approx. = + = 5'42 ft.

|(^p ig^^g)
The
actual calculation gave 5*37 ft.
2. Vessel 15,000 tons, 57'2 tons per inch, 26 ft. 9 in. mean draught.
= +
C.B. below L.W.L. approx.
j(?^ 12 x^) = 11<76 ft '
The actual calculation gave 11*77 ft.
Conditions of Equilibrium of a Body floating freely and at

Rest in Still Water, We have seen that for any body floating
freely and at rest, the upward support of the buoyancy exactly
equals the weight of the body. The weight acts through the centre
of gravity, and the buoyancy acts through the centre of
buoyancy.
If the vessel is in equilibrium, i.e. has no tendency of herself to
move, these two forces which are acting on the ship must act in
the same vertical line. That is to say, the C.G. and the C.B.
must be in the same vertical line.
This is the condition of things in a tug of war. If no movement takes
place, it is evident that both sides are equal, and they are both pulling in the
same line.
We have thus two conditions which are satisfied in the case of

a body floating in equilibrium, viz.
1. The weight of the "body exactly equals the weight
of the water
displaced, and
2. The C.G. and the C.B. must be in the same vertical line.
1
See the author's "Theoretical Naval Architecture.
1 '
208 WAR-SHIPS.
Stable, Unstable, and Neutral Equilibrium, The

equilibrium, however, may be either (a) stable, (b) unstable, or
(c) neutral. These kinds of equilibrium are defined as follows,
viz.
(a) Stable Equilibrium. If the vessel be slightly inclined from

her position of rest she will tend to return.
(b) Unstable Equilibrium.
If the vessel be slightly inclined
from her position of rest, she tends to incline still further.
(c) Neutral Equilibrium. If the vessel be slightly inclined from
her position of rest, she neither tends to return to or to incline still
farther from that position.
These three kinds of equilibrium are seen in Fig. 161, in which cylinders are
shown resting on a smooth table.
(a) In the first case the C.G. is below the centre, and if the cylinder is
FIG. 161.
slightly inclined it is seen that the two forces acting, viz. (1) the weight
through the C.G., and (2) the support of the table through the point of contact,
form a couple which tend to take the cylinder back to the upright. This is a
state of equilibrium which is stable.
(b} In the second case the C.G-. is above the centre, and on slightly inclining
it isseen that the couple acting is an upsetting couple, and the cylinder will
incline still farther. This is a state of equilibrium which is unstable.
(c) In the last case the C.G. coincides with the centre of the figure, and there
will be no tendency to return to or incline further from the original position on
giving a small inclination. This is a state of equilibrium which is neutral. This
is the necessary condition for a billiard ball. The ivory must be perfectly homo-
geneous, so that the C.G. is at the centre of the ball, and the ball must be perfectly
gpherical, so that the support will always act through the centre.
INITIAL STABILITY, METACENTRIC
HEIGHT, ETC. 209
In the case of a ship let Fig. 162 represent the vessel inclined
to a small angle 0. WL was the position of the waterline on the
ship when upright; B, the
position of the C.B. when
upright; and G, the position
of the C.G. The ship as in-
clined has a new waterline,
WL', and the C.B. of the new
displacement will be at B', so
that the buoyancy will now
act through B'. Let the verti-
cal through B' cut the
original
vertical through B in M.
Then in the first case, in Fig. 161, we notice that the couple acting
on the ship tends to bring her back to the upright. In the second
case the couple tends to incline her still further from the upright.
We see that whether the couple is a righting couple or an
upsetting couple depends on the relative positions of the points
G and M.
If G is below M, the couple is a righting couple, and the
ship is
in stable equilibrium. If G is above M, the couple is an
upsetting
couple, and the ship is in unstable equilibrium. If G coincides
with M, the ship is in neutral equilibrium. The point is thus M
seen to be an important point, as its position relative to G deter-
mines the state of the equilibrium in the upright condition. It is
given the name of the transverse metacentre when dealing with
transverse inclinations, and the distance between G and M is
termed the metacentric height.
We thus see that there are three conditions which must be
fulfilled in order that a floating body shall float freely and at rest
in stable equilibrium, viz.
1. The weight of the body must exactly equal the weight of the
water displaced ;
The C.Gr. and the C.B. must be in the same vertical line ; and
2.
The C.Cr. must be below the transverse metacentre.

3.
For angles up to 10 or 15 the intersection of the vertical

through B' with the original vertical through B is practi-
cally at the same point, viz. M. For larger angles this will not
be the case.
Before dealing generally with the question of initial stability,
i.e.
stability at small angles, we shall deal separately with the two
p
210 WAR-SHIPS.
points which determine it, viz. the transverse metacentre and

the centre of gravity.
Position of the Transverse Metacentre of a Ship when
floating at any given Waterline. The point depends solely M
upon the geometrical shape of the underwater body, and its
position can be determined for the ship when floating at any
1
It is determined with reference to the
particular waterline.
C.B., and because of this the position of the C.B. has a distinct
influence on the stability. The distance is given by (Moment BM
of Inertia of waterplane about the centre line) -f- (Volume of
displacement) or BM = y.
The moment
of a figure about any axis is obtained by dividing
the area into a large number of small areas, and multiplying each
by its distance from the axis. The addition of all such products
is called the moment.
The moment of inertia of a figure about any axis goes one step
further. The area is divided into a large number of small areas,
and each of these is multiplied by the square of its distance from
the axis. The addition of all such products is a moment of the

second degree, and is called the moment of inertia?
The expression ^ may be approximately found as follows :
(1) I can be written I

= ?i.L.B 3
where L= length of ship
B= breadth of ship
n is a coefficient
varying with different
shapes of waterplane (for a rectangle
1
See the author's " Text-book of Theoretical Naval Architecture."
2
This definition and the value of the moment of inertia given for a rectangle
may be illustrated by the following approximation to the I of a rectangle 100 ft.
X 20 ft. about the centre line. Divide the breadth into ten strips, each having an
area of 100 X 2 = 200 square ft. the centre of the strips from the centre line are
;
9, 7, 5, 3, and 1 ft. respectively. The I of the half of rectangle about the centre
line is therefore nearly
[200 X 2
(9) ] + [200 X (7) J
2
+ [200 X (5) ]
2
+ [200 X 2
(3) ] + [200 X 2
(1) J
= 200 X 165 = 33,000
and for both sides I = 66,000. The exact value is T15 (100)(20) 3 = 66,666. If the
strips had been taken 1 ft. wide a closer approximation still would have been
obtained,
INITIAL STABILITY, META CENTRIC HEIGHT, ETC. 211
(2) V can be written V = /j.L.B.D

where D = mean draught (ex keel projection
if any)
k is a coefficient of fineness
I rc.L.B 3 B2
so that BM = y^.L.B.D" *'
D
where a is a new coefficient obtained from n and k. It is
found that a varies between narrow limits, being about in ^
battle-ship forms and ^ in cruiser forms. The important point
that this approximate formula emphasizes is, the great influence
of beam on the distance BM, and therefore on the position of M.
This is the reason why battle-ships have to be so broad. It is
necessary to have M above G for stable equilibrium, and in battle-

ships the point G is high owing to the great weights of guns and
Battle Ship
Battle Cruiser
FIG. 163. Shapes of load water plane.
armour carried high up, M

must therefore be high also, and this
is obtained by giving such ships a large beam.
Thus Drake, a cruiser of 14,100 tons, carrying two 9'2-in. guns in shallow
and sixteen 6-in. guns in casemates, is 71 ft. beam.
barbettes
But Duncan, a battle-ship of 14,000 tons, carrying four 12-in. guns in
massive barbettes, with twelve 6-in. guns in casemates, has to be 75 ft.
beam to obtain sufficient stability. (See Fig. 163 for comparison of shapes of
waterplane of a battle-ship and a cruiser.)
A similar cause was in operation resulting in the Lord Nelson being made
79J ft. beam. In this ship placing of the whole armament of four 12-in. and ten
to be higher than in previous
9-2-in guns on the upper deck caused the C.G.
ships, and the metacentre had to be made higher by increasing the beam.
Similarly the Dreadnought with ten 12-in. guns, eight on the upper deck
and two on the had to be made 82 ft. in breadth to get a sufficiently
forecastle,
high metacentre.
212 WAR-SHIPS.
EXAMPLES OF STABLE AND UNSTABLE EQUILIBRIUM.

Anexample of the above principles is seen in the case of a
log floating with
one half its bulk immersed say 20 ft. long
and 18 in. square. It is a matter of experience that such a
will never float with one face horizontal, as Fig. 164, but
log
always with a corner of the section downwards, as Fig. 165. The
following application of the foregoing principles will explain the
reason.
If the log is placed as in Fig. 164, we have the C.B. 4J in.
from W.L., and the C.G. at half depth. To find the position of the
metacentre we use the formula
BM = and I = rV X 240 x IS 3 and V= 240 x 18 x 9

^,
all dimensions being in inches, so that
BM = (^ X 240 x 18 3 ) 4- (240 X IS x 9) = 3 in.

but BG = 4J in.
so that M is 1\ in. below G and the log is in unstable equilibrium,

and cannot float with one face horizontal.
FIG. 164. FIG. 1G5.
however, the log be placed as in Fig. 165, with one corner

If,
downwards, we have the C.B. 4'24 in. from the C.G. and
A X 240 x (18\/2) 3
240 x 18 x 9
= 8-48 in.
so that M is 4*24 in. above G-, and the log is in stable
equilibrium.
We notice that there are two influences at work in obtaining
the stability in the second case, viz. (1) C.B. rises, owing to the
new shape of the displacement, carrying M
with it, and (2) BM
is increased owing to the greater breadth at the waterplane.
Another interesting example of stable and unstable equilibrium in a floating

body is seen in the case of a duck or a swan. Under ordinary circumstances
22t
214 WAR-SHIPS. -
It will, however, be a different point for any other waterline,

because Bwill shift, and the distance BM
will be different. It is
desirable to have some ready means of determining the position
of Mfor any waterline at which a ship happens to be floating.
This is done by constructing a metacentrio diagram. Four or more
parallel waterlines are taken, and for the ship as floating at each
FIG. 167.
of these lines the position of the G.B. is calculated. The waterlines

are set out at the proper distance apart, as Fig. 166, and a line
drawn across at 45. Through the intersections verticals are drawn
and on these verticals are set down
the positions of the respective
centres of buoyancy, BI, B 2 B3 B4
, ,
Thus for the ship shown in
.
Fig. 166 the following were results obtained:

INITIAL STABILITY, METACENTRIC HEIGHT, ETC. 21$
At 30 ft. 6 in. waterline, C.B. is 10'35 ft. below 27 ft. 6 in. waterline
27 ft. 6 in. C.B. is 12-00 ft.
,, 24 ft. 6 in. C.B. is 13-65 ft.
,, 21 ft. 6 in. C.B. is 15-3 ft.
A curve drawn through these points gives the locus of centres

of buoyancy.
In a similar manner the distance BM is determined for each of
the four conditions of draught. Thus at 30 ft. 6 in. waterline, BM
= 14-9 ft. at 27 ft. 6 in. waterline,
;
BM
= 17'3 ft. at 24 ft. 6 in. ;
waterline, BM
= 20*5 ft. at 21 ft. 6 in. waterline,
;
= 24'5 ft. BM
These distances, set up from BI, B 2 B 3 B respectively, give the, ,
spots MI, 2 3 M M M
4 and the curve through these is the locus of
, , ,
FIG. 168.
transverse metacentres. We are thus enabled to determine the
position of the transverse

metacentre for any waterline at which
the ship happens to float. Thus at 25 ft. 3 in. draught, is 6 '4 ft. M
above the 27 ft. 6 in. waterline.
Fig. 167 shows the curves of C.B. and metacentres of a number

of ships all placed together. The sections of these ships are shown
in Fig. 168. The battle-ship is very broad, and the metacentre is
consequently high. The Devastation has a sudden drop as she
lightens, owing to the overhang
of the armour. With the first and
second class cruisers the finer body causes a lift in the locus of the
C.B. The Maine is a typical merchant vessel (she has been
216 WAR-SHIPS.
taken into the service, and converted into a hospital ship). She
is much narrower than the other vessels, and the effect of this is
seen in the low locus of metacentres. (Such ships, carrying great

weights of cargo, have their C.G. low as compared with war-ships
which have to carry heavy weights of guns, armour, etc., high
up). The Waterwitcli (formerly an auxiliary steam yacht, now
a surveying vessel), presents some points of interest. The peg-
top section causes a high C.B. As the ship lightens the M
curve
dips downwards, owing to the small breadth at the waterline.
Position of the Centre of Gravity. We have been dealing
above with the position of the transverse metacentre, but in order
to know anything about the initial stability of a ship we must also
know the vertical position of the C.G. It is possible to determine
this position by means
of direct calculation, and this very laborious
calculation has to be done in the case of a new design. Finding
the position of the C.G. in the various conditions of the ship, deep,
normal, and light, it is possible to arrive at the estimated meta-
centric height in these conditions.
Inclining Experiment. When a ship is completing or finished
"
possible, by means of the
it is inclining experiment," to determine
the metacentric height, and thus to find the position of the C.G.
This experiment isalways carried out on new ships of the Eoyal
Navy, and also after a vessel has undergone extensive alterations
likely to affect the stability. The information obtained from this
experiment forms the basis of the stability calculations for the
completed ship, and the main features of the vessel's stability are
furnished to the Ship's Book in the Stability Statement. The
information thus obtained also enables us to see how far the
estimate of the design has been realized, and gives most valuable
data for use in subsequent designs.
The mainfeatures of an inclining experiment are as follows.
Weights are placed on the upper deck, as Fig. 169, and w tons,
say, is moved across the deck, a distance of d feet. Then the C.G.
of the ship will shift to G', such that
(1)
The ship cannot remain upright as shown, because the second

condition of equilibrium is not fulfilled, viz. that the C.G. and
the C.B. must be in the same vertical. The ship must heel over
until the new C.B., B', comes into the same vertical as the new
C.G., G'. Then the point where B'G-' meets the middle line is M
the transverse metacentre. If 8 is the angle of heel, then
GG '
(2)
Substituting for GG' from (1), we have
= w x d
tan
W x GM
w X d
or GM = W X tan 9 (3)
The only part in this that we do not know is tan 0, and this
-*
m
.
i i
P" i
2i8 WAR-SHIPS. ,
A
vessel displacing 5372 tons is inclined by shifting 25 tons of ballast across
the deck, the mean deflection observed being 10 \ in. in 15 ft.
_,,.
w * i*
~ z;y x. ou o.n
A 9
V^M =ATT . . L _
'
Vorr> x ^ ^ r. f v
The transverse metacentre was measured off the metacentric diagram 31 ft.
above the L.W.L. at the draught given, so that the C.G. of the ship in the
inclined condition was 31 2-9 = 0'2 ft. above the L.W.L.
The ship was incomplete at the time of the experiment, so the weights to
complete and remove were determined, amounting to a net total of 600 tons,
with the C.G. 3'5 ft. above the L.W.L., or 3-3 ft. above the C.G. in the in-
clined condition. The rise of the C.G. due to this addition was
600 * 3 '3
=0-3 ft.
(5372 + 600)
so that the C.G. in the completed condition, 5972 tons, would be 0-5 ft. above
the L.W.L. At (he draught corresponding to this displacement the transverse
metacentre was measured to be 2 '9 ft. above L.W.L, so that the GM in the
completed condition was 2'9

- 0-5 = 2-4 ft.
In this way, using the results of the inclining experiment as

a basis, we are enabled to determine the draught and metacentric
height in any desired condition. The usual conditions are deep

load,normal load, and light, which have been defined in Chapter
XIV. These are shown on the metacentric diagram, as Fig. 166.
In that case we have
Deep load draught 28

;
ft. 8 in., GM = 37 ft.
Normal load draught 27

;
ft. in., GM = 3'4 ft.
Light; draught 25 ft. 3 in., GM = 3'3 ft.
These particulars are given on the stability statement, a

specimen one being given at the end of Chapter XVIII.
In cruisers, in which the coal capacity is relatively very large
and a large proportion of it is above the protective deck, a special
condition looked into is that, supposing the upper bunkers are full,
while the lower bunkers are empty. This would be an extreme
condition,and the ship, unless damaged, would hardly get into a
worse condition of stability than under such circumstances. If
the condition thus found gives too small a GM, a special notation
would be made on the stability statement giving definite instruc-
tions as to the coal stowage. In most cases, however, sufficient
stability is retained even under these extreme conditions, and a

note is made on the statement that, so far as stability is concerned,
the coal may be worked in any manner desired by the commanding
officer.
Tigs. 170 to 174 show metacentric diagrams of some typical
.!
1
z
13
DEEP
220 WAR-SHIPS. -
consumed are in the aggregate below the C.G-. Thus the GM. in
the light condition is found to be less than in the other conditions.
Fig. 170 is the diagram for a first class cruiser, Fig. 171 for a
second class cruiser, Fig. 172 for a destroyer. The diagram for
the Maine, originally a merchant steamer, is given in Fig. 173.
In this ship it has been necessary to provide 1000 tons of perma-
nent ballast to give proper immersion and stability, and this is
so stowed so as to provide stability, when the double-bottom tanks
are empty and also the lower bunkers. In this extreme condition
a GM
of 07 ft. only is obtained but when fully equipped with
;
double-bottom tanks full, a GM

of 2-2 ft. is obtained. Fig. 174
gives the diagram for the Waterwitcli mentioned above. In this
ship, carrying considerable sail power, a good GM is required, and
3*0 ft. is obtained in the deep load condition. To get this it is
necessary to stow 65 tons of permanent ballast.

Values of GM, the Metacentric Height. The amount of
GM given to a vessel is determined by the class of vessel and the
qualities it is desired to obtain. To take two instances. In sailing-
ships a sufficient GM
must be provided to enable the ship to
"
"
stand up under her canvas. In such a ship a small GM would
mean a large angle of heel when sailing, which is undesirable. In
a vessel like the Inflexible, in which the armour-belt only extended
over one-third the length (Fig. 126), a large was provided, GM
viz.8 ft., to enable the vessel to remain upright, even supposing
the unprotected ends open to the sea.
For small angles the moment of the couple tending to right
a ship, i.e. the stability, is W x GM x sin 0, so that if
is large GM
this righting moment is large, and if GM
small, this righting is
moment is small. If GM
is large, the ship comes back to the
upright very suddenly after being inclined, and the ship will have
a quick motion. Such a ship is stiff. If is small, the ship GM
is easily inclined, but returns to the upright slowly with an easy
motion. 1
There are thus two opposing conditions to fulfil in settling the
GM for a war-ship, viz.
1. GM must be large enough to enable the ship to retain
stability after a fair amount of damage.
2. GM must not be so large as to make the ship have violent
1
This is further considered in Chapter XIX. A "
crank " ship easily inclined
is found to "be the steadiest in a seaway.
motions at sea, this being specially important in view of the

fighting of guns.
The following are average values of the metacentric height
given to the modern ships of the British Navy
1
Battle-ships 3 to 5 ft.
Large cruisers ... ... 2 J to 3 J ft.
Small cruisers ... ... 2 to 2 J ft.
Destroyers ... ... about 2 ft.
For battle-ships it is necessary that sufficient GM shall remain

after the unprotected ends are riddled and open to the sea. The
gradual enlargement of the waterplane area protected during the
last thirty years has been the cause of a gradual reduction of meta-
centric heights in these ships. Thus in the Inflexible, with a belt
only one-third the length, a GM
of 8 ft. was provided. In the
Admiral class, the belt being four-ninths the length, the was GM
5 to 6 ft. For modern ships with belt about two-thirds the length,
the GM is 3 J to 4 ft. In a recent ship the effect of riddling the
ends is to reduce the GMby about 2 ft. In the Inflexible this
riddling reduced the GM by 6 ft.
Vessels like destroyers are given a which GM is

relatively
large. These vessels, as the speed increases, form a wave which,
dipping down amidships, causes a considerable reduction of area
and moment of inertia of In consequence of this it
waterplane.
is necessary to make the GM
relatively large. Another reason
for this is seen in the inward heel caused by putting the rudder
over. If the GMwere small, this heeling might be excessive.
For merchant steam-ships, the GM
varies continually owing to
the different nature and disposition of the cargo carried on different
voyages. Ships with cargo of light density frequently go long
voyages with metacentric heights of less than 1 ft., and their
behaviour is reported to be in every way comfortable and safe.
Such ships, however, have to be carefully treated when light, and
frequently require water-ballast in the double bottoms to enable
them to remain upright. Such a metacentric height as is sufficient
in this type of ship would not be permissible in war-ships for the
reasons already stated.
1
In the Triumph and Sioiftsure, recently added to the Royal Navy, the meta-
centric heights are-
Deep load 3-75 ft.
Normal load 3-5 ft.
Light 2-7 ft.

222 WAR-SHIPS.
Sailing ships are given metacentric heights of 3 to 3J ft. to

enable them to "stand up" under their canvas without heeling to
an undesirable angle.
Influence of Coal stowed in Upper Bunkers. 1. The
question of coal in the upper bunkers at the side of war- ships
is important because of the resistance such coal offers to direct
penetration. It has been found that 2 ft. of coal is equivalent

in resisting power to 1 in. of iron. This is specially important
in deck-protected cruisers (Figs. 21, 22, 24, 26), which depend so
largely on the coal above the protective deck for their protection,
and on this account the coal in the upper bunkers at the side
should be the last to be used. All ships of this type in the
Royal Navy have sufficient stability even supposing all the
coal in the lower bunkers burnt out and the upper bunkers
completely full. In the sloops, which are quite unprotected save
by the coal, a division is placed in the bunkers, so that some coal

may remain above the flat as long as possible, in order to retain
itsprotection (Figs. 29, 30).
2. In addition to this direct protection there is the fact that
the bodily sinkage on riddling the side and admitting water

would be less with coal than without it. Every cubic foot of
bunker space, with coal in, contains five-eighths solid space
occupied by the coal, and three-eighths vacant space, into which
water could penetrate. The influence of this in limiting bodily
sinkage has already been seen in Chapter XV.
3. Of greater importance, however, than (1) or (2) is the
influence of the coal in preserving the initial stability, supposing

the side of the ship
j is riddled in way of
the upper bunkers.
(a) With the side
of the ship intact,
FIG. 175. the shaded parts in
Fig. 175 will con-
tribute their full value to the moment of inertia of the water-
plane.
(b) With the side riddled and no coal, the shaded parts will
contribute nothing to the moment of inertia of the waterplane.
(c) With the side riddled and coal in, the shaded parts will
contribute five-eighths their area to the moment of inertia of the
waterplane.
We have seen that the position of the transverse metacentre,

on which the initial stability so largely depends, is directly influ-
enced by the transverse moment of inertia of the waterplane.
Fig. 176 gives the
result of calculations
made on a cruiser 480
ft. X 71 ft. X 26 ft. X
13,000 tons, in which

the sides in way of
bunkers on loth sides of
the ship were assumed
to be gradually riddled. FIG. 17G.
The vessel starts with a

GM of 3'4 ft. If coal is in the upper bunkers the riddling of both
sides for the length of 230 ft. leaves the ship with a GM
of 9 in.,
so that the ship although tender would be stable. If the bunkers
are empty the reduction of GM as the sides are riddled is much
more rapid, and the vessel would become unstable when about
half the length of bunkers was riddled.
Fig. 177 gives the result of calculations made on a cruiser
370 ft. X 57 ft. X 20 ft. X 6160 tons, having one side only
gradually riddled in way of
the upper bunkers, and the
consequent heel corrected
by the admission of water U,"-ij
to the wing compartments
or double bottom respec-
tively. The curve AA
shows that, with coal in,
the riddling of the side for
the whole bunker length
leaves the ship with a GM
of 2|- ft., supposing that water is
admitted to the wings to keep her upright. BB is the curve,
supposing the coal out of the upper bunkers and the heel
corrected by admitting water to the wings. The ship still retains
a GM of 1 ft., supposing the length of 165 ft. is riddled. A
rather smaller GM
is found to result if the heel is corrected by
admitting water to the double bottoms only, curve CO. In these

cases the lowering of M
due to the loss of moment of inertia of
the waterplane is partially compensated for by the lowering of G
due to the admission of water to the wings or double bottoms.
224 WAR-SHIPS.
The following example is introduced to show that a given

quantity of coal on board a vessel is more usefully disposed, as
regards stability, when the side is riddled, when in the upper
bunkers at the side than in the lower bunkers, in spite of the
lowering of the C.G-. of the ship that takes place when the coal is
trimmed down.
EXAMPLE. A box-shaped vessel is 350 ft. x 60 ft. with lungers at the side
amidships 10 ft. wide, 160 ft. long, extending from 14ft. to 26 ft. above the keel.
When these bunkers are full the draught is 20 ft., and the metacentric height
3 ft. Determine the effect on the initial stability
(i.) With sides riddled in way of upper bunkers, coal in.
(ii.) With sides riddled in way of upper bunkers, the coal having been
trimmed lower bunkers 8ft. above keel.
to the
Taking 43 cubic ft. of coal to the ton, the side bunkers will hold about
900 Ions. The displacement of vessel is 12,000 tons.
(i.)
M
With the sides intact is 25 ft. above keel, and G consequently 22 ft. above
keel. In the bunkers five-eighths the space is occupied by the coal, and three-
eighths the space is vacant. When the sides are riddled the vessel sinks
slightly, and the moment of inertia of the waterplane is reduced by the admis-
sion of water to the space in the bunkers unoccupied by the coal, from its
original value of 6,300,000 to 5,540,000. This causes M to drop to 23-2 ft. from
keel, and the resulting metacentric height is 1*2 ft. The vessel is therefore
stable since M is above G.
(ii.)
If now we trim the coal to the lower bunkers, we depress the C.G.
= '
9 ft -> so that the GM is 3<9 ft -* and G is 21>1 ft - aoove keel -
On we
get a greater sinkage, but the chief effect is the
riddling the sides
reduction of the moment
of inertia of the waterplane to 4,274,000, the area ol
the waterplane in the bunkers contributing nothing to the moment of inertia.
This gives the point M
20-36 ft. above keel, or 0'74 ft. below the C.G. The
vessel is thus unstable in the upright condition.
Methods of increasing the Metacentric Height of a

Vessel. 1. If we put ballast, either pig-iron or water, into the
lower part of the ship, there will be two effects, viz.
(a) Increase of draught, and
Depression of the C.G. of the ship.
(5)
The increase of draught, if moderate, will not in general alter
the position of the transverse metacentre very much; whether
it does so or not will depend on the shape of the metacentric
diagram. If w be the added weight, the weight of the ship W
before the addition, and d the distance of the added weight below
the C.G., then ,
.
X
depression of C.G. = , ,
(W -f w)
Thus the addition of the weight of 20 tons, 10 ft. below the
C.G. of a vessel of 1000 tons displacement, will lower the C.G. of

20 x 10
the ship
102Q
= 0-2 ft. In general the GM
will increase by
this amount, but if the M locus in the metacentric diagram slopes

down sharply as the draught increases the increase of GM may be
rather less than thus obtained.
In adding water-ballast to the double bottom of a vessel it is
essential that each compartment should be completely filled so
that the water will act as a solid weight. If a free surface is left
the water can shift over to the side to which the ship is heeling,
and this tends to counteract the increased obtained by theGM
water-ballast. The manholes to double bottoms are always made
with raised coamings, in order to ensure the compartments being
completely filled (see Fig. 48). Air escapes are also provided.
There may be cases where it is undesirable to increase the
draught of a ship by adding ballast, and yet it is necessary to
obtain greater initial stability. In such cases the following method,
or No. 3 below, would have to be adopted.
2. If top weight is taken out of a ship there will be two
effects, viz. (1) decrease of draught, and (2) depression of
the C.G. Thus a ship is of 5000 tons displacement. The effect of
removing two military tops weighing 24 tons, originally 70 ft.
above the C.G., would be to cause depression of the C.G.
xgflQfl 7^r\
= ^'34 ft., and this will be generally the increase
of metacentric height, unless there is something exceptional about

the metacentric diagram.
3. The two previous methods were concerned with lowering
the C.G., the present method deals with the metacentre. It will
be remembered that
the position of the
metacentre is di-
rectly dependent on
the moment of in*
ertia of the water-
plane. If we can
increase this we
raise the metacentre, FIG. 178.
and so increase the
initial stability. This can be done by placing a girdling at the
waterline, over the midship portion of the length, as Fig. 178. This
226 WAR-SHIPS.
adds very draught, but considerably to the moment of

little to tlie
inertia of the waterplane. This method of increasing the stiffness

used to be frequently adopted in the wooden sailing-ships in order
" "
to enable them to stand up better.
An instance of adoption in the Eoyal Navy was in the case
its
of the Sultan. This ship had to undergo an extensive reconstruction,

and it was found that the alterations would leave her with in-
The best way to increase the stability was
sufficient stability.
found by adding a wooden girdling over the midship portion
to be
of the length, as the addition of any weight on board was
undesirable.
Stability when partially Waterborne. An application of the
principles of the present chapter, of interest and some importance,
is seen in the reduction of stability which takes place when a
vessel is partially waterborne. This happens when a vessel is

being docked or undocked, and also if a vessel is run on to a
shelving beach. In Fig. 179, suppose a ship is being docked, and

the water level falls from WL' to W"L". If we suppose a small
inclination 0, the support of the displacement of the zone between
W'L' and W"L", viz. w, which originally acted through 5, the C.G.
of the zone, now acts at the keel, and the buoyancy W w acts in
FIG. 179.
the line BiMi, where MI is the metacentre corresponding to the

waterline W"L".
The original moment righting the ship was W x GM x sin 9,
but the moment now righting the ship is
{(W - w) QMi - w.OG} sin 6

= {(W x GMi) - (w x OMi)} sin 8
since the influence of w is to upset the ship.

It may be shown that the reduction of metacentric height thus
caused is (
^ X Ob j
In the case of a ship being docked, the critical point is reached
when the keel is just taking the blocks all fore and aft, and the
time until this happens is longer in the case of a ship trimming a
great deal by the stern than in a ship on a more even keel. In
such a ship, therefore, the support w may reach a considerable
amount before the ship takes the blocks, after which the shores
can be set up. Just before the shores are set up, there is, there-
fore, a reduction of stability which may be sufficient to render a
is necessary, therefore, when
ship unstable. It docking and
undocking keep the
to ship well under control to prevent any
transverse inclinations while any of the weight is taken by the
blocks.
For ordinary ships the loss of metacentric height thus caused
will not be sufficient to reduce the GM
enough to cause instability,
but it is possible in a ship having large trim and small metacentric
height when being docked.
It is important to note in connection with the
docking of
ships that a ship with small GM
should never be undocked, if,
while in dry dock, any alteration of the weights on board is made
which tends to reduce the metacentric height, unless other weights
are added to compensate. For example, a merchant ship when
light may require water-ballast to keep her upright. If docked in
this condition the ballast must not be removed while in dock
(unless compensation is made), or else it would be found that

when the ship was again afloat she would be unstable.
Influence of Free Water. If water is admitted to a com-
partment so as to completely fill the space, its effect is exactly the
same as that of a solid weight, and the resulting influence on the
stability can be readily calculated. If, however, the water does
not completely fill the compartment, but has what we term a
"
free surface," the influence on the stability is not so easily obtained.
Take, for instance, the conditions shown in Fig. 179 A, with water in
a boiler-room at the level wl when the ship is upright. The addition
of this water, consideredsimply as an added weight, is to pull down
the C.G. of the ship from G to G, and = (w X k) 4- (W
GG w). +
If the ship inclines, the water surface takes up the position
w'l' t and centre of gravity shifts from I to &'.
its If I'm be drawn
vertically, the water indeed acts as if its weight were concentrated

228 WAR-SHIPS.
at the bob 5 of a at the point m.

pendulum bm, suspended The
point m has received the name
" "
virtual centre of gravity because,
although the actual centre of gravity is at I, the effect of the water
on the ship is the same as if it were at the point m. The centre
of gravity of the ship and added water which actually is at G, is
virtually at the point G', owing to the mobility of the water. If

W is the weight of the ship after the addition of tons of water, w
then W X GG' = w x lm>. Now Im is given by i 4- 35w, where
i is the moment of inertia of the free-water surface about its
centre line. The rise of the C.G., viz. GG', is therefore given
by i-f-35.W. This isseen to be directly dependent upon the
moment of inertia of the free- water surface. Take, for instance,
a boiler-room compartment 36 ft. long and 45 ft. broad. The
moment of inertia of the free-water surface if water is admitted is
3
ô X 36 x 45 If the displacement of the ship is 14,000 tons before
.
FIG. 179A.
adding the water, and 12 inches of water be admitted, this weight

46 tons will pull down the C.G. of the ship about (46 X 24) -f-
14,046 = 0*08 ft., which will be the increase of metacentric
height supposing the water solid. Owing, however, to the
3
mobility of the water the C.G. is virtually raised (^ X 36 x 45 )
-f-(14,046 X 35) = 0-55 ft. The net effect, therefore, of flooding
the boiler-room to a depth of 12 in. is to reduce the metacentric
height by 0'47 ft. or, say, 6 in. nearly. By a similar process
of calculation, it may be shown that for depths of water in
the boiler-room below about 9 ft. there is a loss of stability,
but for depths above 9 ft. there is a gain in stability, because
then the effect of the weight is greater than the effect of the
free surface. A middle line bulkhead dividing the water surface
reduces the moment of inertia to one-fourth. With two engine-
rooms each 23 feet wide, if both be flooded, the metacentric
height is reduced until 2 ft. of water is admitted, after which

there is a gain.
Similar principles apply to the case of a double-bottom com-
partment. With such a space 16 ft. long and 16 ft. broad there
is a small loss of metacentric height until the depth of water is
about 1 foot, after which there is a gain.
CHAPTER XVII.
TRIM, MOMENT TO CHANGE TRIM ONE INCH, ETC.
WE have now to deal with inclinations in a fore-and-aft or

longitudinal direction. As the stability of a ship isa minimum
for transverse inclinations, so the stability is a maximum for
longitudinal inclinations. We do not need, therefore, to study the
FIG. 180.
longitudinal stability of a ship to ascertain whether she is safe or

not, as we do the transverse stability, but in order to deal with
questions of trim or forward and after draughts.
If a ship, Fig. 180, is floating originally at a waterline WL,
230
TRIM, MOMENT TO CHANGE TRIM ONE INCH, ETC, 2 3*
and by some means is made to float at the water-line WL', the

centre of buoyancy must shift, owing to the changed shape of the
displacement, from B to B' say. Then the original vertical through

B and the new vertical through B' will intersect in the point M,
which is the longitudinal metacentre. This point is precisely
analogous to the transverse metacentre, the difference consisting
in the direction of the inclination. The distance between the
C.G. and the longitudinal metacentre is the longitudinal metacentric
height.
The point M is determined with reference to the C.B., and the
distance BM is given by the equation
where IQ is the moment of inertia of the waterplane about a

transverse axis through its centre of gravity (this C.G-. is termed
the centre of flotation), and V
is the volume of displacement.
The calculation for IQ is somewhat complicated,* but it may

be approximately written
I= ?t'.L
3
.B (n!being a coefficient)
also V= &.L. B.D (k being the coefficient of fineness)
so that BM = . =r =& .
approximately
y-
K D D
where L is the length of ship between perpendiculars in feet
D is the mean draught in feet
& is a coefficient which does not vary much from 0'075.f
This approximate formula shows the great influence of the length

in determining the position of the longitudinal metacentre.
Longitudinal shift of Weights already on Board. The

trim of a ship is the difference between the forward and after
draughts. Thus H.M.S. Pelorus is designed, under normal load

conditions, to float at a draught of 12 ft. forward and 15 ft. aft,
giving a trim of 3 ft. by the stern.
Change of trim is the sum of the changes of draught forward
* See the author's " Theoretical Naval Architecture."
t In a vessel with full waterplane n' will be large, but k will also be large, the
ship being of full form. If n' is small, /; also will be small. So that the quotient
^-
docs not vary much for ordinary ships.
232 WAR-SHIPS.
and aft. Thus if Pelorus, when floating at 12 ft. forward, 15 ft.
aft, has certain weights shifted on board resulting in a draught

of 12 ft. 10 in. forward and 14 ft. 2 in. aft, she is said to have
changed trim 10 in. + 10 in. = 20 in. Change of trim can be
produced by the fore-and-aft shift of weights already on board,
this being analogous to the inclining experiment, in which heeling
is caused by shifting weights in a transverse direction.
In Fig. 180, let w be a weight on the deck at A when the
vessel is floating at the waterline WL. Now suppose w is moved
forward through a distance d. G, the centre of gravity of the
ship, will in consequence move parallel to the shift of w to G',
such that GG' = Under these circumstances the vessel

W
^ Tr .
cannot float at the waterline WL ;

as in the second sketch in Fig.
180, because the C.G. and the C.B. are not in thesame vertical line.
The ship must adjust herself to the line W'L', as shown, so that
G', the new centre of gravity, and B', the new centre of buoyancy,
are in the same vertical. Then the line through B'G' intersects
the original vertical through BG, in M, the longitudinal metacentre.
If 9 is the small angle of inclination
GG'
=
GM
but also
PL' WW + LU _ change of trim
"length length length
Having thus two values of tan 6 we can equate, so that
change of trim __
=
GG' ==
wX d
length ~GM W X GM
using the value found above for GG r
.
We therefore have
change of trim infect = - -

x
and the change of trim in inches = X x 12
Transposing we have the moment causing the change of trim, viz.
w x d = ^X G1V
x
-rrs
12 x Y~
L change of trim in inches
TRIM, MOMENT TO CHANGE TRIM ONE INCH, ETC. 233
and if the change of trim be 1 in.
Moment to change trim 1 =W-^

x GM ,
in. foot tons
LA X ^-
Ju
where W = displacement in tons

GM = longitudinal metacentric height
L = length between draught marks
In ships of small ratio of length to breadth, like battle-ships
up to Royal Sovereign, the longitudinal was approximately GM
equal to the length. In these ships, therefore, the moment to
change trim 1 in. is very nearly one-twelfth the displacement in tons.
This does not hold so well for more modern ships, and the following
give good approximations to this moment to change trim 1 in. :
For vessels of full form like battle-ships Q nnn x L 2 x B.

y,uuu
For vessels of finer form like cruisers

^T7\r\h
X L2 x B.
where L = length between perpendiculars in feet

B = breadth of ship in feet
The length and breadth do not vary much for considerable changes of
draught ;
the above formulae, therefore, show that the moment to change the trim
1 in. varies very little for the range of draughts at which a vessel is likely
to float.
The following examples will illustrate the use of the above in
determining change of trim :
EXAMPLE. A vessel is floatiny ata draught of 12 ft. 1 in. forward and

14 ft. 10 in. aft. Determine the draughts forward and aft after shifting 5 tons
from forward to aft through 210 ft. The moment to change trim 1 in. is 295
foot tons.
In this case w= 5 tons, d = 210 ft., so that w x d = 1050 foot tons. The
change of trim is therefore -no* ^ i n- This change of trim will cause an
increase of draught aft of If in., and a decrease of draught forward of If in,
So that
draught forward = 12 ft. 1 in. - If in. = 11 ft. 11 in.
draught aft = 14 ft. 10 in. + If in. = 14 ft. llf in.
EXAMPLE. Determine approximately the shift of 50 tons on board the above

ship necessary to bring her to an even keel.
The mean draught is
12 ft. 1 in. + 14 ft. 10 in.

= 1Q
13 ,.
ft.
.,
5
.
in.
234 WAR-SHIPS.
There is therefore necessary a change of trim of
(13 ft,
5J in. - 12 ft. 1
in.) + (14 ft. 10 in. - 13 ft. 5 in.)
= 33 in.
The necessary moment from aft forward is therefore 295 x 33 = 9735 foot
tons, so that 50 tons would need to be shifted = 195 ft., say, from aft forward
^-
in order to bring the ship to an even keel.
Effect on the Trim due to adding a Weight of Moderate

Amount, If a ship sinks from a waterline to a parallel WL
waterline the added buoyancy of the layer will act at the
WL',
C.G. of the layer. If the sinkage is small this point will very
nearly be in the same section as the centre of flotation. If, there-
fore,we wish to add a weight to a ship so that the trim shall not
be changed, the added weight must be in the same section as the
added buoyancy, and if the added weight be not large, it must,
therefore, be placed in the same section as the centre of flotation.
The centre of flotation in ships of the Navy is usually abaft amid-
ships,on the average about one twenty-fifth the length.
is placed on board anywhere else both increase of
If a weight
draught and change of trim must occur. We imagine the weight

is added first at the centre of flotation, by which we can see how
much she will sink to a parallel waterline, and then we shift the
weight to the position given, and determine the change of trim.
The following will illustrate the method of dealing with the
addition of weights on board a ship.
EXAMPLE. A ship is floating at a draught o/20 ft. forward and 22 ft. aft,
when the following weights are placed on board in the positions given, viz.
20 tons, 100 /*. lefore the centre of flotation

45 80
60 50 abaft
30 10
What will le the new draught, the moment to change trim 1 in. being 800 foot
and the tons per inch 35.
tons
The total weight added is 155 tons, and if
placed at the centre of flotation,
the increased draught is
iff-
= 4^ in.
The weights to be moved forward give a moment of (20 x 100) + (45 x 80)
= 5600 foot tons, and the weights to be moved aft give a moment of (60 x 50)
+ (30 x 10) = 3300 foot tons. The forward moment is thus in excess by
2300 foot tons, and this will cause a change of trim of ^p Q = 3 in. nearly, and
the new draught forward is 20 ft. 4 in. + 1^ in. = 20 ft. 6 in., and the new
draught aft is 22 ft. 4* in.
- 1 in. = 22 ft. 3 in.
Strictly speaking, the change of trim ought not to be divided

equally forward and aft. It should be less than half (about 0'46)
aft, and rather more than half (about 0'54) forward. This will be
understood by reference to Fig. 180. Since F, the centre of
flotation, is abaftamidships, WW
is not equal to LL', but rather
less. Unless the change of trim is very considerable, however, the

error involved in taking the half forward and aft is small.
The following example will illustrate how to deal with the
problem of bringing a vessel to such a draught as will allow her to
1
pass over a place like a bar at the mouth of a river. It is assumed
also that no information is available, except the dimensions of the
ship, so that the approximations already given have to be employed.
EXAMPLE. A cruiser 300 ft. x 36J ft. is floating at a draught of 12 ft.

forward and 15 ft. aft. It is desired to bring her to an even ketl at a draught
of 12 ft., in order to pass over a bar at the mouth of a river. Determine

approximately how this could be accomplished.
The mean draught is 13 ft. 6 in., so that the ship would need lightening
18 in. The approximate tons per inch (page 186) would be
^-~
= 18*25
tons, so that 18 x 18*25

= 330 tons, say, must be removed. Suppose this is
done so that the vessel lightens to a draught of 10 ft. 6 in. forward, 13 ft.
G in. aft?"
The approximate moment to change trim 1 in. is Yrfojff x (300) 2 x 36-5
= 300 foot tons, say. As a change of trim of 3G ins. is necessary to bring the
vessel to an even keel, we require a moment of 300 x 36 = 10,800 foot tons.
The 330 tons must therefore be removed i$$ ga = 33 ft., say, abaft the centre of
flotation to give the necessary moment. Taking the centre of flotation as ^ the
length abaft amidships, or 12 ft.,the 330 tons would have to be removed so that
its C.G. was about 45 ft. abaft amidships to give the required draught of 12ft.
forward and 12ft. aft.
(It may be stated that, taking the correct data for the above ship, the answer
would be 350 tons, taken out 42^ ft. abaft amidships, so that the approximation
is a very good one.)
Passage through the Suez Canal. In order to pass through

the Suez Canal it is now necessary that the maximum draught
should not exceed 28 ft., and there is no difficulty in bringing
the deepest draught battle-ships to this draught.

The following were the instructions issued in the case of the
Majestic class, and although the conditions are now more favour-
able as regards draught it is instructive to note what could readily
be done in the way of lightening the ship.
1
If, as would probably be the case, the bar is in fresh water, allowance must be
made for the bodily sinkage in going from salt to fresh water, see p. 189.
236 WAR-SHIPS.,
In order that the vessel may pass through the Suez Canal, the extreme
draught should not exceed 25 ft. 7 in., which may be obtained in the following
manner :
1. Remove all water from the hydraulic tanks, the after fresh-water tanks,
and all except 20 tons from fresh- water tanks forward.

2. Eemove all water from boilers in the after boiler-rooms the water in the ;
other boilers should be to working height.

3. Empty the reserve feed-tanks, with the exception of about 10 tons of
water.
4. Remove all coal except 200 tons in bunkers abreast the four forward
boilers.
5. Remove all and spirits, except 20 tons ; the provisions,
provisions, bread,
etc., removed should beprincipally from the after store-rooms.
6. Remove all officers' stores and slops except 30 tons, removing principally
from the after store-rooms.

7. Remove 55 tons of shell from forward 12-in. shell-room, and an equal
quantity from the after 12-in. shell-room also 40 tons of shell from forward G-in.
;
shell-room, and an equal quantity from the after 6-in. shell- room.
Generally.
1. When the ship is floating at or near a mean draught of 27 ft., a weight of

675 tons added to or removed from the ship will increase or decrease the mean
draught by 1 ft.
2. A longitudinal moment of about 15,800 foot tons will alter trim by 1 ft.,
i.e. if w tons is weight moved, and d feet the distance the weight is moved
longitudinally, then -
QA ^ is the change of trim in feet, or the increase of
lO,oUU
draught the decrease of draught forward, or vice versa.
aft plus
3. A weight
placed on board or removed from about station 93, which is
about the middle of the after boiler-room, will not affect the trim.
4. To ascertain the combined effect on draught and trim of removing a weight
or placing a weight on board, the effect on draught only is first obtained by rule
(1) above, supposing the weight put on board at station 93 ; the effect on the
trim is then obtained by rule (2) above, the distance moved through being the
distance in feet between station 93 and the actual position of the weight.
Note. The removal of any weight before station 68, which is 12 ft. abaft the
forward bulkhead of forward boiler-room, will not diminish the draught aft.
Similarly, any weight removed from abaft station 116, which is abreast the
mainmast, will not decrease the draught forward.
Stability of a Submarine Boat. In a vessel totally sub-

merged the shape of the displacement does not alter for any
inclination (Fig. 181), and therefore the upward force of the
buoyancy must always act through the same point, viz. the centre
of buoyancy. The stability of any angle 9 is x BG X sin 0, W
varying directly as sin 6. It will be a maximum at 90, where
1
will
sin 9 is a maximum, and the angle of vanishing stability
be 180, where sin = 0. W * TE * SuRFACE -; . -
In order to give good sta-
bility, therefore, EG- must
be as large as possible. This
is done by so arranging the
weights and ballast that G

is below B.
Asubmarine boat has,

FIG. 181. Stability of a Submarine.
however, to float on the sur-
face, and then the conditions
of stability are precisely
similar to those of a ship,
the stability near the upright
depending on the distance GM
between the centre of gravity O
LWL DIVING
and the metacentre, and the
position of the latter point
being directly dependent on
the water-line area. The
AWASH
form of the metacentric dia-
2WL
gram as contrasted with that
for an ordinary ship is shown
in Fig. 181 A, which in each
case is calculated for normal
trim. It is seen that the
metacentre for the submarine
isvery low, and as the vessel
deepens in draught from the
awash condition to the diving SUBMAI
condition the centre of buoy-
ancy and the metacentre
gradually approach, this being
caused by the reduction of
the waterplane area. In the
submerged condition there is
no metacentre, and the sta-
blllty, as seen above, IS
^ 18]Ai _ Metacentric
and Submarine.
D iagrflni8 _ Cni iser
directly dependent on the

distance between the centre of gravity and the centre of buoyancy.
1
See next chapter.
23S WAR-SHIPS.
If the boat changes trim in the aiuash condition there is a

considerable reduction of stability, because of the altered shape
of the waterplane. Fig. 18 IB shows the distances between the
centre of buoyancy and the transverse metacentre of a cruiser
and a submarine for varying degrees of fore-and-aft inclination,
6 4-
TRIMMED BY EVEN TRIMMED BY

STERN. KEEL. Bow.
FIG. 181 B. Change of BM T due to Trim.
the distances for the submarine being exaggerated in scale, and
is seen that the metacentre sensibly lowers
it
diminishing the
metacentric height (already small) when the boat is much away
from an even heel. More important still is the effect on the
longitudinal stability. Fig. 181c shows the variation in the
distance between the centre of buoyancy and the longitudinal
metacentre of a cruiser and a submarine for varying degrees of
fore-and-aft inclination, the distance for the submarine being
TRIMMED BY TRIMMED BY
STERN, Bow.
FIG. 181c. Change in BM L due to Trim.
exaggerated in scale. It is seen how the longitudinal metacentre

becomes lowered in the submarine, and in a given case 4 of
fore-and-aft inclination decreased
longitudinalthe metacentric
height by 50 per cent, as compared with that height when on an
even keel. That means, that the moment to change trim 1 inch
was correspondingly reduced, making the vessel very sensitive

to any forces tending to change the trim. It is thus seen that
for a vessel shaped like a submarine any departure from an even

keel in the awash condition is accompanied by a serious diminu-
tion of stability, both transverse and longitudinal.
When proceeding at full speed in the awash condition it is
essential that the freeboard and reserve of buoyancy should be
at theirmaximum, as the ability to resist longitudinal inclinations
decreases rapidly with the diminution of the waterplane area,
and the transverse stability also decreases with deeper draught
due to the same cause.
For longitudinal inclinations the distance between B and G is
the same as for transverse inclinations, so that the case is very
different from an ordinary ship, in which the stability for fore-and-
aft inclinations is very great. In a submarine, therefore, the
stability is the same for all directions of inclination, and such a
boat is exceedingly sensitive to anything tending to disturb the
fore-and-aft position, and every member of the crew must remain
in one position and no weight must be allowed to shift.
(The illustrations, etc., accompanying these notes on sub-

marines have been taken from a paper read before the Eoyal
Society by Sir W. H. White, K.C.B., in 1906, and reproduced
in Engineering for May 25, 1906.)
Change of Heel and Trim after Bilging. When a vessel is
bilged near either end both bodily sinkage and change of trim occur.
This is well illustrated by the case of the Victoria in Chapter XXIV.
In that case the change of trim was so considerable as to bring the
upper deck forward under water, and consequently water gained
access to the ship through hatchways, etc., and the movement
down by the head was greatly accelerated. Fig. 182 has been
drawn for a box-shaped vessel 175 ft. long, 30 ft. broad, 15 ft.
deep, 8 ft. draught, before damage. If an empty compartment
between bulkheads 25 ft. and 55 ft. from the bow is laid open to
the sea the vessel will float at a draught of 13 ft. 5 in. forward
and 6 ft. 8 in. aft. It is seen that the stem head is quite close to
the water, and although the loss of buoyancy is not very consider-
able, yet this, with the change of trim, causes a dangerous condition.
It is thus seen to be most important to carry watertight transverse
bulkheads well above water. Figs. 52 and 54, which show the water-
tight subdivision of a battleship and small cruiser respectively,
show that most of these bulkheads are carried to the upper deck.
240
WAR-SHIPS.
Calculations are sometimes made to determine the effect of
opening up compartments to the sea, as might possibly be caused
Fio. 182.
by the ram, or torpedo, or by collision. Thus in a recent battle-
ship, ifa boiler-room amidships were filled, together with the

double bottom compartments right in to the vertical keel, and also
the upper and lower bunkers (filled with coal) open to the sea, the
ship would sink bodily about 20 in. and heel about 5 degrees.
The metacentric height on which the stiffness depends increases,
however, by about 6 in. If the forward boiler-room is taken, how-
" "
ever, there is also a change of trim which in this case would be
about 5 ft.
If, again, one engine-room be filled with the compartments

abreast also flooded as above, the bunkers being open to the sea
but full of coal, there is a bodily sinkage of about 23 in., a heel
of about 15 degrees, coupled with a change of trim by the stern of
about 3 ft., and an increase of metacentric height of about 6 in.
If, however, there is no coal in the bunkers the above figures are
increased to 32 in. bodily sinkage, 29 degrees heel, and a trim
by the stern of 39 in., with a decrease of metacentric height of a
foot. This latter result illustrates the great influence which coal
has in preserving stability when water is admitted to a bunker.
This has already been considered at some length in Chapter XVII.
Besides the trimming tanks provided at the ends of a ship
for correcting trim, the submerged torpedo-rooms are compart-
ments which can rapidly be flooded through the torpedo tubes.
Thus in a battle-ship the flooding of a broadside submerged torpedo-
room and the spaces adjacent leads to a bodily sinkage of about
10 in. and a change of trim of about 6 feet by the bow or stern
as the case may be. Heel may be corrected by the admission
of water to wing compartments, which are provided with flooding
arrangements for this purpose.

Information for use when docking in a Floating Dock.

When a ship is to be docked in a floating dock, especially if the
weight is close to the lifting capacity of the dock, it is necessary
to place the ship in the dock so that its C.G. is at the centre of
length of the dock, in order thatwhen lifted the dock shall be on
an even keel.
Information in the following form is now provided on the
docking plans of all H.M. ships to enable the position of the C.G.
to be closely approximated to knowing the draughts forward
and aft.
CHAPTEK XVIII.
STABILITY AT LARGE ANGLES OF INCLINATION.
WE have seen that the stability of a ship at any angle is the effort
she makes to return to the upright when put over to that angle.
For small angles of inclination, up to 10 to 15, this depends
directly on the metacentric height.
Thus at 10 the Eoyal
Sovereign, with 3J ft.
and GM
14,150 tons displacement, will
have a righting moment of
14,150 x 3-5 X sin 10 = 8,600 foot tons.
It is however, for a vessel to have sufficient meta-

possible,
centric height but insufficient stability at large angles. This was
specially brought
out in the investigations which followed the loss
of H.M.S. Captain. Metacentric height alone, apart from other
considerations, principally free-
board, will not ensure a vessel
having sufficient stability, and
special calculations are necessary
to determine the righting moment
at large angles of inclination.
Curve of Stability. Take a
vessel inclined to a large angle 0,
as Fig. 183. The upward force
of the buoyancy acts through B',
the new
centre of buoyancy, and
the couple tending to right the
FIG. 183. ship is X GZ, GZ being theW

righting lever. The length of this
righting lever will depend on how far the centre of buoyancy shifts
out, and this length will vary for different angles. Thus for
H.M.S. Captain the following values of GZ were calculated, viz.
7, 41 in. 14, 81 in. 21, 10f in. 28, 10 in. 35, 7f in.; 42,
; ; ; ;
242
STABILITY AT LARGE ANGLES OF INCLINATION. 243
5J in.54J, zero. A convenient way of representing these results

;
is draw a base line to represent angles of inclination and set up

to
as ordinates the lengths of GZ as found. A curve drawn through
the spots thus obtained is a curve of statical stability. The curve
for the Captain is in Fig. 184.
Fig. 185 shows a curve of stability constructed as above. The
angle at which GZ
obtains its maximum value is termed the angle
of maximum stability (in this case 47). The angle at which the
M ON ARCH -
FIG. 184.
curve crosses the base line (in this case 77) is termed the angle of
vanishing stability, or the range of stability. Up to this angle the
vessel possesses a righting lever which will take her back to the
upright. At 77 the ship is in equilibrium, the C.G. and C.B. being

in the same vertical, but this equilibrium is unstable, and a small
inclination either side of 77 will take her away from that angle ;
if to 75, say, she will go back to the upright; if to 79, say,

she will
capsize.
244 WAR-SHIPS.
In striking contrast to the curve for the Captain is that for the
Monarch (Fig. 184). In this case the angle of maximum stability
is not reached until 40 as against 21 in the Captain, and the
value of the maximum GZ is about twice as great. The stability
does not vanish until the large angle of 70 is reached. The reason
for the difference between the two ships is seen by comparing the
sections. The Monarch had high freeboard, which pulls out the
centre of buoyancy as the ship heels over. The Captain had a
low freeboard, giving a curve of stability which was dangerously
small for a ship carrying a large amount of sail.
In considering a curve of stability certain assumptions have to
a FT
MAXIMUM STABILITY 47
90^
FIG. 185.
be made, which are necessary, in order to make the calculations at

all feasible.
1. The
sides and deck are assumed to be watertight for the
range over which the curve is drawn. Thus all sidelights and
gun-ports are supposed to be closed below the upper deck If the
effect of the forecastle or poop is included, any openings in these
superstructures are supposed to be closed.

2. The C.G. of the ship is taken in the same position in the
ship throughout the inclination, i.e. it is supposed that no shift of

weights takes place.
The important features of a curve of stability are :
1. Inclination which the curve has to the base line at the start.
This inclination depends directly on the metacentric height.
2. The angle at which the maximum value of the righting arm
occurs and its value at that angle.

3. The range or the angle at which the curve crosses the base
line, and the vessel becomes unstable.

Curve of Stability by variation of Beam, Free-
Effect on a
board, and Position of C.G. In order to illustrate these points
a box-shaped vessel is taken having a breadth of 50^ ft., draught
of 21 ft., freeboard of 6 ft., and metacentric height of 2 '6 feet.
8 FT
IFI
'
80 90
A- -J3- -C- -D-

2) I
2 46 WAR-SHIPS.
ship has therefore the same GM

as at first, viz. 2 '6 feet. The
freeboard, however, has the effect of lengthening out the curve
enormously, and the range is increased from 39 to 84.
3. If now we take the ship as
represented by C, assuming
that the C.G. is raised 1 ft., making the GM
1/6 ft., the curve of
stability is given by C, the range being reduced by about 10.

The curve C may be regarded as giving the double effect of
increasing the freeboard, viz. rise of C.G. as well as increasing
the freeboard.
We see from these examples
(a) An increase of beam increases the initial stability, and
IF!
ao w 40.
FIG. 187. Stability curves, H.M.S. Furydice.
therefore the slope of the curve near the origin, but does not
greatly influence the area enclosed by the curve or the range.
(5) An increase of freeboard has a most important influence
in lengthening out the curve and increasing the area enclosed.
(c) An
alteration in the position of the C.G. influences
both the initial stability and the area and range of the curve
of stability. If the C.G. is raised to G', say, then the ordinate
of the curve at angle 9 is lessened by GG' sin 9. If the C.G.
isdepressed to G", say, the ordinate of the curve at angle 9 is
increased by GG" sin 9.

Dynamical Stability. The work necessary to be done on a
ship to force her over to a given angle is termed the dynamical
stability at that angle. It can be shown that a measure of the
dynamical stability at any angle is obtained by the area of the

curve of statical stability up to that angle. This is illustrated by
Fig. 187, in which the curves of stability of the Eurydice are
shown. When the sides of the ship were watertight to the upper
deck, the curve of stability enclosed a large area. When however
MlANTONOMOH.
SFI
2FX
70
FIG. 188. Cunres of stability.
the ports were open, there was a sudden drop at 17, and the
ordinates of the curve decrease until the vanishing point was reached
at 40. The ratio of the areas enclosed is about 1 to 3 ;
i.e. with
the sides intact, about three times as much work would have
FIG. 189. Curves of stability.
to be done by the wind and the heave of the sea to capsize the
ship as would be necessary supposing the ports open.
It is thus seen that the area enclosed by a curve is of im-
portance as well as the range of stability.

Curves of Stability. In Figs. 188 and 189 are given some
typical curves of stability.
2 48 WAR-SHIPS. .
The Miantonomoh was an American monitor of low free-

board, viz. 3ft., and great GM, viz. 14 ft. The curve rises with
great steepness at the start, and the adverse influence of the low
freeboard is counteracted by the great metacentric height.
The Royal Sovereign is a typical British battle-ship, with a GM

of 3J ft, and high freeboard, viz. 17 ft. The area enclosed by the
curve is considerable, and the range is 63. The Triumph and
Swiftsure, bought into the Eoyal Navy, have a curve of stability
very similar to Eoyal Sovereign.
The Magenta is a French battle-ship, with a of 2'3 ft., and GM
l
high freeboard of about 16 ft. The low curve for this vessel is
caused by the small GM, and also by the great fall in of the side
above water, which is a notable feature of ships of the French
Navy. Fig. 132 shows a typical section of one of these ships, in
DECREES
FIG. 190. Stability curves.
which the breadth at the deck is only 48 ft., while the breadth at
the waterline is 66 ft.
The Eoyal Arthur is a typical British cruiser. For this ship

two curves are constructed (a) assuming the forecastle open, and
(6) assuming the forecastle closed. It is seen that without the
forecastle the range is large, viz. 82, and the effect of the fore-
castle, if watertight, would be to increase the range to over 90.
Destroyers have good curves of stability, as they have a

relatively large GM
and great reserve of buoyancy. The curve
shown is for the Daring, in which the range is over 90.
Any curve of stability of a ship is drawn for one particular
displacement and position of the C.G., so that separate curves
have to be drawn for the various assumed conditions of the ship,
viz.deep load, normal load, and light, Fig. 190. The angles
of maximum stability and vanishing stability in the two former
1
Marine Francatse, May, 1895.
conditions are supplied to the Commanding Officer in the stability

statement already mentioned. If the ship has a poop or fore-
castle, or both, the effect of these is usually noted on the statement.

A specimen statement is given below. (It is suggested that the
ANGLES OF INCLINATION.
Fro. 191.
particulars of the stability of the ship in which officers are at
present serving might be added.)

Effect of Tumble Home. We have seen the importance
of freeboard or reserve of buoyancy in giving ships a curve of
stability with good range and enclosing a sufficient area. In the
case, however, of some foreign ships, good freeboard is associated
with a great tumble home, or fall in of the sides (see fig. 132).
250 WAR-SHIPS. ,
This acts prejudicially in regard to the stability, as the

following
example will show :
Two vessels have been taken, Fig. 191, both with 13 ft.
draught, 9 ft. freeboard, 32 ft. beam, and 2 ft. metacentric height.
In one however, the side falls in as shown to the deck, where
case,
the breadth is 24 ft. This results in a much lower curve of
stability than is obtained without the fall in we have seen that
;
the area enclosed by a curve of stability is of great importance.

The corresponding metacentric diagrams are also interesting.
In the second case, directly the side falls in the metacentric curve
drops as the draught increases. Such a ship if damaged so that

a bodily sink age results might very possibly lose her initial
stability, owing to this drop of the metacentre as the draught

increases, especially if the ship was damaged in the neighbourhood
of the waterline.
Destruction of Upper Works. A war-ship should have
sufficient stability for manoeuvring and fighting supposing the
unarmoured portions of the side are open to the sea, and in this
connection height of armour above water is of great importance
as this fixes the height of the freeboard under such conditions.
It has been calculated that the French ship Carnot, with the thick
belt 2 ft. above water, has only a
range of stability of 6| degrees
when the unarmoured portions of the side are open to the sea.
Such stability is quite insufficient for safety. Eecent French
designs, as e.g. the Eepubliqut, have recognised this serious weak-
ness and provide for the armour being much higher than formerly;
in the case of the Repullique, this being 7 ft. 6 in. It will be seen
by reference to Fig. 131 that all British designs since the Royal
Sovereign carry the belt high above water, enabling sufficient
stability to be maintained, even supposing the whole of the
unarmoured side is penetrated and open to the sea when the
vessel inclines.
H.M.S.
STATEMENT OF METACENTRIC HEIGHTS AND STABILITY, BASED ON AN INCLINING

EXPERIMENT MADE ON THE SHIP ON MARCH 27; 1900.
Conditions. Feet. Remarks.
A. The ship when fully equipped, with reserve

feed-tanks empty, with 300 tons of coal
in lower bunkers and 300 tons of coal in
upper bunkers, at a mean draught of 20
ft. 6 in. has a metacentric height of
B. Theship when fully equipped, with fresh-

water and reserve feed-tanks and all
bunkers quite full, i.e. with 1085 tons
of coal on board, at a mean draught of 21
ft. 10 in. has a metacentric height of ... 2-4
When lightened to a mean draught of 18 ft. 3

in., or when boilers are full to working
height, engine condensers and feed-tanks
at working height, and all coals, water
(including reserve feed), provisions,
officers' stores, and one half warrant
and engineers' stores consumed,
officers'
the metacentric height is about ... 20
Forecastle.
Open. Closed.
Deg.
The angle which the ship reaches her
at
maximum above condition
stability in the
A, and beyond which the righting force
diminishes, is about ...
The angle at which the ship reaches her

maximum stability in the above condition
B, and beyond which the righting force
diminishes, is about 4T. 45
The angle at which her stability entirely
vanishes in the above condition A is
about 80
The angle at which her stability entirely
vanishes in the above condition B is
about 80
NOTE. So far as stability is concerned the coal may be worked in any manner
desired by the Commanding Officer.
CHAPTER XIX.
THE ROLLING OF SHIPS.
Rolling in Still Water. Boiling in still water is of no

immediate practical importance, because under ordinary circum-
stances a ship will not roll in still water. It is, however, necessary
to study the subject, because it is only when the conditions
operating in this case are understood that we are able to extend
the inquiry to the more difficult case of rolling among waves.
If a ship, floating upright in stable equilibrium in still water,
is inclined to a certain angle from the upright, the couple
tending to take her back to the upright is W
x GZ. If the ship
is released she will acquire angular velocity, passing through the
upright to an angle on the other side rather less than 0. At this

new angle, 0', say, she will have a couple tending to take her back
to the upright, and so the ship, once being inclined and released,
will continue to oscillate through smaller and smaller arcs of
When the
oscillation until she finally comes to a position of rest.
ship is inclined to the angle as above, it is necessary to do work

on the ship to effect the inclination, and this work is stored up in
the ship as potential energy, or energy due to position. When the
ship is released, this energy becomes converted into other forms of
energy, and, if no resistances were operating tending to stop the
motion, the ship, when passing through the upright, would have
kinetic energy, or energy due to motion, exactly equal to the original
potential energy.
This conversion of energy of one form into energy of other forms is well seen
in the case of a stoneon the top of a house. In some way work has had to be
dono to get the stone there, and the stone, in virtue of its position, has stored up
in it a certain amount of If the stone is released, it will reach
potential energy.
the ground with a certain velocity, and (neglecting the friction of the air) the
kinetic energy then possessed is equal to the original potential energy. When
stopped by the ground both potential energy and kinetic energy disappear, but
the energy is not lost, but is dissipated into the form of heat energy.
252
THE ROLLING OF SHIPS. 253
In the case of a ship rolling unresistedty the energy is alter-

nately potential and kinetic at the extremity and middle of each
roll, and the rolling would go on continuously but when resistances
;
operate, the energy gets drained away from the ship and becomes
finally dissipated by imparting heat and motion to the air and
water surrounding the ship.
Unresisted Rolling in Still Water. It can be shown that
for unresisted rolling in still water the period of a single oscillation
(from port to starboard, or vice versa) is

very nearly given by
T =
V
0-55 X /Jl!_ seconds
GM
aivr
where GM is the metacentric height and k is obtained from the

following definition :
The moment of inertia of a solid body about any axis is found

by adding together the product of each weight making up the
body and the square of its distance from the axis. (This is
analogous to the moment of inertia of a plane area about an axis,
dealt with in Chapter XVII.) If for a ship this axis is the axis
of oscillation, W the weight and I the moment of inertia about
the axis, then Te is such a quantity that I = W
X &2 and k is termed
,
the radius of gyration.

The calculation for k is a
most laborious one, but it has been
done in a few and having also the metacentric height, an
cases,
estimate could then be made of the time of oscillation from the
above formula. Practical agreement was found to exist between
the actual and the estimated times of oscillation, even although
the rolling could not have been unresisted.
The formula shows that to make the period long, i.e. to increase
the time of oscillation, it is necessary to
(1) increase the radius of gyration, and, or

(2) decrease the metacentric height.
(The longer the period of a ship the more likely is she to be steady
in a seaway.) Of these two, the first is of the lesser importance,
because the distribution of the weights is governed by other
features of the design than the desirability of obtaining a long
period. The formula, however, shows clearly that winging weights,

i.e.
placing them at the sides, operates in the direction of increasing
the period, although any practicable shift of weights on board a
254 WAR-SHIPS.
war-ship can otily have a small effect on the period. We should

expect, therefore, to find that an
armoured ship would roll more
slowly than an unarmoured ship of about the same displacement
and metacentric height, and this is confirmed by experience of
ships in the Eoyal Navy.
A very considerable effect in lengthening the period is obtained
by reducing the metacentric height. Thus in the Eoyal Sovereign,
in which ship the period is 8 seconds and the GM
about 3J ft.,
suppose the GM is reduced to 3 ft., without altering the radius
of gyration. Then we should get a period of 8*64 seconds, or an
increase of 8 per cent.
An interesting application of the above principles is found in the current

practice of many merchant vessels. In many trades, voyages have to be under-
taken with little or no cargo, because of the absence of return freights. It is
necessary, for seaworthiness and proper immersion of the propellers, to sink the
vessels by means This has usually been placed in the double-
of water-ballast.
bottom compartments. This, however, frequently pulls down the C.G. of the ship
so far as to give the ship a large GM. This causes a very quick period, and in
some cases has not merely rendered the ship uncomfortable, but actually
this
unsafe. In ships, therefore, it is the practice to provide tanks in the 'tween
many
decks and hold at the sides, and even on the upper deck. These tanks below
are frequently large enough to hold ordinary cargo when necessary, but for
"light" voyages they can be filled with water. The weight thus added, while
giving sufficient immersion, does not produce excessive GM, and being at the
sides tends to lengthen the period by increasing the radius of gyration.
The assumption used in obtaining the above formula for the
period from side to side, viz. T = Q'55\/ A_J is, that the righting
GM
lever varies directly as the angle or GZ = GM
x 0, i.e. it assumes
that the curve of stability is a straight line up to the angle
considered. Under this condition large and small inclinations will
be performed in the same time. A ship rolling in this manner is
said to be isochronous.
Although the various assumptions made in obtaining the above

formula are not strictly true, yet it is found by actual experiment
that, within angles of 10 to 15 of the vertical, ships are very
nearly isochronous in their rolling. This is the case although the
ship experiences resistances which eventually bring her to rest.
The following are the approximate periods of some typical
ships, i.e. the time from port to starboard, or vice versa,
( A vessel of large GM., viz. 8 ft., causing a

5 Sees. < small period. This was the cause of the
Inflexible,
(
introduction of water chambers (see later).
Royal Sovereign \
(
shi P s of moderate GM - ab ut 3J ft., and
., . ,. > 8 sees. < great moment of inertia, due to beam and
Majestic \ \ armour at BideB.
Powerful 7 sees. \
ArTOQCtnt 6 Sees* >

Protected cruisers with no side armour.
Pelorus _
5k
^ sees. )
Gunboats and I
Small period due to (a) small moment of
j
Destroyers I
{
inertia; (b) relatively very large GM.
Resisted Rolling in Still Water. Under the actual con-

ditions under which a ship will roll in still water, resistances to
the rolling are set up which drain the ship of energy and which
sooner or later will bring her to rest. These resistances may be
classified as follows :
1. Friction of water on the ship's surface.

2. Effect of sharpness of form of ship's section.
3. Effect of bilge keels or keel projections (if any), including
the flat portions of the ship.
4. Creation of waves on the surface.
5. Air resistance.
6. Use of water chambers.
1. Friction. This cannot be of great amount in ordinary
is kept smoothly painted to reduce the
ships, because the surface
resistance when steaming to the smallest possible amount.
2. Form of Section. In a ship of circular section the relative
velocity of the water and the surface of the ship is the same at
all points of the section. In a ship of sharp form at the bilge,
however, the water at the corner gets a motion opposite to the
ship, and having to slip past
the bilge, the effect both as regards
friction and on bilge keels is greater in a sharp bilge than in a
rounder form of section.
3. Bilge Keels. The reason of the great extinctive effect of
bilge keels in reducing rolling has been imperfectly understood
until recently. The explanation is of considerable difficulty, and
the following remarks do not pretend to completely deal with the
subject :
(a) A bilge keel is like a flat surface passing through water

broadside on. The laws governing the resistance of such flat
surfaces have been investigated, but in applying them to the case
256 WAR- SHIPS.
of a ship it has been found that the extinctive effect observed

could not thus be fully accounted for.
(b) A
further influence has been suggested by Prof. Bryan,
F.K.8. Consider the flow of water round a right angle as Fig. 192.
The water next the surface has to suddenly
change its direction at B. This causes a
M \ diminution of speed up to the point B, where
u \ .
v^ ^ it must be zero. The other streams of water
-^^^- A are deflected, and along AB we get a diminu-
FIG 192
^ on ^ vel c ity f th e stream lines. This
falling off of speed is accompanied by an
increase of pressure, both along and BC. 1 If, for instance, AB
a boat's rudder is suddenly put over to right angles, the leverage
of the water pressure on the rudder about the axis of rotation
of the boat is small, but the turning effect on the boat is con-
siderable. caused by the pressure on the deadwood of
This is
the boat, which has a considerable leverage about the C.G-. of the
ship.
In the case of bilge keels projecting from the surface of a ship,
suppose the ship is rotating clockwise, as in Fig. 193. The relative
velocity of the ship and the
-|
Û i water along A2 Ai has to be
brought to zero at AI, and
there is caused an increase
of pressure along A 2 Ai.
This results in a resultant
force P acting as shown,
a similar force Q being
Flo 193
found on the other side of
the ship. These forces both
act in such lines that they give a moment tending to stop the
rotation. This effect will be more pronounced as the section of
the ship sharper, because of the greater relative velocity of the
is
water past the bilge as compared with a round section.

4. Wave Formation. At each roll of the ship a wave is
The relation between speed and pressure in flowing water, by which if speed
1
diminishes the pressure increases, and vice versa, is noticed in water-pipes. If a

tap
is turned off suddenly, as in the old-fashioned
taps, a knock is heard in the pipes,
caused by the sudden rise of pressure consequent on the motion being stopped. If
as in taps now in domestic use, the water is turned off gradually, the rise of
pressure
is not so sudden and the pipes are not so severely strained.
created on the surface of the water at each side ; this wave passes
away from the ship, and requires energy spent to create it. A wave of
very small height represents a large amount of energy, and the drain
on the ship's energy is a distinct resistance reducing the rolling.
5. Air Resistance. The resistance of the air must be quite
small under ordinary circumstances, but it may be made consider-
able by the use of steadying sails. It is well known that sails
have a great steadying on a ship's rolling.
effect
6. Water Chambers. In the Inflexible and following ships

a large metacentric height was an essential feature of the design,
because it was necessary to provide such stability that the vessels
should be able to stand upright, even supposing the unarmoured
ends were completely riddled. It was known that this would
cause a short period and quick rolling motion. This is an
undesirable quality in any ship, and especially in a war-ship. The
bilge keels could only have been of limited size because of the
great beam of the ship causing difficulties in docking. On this
account it was proposed and approved to fit athwartship chambers
containing loose water. This water passes from side to side as

the ship rolls, and causes waste of energy. This must be taken
out of the ship, and so lessens the rolling. These water chambers
were found to fulfil their purpose in diminishing the rolling, but
the system was ultimately abandoned on account of the noise of
the water rushing from side to side, and because the spaces were
required for other purposes. As we have seen, there has been a
gradual increase of waterplane area protected by armour in battle-
ships since the Inflexible, so that the riddling of the ends has a less
proportionate effect. On this account metacentric heights have
been diminished from 8 ft.in that ship to 3|- to 4 ft. in more
recent ships. This has resulted in longer periods, so that the
conditions of rolling are quite different, and the steadying effect of
water chambers has not been required.
Bilge Keels. Fig. 194 shows several forms of bilge keel as
fitted to ships of the Navy. The standard form for steel ships
is made of two 17J-lb. (fa in.) plates connected to the ship's bottom
as shown, the space between being filled in with light wood. The
projection in battle-ships is inconvenient in connection with dock-
ing, and the breadth
of the bilge keel is made somewhat less amid-
ships than forward and aft where the ship gets narrower.
For smaller ships the bilge keel can be formed as shown
the figure gives the construction in a torpedo gunboat.
258 WAR-SHIPS. ,
For sheathed ships the keel is formed of a single steel plate

connected to the bottom by double angles, and supported at
intervals by brackets. This is cased in with teak as shown.
All vessels in the Navy, including the latest destroyers, are
fittedwith bilge keels. The keels usually extend over rather less
than half length. In ships of very full midship section, as Fig. 13 A,
the projection of the bilge keel amidships is small to facilitate
docking, but forward and aft it is increased in depth as the form
of the ship gets finer.
Rolling among Waves. In dealing with this subject it is
important to note that a wave is not the passage of water but the
passage of motion. The motion of the particles of water composing
a wave is quite small, as may be noticed by watching a piece
ofwood among waves. The wave profile is seen to move along
with considerable speed, but the wood sways backwards and for-
SMgATHINS
SHEATHED
SHIR
BATTLESHIP
FIG 194. Bilge keels.
wards about a mean position. As a matter of fact the particles of

water are moving in circular orbits, the radius of these orbits
decreasing with the depth, so that at a moderate depth the water
has a very slight motion. In waves, therefore, we find that the
force of gravity is modified because of the centrifugal force set
up by the orbital motion of the water.

It is well known that a can of water can be
swung right round without any
water spilling. When at the highest point the weight of the water acts down,
but the circular motion gives rise to a centrifugal force acting outwards, and so
long as this latter force is greater than the weight no water will be spilled. If
the motion is slowed up a point would be reached when the
weight would be
greater than the centrifugal force, and the water would be spilled. It is the
centrifugal force being greater than the force of gravity which keeps the car on
"
the rails in looping the loop,"
At the crest of a wave the centrifugal

force on the particles
of water acts upwards against gravity, at the trough it acts
downwards with gravity. The apparent or virtual weight of a
body will therefore be less in the crest than the actual and more
in the trough than the actual. This apparent weight may be
20 per cent, more or less than the real weight, according as it is
in the trough or the crest.
This has been frequently verified by experience. If a spring balance is used,

the indications on the dial for a given weight will be found less when the vessel
is on the crest, and more when in the trough of a wave. This is also the
explanation of the well-known phenomena of the tenderness of sailing boats on
the crest of a long smooth wave. The virtual weight is considerably less than
the actual, and consequently the righting moment is less than in still water.
The wind moment is not affected in this way, and so on the crest of a wave, a
boat, of sufficient stiffness in still water, is liable to be blown over to a large
angle and possibly to capsize.
The effect of the centrifugal force at other portions of the wave

is to alter the amount of the virtual gravity, and to cause it to
LENGTH
FIG. 195. Wave profile.
act in a line of action perpendicular to the wave slope at any

particular point. In considering rolling among waves we do not
usually consider the variation of the virtual weight, but we must
consider the variation of the line of action of the virtual buoyancy
and gravity. This line of action will have its maximum inclination
at about quarter the length of the wave from crest to crest or
trough to trough. A
small raft, as in Fig. 195, will always tend
to keep normal to the wave surface; this normal is termed the
virtual upright at any particular instant.
If now we take a ship floating broadside on to a series of
waves (supposed long in comparison with the size of the ship), we
shall have the set of forces as shown in Fig. 196. The inclination
of the wave normal to the vertical is 9, and this wave normal is
the virtual upright. If the ship is as shown the righting force is
not due to the angle 9 +

9', but to the angle 9'.
260 WAR- SHIPS.
If the ship has a very quick period compared with that of the
wave, she will quickly come to the virtual upright, and so will
take up the motion of the wave. This would be the case in a
raft as in Fig. 195,and it is found that a ship of very short period
does not roll very much or ship much water when rolling among
waves, because she always keeps the deck parallel to the wave
surface.
If a ship has a long period compared with that of the wave, the
ship, at any particular instant, as in Fig. 196, does not come to
the virtual upright with any sudden-
ness,and the wave profile passes on
and soon acts in the contrary direction.
The ship therefore remains steady,
never heeling to large angles. This
quality of remaining nearly upright
when among waves is termed steadiness,
and obtained in ships with a long
is
period. We have seen above that a

FIO 196 long period is mainly obtained by
giving a small metacentric height.
Such a ship is crank, i.e. is easily inclined by external forces,
but in a seaway is most likely to be exceedingly steady.
If, however, a ship has her double period (port to port or vice
versa) equal to the period of the waves (time the length of the
wave is traversed), we have a serious state of things. This timing
istermed synchronism. As each wave passes the ship, an impulse
is given timing with the period of the ship herself, and the
tendency of this is to produce larger and larger angles of oscillation.
was perfectly isochronous for large as well as small
If the vessel
angles, and no resistances were acting, such a system of waves

would inevitably capsize her. The actual conditions operating,
however, are as follows :
1. As large angles are reached a ship departs from isochronous

1
rolling, and the condition of synchronism with the wave is not
fulfilled.
Eesistances operate, and, especially in a vessel with bilge

2.
keels, the energy imparted by the wave is soon absorbed by the

energy taken out of the ship by the various resistances. When
1
For a simple pendulum swinging 30 each side of the vertical 1-7 per cent,
increase of period is noticed as
compared with a. gmalj oscUUtion; for 60. 7'3
per cent.
these are equal no further increment of rolling can take

place.
3. A succession of waves of precisely the same period is a very
unlikely occurrence.
It has often been noticed that ships with a great reputation
for steadiness at sea occasionally roll heavily. This is doubtless
caused by the fact that a succession of waves has been met with
having a period approximately synchronizing with the double
period of the ship.
In a ship thus rolling heavily a slight alteration of the course
would be sufficient to destroy the synchronism, since what affects
the ship is the apparent period of the waves, and if the ship's
course be taken obliquely to the wave advance, the synchronism
is at once destroyed.
The longer the period of a ship the less chance there is of
meeting synchronizing waves. A series of waves of 16 seconds

period is quite exceptional, so that the battle-ships of the British
Navy having 16 seconds for their double period should be very
steady, and this is borne out by actual experience. Atlantic
storm waves have periods about 10 seconds, and it is only the
smaller vessels of the Navy which have their double period as low
as this see above for periods of some typical ships.
;
Observations of Rolling. 1 If a ship is rolling in still water,

and a pendulum could be suspended at the centre of oscillation,
then the point of suspension of such a pendulum would have no
motion, and the pendulum would always remain vertical the ;
angles indicated would therefore give the angles of oscillation of

the ship. If, however, the point of suspension is somewhere else,
then as the ship rolls this point has motion and the pendulum
does not give the true vertical. If one takes a fishing-rod, for
instance, with a few feet of line, it is evident that, if the rod is
swayed backwards and forwards, the line does not remain vertical.
The same state of things obtains on board a ship the pendulum ;
does not hang vertically, and the angle it indicates will be in

excess or defect of the true angle to which the ship rolls, unless it
happens to be suspended at the centre of oscillation. If the point
of suspension is above this centre the angle indicated will be in
excess, if below, the angle will be in defect. (It is usually
assumed that the centre of oscillation is near the C.G. of
In Sir W. H. White's
1
"Manual of Naval Architecture" a whole chapter is
devoted to this subject.
262
the ship.) In the above illustration of the fishing-rod, if the

line is very long, the motion of the rod does not have any sensible
effect on the line, so that for all practical purposes the line will
remain vertical. This is the principle which has been effectively
employed in instruments for measuring rolling, viz. that a
pendulum of very long period is not appreciably affected by the

motion of the ship, but will maintain itself practically vertical as
the ship rolls.
When a ship is rolling at sea, an ordinary pendulum is still
less likely to give correct angles, as in addition to rolling, the ship

has a bodily movement among the waves.
Mr. Froude's apparatus for rolling records is most valuable
when accurate observations are desired, as it shows automatically
the rolling of the ship together with a time record. It is, however,
too elaborate an instrument for ordinary use. It primarily depends
on a heavy wheel, so weighted that its C.G. is very close
to the axis of suspension (J^Q-Q in. below), giving a very long
1
period for a single oscillation, viz. 34 seconds.
An
instrument, Fig. 197, which is very simple and which has
given admirable results has been devised by Mr. Mallock (I.N.A.,
1901). It gives a pen-
dulum of very long
period, but the instru-
vi ment is of small dimen-
sions. A paddle P is
supported on delicate
pivots and is enclosed in

a box containing fluid.
The paddle is made of
the same density as the
fluid, and in this way
the friction on the pivots
is very small (buoyancy
FIG. 197. Mallock's rolling indicator.
practically equals the
HH, Glass front B, Hollow boss; I, Pointer;
;
F, Filling hole and expansion box. weight). Any rotary

motion of the outside
case is not communicated to the mass of the fluid, and in the
interior of the box the fluid is practically at rest when the box is in
motion. The paddle is adjusted so that its C.G. is just below the
This interesting as analogous to the case of a ship vrith great moment of
1
is
inertia and small metacentrio height, both of which conduce to a long period.
axis. When free it has a complete period of 4 seconds, but

when enclosed in the box the complete period is between 30 and
40 seconds. The paddle therefore remains practically vertical, and
the box being attached to the ship, the pointer I will show on the
paddle the angle of This rolling indicator is now issued to
roll.
1
ships of the Royal Navy.

The method of observing angles of roll by the use of battens is
very simple. It can, however, only be used when the horizon or
some fixed object (as a star) is visible. The battens are arranged
so that they can be rigged up on the fore bridge as Fig. 198.
Fia. 198.
There are two battens on which a scale of angles is painted, the
zero corresponding to the horizontal when the ship is upright. A

bracket is placed at the middle line (attached to the chart house,
or distant object can be
say) having a horizontal slit. The horizon
sighted through this slit at the extremity of each roll, and the
Another is a metal
angle can be noted. fitting that is supplied
semicircle graduated in degrees, with a movable pointer pivoted
at the centre which can be sighted to the horizon, and thus the
angle
& of heel read off. The time of each roll should also be noted
by another observer. Forms are issued to ships of the Eoyal
Navy giving detailed instructions.
i " Mr. Mallock's instrument

is exceedingly simple, it is always in place, it may
be put anywhere, always measuring the angle of heel and is ready to be

it is
observed. I have invariably heard it spoken of very highly by those who have
used it." Mr. (now Sir) Philip Watts, F.R.S. (I.N.A., 1901).
264 WAR- SHIPS.
Pitching. The expression for the period of pitching of a ship

is of similar form to that for rolling, only we have to use ki, the
radius of gyration of the vessel about a transverse axis through
the centre of gravity of the vessel, and GM
L the longitudinal
,
metacentric height. This period is
It is thus seen that increase in k and decrease in GM L will

tend to lengthen the period of pitching. The former is caused
by heavy weights near the extremities, and other things being
equal a ship with her heavy weights near the centre of length
will have a shorter period and be a drier ship than a vessel with
the weights towards the ends. It is understood that the Renown
"
has a reputation in the service for being a " comfortable ship at
sea, and this is probably due, at any rate partially, to the closeness
of the barbettes containing the 10-in. guns, to which must of
course be added the good freeboard.
CHAPTER XX.
THE TURNING OF SHIPS.
the rudder of a ship moving ahead is put over, a force is

brought into existence causing the ship to (1) heel, (2) turn, (3) to
slacken in speed, and (4) to have side movement or drift. The
rudder, being placed obliquely to the middle line of the ship,
causes the streams of water flowing aft to be deflected, and this
causes a force to act upon the rudder, as P, Fig. 199. The value
of this normal force depends upon the area of the rudder, the
square of the speed of the water meeting the rudder, and the angle
to which the rudder is placed. In a sailing-ship the speed of
water meeting the rudder is rather less than the speed of the ship,
because the friction of the ship's surface causes a layer of water to
be dragged along in the direction of the ship's motion. The rudder
of such a ship is thus not passing through still water but through
water which has a forward motion. The steering of a sailing-ship
depends on the motion of the ship, and such a ship loses her power
of steering as she loses way. With a screw-steamer, although
there is the same frictional wake, yet the action of the propellers
send a stream of water astern, and such a ship has steerage directly
the engines are working, before she attains any motion at all. A
ship with a very full stern is likely to steer badly, as the water
does not get a clean flow past the rudder, which is necessary in
order to get the normal pressure required. (See discussion of the
steering of Agamemnon, United Service Institution, 1887-88.)
In Fig. 199, P be the normal pressure acting on the rudder
let
at C. of the ship, G, introduce two equal and opposite
At the C.G.
forces, P, in a line parallel to the line of action of P. Then we
have acting on the ship
(i.)
A couple tending to turn the ship, as shown, of magnitude
PxDG; and
(ii.)
A force, P, acting in the line EG.
265
266 WAR-SHIPS.
This force, P, will have a transverse component, FG-, P cos 0,

to starboard, and a fore-and-aft
tending to move the ship bodily
component, EF, P sin 0, tending to stop the ship. The force
since the resistance of the
causing side motion has small effect,
ship to this motion is very great. The fore-and-aft component
has, however, a sensible effect in checking the speed when turning
In a ship with a deadwood there is a side pressure due

to the slackening of the stream lines on putting the rudder over.
This side pressure on the deadwood has a considerable leverage
to turn the ship (see Fig. 192, and note on action of bilge keels,
p. 226.)
Heel caused by putting Rudder over. On first putting a
rudder over, the force P has a tendency to cause heeling inwards.
This inward heeling is specially
felt in vessels like destroyers, in
which the rudder area is relatively
large. In a full-sized ship, how-
ever, this inward heeling tendency
is only of short duration, as when
the ship gets on the circle the

FIG. 200. centrifugal force comes into action,
and when, as is usually the case,
the C.G. of the ship is above the centre of pressure of the water
on the outward side (centre of lateral resistance), there is a
couple, as shown
in Fig. 200, tending to heel the ship outwards.
This heeling tendency is resisted by the stability of the ship,
and it can be shown that the vessel will take up an angle of heel 0,
given approximately by
sin
V 2
-*( GM
THE TURNING OF SHIPS. 267
where d is the distance in feet between the C.G. of ship and

the centre of lateral resistance ;
V is speed in knots on the circle ;
R is radius of turning circle in feet ;
GM isthe metacentric height.

The above shows the qualities of a ship which affect the heeling
when on the circle, viz.
(i.) depends directly as the square of the speed ;

It
(ii.) depends inversely as the radius of the turning circle; and

It
(iii.) It depends inversely as the metacentric height.
Thus a ship of high speed and small GM, turning in a small

circle, might possibly heel to a considerable angle, sufficient to
prevent the guns being laid horizontal on the inner side.
In the case of Yashima, 1 a Japanese battle-ship, the outward
heel at full speed was 8f at 10 knots only 2.
, This ship had a
very large rudder, and turned in a very small circle. On first
putting the rudder over there was an inward heel, but when on
the circle the inward heel, due to the pressure on the rudder, was
overcome by the outward heel, due to the centrifugal force.
In destroyers, where the distance of the C.G. from the centre
of lateral resistance is not great, the outward heeling tendency due
to centrifugal force on the circle may be overcome by the inward
heeling due to the rudder pressure. If the helm in such a ship

were suddenly righted, the inward heeling tendency, due to the
rudder, would be suddenly withdrawn, and the ship might give a
dangerous lurch outwards. Under these circumstances the rudder
should be righted gradually, so that as the rudder pressure is with-
drawn the ship may come off the circular path. The above is one
of the reasons for giving destroyers a relatively large metacentric
height, in order to provide for their safety when manoeuvring.

Pivoting Point and Drift Angle. In a ship turning in a
circular arc the centre line of the ship points inside the circle, so
that the thrust of the propellers is delivered in a direction oblique
tothe motion of the ship. This, together with the drag of the
rudder,is the reason of the reduction of speed always experienced
when turning. (In a recent case the speed was reduced from 17 J
knots,on putting the rudder over, to 8 knots when on the circle.)
If, any instant, the ship is as shown in Fig. 201, GiGG2 being
at
the path of the C.G. and being the centre of the path, then GT
being the tangent to the path at G, the angle PGT is the drift
1
See a paper by Mr. (now Sir) Philip Watts, F.R.S. (I.N.A., 1898).
268 WAR-SHIPS.
angle at the point G. At the point P, where OP is drawn perpen-

dicular to the centre line of the ship, there is no drift angle, as the
tangent to the circle through P is the centre line of the ship.

The motion of any point in the ship is instantaneously in the
direction of the tangent to the circle that point is turning in. At
the point P this tangent is the centre line of the ship. At the
point 6, for instance, the motion in the direction "be may be resolved
into its components, Id in the direction of the keel, and le in an
athwartship direction. All points abaft P therefore will, relative
to P, move to port, and allpoints forward of P will move to star-
board in Fig. 201. This is why, to an observer on board, the ship
appears to be turning about the pivoting point P.
Path of Ship when turning. When the rudder of a ship is
put over, the ship commences to turn in a spiral path, as Fig.

202. By the time she has gone through eight points the path is
approximately circular. The distance from the position at which
the helm is put over to the position when she is at right angles to
her original course is termed the advance, and the distance from
the original course to the position of the ship when she has turned
through sixteen points is termed the tactical diameter}- The path
1
The tactical diameter is frequently measured as in Fig. 202, but strictly this
should be as defined above, and shown in Fig. 202A. In ships turning in a small
circle there is an appreciable difference in the two distances.
swept out by the stern will have a greater diameter than this.
This must he allowed for when considering the room a ship can
turn in (see Fig. 202 A).
The features of a ship which influence the turning are
1. Time of putting the helm over.
2. Angle of helm.
3. Size of rudder.
4. Moment of resistance of underwater body of ship to turning.
5. Moment of inertia of the vessel.
1. In modern ships with steam steering gear the time of
putting the rudder hard over is a matter of seconds only. The

shape of the rudder is of importance in this connection ; a rudder
that is balanced has the centre of pressure close to the axis, and
offers a small resistance only to being put over. Such a rudder
can be got over more quickly than one hinged on the fore side
(see rudders
in Figs. 71 to 76). The more quickly the rudder is
its turning effect comes into operation.
put over the sooner This
is illustratedby the paths on turning of Orlando and Astrcea (Fig.
202). Although the former ship has a smaller tactical diameter,
yet it is longer in getting into the circle because of the fact that
the rudder was not balanced. If there is a difference in the time
of getting the rudder over at high speeds as compared with low
speeds, the tactical diameter at the higher speeds will be greater

than at the lower speeds. Usually, however, with steam steering
gear, the path on turning is practically constant for all speeds.
The usual maximum angle of helm in ships of the Eoyal
2.
Navy is 35. The tactical diameter will vary approximately

inversely as the angle of helm, so that a vessel may be made to
turn in a path greater than that with the maximum helm angle by
using a smaller helm angle. Ships of different type may thus be
made move through similar arcs by determining the helm
to
angles beforehand by experiment.

3. The size of the rudder has a direct influence on turning,
because the pressure P depends directly on the rudder area. This

area is expressed as a fraction of the area of the immersed middle
line plane of the ship. For large ships in the Navy this ratio
is from ^
to g\j. Eecent battle-ships and cruisers have a ratio of
5^, about. This, however, has been considerably exceeded in the
Dreadnought with twin rudders. The Yashima mentioned above
has the ratio ^. In a typical destroyer the ratio was ^.
4. The resistance the ship offers in turning depends on the
270 WAR-SHIPS.
"
shape of the underwater body and the position of the pivoting
point." This pivoting point is usually forward of amidships
'"
w. &~-A*$*6eA 38O FEET P,F?
;R
'
" '
.'...--.y
_Wj ;
. 202.
when going ahead in some ships it is right up at the bow. The

;
moment of resistance offered by a portion of the surface will vary

roughly as the cube of distance from the pivoting point, and as
its
the cosine of the apgle it makes with the vertical. The flat
portions of a ship at the after end are therefore best adapted for
offering effective resistance to turning, and on this account the
flatportions at the stern of recent ships are cut away in order to
improve the turning (see Figs. 72 to 76 for examples). In large
cruisers the stern is cut right up, with an underhung balanced
rudder. In the fourteen battle-ships of Formidable and Duncan classes
the stern is cut away as shown in Fig. 75, being brought down at
the sternpost to take the blocks when docking. In King Edward
VII. (Fig. 76) this cut away is associated with a partially balanced
rudder. The cut up at the bow
is of little value in influencing
the turning when going

ahead, but has some influence when going
astern. A
ship trimming by the stern more than usual will have
a larger tactical diameter in that condition, and the converse will
be the case if she trims more by the head. A
short ship will turn
more readily than a long ship, on account of the less resistance
offered to the turning.

5. A ship with heavy weights at the extremities will turn
more slowly than a ship of the same size and weight, etc., with
the weight concentrated more amidships, and when once turning
will be more difficult to get back to the straight again. This is due
to the greater moment of inertia of the ship about a vertical axis
in the former case.
Suppose two balls each weighing 1 Ib. are fastened on a stick 12 in. apart,
and two other balls of the same weight are fastened on a stick 60 in. apart. It
is readily seen that the latter system is more difficult to start rotating about an
axis in the middle perpendicular to the stick than the former, and when once
in motion will be more difficult to stop. This is due to the different moments of
2
inertia,. In the first case it is roughly 2[1 x (^) ] (i.e. weight multiplied by
in the second it is roughly 2[lx(f) 2 ], the ratio being
square of distance),
(D :
(), r 1 ' 25 -
comparison between the turning circles of Orlando,

The
Astrcea, and Arrogant (Fig. 202) illustrates the above principles
very clearly. The Orlando is 300 ft. long,
with a rectangular
rudder. The Astrcea is 320 ft. long, with a balanced rudder and
no cut up at the stern. The Arrogant is also 320 ft. long, but has
two balanced rudders with considerable cut up at the stern.
The Orlando, although shorter than the Astrcea, does not get
272 WAR- SHIPS.
into the circular path so soon as the Astrcea, on account of
the type of rudder, which is not balanced. She turns, how-
ever, in a smaller circle than the Astrcea on account of the lesser
length.
The difference between the Astrcea and the Arrogant, both of
the same length, is very marked.
The advance of Arrogant is 350 yards, of Astrcea, 440 yards.
The tactical diameter of Arrogant is 380 yards, or 3'6 lengths.
Astrcea is 650 yards, or 61 lengths.
This great difference is due to two causes, viz. the double
rudders of Arrogant and the large cut up at the stern of that
ship. The ships of the Arrogant class were specially designed as
fleet cruisers, and this great turning facility was made a feature
of the design.
The following comparison between the turning of the Diadem
and Cressy illustrates also the influence of the shape of the stern.
The Diadem is 435 ft. x 69 ft. X 25 ft. x 11,000 tons, with a
stern like Edgar in Fig. 71. The Cressy is 440 ft. x 69 J ft.
X 26 ft. X
12,000 tons, with a stern shaped as shown in Fig. 73.
The rudder is rather larger than in
of the Cressy
Diadem, but
the ratio of rudder area to immersed middle line plane is the same
in both cases.
of the turning trials of Topaze and Amethyst. These were cruisers

of similar dimensions, viz. 360 ft. length and 40 ft. beam, mean
draught, 3000 tons displacement. The shapes of the sterns were

similar, except that in the Amethyst there was a middle line
propeller immediately in front of the rudder, the Topaze being
YARDS.
FIG. 202A.
a twin screw vessel. The rudders of the two ships were of the
same balanced type and of much the same area.
The turning circles are given in Fig. 20 2 A, from which it is seen
that the twin screw ship has a tactical diameter of 870 yards and
the triple screw ship has a tactical diameter of 550 yards or 7*25
and 4' 6 times the length respectively. It is also seen that the
latter ship gets into the circular path much sooner than the former
T
274 WAR-SHIPS.
ship. All the conditions are practically identical except that the
ship with the smaller circle has a propeller operating immediately
on the rudder.
Turning of a Twin Screw Ship. The above discussion deals
with the turning of ships under the action of the rudder alone. A
twin screw vessel, however, may be made to turn in a smaller arc
by the use of its screws in association with the rudder. The
engine on the side to which the helm is put would be worked
ahead, and the other worked astern. This power of turning in the
smallest possible circle may be of great value in special circum-
stances to avoid collision. It is found that the advance and
tactical diameter with one screw ahead and one astern are some-
what less than with both screws ahead. The time taken is
generally greater.
Twin screw
vessels have a great advantage over single screw
ships because of the possibility of steering by the screws alone
by varying the revolutions. Several battle-ships have gone long
voyages without a rudder at all, the steering being done by the

twin screws.
Turning Trials. Systematic turning trials are carried out on
allH.M. ships, and a record is kept in the ship's book for the
information of those officers who have subsequently to navigate
the ship. There are two sets of trials the first those carried out
;
during the official steam trials of the ship when she is in the
builder's hands, and secondly a series of turning trials at 12 knots
and 6 knots, carried out when the ship is in commission.
1. Trials before Commission. These trials determine the time
of turning through 4 and 8 points, the advance and the tactical
diameter under the following condition, viz. :
Both engines full natural draught power.

Both engines 12 knots.
One engine ahead and one astern, 12 knots.
In each case the ship is turned to starboard and port.
The following is the method adopted in the above trials to
determine the path of a vessel when turning. Two points are
selected near the ends of the ship, the distance apart being known,
and at each of these positions a graduated horizontal circle is set
up with a pointer movable in a horizontal plane, having sights
which can be kept bearing on any given object as the ship swings
round. Two weighted rafts with a flag attachment are dropped
overboard, about a mile or so apart. It is assumed that these
remain practically stationary. The ship is brought up to one of

these rafts, at the speed desired, so as to pass the raft as near as
can be judged at the distance of the radius of the turning circle
expected. Shortly before coming broadside on a signal is made,
when the course is noted, the time is taken, and the rudder is put
hard over. At the signal the angles OAB, OBA
(Fig. 202B), are
recorded. These angles with the known distance, AB, fully
determines the triangle, OAB, and consequently the position of
the ship relative to the raft. A similar signal is made after
TACT CAU
i Dl AM ETER
Fro. 202B. Construction of path of turning.
turning four points and corresponding observations taken. This is

repeated at every four points, until the ship has turned the
complete circle. The nine triangles are then set out on a
convenient scale as Fig. 202s, and the path of the ship drawn in.
The next circle is taken round the other raft, in approaching which
the vessel is brought on a straight course and so on until the
whole series is completed.
Knowing the interval of time between each position of the
ship a calculation can be made of the speed, which rapidly drops

27 6 WAR-SHIPS.
owing to the causes already referred to. In a recent vessel with

double rudders starting at 24} knots, the speeds at each of the
four-point intervals worked out as follows 19-6, 17-8, 16-3, 14-4,
:
12-3, 107, 10-3, 10-3 knots respectively, After turning sixteen

points, with the course reversed, the speed had dropped to
i.e.
nearly one half the speed corresponding to the revolutions at

which the engines were running.
The following is the record as obtained for a battleship :
The object of (5) and (c) of the last section is to ascertain

whether the ship can best avoid an object right ahead (as shallow
water or another ship) by reversing with all steam at command,
or by turning with both screws ahead, or with one screw reversed
as in section (1),
down that each section should be completed in a day,
It is laid
and possible all the four sections should be undertaken with
if
the ship in similar conditions of trim, in similar weather, and in

water over 20 fathoms deep. Full instructions how to proceed
with the trials are contained in the form No. S. 347.
In using a range-finder for getting the distances of the buoy
from the ship, notice must be taken of the lower limit of the
range-finder, so as not to go too close to the buoy.
CHAPTER XXI.
THE RESISTANCE AND PROPULSION OF SHIPS.
Resistance. The resistance opposed to a ship when moving

through water is much more complex than the resistance offered
to the motion of a train, say. In first considering the subject, we
must leave out of account the disturbance caused by the propelling
agent, usually the screw propeller, and imagine that the ship is
towed through the water by some other ship. This has actually
been done by experimenters on the subject, the most notable series
of experiments being those carried out by Mr. W. Froude on
H.M.S. Greyhound in 1871. Mr. Froude had the ship towed by
H.M.S. Active, as in Fig. 203, to avoid any disturbance due to the
wake behind the latter ship. The tow-rope was connected on the
FIG. 203.
Greyhound to a dynamometer, to register the strain, and it was

which was overcoming instant by instant the resistance
this strain
offered by the water to the onward motion of the Greyhound.
The experiments were carried out over a wide range of speed, and
as a result Mr. Froude had a record of resistances at various
speeds. When such a record as this is obtained, it is convenient

to represent it graphically by drawing a base to represent speeds,
and erect ordinates to represent the resistances. The spots thus
obtained enable a curve to be drawn showing resistance on a speed
base. The curve obtained for the Greyhound is shown by AA
in Fig. 204, and it is very suggestive. We notice that the
278
THE RESISTANCE AND PROPULSION OF SHIPS. 279
resistance does not increase regularly as the speed increases, but

the rate of increase is much more rapid at high speeds than at
low speeds. Thus to increase the speed from 7 to 8 knots an
extra resistance of 1500 Ibs. has to be overcome, whereas to in-
crease the speed from 11 to 12 knots an extra resistance of 6000
Ibs. has to be overcome, or four times as much for an increment
of 1 knot.
This agrees with our experience. We know how much more
difficult it is to increasethe speed of a ship by a knot, say, near
the top speed than at the lower speeds. Fig. 205 shows the curve
of I.H.P. on base of speed of H.M.S. Drake, and the following
FIG. 204.
shows in tabular form the increase of power necessary for each

2 knots from 10 to 24 knots :
Speed in knots . .
280 WAR-SHIPS.
horse-power. This great increase of power necessary for high

speeds is due to the great increase of resistance.
The rate at which resistance increases as speed increases is
therefore a matter of great importance. Mr. Froude found in the

Greyhound that up to 8 knots the resistance was varying as the
30.000 I.H.P.
25000 l.H.R
20.000 l.H.R
15,000 l.H.R
. o;
10,000 I.H.P 100 REV
no**..
S.eo'o l.H.R' BORtv
~.SPEED IN KNOTS
Fio. 205.~
square of the speed. That is to say, if Bx is the resistance at

speed Yj and K2 is the resistance at speed 2, then
2
K, :
E, : :
(V,) :
(V2 )<>
Mr. Froude found, as the curve in Fig. 204 indicates, that the
resistance variedmuch faster than the square of the speed at the

higher speeds. Indeed, at 12 knots it was varying as the fourth
power of the speed, a very high rate of increase. In the case of

the Drake the resistance is varying nearly as the third power of
the speed between 23 and 24 knots.
Effective Horse- power. If we know the resistance of a ship
which being towed at any given speed, we can determino the
is
1
horse-power that is being transmitted through the tow-rope to
overcome this resistance. the effective horse -power, some-
This is
times called the tow-rope horse-power. This horse-power is a
very different thing" from the power exerted by the vessel's own
engines or the indicated horse-power (I.H.P.).
In any general case, if B is the resistance in pounds, the V
speed in knots (1 knot is a speed of 6080 feet per hour), then-
Work done per minute =R X -- --

(
^ J
foot-lbs.
/VX6080N
, ,
and horse-power
H
--- 60 )
x V).
This is the effective horse-power (E.H.P.). Mr. Froude was

thus able at once to turn the resistance of the Greyhound at any
speed into E.H.P. He found a striking difference between the
E.H.P. thus obtained and the I.H.P. which had to be exerted by
the vessel's own engines in order to get similar speeds. Thus at
10 knots the E.H.P. worked out to 380, and for this speed the
I.H.P. necessary was 786, giving a ratio of E.H.P. -f- I.H.P. of
|!-g-
= 0*42. This was a most important result, showing that of
the power exerted at the vessel's own engines the large proportion
of 58 per cent, was wasted so far as the ship was concerned.
Mr. Froude, on arriving at these results, was led to make further
investigations in order to look into the cause of this great loss
of power.
The usual value of this ratio in vessels of the Eoyal Navy is
from 45 to 50 per cent., rising to55 per cent., or higher in some
1
The work done by a force (like the strain in the tow-rope), acting through a
certain distance, is given by the product of the force and the distance through
which it acts. This isindependent of the time taken. The power exerted takes
into account the time in which the work is done. The unit of power used is the
horse-power, which ia defined as 33,000 foot Ibs. of work performed in one minute.
2 g2 WAR-SHIPS.
few cases. This ratio is termed the propulsive coefficient, being the
ratio of the horse-power usefully employed to the horse-power
actually exerted.
We have been considering above the total resistance experienced
by a ship on being towed through water, it is necessary now to
inquire how this total resistance is made up. It may be divided
into four parts, viz.
1. Eesistance due to the friction of water on the surface.
2. Eesistance due to the formation of eddies.
3. Eesistance due to the formation of waves on the surface.
4. Eesistance of the air.
Frictional Resistance. This can be directly calculated

1.
from data obtained from a series of experiments made by Mr.

Froude on boards coated with various surfaces. For an ordinary
smooth surface like a vessel coated with paint the resistance
in pounds is given by
E=/.S.V r83
where S is wetted surface in square feet ;
V is speed in knots ;
/ is a coefficient.
This coefficient varies according to the length of the surface, being

greater for short than for long surfaces. For ships its value does
not vary much from 0*009. For short models such as are used in
the experimental tank the value of / is greater. notice that We
this resistance for a smooth surface varies at a rather lower rate
than the square, viz. 1'83. If, however, the surface is rough, like
sand, the coefficient / is twice as great, and the power of the speed
rises from T83 to 2*0. This illustrates the fact that in order to
keep the resistance as low as possible, and so economize horse-
power, and therefore coals, it is necessary to keep the bottom
clean by periodical docking. On this account, also, ships which
are employed on distant service, with the probability of remaining
at sea for long periods, are sheathed with copper to prevent fouling.
The frictional resistance is of importance at all speeds, but at
low speeds it accounts for the bulk of the resistance. For a
torpedo-boat destroyer, which has an abnormally wide range of
speed, at 12 knots the frictional resistance is 80 per cent, of the
total; at16 knots, 70 per cent. at 20 knots, 50 per cent. and at
; ;
30 knots, 45 per cent.

2. Eddy Resistance. This is due to the eddies formed behind
a blunt ending to the underwater body. Ships built as formerly,

with very full sterns and thick sternposts, experienced this
resistance to a large extent, but in modern ships of finer form it
does not exist to any appreciable degree. Every care is taken to
avoid any abrupt terminations which might cause eddy making.
One instance of this is seen in the shaft brackets the section of
;
the arms is madeas in Fig. 82, taken to a small radius at the after
end, so that no eddies are caused at the rear.
3. Wave Resistance.
It is this form of resistance which
becomes of the greatest importance at high speeds, and it is because
of the rapid growth of this resistance that it becomes increasingly
difficult toobtain these high speeds in full-sized ships.
When is towed through water there are two
a ship separate
and distinct series of waves brought into existence, viz. those formed
at the bow and those formed at the stern. Each of these series
consists of (a) a series that diverge with their crests sloping aft,
and (b) a series of transverse waves whose crests are nearly
perpendicular to the middle line of the ship.
The diverging waves both at the bow and the stern at once
pass away from the ship. The transverse waves of the bow series
are of the most importance, and the interference of these waves
with the corresponding waves at the stern causes considerable
variation of the resistance. If a crest of the bow wave series
coincides with a crest of the stern wave series there is an increase
of wave resistance. A decrease
is found to result if a crest of the
bow wave with a trough of the stern series.

series coincides
When a ship maintains a steady speed, say 15 knots, the
accompanying series of transverse waves also has a speed of 15
knots. Such a series has a definite length, viz. 126 ft. from crest
to crest, or trough to trough. At 10 knots the length would be
56 ft. It is thus possible to make an estimate of the speed at
which a ship is travelling by observing the length from crest to crest
of the wave along the side of the vessel. (If V be the speed in knots
and L the length of wave in feet, then V =
1'33\/L.) As speeds
therefore increase, the accompanying wave system gets longer,
increasing in length as the square of the speed. In the case of

small vessels, like destroyers, travelling at the high speed of
30 knots, say, a wave is created longer than the ship, and she lies
on the back slope of a wave of her own creation (see Fig. 207B).
We thus see that as speeds increase, the length and height of
the waves formed must also increase very rapidly, and consequently
28 4 WAR-SHIPS.
the energy required to maintain them. The resistance thus caused

varies at a higher rate of the speed than the square, and it is not
calculation its amount for any
possible to determine beforehand by
given case. We
have to rely on the results of model experiments,
or on the trials of ships of similar form, when estimating the power
necessary for a new design of high speed.

4. Air Resistance. This is a subject about which our
knowledge is very scanty. It became of less importance than
formerly with the abolition of sails, but in the present high-speed
ships it must be of appreciable amount. If a ship is steaming
against the wind, the relative velocity is the speed of wind plus
speed of ship, and this gives rise to considerable resistance.
Corresponding Speeds. We
continually use the terms high
speed and low speed, as applied to certain ships, but these terms
are strictly relative. What would be a high speed for one ship
might very well be a low speed for another. Thus 15 knots is a
high speed for a ship 150 ft. long, but quite moderate for a ship
500 ft. long. To obtain a real measure of speed in any case we
y
find its ratio to the square root of the length, viz. ?-=.
VL
y =
where T~ 0'5 to 0*7 the ship is being driven at a moderate
economical speed ;
y =O f
TO we have
j=^
7 to the speed of mail steamers
\/L
and battle-ships ;
Y= =
1/0 to 1*3 we have cruiser speeds.
Beyond this we cannot go

in full- sized vessels under present con-
ditions, because not possible to get enough engine and boiler
it is
power into the ships. It can be done, however, in destroyers by

using very fast-running engines and forced draught to boilers, and
in these ships
Y= = T9 to 2*5. Such speeds as these are exces-
\/ L
sive, and require a great expenditure of horse-power to obtain.
"
When these excessive speeds are reached, although the horse-power required
isvery great, yet the resistance does not vary at so great a rate as is the case at
lower speeds. The following figures show how the total resistance varies in a
typical destroyer :
Up to 11 knots it varies as the 2nd power nearly.

At 16 3rd
From 18-20 33rd
At 22 2-7th
At 25 2nd
At 30 2nd
If we took the destroyer as a model, and took a vessel of the same form but
14,100 tons displacement, 25 knots of the destroyer would correspond to 47 knots
of the larger vessel, and this vessel would not reach the condition where further
increments of speed are obtained with comparatively moderate additions of
power until she reached 47 knots, which is an impossible speed for such a vessel
under existing conditions" (Sir W. H. White's Address to British Association,
1899).
As speeds have continually increased in the mercantile marine,

the lengths have also shown a corresponding increase, one result
which
V within moderate limits. In
of is to keep the ratio
vL
^=-
cruisers, however, this continual increase of length has been un-

desirable as speeds have increased and the ships are only of
40,000
80.000
gftOOO
10,000
Z. 23. 24.
-KNOTS.
FIG. 206.
I.H.P. of ships of 9000 tons displacempnt, but of varying lengths.
"
(From Sir Philip Watts's article in Encyclopaedia Brttannica")
sufficientlength to obtain the qualities desired, and the extra

power rendered necessary by limiting the length has had to be
accepted. If lengths had been increased the economy of propulsion
obtained would have been discounted by losses and disadvantages
1
in other directions.
1
See also Chapter XXII.
286 WAR-SHIPS.
The curves in Fig. 206 illustrate the importance of length.

The diagram represents approximate curves of horse-power on
base of speed for vessels of 9000 tons displacement of varying
lengths, each having a coefficient of fineness of 50 per cent. We
see from this the
excessive horse-power required for the short
ship as compared with the long ship of the same displacement
when high speeds are reached. Indeed, a speed of 25 knots
would be impossible for the 350 and 400 ft.
ship, because of the
excessive horse-power necessary.
The following table illustrates very forcibly the value of
length when speeds are being reached which are high for the
ship :
Speed in knots . . .
If we are designing a ship 500 ft. long of similar form, the speed cor-
responding to the 20 knots of the 400-ft ship is

20^/^9 = 22'36
knots.
Froude's Law of Comparison. The resistances other than
frictional of similar ships, or of a ship and her model at corre-
sponding speeds, are connected by a most important law, called
the law of comparison, viz.
If the linear dimensions of a vessel le I times those of the model

(or model ship), and the of the latter at speeds FI, 2 , F"8,
resistances, F
etc., are RI, R& R3) etc., then at the corresponding speeds of the ship,
Vi\/l> V*\fi> PffvA e te-> tfie resistances of the ship mil le Rf,
RJP, Ef, etc.
The law of comparison would apply to the frictional resistance

varied as the square of the speed, and if the coefficient were
if this
the same for long and short surfaces. Neither of these conditions
are actually fulfilled, as we have seen, so the law cannot strictly be
applied to frictional resistance.
In Mr. Froude's experiments on the Greyhound, mentioned
above, experiments were also made with a model of the ship at
the experimental tank to determine the resistance. It was found
that, deducting the frictional resistance of the ship and the model
from the total in either case, the
remaining resistance of the ship
compared with that of the model in accordance with the law of
comparison as above stated. That is, the length of the model
being one-sixteenth that of the ship, at speeds of the ship \/16, or

four times that of the model, the resistances of the ship other
8
than frictional were practically 16 or 4096 times that of the ,
model.
It is the established practice of the British Admiralty to have
models made and run in the experimental tank in order to deter-
mine the resistance of the ships of the Koyal Navy, and most
valuable data is obtained in this way to determine the power
necessary when considering new designs. Alternative forms can
also be readily tried in order to determine the best possible form
of underwater body for the desired speed. It may be mentioned
that this best form cannot always be adopted, because of the
conditions of stability which must necessarily be satisfied.
have seen that E.H.P. = ^J^(R x V), so that we can use
We
E.H.P. in the law of comparison instead of resistance as follows :
If EI, 2 E
E%, etc., are the effective horse-powers (other than
,
288 WAR-SHIPS.
F"2 , F"3 etc., of the model (or model

frictional) at speeds Fi, ,
ship), then
the E.H.P. of the ship whose linear dimensions are I times those of
the model, at the corresponding speeds V\\/l, V^l, V^/l } etc., are
3 3 B
(JSi X Z X \/T), (E2 x
I X -x/0, (^3 X 1 X >/0, 6fo
Methods of Estimating I.H.P. There are a number of
methods in vogue for estimating I.H.P. the following are in use ;
when an experimental tank is available :
A model in paraffin wax is made at the tank and is towed at

the series of speeds corresponding to those of the ship. From the
resistances thus obtained a curve of E.H.P. on base of speeds can
be drawn for the ship by using the law of comparison, after
making the proper corrections for frictional resistance.
To convert the E.H.P. thus obtained into I.H.P., data obtained
Tj1
TT T)
as to the value of the ratio or propulsive coefficient,

j- ^p ',
in previous ships has to be used. As already stated, the usual

value of this in ships of the Eoyal Navy is from 45 to 50 per cent.
The value of the propulsive coefficient will not be constant for the
same ship for all speeds, because the propellers are designed to be
most efficient at the top speed, and the friction of the engines is
relatively lesswhen working at the highest revolutions.
In the early stages of the working out of a design, when it is
not in a sufficiently settled state to have a model run, preliminary
estimates may be made of power by using the information obtained
from previous models or from the trial data of actual ships. The
latter would have to be adopted if an experimental tank was not
available.
The following example will illustrate how one would utilize the
data already obtained from model experiments to make a prelimi-
nary estimate of power :
it is desired to design a
Suppose battle-ship of 12,000 tons of 20 knots speed.
To get this speed one would need a fine form ; a good breadth is necessary for
stability purposes, and a moderate length is desirable for handiness. have We
data concerning the E.H.P. of a vessel 320 ft. x 57 ft. x x 5150 tons, and
19-5 ft.
this appears to have a form that is desirable. We therefore have the following
calculations :
Ratio of displacement
= 3/2W=
= Z
3 = ^^ = M5. 2-33,
so that I 1-325, and jr=
The dimensions of the 12,000-ton ship would therefore be
Length (320 x 1-325), breadth (57 x 1-325), draught (19-5 x 1-325),

or, 425 ft. x 75J ft. x 26 ft. x 12,000 tons, say.
The speed of the smaller ship corresponding to 20 knots of the 12,000-ton
ship is = 17-4 knots. At this speed the E.H.P. of the 5150-ton ship is
-|^r
found to be 2960, so that the E.H.P. of the larger ship at 20 knots is
2960 x 2-33 x 115 = 7940.

If we assume a propulsive coefficient of 47^ per cent,, the I.H.P. would be
7940 x
We
^ = 16,750 I.H.P.
therefore have a ship
425 ft. x 75| ft. x 26 ft. x 12,000 tons 16,750 I.H.P., 20 knots.
;
(In the above calculation we have made no correction for frictional resistance,
assuming it to vary according to the law of comparison. This gives the result
for the larger ship a little in excess, so that we are somewhat on the safe side.)
The following example will illustrate how we have to proceed

when using the trials of a previous ship as data :
" "
ff.M.S. Edgar is 360 ft. x 60 ft. x 23f ft. x 7390 tons, and when tried
obtained 10, 14, 18, 20 knots with 1000, 8000, 7500, 11,000 I.H.P. ship 'of A
similar form, 11,000 tons, is being designed^ and it is desired to have a speed of
21 knots. Make an
estimate of the power required.
(11 000^
The ratio of displacement is Z3 =
/.room = 1*49, so that ratio of linear
dimensions is
/(l'49)
= M4.
The new ship similar to Edgar is accordingly
410 ft. x 68i ft. x 27 ft. x 11,000 tons.
The ratio of corresponding speeds is Jl'14 = 1'07, so that to 21 knots of

/o-l \
new = 19'65 knots of Edgar. By putting curve in to

ship corresponds TT^
scale we find that 10,400 I.H.P. is required for Edgar's 19'65 knots.
For the new ship's 21 knots we therefore require 10,400 x 1*49 x 1'07 =
16,600 I.H.P.
We tacitly assume that we shall obtain the same efficiency of propulsion as
was obtained model ship.
in the
Components of I.H.P. We have seen above that of the

power actually exerted at the engines of a vessel, only about one-
half is usefully employed in overcoming the resistance experienced
by the ship passing through the water. It will be of interest to
note briefly the reasons for this difference.
The thrust necessary to overcome the resistance of a ship is
developed by the projection stern ward by the propeller of a column
of water, and the reaction constitutes the forward thrust which is
transmitted to the ship at the thrust block. vessel moving at A
speed t say,Vis accompanied by a belt of water drawn along by
the friction of the ship's surface. This forward moving water is
called the frictional wake. The motion of this wake at the stern is
complex and variable, and gradually disappears as we go away from

U
-
- - : : .
.
of restance in both
'
Ali
HUi
THE RESISTS "F> PROPULSION OF SHIPS. 291
n the above we see tat 15 per cent, of the

expended power is
Be machinery beforeit reaches the propeller.

This will be
leded at low speeds because then the dead load friction is
ly of greater irnprtance. Also, seeing that the propeller
signed to be most eicient at the highest speed, it is evident
i these two causes tit the propulsive coefficient will be less
he lower speeds thaat the higher speeds.

Progressive Spec Trials. The trials now required to
contract conditins of ships of the Koyal Navy are to
ne power and nc speed. In the design of a ship a certain
er is specified, andif this power is obtained for the given
th of time and uner the specified conditions, the failure of
aip to obtain the peed would be the responsibility of the

gner and not of thtcontractor.
[ u order to determie the real speed and to obtain data for future
itis necessary to an trials on the measured mile or over a
**se of known is not always done with

Icngh. This every
>,bat usually with one ship of each class, and for this
t least
>a series of progrenve trials is run at varying speeds in order
>btain a completecurve of power on base of speed, as e.g.
I.S. Drake (Fig. 20). For these speed trials there can obviously
mly one standard/iz. with the highest possible efficiency. The

om should be ckn, the weather favourable, the engines and
ers in perfect ordc, the best coal and most skilful supervision
stoking. If any f these conditions are not satisfied, then we

have no scientifi knowledge of the performance, as the in-
nee of the various factors affecting the ship is unknown. It
owever, not unusil to find in the history of a ship, subsequent
ier trials, that, wbn the engines have got into perfect working
?r,the ship ha done

j&f than even whe tried *
t er the conditions above .
6
itioned.
In running spec trials
icasured mile corse is
ally used, on whii, after

jful survey, twoâirs of
ts, AB and O, have F, o 07
n placed, as Jg. 207,

MHT a nautical nle (6080 ft.) apart. Two buoys mark a course
right angles to le posts, as shown (or the course is on a given
290 WAR-SHIPS.
the ship's surface, but we may regard it as equivalent to a uniform

current of water having a forward velocity n V, say. The pro-
peller, therefore, is not working
in still water and passing through
the water with speed F, but is working in water having a speed
relative to the ship of Vn V. The propeller derives increased
thrust from this cause, and the gain
will be the greater for a single
screw ship than for a twin screw ship, because the frictional wake
is of greatest influence near the middle line, where the single screw
is placed.
The resistance caused by the deflection of water at the forward
part of a ship is nearly all returned to the ship in the forward

thrust obtained by the closing in of the water at the stern. (In
the ideal case of a body moving wholly immersed in a frictionless
fluid, the net resistance in the direction of motion is zero.) Any-
thing, therefore, which interferes with the natural closing in of the
stream lines at the stern will be a cause of resistance. The presence
of the propeller at the stern of a ship interferes in this way, and
is a distinct cause of resistance. This augment of resistance, as it
is termed, will be greater for a single screw ship than for a twin
screw ship, because in the latter case the screws are further away
from the hull.
We notice that in a single screw ship, although the gain due
to the frictional wake is greater than in a twin screw ship, yet the
loss due to augment of resistance is also greater, and it is found
that under ordinary circumstances the gain due to wake is practi-
cally equal to the loss due to the augment of resistance in both
single and twin screw ships.
The various items which make up the I.H.P. may be stated
approximately as follows, in the case of a twin screw vessel with
fast-running piston engines of high pressure :
Per cent
Dead load friction, due to dead weight of working
parts, friction of packings, bearings, etc. ; approximately

constant at all speeds 7
and speed of engines .......

Working load friction, varying with thrust of propeller
Working of air pump off the main engines . .

7
1
Loss at propeller, due to slip, friction of blades and
augmentation of resistance, allowing for the gain due to
the frictional wake 33
Effective horse-power expended in overcoming the net
resistance of the ship to the onward motion . . .52
100
In the above we see that 15 per cent, of the power is expended

in the machinery before it reaches the propeller. This will be
exceeded at low speeds, because then the dead load friction is
relatively of greater importance. Also, seeing that the propeller
is designed to be most efficient at the highest speed, it is evident
from these two causes that the propulsive coefficient will be less
at the lower speeds than at the higher speeds.
Progressive Speed Trials. The trials now required to
satisfy contract conditions of ships of the Eoyal Navy are to

determine power and not speed. In the design of a ship a certain
power is specified, and if this power is obtained for the given
length of time and under the specified conditions, the failure of
a ship to obtain the speed would be the responsibility of the
designer and not of the contractor.
In order to determine the real speed and to obtain data for future
use, it is necessary to run on the measured mile or over a
trials
course of known length. is not always done with

This every
ship, but usually with at least one ship of each class, and for this
ship a series of progressive trials is run at varying speeds in order
to obtain a complete curve of power on base of speed, as e.g.
H.M.S. Drake (Fig. 205). For these speed trials there can obviously
be only one standard, viz. with the highest possible efficiency. The
bottom should be clean, the weather favourable, the engines and
boilers in perfect order, the best coal and most skilful supervision
and stoking. If any of these conditions are not satisfied, then we
can have no scientific knowledge of the performance, as the in-
fluence of the various factors affecting the ship is unknown. It
is, however, not unusual to find in the history of a ship, subsequent

to her trials, that, when the engines have got into perfect working
order, the ship has done

better than even when tried *x
under the conditions above it
mentioned.
In running speed trials
a measured mile course is
usually used, on which, after

careful survey, two pairs of Q^. L . _ SHIP*.
posts, AB and CD, have Fio. 207.

been placed, as Fig. 207,
exactly a nautical mile (6080 ft.) apart. Two buoys mark a course
at right angles to the posts, a shown (or the course is on a
given
294 WAR-SHIPS.
near the Clyde, where the depth of water

trials is at Skelinorlie,
isover 40 fathoms, and the course is sheltered. This course is
used whenever possible.
Figs. 207A and 20?B are results obtained from special trials
carried out on H.M.S. Cossack, a torpedo-boat destroyer, 270 ft.
long, 26 beam, and 836 tons displacement. In deep water at
ft.
Skelmorlie the curve of power on base of speed is shown in Fig.

296 WAR-SHIPS.
speed obtained without increase of power.

is At 26 knots the
power the
is same, and beyond this speed the boat gets a better
in shallow water than in deep water for the same power,
speed
and at 34 knots the power required on the Maplins is 11 per cent,
less than that required at Skelmorlie. It will be seen by refer-
ence to Fig. 207s that the stern wave had practically disappeared
and indeed this was the case on the Maplins at speeds
at 35 knots,
above 28 knots. It is thus seen that at the lower speeds the
wave system becomes exaggerated in shallow water leading to an
abnormal increase of power, but a radical change takes place in
the wave formation at the higher speeds involving a considerable
decrease in the necessary power. (The above is taken from a
paper by Sir P. Watts, K.C.B., etc., I.KA. for 1909.)
In running progressive trials it is necessary to have the ship
as near to the normal load draught as possible. When a ship
is deep or light the horse-power required for a given speed will
vary approximately as the displacement.
From the information obtained on such a series of trials a
curve of I.H.P. on base of speed can be constructed, as Fig 205,
and then it can be at once determined what I.H.P. is required for
a given speed, or what speed can be got for a given I.H.P under
trial conditions. The curve of revolutions on base of speed can
also be drawn as shown, and such a curve, showing revolutions
necessary for any desired speed, is found very useful in the subse-
quent history of the ship. Eecords of all trials are kept in the
ship's book for the information of the officers commanding.
Power Trials. The following series of trials are now carried
out in ships of the Royal Navy.
1. Battle-ships, first class cruisers, second class cruisers.
(a) Preliminary trial at sea.
(6) 30-hour trial at about one-fifth full power.
(c) 30-hour trial at about 70 to 75 per cent, full power.
(d) 8-hour trial at full power.
(e) Trial after opening up (24 hours).

2. Third class cruisers.
(a) Preliminary trial at sea.

(b) 30-hour trial at half forced draught power.
(c) 8-hour trial at authorized natural draught power.
(d) 4-hour trial at full forced draught power.
(e) Trial after opening up (24 hours).
It is important to note the conditions under which the
machinery and boilers are designed to exert the maximum power.

It is for a certain limited time only in each case. The Drake, for
instance, is not a 23|-knot ship in the same sense as the Atlantic
liner Kaiser Willielm der Grosse is a 22|-knot ship. The I.H.P.
in each case is about the same, viz. 30,000, but in the latter case
the machinery is designed to maintain the speed across the Atlantic,
and the weight involved is nearly double that for the Drake of
the same power. In fact, such fast liners are practically only able
to carry themselves and the necessary coals across, the amount
of deadweight cargo they can carry being very small. The Drake,
on the other hand, carries a large weight of armour and armament,
but the machinery is only intended to attain 30,000 I.H.P. for a
period not exceeding 8 hours. In a vessel like Pioneer, for instance,
the full power is 7000, but this is obtained by forcing the boilers,
and cannot be maintained for a longer period than 4 hours the ;
authorized natural draught power is 5000, and this is intended for

a period not exceeding 8 hours.
In dealing with ships of the Eoyal Navy the following terms
are in use, viz.
(a) The authorized full power is taken as the unit.

"
(6) With all despatch," four-fifths the unit, for 30 hours (this
is when great urgency is necessary).
"
(c) With despatch," three-fifths the unit ; this should not be
materially exceeded when the period of steaming exceeds 30 hours.
"
(d) With moderate despatch," two-fifths the unit.
Ordinary speed, one-fifth the unit.
(e)
" "
(/) Most economical speed as determined by trial and in
accordance with the varying conditions of wind and weather.
The maximum speed at which the Drake, for instance, could
proceed so long as her coal lasted would be about 21 knots, corre-
sponding to 18,000 I.H.P. The speed at starting, with bunkers
full, would be something less than this, but it would get greater as
the ship lightened.
CHAPTER XXII.
THE DESIGN OF WAR-SHIPS.
THE which a ship is intended to be employed has

service for
manifestly the predominating influence on her design. The duties

which ships of a navy like the British Navy have to perform are
so varied, that no single ship could possibly combine all the
qualities needed in war-ships. Very high speed, heavy armour
protection and powerful armament can only be embodied in one
design if the ship is of very large dimensions. A compromise
is necessarily effected, and if we sacrifice some protection and
guns to obtain high speed and large coal or oil capacity we
have a cruiser; if we have less speed and pay most attention
to protection and armament we have a battle-ship. If we
want a ship that shall be able to keep the sea for long periods
without docking, we must have a vessel sheathed with wood and
copper, and in doing so we have to accept some increase of cost
and decrease in measured-mile speed, as compared with a vessel
with an ordinary steel skin. Again, a vessel intended for coast
defence would need only a moderate coal capacity and a small
draught of water. There are navies in which such a type of ship
would be valuable; the construction, however, of battle-ships of
small size has been discontinued in the British
Navy some years. for
The design of a war-ship would be an almost impossible task
apart from experience and data obtained from previous ships.
When the main features which it is desired to embody in a new
design are given by the authorities, it is the function of the
naval architect to work out such a design as shall satisfactorily
embody those features. Experience in the specialities of war-ship
design a necessary qualification, as the conditions to be satisfied
is
are altogether different from those in the case of merchant steamers.

There are many qualities which to a greater or less extent
must be found in any war-ship design. Some of these are
1. Strength, both structural and local. We have already
discussed this at some length.
2. Stability. This is a vital quality. A war- ship must have
298
THE DESIGN OF WAR-SHIPS. 299
sufficient stability left after sustaining a reasonable amount of
damage. It is on this account that the rnetacentric heights given

to war-ships are greater than obtain in merchant steamers. The
stability at large angles also requires careful consideration, be-
cause of the high position of the O.G. of ship. The question
of the most economical propulsion frequently has to go into the
second place in order to obtain a proper amount of stability.
3. This depends on the intended service of the ship.
Speed.
Speeds have considerably increased during recent years, this
having been rendered possible by the use of watertube boilers,
with high steam pressures and high revolutions, and also the
adoption of turbine machinery.
4. Handiness. The influence of the shape of the stern and
the rudder on turning have already been discussed. A short ship
also is handier than a long ship, other things being the same.
5. Habitdbility. This is important because of the necessity
of keeping the crew in a good state of health. A
high freeboard
ship has a great advantage over a low freeboard ship in this
respect, the living spaces being much more airy and light.
6. Convenient transport of coal and ammunition.
7. Economy of first cost and maintenance. These two things
are sometimes opposed. Thus a steel ship will be cheaper than a
sheathed ship because of the cost of the sheathing and the metal
ste,m, etc., necessary. The cost of maintenance of the sheathed
ship, however, will be considerably less than the steel ship, because
it will not foul so quickly or require such
frequent docking.
8. Length of vitality. The amount of coal, ammunition, etc.,
carried by a ship will determine how long she can remain efficient
as a fighting machine. The coal will determine the radius of action.
9. Slowness of destruction. This includes protection by armour
and decks, and the provision of minute subdivision.
10. Armament. Being the available provision for attack
guns, torpedo equipment, ram.
Of the first stages of a design, Sir William White says
(" Manual of Naval Architecture ")
" Iu the
preliminary stages the processes are necessarily tentative and subject
to correction. The various features of the design are, to a large extent, inter-
dependent. At the outset the dimensions, form, and displacement are undeter-
mined. Yet upon them depend the power which the engines must develop to
give the desired speed, the weight of the hull, and the weight of certain parts of
the equipment. In the finished ship the sum of the weights of the hull structure,
propelling apparatus, equipment, coals, and load must equal the displacement to
3 oo WAR-SHIPS.
the specified load-line. Apart from experience, a problem involving so many

unknown yet related quantities could scarcely be solved. On the basis of
experience, recorded data, and model experiments it is dealt with readily.
Approximate dimensions and forms are first assumed. The weight of hull is
then approximated to for the system of construction adopted and the type of
ship. An estimate of the probable engine power is made, either on data obtained
from the steam trials of previous ships or from model experiments. The weight
of the engines and boilers is then ascertained for the horse-power, and the rate of
coal consumption per hour calculated on the same basis, while the total weight
of coal for the intended steaming distance at the desired speed is readily deduced.
Adding together these first approximations to the weights of hull, equipment,
machinery, and coal, and to the total adding the load stipulated to be carried, a
grand total is reached which should equal the displacement provisionally assumed.
If the sum total is in excess or defect of the provisional displacement, corrections
must be made on the dimensions originally assumed, with a view of obtaining a
balance. For these corrected dimensions a fresh series of approximations is
made to the weights of hull, equipment, machinery, and coal. A balance
between the grand total of weights, and the displacement corresponding to the
form and dimensions, is ultimately obtained. When no large departure from
previous experience or precedent is made, this preliminary work is rapidly
performed. Under other circumstances, the selection of the most suitable
dimensions and form may involve the consideration of many alternatives."
l
The total displacement of a completed design is made up of
the following items, viz.
1. General equipment.
2. Armament.
3. Machinery.
4. Engineer's stores.
5. Coal.
6. Armour and protection to hull.
armament.
7. Hull, including structure and fittings.
8. Board margin.
General equipment.
1. This includes fresh water provisions ;
bread and officers' stores (including ward-room

(including spirits) ;
and gun-room stores and paymaster's slops) ; officers, men, and

effects ;
anchors ; masts, rigging, etc. ; boats ; warrant-
cables ;
officer's stores and net defence.

1
These weights depend largely on
the type of ship and on the complement. The intended service of
the ship has an influence, as the weight of stores allowed would
be greater or less according as the vessel is intended for distant,
isolated service, or with a fleet not far from a dockyard.
2. Armament. The weight of this can be very closely estimated
1
See p. 298 for some specimen weights.
when the detail of the armament is settled. An important point

in connection with this is the number of rounds taken per gun.
Machinery. When the I.H.P. is provisionally settled, an

1
3.
estimate is prepared by the Engineers of the necessary weight.

This will depend on several things, e.g. the type or types of boiler
to be used; the revolutions and stroke of the engines and
the speed of the pistons; the degree to which the boilers are
to be forced. For the largest set of reciprocating engines fitted
(15,000 I.H.P. ), 120 revolutions is the maximum; as engines of
lower power are reached higher revolutions are possible ; thus the
engine for 6250 I.H.P. has 180 revolutions, and for 4900 I.H.P.
has 250 revolutions. Destroyer engines are faster still. The
adoption of watertube boilers has had a great influence on design
in recent years, enabling a larger power to be developed on a given
weight than formerly. The employment of turbine machinery

has had great influence on the conditions of recent design. See
description of Dreadnought in Appendix.
4. Engineer's stores. The allowance of these stores depends on
the intended service of the ship as well as on the power.
Fuel.
5. It is the practice of the designed displacement of
H.M. ships to include a certain weight for fuel. This is called the
legend weight. Thus the Eoyal Sovereign has 900 tons, as also
Majestic, Formidable, Duncan, Lord Nelson and Dreadnought
classes. The total capacity available for coal is considerably more
than being over 2000 tons in most recent battle-ships, and
this,
2700 tons in the Dreadnought. All the official steam trials to test
the speed are carried out at the draught corresponding to the legend
condition. Sometimes, however, trials have been carried out before
the completion of the ship to determine the acceptance of machinery
from the contractors. These trials are simply for I.H.P., and not for
ballasted to her normal
speed, and then the ship is not necessarily
load line, so long as proper immersion is given to the propellers.
6. Armour and deck protection. The weight devoted to pro-
tection be divided into
may
(a) Vertical armour for the protection of the buoyancy and
stability ;
(b) Vertical armour for the protection of the armament ;

and
(c) Deck protection.
The percentage of each of these of the total weight of protection
in a recent battle-ship was as follows (a) 38 per cent. (b) 34 per
:
;
i
See Chapter XXI.
302 WAR-SHIPS.
cent. (c) 28 per cent.

;
This shows clearly how large a proportion
is devoted to the effective protection of the armament. This
includes barbettes and casemates, but not gun shields, which are
taken in the armament. The barbettes especially are well pro-
tected. Thus in the Duncan the side armour is 7 in., but the
barbette armour is 11 in. The reason is that each barbette repre-
sents such a large proportion of the fighting power of the ship. A
single shot piercing the belt might not be a serious matter, but one
shot through the barbette armour would probably cripple nearly
one-half of the ship's fighting capacity.
A feature of modern battle- ship designs has been the larger
area covered with armour than formerly, with a corresponding
reduction of thickness. This has been fully dealt with in Chapter
XIII. The improvements made in the quality of armour has also
had a great influence on cruiser designs. Up to the Diadem the
protection was considered to be best obtained by a thick protective
deck at the waterline, as Figs. 21 and 22. Owing, however, to the
introduction of Krupp armour, this system of protection was modi-
fied by the adoption of a broad patch of 6-in. armour in the Cressy,
over about half the length, in association with thick decks and bow
protection (see Figs. 136 and 137). Armour protection has been
adopted up to the present time for first class cruisers.
7. Hull. The weight devoted to the hull comprises

(a) Weight of the structure and ;
(b) Weight of fittings, etc., not contributing to the structural

strength.
The total weight of hull in large ships varies from 35 to 40 per
cent, of the total displacement, and it is only by a most careful
arrangement of the material, combined with high-class workman-
ship, that the weight can be brought as low as this. The corre-
sponding weight in large merchant steamers is considerably in
excess of the above figure this is very notable when we consider
;
that about one-half of the weight of hull in a war-ship is concerned

with fittings, etc., which do not contribute primarily to the structural
strength.
Economies of weight during building are important to keep
weights down, and a great deal can be done in this direction with
no loss of strength. Lightening holes are largely employed.
The lightened liner, shownin Fig. 47, is an instance of reduction
of weight with no reduction of strength. Such small savings
of weight do not appear to be in themselves of much value,
but in the aggregate they amount to a considerable saving of

weight.
Considerable reductions of weight have been effected in recent
ships by lessening the duplication, etc., formerly fitted. One
instance of this is seen in the abolition of relieving tackles to the
steering gear. The stockless anchors now adopted have enabled
the large weight formerly devoted to catheads, bill-boards, etc.,
to be saved. A
large amount of weight has been saved in
recent shipsby the omission of teak linings and light boxes to
magazines, omission of wood decks, simplification of pumping
arrangements, etc.
8. Board margin. This a weight provided for at the time of

is
the design to cover alterations or additions made during the
progress of the building of the ship. Any weight thus required,

not provided for in the original legend of weights, has to be taken
out of the Board margin, and specially submitted for approval
to the Board of Admiralty.
Influence of Weight saved or added on a Design. 1 It is
worth noting that a weight saved in any way has an influence on
a design far greater than is given by the number of tons thus
saved. Thus, suppose in any part of a design, say the equip-
ment, 50 tons can be saved. The influence of this 50 tons less is
felt in all parts of the design. The ship thus lightened requires
less I.H.P. for the same speed the engines, etc., thus weigh less
;
and require a smaller complement. A smaller-sized ship will then

be sufficient, which will weigh less than before and require less
I.H.P. These things thus act and react upon one another, and, as
a final result, we should find that by saving 50 tons on the equip-
ment a saving on the whole design, amounting to 100 tons or more
would be possible.
As an extreme case, it is calculated that the adoption of two
5 6 -ft. steam pinnaces (each 18 tons), now used instead of the two
37-ft.steam pinnaces (each 9 tons) formerly supplied, has caused a
weight of about 150 tons to be added to the total weight of the
design of large ships. The actual additional weight of the boats
carried is only 18 tons. But the heavier boats require strong
masts, derricks, and steam, electric, or hydraulic boat hoists for
lifting, and special stowage; so that the result is an additional
weight of 70 tons to be carried about. The influence of this is to
1
See a paper on The effect of modern accessories on the size and cost of war-
;'
ships," by Mr. Whiting, Assistant-Director of Naval Construction (I.N.A., 1903).

304 WAR-SHIPS.
necessitate a design to be about 150 tons heavier than would have

been necessary had the lighter boats been carried.
Stability of a Design Besides fixing on the legend of
weights, we need concurrently to make a calculation for the C.G.,
both in a vertical and horizonal direction. The first is required
to obtain sufficient stability, and the second to see that the
vessel when complete shall float at the desired trim.
The type of calculation, as prepared for a small cruiser, is
shown in the following table :
Items.
THE DESIGN OF WAR-SHIPS. 3<>5
order that the vessel may float at the required draught and trim.
Also, the transverse metacentre must be 2'9 ft. above L.W.L. to
get a metacentric height of 2 ft.

This determines the initial stability. Calculations must, how-
ever, be made to determine the " curve of stability," to see tjiat
the ship has sufficient stability at large angles of inclination.
The detailed calculation of the weight and position of the
C.G. of hull is a long and complicated operation, and although
this is usually done as the design proceeds, yet in the early
stages it is necessary to make an approximation. For this
the information obtainable from the inclining experiments of
It is the practice now to incline the
previous ships is invaluable.
ships of the Navy, not only to ascertain the stability of the ships
themselves, but also to afford data for future designs. The D. 284
form filled up by the dockyard, and the D. 211 form filled up by
the ship's officers, are also of extreme value in affording information
as to weights to the designing staff at the Admiralty.
The tendency of modern designs has been to lead to increased
weight above, by the adoption of an armoured battery and heavier
guns on the upper deck, protected by heavy shields. This tendency
necessitates the stability being carefully considered, because of the
higher position of .the C.G. thus caused. The influence of this is

seen, for example, in the breadth given to the Duke of Edinburgh
as compared with the Drake of greater displacement.
Drake, 500 ft. x 71 ft. x 26 ft. x 14,100 tons; 2 9'2-in. guns
with shields ; 16 6-in. guns in casemates.
Duke of Edinburgh, 480 ft. X
73J ft. X 27ft. X 13,550 tons;
6 9'2-in.guns with shields 10 ;
6-in. guns in battery.
It will be remembered that in Chapter XVII. we saw that the
breadth of ship has a great influence on the position of the trans-
verse metacentre. In a ship where the conditions of the design
lead to a high C.G., the transverse metacentre must also be high to
get sufficient metacentric height.
Horse-power and Speed. The methods adopted to obtain
an estimate of the horse-power necessary for any desired speed
have been dealt with in Chapter XXII. The length of a ship in
relation to the speed to be attained has a most important influence
on economical propulsion. To say that a ship has a speed of
15 knots, say, does not/ convey any correct idea as to whether the
speed is high or not for the ship, unless it is coupled with the size
or length of the ship. Fifteen knots would be a high speed for a
x
3 o6
WAR-SHIPS.
vessel 150 ft. long, but quite a moderate speed for a vessel 500 ft.
long. A measure of speed is obtained by comparing it with the
When
y=-
square root of the length. this ratio is above unity we
\/L,
have a speed which is high for the ship, and which requires a
very
large expenditure of horse-power to obtain. It is interesting to
note that, in the Atlantic liners, as speeds have gone up, so
lengths
V=
have increased, keeping the ratio nearly constant. Thus we
have-
Ship.
308 WAR-SHIPS.
In the latter case it is seen that the ship has only a slight
armament of 12-pounder and smaller guns, with only 150 tons of
coal at the designed draught. The machinery is 17,000 LH.F. for
the 25 knots desired. Thus the ship is only able to carry herself
and her machinery with a comparatively small load, in order to
reach the high speed of 25 knots. In the former case the ship
is able to carry an armament of 4-in. guns, with a large amount
of coal, but this increase of load carried can only be obtained by
having the lower speed of 21f knots. It is seen that an increase

of speed from 21f to 25 knots, or an increase of 15 per cent.,
means an increase of nearly 75 per cent, on the power.
Distribution of Weights. Detailed weights are not pub-
lished for ships of the Eoyal Navy, but Sir Philip Watts, K.C.B.,
gave the following figures for the percentages of weight of a 1905

battle-ship before the I.N.A. in July, 1905, viz. :
Equipment, 4 per cent. ; armament, 19 per cent. propelling

;
machinery, 10 '5 per cent. ; coal, 5 '5 per cent. ; armour, 26 per
cent. hull, 35 per cent.
;
The following table gives typical weights for ships of the

Japanese Navy, from a paper read by the late Admiral Sasso, the
Director of the Naval Construction Corps.
CHAPTER XXIII.
NOTES 1 ON THE LOSS OF H.M.S. " VICTORIA" (June 22, 1893).
CIRCUMSTANCES leading up to the collision :
The fleet was proceeding in two lines, the Victoria leading the
starboard column, the Camperdown leading the port column. The
ships were proceeding at a speed of 8-8 knots, the two lines being
1200 yards apart. At the time of the signal to turn, the helm of
" "
Victoria went hard-a-starboard 35 (corresponding to a tactical
diameter of 600 yards). The helm of Camperdown was put at 28
(corresponding to a tactical diameter of 800 yards). When the ships
had turned through eight points, it was recognized that collision
was inevitable, and the port engines of Victoria and the starboard
engines of Camperdown were reversed, but this had little effect, as
the collision took place one minute later. The speed of the ships
at the time of the collision was from 5 to 6 knots.
The Camperdown struck the Victoria nearly at right angles,
about 65 ft. abaft the stem. The blow was just before important
transverse bulkheads (Fig. 208). Both the ships turning rapidly
at the time of the collision caused the sterns to swing together,
and considerably widened the breach in the side of the
this
Victoria. This, together with the hole caused by the original
blow, destroyed the connections of the bulkheads above-mentioned
with the side of the ship. The value of these bulkheads was thus
completely destroyed, and the compartments on either side were
thrown open to the sea.
For the first minute after the collision the two vessels were
locked, and during this time the Victoria heeled slightly to star-
board and settled a little by the bow. After the Camperdown had
cleared, the Victoria continued to settle by the bow and to increase
her heel to starboard. These movements proceeded gradually for
1
See also end of Chapter VI regarding the question of watertight doora.
3 ro WAR-SHIPS.
about ten minutes, when a sudden lurch to starboard took place.

The ship turned bottom up and finally sank by the head.
It will be desirable to briefly describe the main features of
the Victoria in so far as they bear on this question. The ship
was built to carry two 111-ton guns in a turret forward, and
PROFILE ,u
/
I
*/* |^^
NOTES ON THE LOSS OF H.M.S. "VICTORIA." 311
had ports for the 6-in. guns. With these doors and ports closed,
the freeboard near midships was about 18 \ ft., with them open
the freeboard was about 12 ft.
At the time of the collision a number of watertight doors
forward were open, the turret ports and scuttle on upper deck
Fio. 209.
were open, and the doors in the front of the battery and the gun-
ports in battery side were also open.
For convenience it will be desirable to consider separately the
two movements which actually proceeded simultaneously (Fig.
209), viz.
1. Depression of the bow.
3i2 WAR-SHIPS.
2. Heel to the wounded (starboard) side.

1. Depression of the low. About four minutes after the collision
the bow had sunk enough stem-head under water.
to bring the
The forward part of the deck then became submerged, allowing
water to pass down the scuttle on the upper deck and then down
the turret ports, the water level reaching to the sill of the door
in the battery front starboard, and the bottom of the foremost
gun-port. At this moment the stem-head was depressed about

23 ft. normal position.
from its
2. Heel
starboard.
to Accompanying this depression by the
bow, a gradual heel to starboard took place, until a heel of about
20 was reached. At this time the door in the battery front and
the foremost 6-in. gun-port were just awash.
Both the above movements were accelerated by the motion
ahead of the Victoria, the ship being made to steam slowly
towards the land with the helm hard-a-starboard.
At this time, at the heel of 20, a sudden lurch to starboard
took place, and the vessel capsized and went down head first.
Calculations have been made, using for data the observed con-
ditions, to investigate the cause of the sudden lurch that was
observed; the suddenness of this lurch was without doubt the
cause of the great loss of life.
The loss of buoyancy caused by the opening up of the com-
partments forward, which were inevitably flooded by the collision,
and those flooded subsequently through open doors and hatchways,
caused a change of trim of 29 ft., or 23 ft. depression forward and
6 ft. lift of the stern. Taking the ship in this condition, but
supposing the turret ports, battery doors, and gun-ports closed, the
vessel would have a positive metacentric height of ft. In Fig. ^
208 the shape of the intact waterplane is shown by I, and the
shape of the waterplane area under the above condition is shown
by II. In Chapter XVI. it is seen that the distance between the
centre of buoyancy and the transverse metacentre is directly pro-
moment of inertia of the waterplane at
portional to the transverse
which the ship is floating lBM = ^).

Under the conditions
assumed above, the waterplane area would have a transverse

moment of inertia of 3,888,000, giving such a position of the
transverse metacentre that a metacentric height of ft. was
^
retained. The ship under this assumed condition was in a con-
dition of stable equilibrium.
"
NOTES ON THE LOSS OF H.M.S. VICTORIAS 313
Taking now the actual state of the ship with the turret ports,
the battery doors, and the gun-ports open, the waterplane would be
suddenly reduced to the shape shown by III in Fig. 208. This

area only has a transverse moment of inertia of 2,783,000, lowering
the transverse metacentre and bringing it below the C.G. This
gave the ship a negative metacentric height of 1*8 ft., which rendered
the ship unstable, and so she capsized.
Condition of ship.
314 WAR-SHIPS.
the compartments produces great change of trim or serious,

heeling.
If such precautions are not taken, the virtual height of free
board is reduced to the height of sills or doors, and the presence
of the superstructures, when water is not excluded from them
does not assist either buoyancy or stability to any sensible
extent.
For a detailed account of the above, see
(i.) "Ironclads
in Action." H. W. Wilson.
" Admiral
(ii.)
Life of Tryon." Admiral Fitzgerald.
(iii.) Brassey's Naval Annual, 1894.
(iv.) Parliamentary Paper, No. C. 7208, of 1893.
(v.) Engineer November 10, 1893,
',
APPENDIX A
QUESTIONS.
CHAPTER I.
" local " strains as

Distinguish between the terms "structural" and
1.
applied to a ship. Enumerate a number of "local" strains. Have any of

these local strains been sufficiently great in your experience to cause
damage ?
2. State the reasons for the superior efficiency of a I beam of steel to a
solid rectangular beam of wood.
3. How may a ship be compared to a beam, and what parts of the structure
are most efficient from this point of view ?
4. How are the longitudinal strains on a ship's structure made the
subject of calculation ?
5. Why is the structure at the keel and at the upper deck
considerably
stronger in a long cruiser than in a battle-ship of the same total displace-
ment?
6. Why is it that the boat deck and the topside plating adjacent are not
made an integral part of the structure in a ship having a boat deck ?
7. To what special sort of strain are the flat portions of a ship forward
specially liable ? do you consider that this straining action is less

Why
in evidence in war-ships than in merchant steamers ?
8. Why is it possible to build a steel or iron ship considerably lighter
than a ship of the same size built of wood ?
9. Why must special attention be devoted to the strength of the upper
deck and structure adjacent in a vessel of large proportion of length to

depth ? From this point of view, show that the method of protection
of large cruisers as Fig. 138, is likely to prove more economical as regards
weight of hull structure than that adopted in, say, the cruisers of the Edgar
class.
Suppose .one had a vessel 300 ft. long, the structure of which had
10.
proved sufficiently strong, and a vessel of the same depth, but 360 ft. long
were required. Discuss generally what portions of the structure would have
to bo strengthened to ensure the new vessel being sufficiently strong.
11. Indicate how the inspection and maintenance of a ship influences
the design of the structure.
315
3 i6 APPENDIX.
CH4PTER II.
1. State the qualities of "mild steel" that make it a suitable material

for shipbuilding purposes.
2. Compare in tabular form the tests laid down for
" mild " rivet
steel,"
steel," "cast steel."
3. What tests are necessary in a steel casting beside those relating to the
strength and ductility of the material? Why are such tests of great im-
portance for castings of steel ?
4. Compare the tests for "mild steel" and those for the special steel
used in cruisers and destroyers.
6. It is laid down that holes in high-tensile steel must be drilled, and not
punched. Why is this ?

6. Describe the process of "pickling" steel plates. What trouble would
you expect to arise in a ship's structure from the steel of which the "mill
"
scale had not been removed before painting ?
7. Draw out to a large scale the section of a zed bar, a tee bulb, an
angle bulb, and a I bar. State places in your present ship in which these
sections are used.
8. A
flange is frequently used on the edge of a plate instead of an angle
bar for connection purposes. What advantage is thereby secured? State
places in your present ship where this is done.
9. Describe the most ordinary form of rivet used in ship work, and show
how such a rivet is used for the outer bottom plating where the outside
surface must be flush.
10. What meant by the pitch of rivets? State the amount
is of this
pitch for rivets diameter where the work has to be watertight.

-in. What
pitch would be used for -in. rivets for internal work not watertight ?
-4ns. 4 to 4| in. 6J to 6 in.
;
11. When your ship is next in dry dock, examine the "lap" caulking
and the " butt" caulking of the outer bottom plating.
12. State the various advantages that result from ordering plating by the
weight required per square foot rather than by thickness.

13. Taking the area of the outer bottom plating of a vessel as 30,000
square ft., estimate the saving of weight, if the steel plating is ordered
20 Ibs. per square ft. instead of | in. thick. What further saving would
be possible if the manufacturer sends in all the plating down to the limit
allowed, viz. 5 per cent, under? Ans. 5*4 tons; 13'4 tons.
14. If the area of the outer bottom plating (specified of 15 Ibs.) is
20,000 square ft., what variation of weight is possible, in view of the

latitude allowed to the manufacturer ? Ans. About 13| tons.
15. What is annealing ? What is the effect of annealing on a plate which
has had a large number of holes punched in it ?
CHAPTER III.
1. Distinguish between "bracket frame" and "solid plate frame."

Where are these frames used in a large armoured ship ?
APPENDIX. 3t7
2. Describe generally the construction adopted in battle-ships below

armour within the limits of the double bottom.
3. What is a "floor-plate"? Sketch and describe the framing of a
battle-ship before and abaft the double bottom.
4. State the advantages of having a double bottom to a ship. is it Why
not possible to provide a double bottom in the smaller ships of the Royal
Navy ? Discuss the question of fitting a wing bulkhead below the armour
deck in a large cruiser.
5. Why is the "transverse" system of construction more suited to the
"
ends of a large ship than the "longitudinal system ?
6. Draw out a table with rough sketches showing the supporting frames
for 4-in., 6-in., 9-in. armour. Why is a rigid support of great importance
behind armour ?
7. Draw in outline midship section, describe and compare the main

features of (1) Admiral class, (2) fioyal Sovereign, (3) Majestic, (4) King
Edward VII.
8. Draw in outline midship section, describe and compare the main
features of the first class cruisers (1) Edgar, (2) Diadem, (3) Cressy, (4)
Monmouth.
9. Draw in outline midship section, and compare the main features of
two second class cruisers. Show how the intended service has had a distinct
influence on the design.
"
10. Draw in outline the midship section of a sloop," and point out what
provision is made in such a vessel for protective purposes.
11. State in general terms the distinction between a battle-ship and a
first class cruiser. Compare H.M.S. Triumph with H.M.S. Duncan and
H.M.S. Cressy, and state in what category you consider she should be
placed.
12. Discuss the question of working zed bars for the framing of ships
instead of two angles riveted back to back from the point of view of (i.)
economy of weight, (ii.) saving of cost.
13. Discuss the importance of avoiding discontinuity of strength in a
ship's structure.
14. Name typical vessels of the Royal Navy which are sheathed with wood
and copper. Whyhave all the sloops and most of the second class cruisers
built in recent years been sheathed ?
15. In going through the double bottom of a vessel it will be noticed that
the non-watertight longitudinals have no lightening holes in certain frame
spaces. Trace the reasons for this.
16. Taking the length of the double bottom of a battle-ship as 250 ft.,
and the frame spacing 4 ft., make an estimate of the saving of weight if an
oval manhole 23 in. x 15 in. is cut in every frame space in the non-water-
tight longitudinals. -4ns. 5 to 6 tons.
17. What advantages beside saving of weight are obtained by cutting

lightening holes in a longitudinal girder ?
18. State places in your present ship where you have noticed holes cut
for lightening purposes. Do you consider any weakness has resulted from
the removal of this material ?
19. What is a middle-line keelson ? When this is intercostal, and the floor
3l8 APPENDIX.
plate is continuous from side to side, how is proper continuity of longitudinal

strength obtained ?
CHAPTER IV.
1. Draw out the sections of the beams used in your present ship for the
various decks and platforms. State the spacing of the beams amidships and
at the ends.
2. Under what circumstances
are zed bars likely to be advantageous for
beams ? do you consider it would have been inadvisable to form the
Why
beams to the decks of your ship of zed bars ?
3. Why is the connection of beams to frames of great importance ? Draw
out two such connections as fitted in your present ship.
4. What is a carling ? State places in your ship where carlings are
fitted.
5. should the upper deck of a vessel be given a round up while the
Why
deck is worked level ?
platform
6. Why would a log of timber wedged up between-decks be inefficient as
a pillar from a structural point of view ?

7. Name places in your ship where the pillars have had to be made
"
portable."
In sailing ships of the mercantile marine in which only one bulkhead
8.
(the collision bulkhead) is fitted, it is laid down in "Lloyd's Rules" that

the beam arms have to be three times the depth of the beam. Discuss the
reasons for this, bearing in mind the ordinary practice in ships of the Royal
Navy.
9. If the upper deck of your ship is completely plated, trace the bolt
fastenings to the planks of the wood deck. See also where these bolts are
placed to take the butts. If the deck is not completely plated, see how the.
butts of the planks are fastened.
10. Why are pillars always made hollow in vessels of the Royal Navy ?
11. Notice in your ship the arrangement of butts of "sheer strake,"
"stringer plate," and the structure adjacent. See how a good shift of butts
is obtained. Why is this of great importance ?
12. What do you

consider the advantages and disadvantages of corticine
as a covering for decks in place of wood ? How is this corticine secured ?
13. Describe the operation of "caulking" as applied (a) to steel plating,
(6) to a wood deck. In both cases state what preparations are necessary
before the caulking is started. What essential quality is caulking intended
to secure ?
CHAPTER V.
1. What special advantages attach to the use of steel instead of iron for
the skin plating of a vessel ?

2. Trace in Fig. 46 how the
strength of the structure is made as uniform
as possible, no special frame space being a special place of weakness, but a
good shift of the butts being obtained.
APPENDIX. 3I9
"
3.What is the "raised and sunken system of working the outer bottom
" "
plating ? What is the function of the liner ?
" Does the lightening in any way injure
4. Sketch a lightened liner."
the strength of the structure ?
5. State places where the outer bottom plating of a large ship is doubled.
For what purpose is this done in each case ?
6. Why is the transverse section of a ship in way of a watertight bulkhead
a line of weakness ? Show how this weakness is
compensated for.
7. Sketch the form of bulkhead liner adopted in your present ship.
8. State the requirements necessary in a watertight manhole to a double-
bottom compartment. Draw out to a large scale the plug fitted in the
cover, and state its uses.
9. Make a tabular statement giving the thicknesses of keel, outer bottom
plating, and sheer strake in (a) a battle-ship, (6) a second class cruiser, (c) a
third class cruiser, (d) a sloop, (e) a destroyer.
10. What is a
" liner " ? What Would
purpose does it serve ? there be
any objections to making such a liner of iron instead of steel if desired on
the score of cheapness?
CHAPTER VI.
1. What is a "collision" bulkhead? Why is this bulkhead of great
importance 1
How far is the collision bulkhead from the forward perpendicular

2.
in your ship? How does this compare with the distance required for
"
merchant ships, built to the rules of Lloyd's Register," viz. one-twentieth
the length ?
3. Why do you think the rules of the Registration Societies specify the
minimum distance of the collision bulkhead abaft the stem ?
4. State the special advantages attendant on the provision of a watertight
bulkhead between the engine-rooms.

6. For what reason is the stiffening adopted for the main bulkheads of a
large ship of so strong a

character ? Why can the bulkheads at the ends of
the machinery space be safely built with considerably less stiffening than
those between the boiler-rooms ?
6. How are compartments of a ship tested for watertightness while
building ? What does this testing ensure besides watertightness 1

7. Write out the instructions regarding the tests required for watertight
compartments of ships in commission. (See Admiralty Circular, S. 32111/1903,

January 29, 1904.)
8. What is a
" cofferdam " bulkhead ? Is one fitted in
your present ship 1
If so, state the means of access to the space between it and the collision
bulkhead.
9. When a magazine is placed next to a boiler-room it is usual to provide
an air space between to prevent the magazine becoming he.ated. Is this the
case in your present ship ? If so, how is access obtained to the space ? Why
issuch access necessary 1
10. Trace the means of escape under protection fitted in your ship from
the submerged torpedo-rooms. Why is such escape necessary ?
320
APPENDIX.
11. Make a list of the horizontal sliding watertight doors in your ship.
In each ease see if it would have been practicable to fit a vertical sliding
door instead.
12. Make a list of the most important vertical sliding doors in your ship.
In each case state why a hinged door would have been undesirable.
13. How many double-bottom compartments (excluding the wings and
reserve feed spaces) are there in your present ship ? What is the water
capacity in tons of the largest, and what of the smallest ? What is the total
capacity in tons of this double bottom available for the introduction of
water as ballast ?
14. Investigate the means of access in your present ship to

(i.) All wing compartments ;
(ii.) Spaces below provision rooms, etc., forward and aft;
(iii.) Spaces before the collision bulkhead ;

(iv.) Inside of masts ;
(v.) Watertight compartments at the sides of submerged torpedo -room

if fitted.
Why means of access provided to such spaces not used for stowage
are such
purposes ?
15. Describe with outline sketches the three types of watertight door
fitted in ships of the Royal Navy. What are the conditions governing the
adoption of each type ?
16. Why is it essential that when name plates or fittings are removed
from bulkheads that the holes left should be immediately filled up by tap
rivets? (See Admiralty Circular, S. 32111/1903, January 29, 1904.)
CHAPTER VII.
1. Sketch and describe the stem and its supports for a battle-ship. Draw
special attention to the provision that is made to withstand the side bending
that is likely to take place when the ships swing together after ramming.
2. Sketch and describe the stem for a sheathed vessel.
3. What
are the special functions of the sternpost of a twin-screw
vessel? Sketch and describe such a sternpost with the connection of the
adjacent structure.
4. What special features do the sternposts of modern cruisers
possess ?
Make a sketch of one such sternpost. What is specially necessary for the
support of the rudder?
5. What information is supplied to your ship with respect to the
rudder ?
6. Make a sketch of a shaft bracket. Indicate the exact nature of the
strains that this fitting has to withstand. What is it made of in your ship ?
7. Make a sketch of rudder of either
(i.) A battle-ship ; or,

(ii.) A cruiser.
8. Why is it necessary to make the diameter of the rudder of a cruiser
of considerable size, in view of the fact that the twisting moment is small,
even at high speeds, because the rudder is balanced 1
APPENDIX. ,
2I
9. Compare the pressure per unit area on a rudder in
(i.) A 10-knot cargo steamer ;
(ii.) A 19-knot battle-ship ;
(iii.) A 23-knot cruiser ;
(iv.) A 30-knot destroyer;

(v.) A 3G-knot destroyer.
Ans. 1:3-6: 5-3 : 9-0 : 13-0.
10. Indicate in detail how the weight of a rudder is taken in

(i.) a battle-
ship like Duiican and (ii.) a cruiser with a balanced rudder.
11. State in detail what would need to be done to remove the rudder of
your ship. How can be lifted and withdrawn for examination and
it
repair ?
12. Show how
the hole in the sternpost for the reception of the rudder
head is made
watertight.
13. Sketch the outline of the sternpost and rudder in Edgar, Cressy,
Duncan, King Edward VII., and Lord Nelson.

14. Name any
ships you know of in the Royal Navy which have two
rudders. Sketch the shape of the stern of one of these ships.
15. State the essential conditions to be satisfied in designing the stern of
a large war-ship.
CHAPTER VIII.
1. State what is meant by a "compensating" steering gear. How is

"
compensation obtained in Rapson's slide," "Harfield's gear," and "Ollis's
"
gear ?
2. What special advantages attach to the use of the screw steering gear ?
3. What advantage is there in a compensating steering gear in view of
the desirability of being able to steer the ship by manual power ?

4. In Fig. 86 of Harfield's gear it will be noticed that the diameter of
the spindle to the forward cross-head is considerably less than the diameter
of the rudder-head. Why are these diameters so different, seeing that both
have to take the same twisting moment ?
5. In Harfield's and Rapson's slide steering gears (Figs. 86 and 84), why
are not the tillers keyed direct on to the rudder-head without the use of the
connecting rods ?
" "
6. Is a Fayrer's brake fitted to the hand-wheels of your ship ? If so,
what purpose does it serve ? If not, why has it been dispensed with ?
7. Why is it essential that steering gear should be under perfect control
when changing from steam to hand, or from one steam engine to the other ?
8. What alternative methods of steering your ship are possible, supposing
the steering gear aft and the rudder are in order ?
9. Supposing the rudder of your ship carried away/ how could you steer
the ship?
10. Make a list of the positions from which your ship can be steered,
supposing the steam steering engines are available.

322 APPENDIX.
CHAPTER IX.
1. Make
a list (with capacities, if possible) of all the pumps available in
your ship for dealing with a leak.

2. Show that the initial rate of inflow of water through a hole d ft.
below the surface is about 14 ,^/d tons per minute for every square foot of
area of the hole.
Explain how the centrifugal circulating pumps in the engine-rooms
3.
may be made available for dealing with inflow of water into a ship.
4. Sketch the "main drain" as fitted to a modern ship. Show how
water is prevented from passing away from the engine-room or from one
boiler-room to another through this main drain.
5. How is water got rid of from (i.) wings, (ii.) barbettes, (iii.) submerged
torpedo-rooms, (iv.) chain lockers ?
G. Sketch and describe a short portion of the "main suction." What

special type of valve is necessary for the suction from a double-bottom
compartment, and why ?
7. Suppose it has been necessary to flood a magazine in your ship. How
would the water be got rid of ?
8. Sketch in outline and describe the construction and working of a
Downton pump.
9.Sketch and describe a Kingston valve. How many of these valves
are fitted to your ship
1 Make a list stating the purpose of each of them.
10. How is water collecting on the top of the inner bottom got rid of ?
11. State exactly what you would do to pump out water in, say, the
forward boiler-room, supposing steam is not available.
12. Trace whab it would be necessary to do to flood the largest magazine
forward in your ship.
13. Why are valves for opening a magazine to the sea kept locked?
Where are the keys kept ?
14. Sketch an air escape as fitted to (i.) a magazine, (ii.) a wing

compartment.
"
15. Sketch a portion of the "fire-main," with a specimen rising main."
What is the fire-main used for ?
16. How would you get water for wash-deck purposes or for fire, in the
event of the steam pumps not being available ?
17. Sketch and describe the fittings in your ship for flooding magazines,
etc., when the ship is in dry dock.

18. The " main drain " was formerly led through the double bottom.
What advantages result from the later practice of leading themain drain
above the inner bottom ?
APPENDIX. 323
CHAPTER X.
1. ventilation for an ordinary building and a ship.

Compare the
2. continuous and efficient ventilation of coal-bunkers of great
Why is
importance ? Write out the regulations regarding the ventilation, etc. , of

coal-bunkers contained in the " Steam Manual."
3. Sketch and describe the ventilation of (1) an upper bunker, (2) a lower
bunker.
4. Describe generally the system of ventilation with large steam-driven
fans. What disadvantages are connected with this system ?
5. Describe generally the present system of ship ventilation.
G. Describe, with sketches, how one of the largest magazines forward in
your ship is ventilated. Why are ventilation exhausts fitted to magazines

and not to shell-rooms ?
7. Write out the regulations contained in the "Gunnery Manual"
regarding the ventilation of magazines.
8. Why is it undesirable to paint the ends of magazine exhausts and air-
escape pipes ?
9. Describe, with sketches, the ventilation in your ship of (i.) sick bay,
(ii.) spirit room, (iii.) engineer's workshop, (iv.) capstan engine compartment.
CHAPTER XL
What is rust, and under what conditions is it formed ?
1.
Describe the process of "pickling." What trouble would you expect

2.
to arise in a ship from the steel of which the mill scale had not been removed
before painting ?
3. Why is it essential to keep the steel of a ship always well painted ?

4. What is a zinc protector ] What object does it serve? Why must
zinc protectors never be painted ?
5. State places in your present ship where you have observed zinc
protectors fitted.
6. What is fouling? How is fouling prevented in (i.) a wood ship, (ii.)
g, steel ship ?
7. Sketch and describe the present system of sheathing vessels with wood
and copper. What service are such vessels specially suitable for Name 1
vessels of the Royal Navy thus sheathed.

8. What is exfoliation ?
For what purpose is cement fitted in ships ? Under what circumstances

9.
would cement prove detrimental rather than beneficial ?
10. What is cork cementing ? What is the object of this process ?
11. State the regulations laid down in the

" Steam Manual "
respecting
the inspection of the structure of H.M. ships.
324 APPENDIX.
CHAPTER XII.
1. Describe with outline sections the armour or other protection in (1) a

battle-ship, (2) a deck-protected cruiser, (3) a sloop.
2. Discuss briefly the circumstances which led up to the particular
arrangement of protection adopted in the Inflexible.

3. What is compound armour, Harveyed armour, and Krupp armour.
Compare the relative efficiency of these types of armour.

4. Trace the reasons which led to the adoption of armour for the first
class cruisers of the Cressy class.

5. Describe the system of protection adopted in (i.) Royal Sovereign,
(ii.) Majestic. Compare and contrast the two systems.
6. For what purpose is teak backing fitted ? What is the thickness now
adopted ?
7. Sketch an armour bolt for hard-faced armour. What is the object of
the sleeve, the india-rubber washer, and the cup washer ?
rt "
8. What is meant by the term figure of merit as applied to armour ?
9. Describe the armour protection, etc., of a vessel of the Duncan class.
10. What is a cofferdam ? State places in your ship where cofferdams
are fitted.
11. What is the value of coal as a material for purposes of protection ?
What vessels depend wholly on their coal for protection ?

12. In the Duncan the barbettes are 11 in. in thickness, while the side is
only 7 State the reasons for this difference.

in.
13. Compare the method of protecting the gun mountings for heavy guns
in (1) Admiral class, (2) Royal Sovereign, (3) Duncan.
14. Make out a list of the armament in your present ship, stating in
each case how the rear of the gun, and how the mounting, etc., is under
protection.
" armour
15. What are gratings"? Why are they fitted ? State places
in your present ship where you have noticed armour gratings fitted.
CHAPTER XIII.
1. A hollow pillar is 4 in. external diameter and in. thick. What is

the sectional area, and what would be the weight in pounds of 10 ft. of this
pillar, if of wrought iron (480 Ibs. to cubic foot) ?
Ans. 4'27 square in. 142 Ibs. ;
2. A wrought-iron armour plate is 15 ft. 3 in. long, 3 ft. 6 in. wide, and
4| in. thick. Calculate its weight in tons. Ans. 4*29 tons.

3. Steel plates (490 Ibs. to the cubic foot) are demanded 400 Ibs.
armour
per square foot instead of 10 in. thick. What is the saving of weight per
1000 square ft. of surface of this armour? Ans. 37 tons.
4. A mast, 90 ft. in length, and 3 ft. external diameter, is composed ot
20-lb. plating, worked flush jointed (as Fig. 8) on 3 tee bars, 5 in. x 3 in. x 15^
Ibs. per foot. Estimate the weight. Ans. About 9J tons.
5. A curvilinear area has the following ordinates at equidistant intervals
of 18 ft, viz. 6-2, 13-8, 21'9, 26'4, 22'35, 1470, and 7'35 ft. Assuming
that Simpson's first rule is correct, find the percentage of error that would
be involved by using the Trapezoidal rule. Ans. 1'2 per cent.
APPENDIX. 325
6. The semi-ordinates of the load water plane of a vessel are 0'2, 3'6, 7 '4,
lO'O, 11-0, 107, 9'3, 6*5, and 2'0 ft. respectively, and thay are 15 ft. apart.
What is the area of the load water plane ?Ans. 1808 square ft.
7. The apart have the following
vertical sections of a vessel 10 ft.
areas 10, 50, 60, 70, 50, 40, 20 square ft.

: Find the volume of displacement.
Ans. 2966 cubic ft.
8. In a given ship, pillars in the hold can be either solid iron, 4| in.
diameter, or hollow iron, 6 in. diameter, and \ in. thick. Estimate the
saving of weight for every 100-ft. length of these pillars, if hollow pillars
are adopted instead of solid, neglecting the effect of the solid heads and
heels of the hollow pillars. Ans. T35 tons.
9. A bunker, 24 ft. long, has a mean section of the form of a trapezoid,
with parallel sides 3 ft. and 4'8 ft., 10*5 ft. apart. Find the number of tons
of coal contained in the bunker, at 43 cubic ft. to the ton.
Ans. 22-8 tons.
CHAPTER XIV.
What
the displacement quoted in the
" "
1. is Navy List ?
2. Explain what is meant when a ship's dimensions are stated as follows
in the Navy Estimates. Length, 300 ft. ; breadth, 36 ft. 6 in. ; mean draught,
13 ft. 6 in. ; displacement, 2,135 tons.
3. State the draughts of your ship in three distinct conditions of
loading.
Define "deep load," "normal load," "light."
4.
What
is "gross register" and
" nett
5. register" as applied to the
tonnage of war-ships ? Add to the comparative table of tonnage in Chapter
XV., the various tonnages for your present ship as given in the Ship's Book.
6. Compare the methods of referring to the tonnage of
war-ships and
merchant ships, and explain why it is that the great size of modern liners, as
compared with the largest war-ships, is not appreciated because cf the different
systems of stating the tonnage.
7. Write out the regulations contained in the King's Regulations
regarding the statements to be given of the tonnage of H.M. ships.
8. Why do ships of the Royal Navy have to be measured by the Board
of Trade officers for tonnage ?
CHAPTER XV.
1. Explain and justify the statement, "the weight of a vessel is equal to
the weight of the water displaced."
2. A vessel at Gravesend, where the water weighs 63*7 Ibs. per cubic ft.,
istound to displace 60,500 cubic ft. What is her weight in tons ?
Ans. 1720 tons.
3. What is the weight of a box-shaped vessel 150 ft. long, 17 ft. broad,
which is found to float in salt water at a draught of 6 ft. 3 in.

Ans. 4682 fcon3 '
4. A cylinder is 500 ft. long, 20 diameter, and floats with the axis in
ft.
the water surface. Find its weight when floating thus in salt water.
Ans. 2244 tons.
326 APPENDIX.
5. Stats, if possible, the tons per inch of your present ship, and compare
it with the approximation given in Chapter XVI.

6. Bilge keels are to be fitted to a ship whose tons per inch is 48. The
estimated weight of the bilge keels is 36, and the volume they occupy is
840 cubic ft. What will be the increase of draught due to fitting these bilge
keels ? Ans. \ in.
7. What is the "coefficient of fineness?" Determine its value for the
following ships ;
(i.)
H.M.S. King Edward VII., 425 ft. X 78 ft. x 26 ft. 9 in. x 16,350 tons,
(ii.) SS. Umbria, 500 ft. x 57 ft. x 22 ft. 6 in. x 9860 tons,
(iii.) H.M.S. Duke of Edinburgh, 480 ft. x 73* ft. x 27 ft. x 13,550 tons,
(iv.) H.M.S. Diadem, 435 ft. x 69 ft. x 25 ft. 3 in. x 11,000 tons (keel
projection 8 in.).
(v.) S.S. Lusitania, 760 ft. x 87 ft. 6 in. x 32 ft. 6 in. x 36,840 tons,
(vi.) H.M.S. Dreadnought, 490 ft. x 82 ft. x 26 ft. 6 in. x 17,900 tons,
(vii.) H.M.S. Invincible, 530 ft. X 78 ft. 6 in. X 26 ft. x 17,250 tons. Ans.
(i.) 0-645 ; (ii.) 0'538 ; (iii.) 0'5 ; (iv.) 0'52 ; (v.) 0-597 ; (vi.) 0'588 ;
(vii.) 0-558.
" "
8. What is tons per inch immersion 1 Find the value of this for a
box-shaped vessel 150 ft. long, 30 ft. broad. Ans. 107 tons.
9. Draw
out a curve giving the displacement of the vessel in the previoua
question for all draughts up to 12 ft.
Ans. Curve is a straight line from at zero draught to 1543 tons
at 12 ft. draught.
10. A vessel of 14 mean draught
has the following displacements at
ft.
level lines 2 apart, viz.

2118, 1682, 1270, 890, 553, 272, 71.
ft. Draw out
the curve of displacement and state (i.) the displacement at a draught of
12 ft. 3J in. forward, 13 ft. 6J in. aft. ; (ii.) the mean draught when the ship
displaces 1750 tons ; (iii. ) the tons per inch immersion at 13 ft. draught.
Ans. (i.) 1880 tons ; (ii.) 12 ft. 4 in. ; (iii.) 181 tons.
11. The tons per inch of a vessel at waterlines 2
apart are 19*45, 18*51, ft.
17-25, 15-6, 13-55, 10-87, and 6 '52, the lowest waterline being 18 in. above
the underside of flat keel. Draw the curve of tons per inch immersion to
scale and estimate the number of tons necessary to sink the vessel from
a draught of 12 ft. to a draught of 13 ft. 6 in. Ans. 344 tons.
12. The area of a waterplane at which a ship floats is 6300 square ft.
What will the sinkage be if 45 tons be placed on board ? Ans. 3 in.
CHAPTER XYI.
1. Define "centre of gravity," " centre of buoyancy."

2. Write down the three necessary conditions that must be fulfilled for a
ship to float freely and at rest in stable equilibrium.

3. Define stable and unstable equilibrium.
4. What is the transverse metacentre ? How does its position with

reference to the C.G. determine the stability of a ship for small angles 1
5. Show how the position of the metacentre with reference to the C.B.
depends on (i.) the shape of the waterplane and (ii.) the displacement.
APPENDIX. 327
6. What is the metacentric ? State this length for your ship in the
height
deeply laden condition.
7. What is a metacentric diagram ? Of what use is such a diagram when
made?
8. Indicate the reason that a log of timber half immersed will not float with
one face horizontal or with the axis vertical.
9. What is the What knowledge of the ship does
inclining experiment ?
one obtain from this experiment ?
10. Show that if a deflection a is observed in a length I of a pendulum,
after traversing w tons through d ft. across the deck of a vessel tons W
displacement, the metacentric height is given by
GM = w x d x I
W x a
11. A vessel of 1722 tons displacement is inclined by shifting 6 tons of

ballast across the deck through 22| ft. A mean deviation of 10.} in. is
obtained with pendulums 15 ft. long. The transverse metacentre is 15 '28 ft.
above keel. Show that the C.G. of the ship is 13 '95 ft. above keel.
12. Discuss the importance of coal-bunkers above the protective deck
of a deck-protected cruiserfrom the point of view of stability.
13. What has girdling at the waterline on the stability of a ship ?
effect
14. Explain how it is that a ship when partially waterborne, as when
being dry docked, suffers a reduction of stability. Under what particular

combination of circumstances may this reduction be sufficient to cause
instability ?
15. Avessel of 1792 tons displacement is inclined by shifting 5 tons
already on board transversely across the deck through 20 ft. The end
of a plumb-line 15 ft. long moves through 5J in. Determine the metacentric
height. Am. T91 ft.
16. Atwo-masted cruiser of 5000 tons displacement has its C.G. 2 ft.
above the waterline. It is decided to add a military top to each mast.
Assuming the weight of each top with its guns, men, and ready ammunition
supply to be 12 tons, with its C.G. 70 ft. above the waterline, what will /be
the effect of the change on the metacentric height ?
Ans. Reduce about 0*3 ft.
CHAPTER XVII.
1. Define longitudinal metacentre. Write down and explain the formula

for obtaining its position with reference to the C.B.
2. Define trim, change of trim. Write down an approximate formula,
suitable to your ship, for the moment to change trim 1 in.
3. Make an estimate of the change of trim in a ship of Pelorus class
due to moving 10 tons through a fore-and-aft distance of 200 ft.
Ans. About 7 in.

4. State the trim of your ship in the deep load condition. What is the
trim in the most usual seagoing condition ?
5. Are any trimming tanks fitted in your ship? If so, explain under
what circumstances you would fill them.
328 APPENDIX.
6. Show why it is that many ships floating on an even keel will go down
by the head if a weight is placed at the middle of the length.

7. Explain how it is possible to determine the position of coals on board
a ship such that the draught aft will not vary whether they are in or out.
8. Find the moment to change trim 1 in. of a vessel 400 ft. long, having
given the following particulars Longitudinal metacentre above C.B., 446ft.

:
;
distance between O.G. and C.B., 14 ft.; displacement, 15,000 tons.

Ans. 1350 foot-tons.
9. A log of fir, specify gravity 0'5, is 12 ft. long, and the section is a
square of 2 ft. side. Find the longitudinal metacentric height when floating
in stable equilibrium. Ans. 16 '5 ft. nearly.
10. H.M.S. Hermes is 350 ft. x 54 ft. x 5600 tons when floating at a
draught of 19 ft. 6 in. forward, 21 ft. 6 in. aft. It is desired to take her
over a bar at which the depth of water is 19 ft. Approximate to the weight
necessary to remove, and where the C.G. of this weight should be.
Ans. 570 tons removed from the ship so disposed that the C.G.
is about 40 ft. abaft of amidships.
"
11. A vessel whose moment to change trim 1 in. " is 73 foot-tons, floats
at a draught of 6 ft. 6 in. on an even keel. Determine the draught forward
and aft if a weight of 5 tons is moved aft through a distance of 135 ft.
Ans. 6 ft. 11 in. forward 6 ft. 10| in. aft.
;
12. A battle-ship 400 ft. x 75 ft. floats at a draught of 26 ft. 3 in.
forward, 27 ft. 3 in. aft, and displaces 15,000 tons. Make an estimate of
the new draught if 500 tons of coal is added 75 ft. before amidships.
Ans. Forward, 28 ft. 6| in. ; aft, 26 ft. 8J in., about.
13. At about what station of your ship would the effect of added weight
be to increase draught bodily but not to change the trim ?
14. What relation exists between the transverse and longitudinal stability
of a wholly submerged body ? Discuss the question of submarine navigation
from the point of view of longitudinal stability.
CHAPTER XVIII.
1. What
are the important features of a curve of stability ?
2. Why specially desirable to have the curve of stability of a sailing-
is it
vessel enclosing a large area ?

3. What is the angle of maximum stability of your ship in the "deep
"
and " normal load
"
conditions ? What is the
" "
load range of stability in
both of these conditions ?
4. A vessel of 5000 tons displacement has a metacentric height of 2*0 ft.
What effort does she make to return to the upright when inclined to (i.) 3,
(ii.) 6, (iii.) 9? Ans. 523 1045 1564 foot-tons.
; ;
5. A vessel's curve of stability has the following ordinates at angles of
15, 30, 45, 60, 75, viz. 0'51, 0'97, 0'90, 0'53, 0.08 ft. respectively.
Estimate the influence on the range of stability caused by lifting the C.G.
of ship 0'2 ft. Ans. Reduce nearly 6.
6. Draw out curves of stability of two ships, one having a great meta-
centric height and one having a moderate metacentric height.
APPENDIX. 329
7. What information is supplied to your ship as to the conditions of

stability in various conditions ?
8. What is the effect of excessive
" tumble home
"
on the stability of a
ship at large angles ?
9. State the various assumptions which have to be made when making

the stability of a ship at large angles the subject of calculation. Do you
think any of these assumptions are unreasonable ?
10. In the refitting of the Royal Sovereign class with casemated guns on
the upper deck, it has been found undesirable to make any alteration to the
turret ship Hood. What reason do you think has led to this action in this
special ship of the class ?
CHAPTER XIX.
Quote a formula giving the period of oscillation of a ship

1. when rolling
unresistedly in still water.
2. What is "isochronous" rolling?
3. What is the " period " of your present ship 1
4. Why was that the Inflexible had the short period of 5J seconds
it ?
Why is such a short period undesirable ?

5. Explain generally the action of bilge keels in reducing rolling.
6. Sketch the standard form of bilge keel as fitted to large steel ships of
the Royal Navy.
7. Why is it that a ship of large metacentric height is likely to be a bad
sea-boat, and a ship with small metacentric height a good sea-boat ?

8. In a ship rolling among waves, what is meant by the " virtual upright "
at any particular instant ?
9. A ship of 6 seconds' period is placed broadside on to a regular series of
waves having a period of 12 seconds. What will happen 1
10. Justify the statement that a " crank ship is likely to be exceedingly
steady in a sea-way."
11. What is synchronism ? If the ship in question 9 is under control,
what could the commanding officer do to destroy the synchronism ?
12. Explain the principle of Mallock's rolling indicator.
13. Why is a pendulum untrustworthy in regard to giving indications of
the rolling of a ship ?
CHAPTER XX.
1. Compare the "balanced" rudder on the turning of a

influence of a
" unbalanced
"
rudder of the same area.
ship as compared with an
2. In some ships it is noticed, that on first putting the rudder over, the
ship bends away from the straight before taking the spiral course. Give an
explanation of this.
3. Explain the features of the Japanese battleship, Yashima, which caused
her to have an outward heel of 8f when on the circle at full speed. Have
you noticed any similar phenomenon in your experience ?
4. What is the "pivoting point" of a ship when turning? Can you
locate this point in your present ship ?
y 2
330 APPENDIX.
5. What difference would you expect to find in your present ship if
turning at full speed (a) with steam steering engine in use, (6) with
hand-wheels only available ? Also what difference at 10 or 12 knots under
these circumstances ?
6. Show that the influence of the after deadwood of a ship is favourable
to turning when the rudder is first put over, but unfavourable when on the
circle.
7. Enumerate all the advantages you think a twin-screw ship possesses in
comparison with a single-screw ship.

8. State for your present ship how an object ahead could be best avoided,
either by
(a) Turning with full helm, both screws ahead ;
(6) Turning with full helm and one screw astern ;

or
(c) Reversing both engines.
CHAPTER XXI.
1. What is effective ?torse-power ? The Greyhound was towed at the rate

of 845 feet per minute, and the horizontal strain on the tow-rope, including
an estimate of the air resistance of masts and rigging, was 6200 Ibs. Find
the E.H.P. at that speed. Ans. 159 H.P. nearly.
2. Suppose we took a destroyer of 250 tons displacement and 27 knots
speed as a model, and designed a vessel of 10,000 tons displacement of

similar form. At what speed of this vessel could we compare her resistance
with that of the model at 27 knots 1 (The ratio of length will be ^TcCHm}
Ans. 50 knots.
3. A
vessel of 7000 tons requires 10,000 I.H.P. to drive her 20 knots,
and the I.H.P. at that speed is varying at the fourth power of the speed
Find approximately the I.H.P. necessary to drive a similar vessel of 10,000
tons at a speed of 21 knots. Ans. 16,000 I.H.P.
4. Make out a table for your present ship,
giving speeds and corresponding
horse-power. Construct a curve of power on base of speed, as Fig. 205, and
make out a table, as for Drake, in Chapter XXII., giving increments of power
for every knot from 10 knots to the highest speed.
5. In the previous question, do any of the spots come manifestly above or
below the general run of the curve ? Are any of these due to (a) foulness of
bottom, (6) bad weather, (c) running in shallow water, or (d) difference of
draught.
6. Enumerate the various factors which make up the difference between
the E.H.P. and the I.H.P. of a ship.
7. Explain the reasons why it is of great importance to keep the bottoms
of war-ships clean. How
often is your present ship put into dry dock ?
8. If it were possible to instal enough power to drive a submarine at high
gpeed, would you not expect that the speed below water would be greater
than the speed on the surface, in view of the absence of wave- making
resistance 1
9. Which is the more economically propelled at full speed, a duck or

APPENDIX. 331
10. Suppose one engine of Drake broke down, at what speed do you
estimate she could go, (a) "with all despatch," (I) "with despatch,"
" with moderate ?
"
(c) despatch
Ans. (a) 181-19 knots (1) 17-17 knots ; (c) 141-15 knots.
;
11. A
twin-screw Atlantic liner in mid Atlantic breaks one propeller
shaft, and it is stated that she finishes the voyage with one engine at three-
quarter speed. How do you account for a falling-off of only a quarter the
speed, with a reduction of one-half the power ?
12. Calculate the value of the coefficient V -r V/L in the following ships,
viz. :
Lord Nelson 410 ft, 18 knots. Ans. 0*89.

Dreadnought 490 ft., 21 knots. 0'95.
Invincilh 530 ft, 25 knots. 1-09.
Swift 345 ft, 3G knots. 1'94.
Mohawk 270 ft.,
33 knots. 2-01.
13. If a speed of 35 knots be required in a cruiser and a ratio of V -* VL

of 1'2 is
accepted, what is the length of ship required? Ans. 850 ft.
CHAPTER XXII.
1. Discuss in general terms the difference between the conditions under
which a naval architect has to design and build a war-ship and an architect
to design and build a large building, or a civil engineer a bridge.
2. Discuss in general terms the conditions under which a war-ship designer
has to work as compared with a naval architect designing large steamers of
the Mercantile Marine.
3. State the advantages of having a forecastle in a war-vessel designed
for high speed.
4. What advantages from a gunnery point of view are found in a war-
ship of high freeboard, in addition to the advantages of comfort and

seaworthiness ?
5. Make a comparison between the dimensions and particulars of H.M.S.
Drake and Duncan of nearly the same displacement, and indicate as far as
you can the reasons for the differences in each case.
G. State the reasons which have caused large Atlantic liners to be
considerably larger than the largest cruisers.

7. Discuss the reasons for the different type of war-ships introduced with
the Dreadnought and Invincible as compared with their immediate predecessors
Lord Nelson and Minotaur.
APPENDIX B
MEMORANDUM EXPLANATORY OF THE PROGRAMME OF
NEW CONSTRUCTION FOR 1905-1906, WITH DETAILS NOT
INCLUDED IN THE NAVY ESTIMATES FOR 1906-1907.
PARLIAMENTARY PAPER No. CD. 3048/1906.

THE Board of Admiralty, having approved of the leading principles to be
embodied in the various types of vessels, appointed a Committee of Naval and
Scientific Experts of large and varied experience as an advisory body to assist
them in coming to a conclusion upon various important features in connection
with these types.
In the instructions to the Committee it was clearly stated that it was
"no part of the function or purpose of the Committee to relieve the
Director of Naval Construction of his official responsibility for the design of
the slu'ps."
Having carefully examined and discussed various types of battleships and
armoured cruisers, and having had six alternative designs of each class under
consideration, the Committee unanimously decided to recommend the designs
which, after the conclusions of the Committee had been further discussed
with the Commanders-in-Chief of the Channel and Atlantic Fleets, in
January, 1905, and concurred in by them, were ultimately adopted.
The principal dimensions of the designs are shown in the accompanying
Table.
The principal features of the Dreadnought design, which has now been
built to, are as follows :
Armament.
Ten 12-in. guns and 27 12-pr. Q.F. anti-torpedo boat guns and 6
submerged torpedo tubes.
In arranging for a uniform armament of 12-in. guns it became at once
apparent that a limitation to the number of guns that could be usefully
carried was imposed by considerations of the blast effect of the guns on the
crews of those guns adjacent to them. It is obviously uneconomical to place
the guns in such relative positions that the blast of any single gun on any
permissible training should very seriously hamper the use of one or more of
the remaining guns.
While it is recognized that broadside fire is held to be the most important
in a battleship, all-round fire is also considered of
great importance, since it
lies in the power of an enemy to force an opponent, who is anxious to
engage, to fight an end-on action.
In the arrangement of armament adopted, 6 of the guns are mounted
in pairs on the centre line of the ship the remaining 4 guns are mounted
;
332
APPENDIX. 333
in pairs on the broadside. Thus 8 12-in. guns (80 per cent, of the main
armament) can be fired on either broadside, and 4, or possibly 6, 12-in.
guns (or 60 per cent, of the main armament) can be fired simultaneously
ahead or astern.
In view of the potentialities of modern torpedo craft, and considering
especially the chances of torpedo attack towards the end of an action, it
is considered
necessary to separate the anti-torpedo boat guns as widely as
possible from one another, so that the whole of them shall not be disabled
by one or two heavy shells. This consideration led the Committee to
recommend a numerous and widely distributed armament of 12-pr. Q.F. guns
new design and greater power than those hitherto carried for use against
of a
torpedo craft.
In order to give the ship good sea-going qualities and to increase the
command of the forward guns, a forecastle is provided, giving the ship a
freeboard forward of 28 ft. a higher freeboard than has been given to any
modern battleship.
The main armour belt has a maximum thickness of 11 in., tapering to
6 in. at the forward and 4 at the after extremity of the vessel; the
in.
redoubt armour varies in thickness from 11 in. to 8 in. ; the turrets and fore
conning tower are 11 in. thick, and the after conning tower is 8 in. thick ;
the protective deck varies from 1| in. to 2| in. in thickness.
Special attention has been given to safeguarding the ship from destruction
by underwater explosion. All the main transverse bulkheads below the
main deck (which will be 9 ft. above the water-line) are unpierced except
for the purpose of leading pipes or wires conveying power. Lifts and other
special arrangements are provided to give access to the various compartments.

Mobility of forces is a prime necessity in war. The greater the mobility
the greater the chance of obtaining a strategic advantage. This mobility
is represented by speed and fuel endurance.
Superior speed also gives the
power of choosing the range. To gain this advantage, the speed designed for
the Dreadnought is 21 knots.
The question of the best type of propelling machinery to be fitted was
also most thoroughly considered. While recognizing that the steam turbine
system of propulsion has at present some disadvantages, yet it was deter-
mined to adopt it because of the saving in weight and reduction in number
of working parts, and reduced liability to breakdown its smooth working,
;
ease of manipulation, saving in coal consumption at high powers and hence

boiler-room space, and saving in engine-room complement and also because
;
of the increased protection which is provided for with this system, due to
the engines being lower in the ship advantages which much more than
:
counterbalance the disadvantages. There was no difficulty in arriving at a

decision to adopt turbine propulsion from the point of view of sea-going
speed only. The point that chiefly occupied the Committee was the question
of providing sufficient stopping and turning power for purposes of quick and
easy manoeuvring. Trials were carried out between the sister vessels Eden
and Waveney, and the Amethyst and Sapphire, one of each class fitted with
reciprocating and the other with turbine engines; experiments were also
carried out at the Admiralty Experimental Works at Haslar, and it was
334 APPENDIX.
considered that all requirements promise to be fully met by the adoption of

suitable turbine machinery, and that the manoeuvring capabilities of the ship,
when in company with a fleet or when working in narrow waters, will be
quite satisfactory.
The necessary stopping and astern power will be obtained by astern
turbines on each of the four shafts. These astern turbines will be arranged
in series, one high and one low pressure astern turbine on each side of the
ship, and in this way the steam will be more economically used when going
astern, and a proportionally greater astern power obtained than in the Eden
and Amethyst.
The ship has a total coal-bunker capacity of 2700 tons, and with this
amount of coal she will be able to steam about 5800 sea miles at economical
speed, and about 3500 sea miles at 18J knots after allowance has been made
for bad weather and for a small amount of coal being left in the bunkers.
Stowage for oil-fuel has been arranged for, but oil-fuel has not been taken
into account in estimating the radius of action, which, of course, will be
greatly increased thereby.
Considerable attention has been devoted to the arrangements for the
accommodation of the officers and men. In view of the increasing length
and greater power of modern ships the usual position of the Admiral's and
Captain's quarters right aft is becoming more and more open to objection.
Up to the present the principal officers have been berthed at the furthest
possible distance from the fore bridge and conning tower, where their most
important duties are performed. It has been decided that in this ship the
Admiral's and Captain's quarters shall be placed on the main deck forward,
near the conning tower; also that the officers' quarters shall be placed
forward, both on the main deck and on the upper deck, in the fore part of
the ship. Ample accommodation for the remainder of the crew is available
on the main and lower decks aft.
"Invincible" Class.
The principal dimensions of this class will be found below.

INDEX
ADMIRAL armour of, 154
class, Butt fastening for wood deck. 51
metacentric height, 221
, Butts, laps, etc., of plating, 18
Agamemnon (new), 165
Air escapes for flooding, 120
Air plugs in manhole covers, 59, 120
CALCULATIONS for area and volume, 186
Air pressure test, 71 Canopus class, protection of, 160
Capacity of pumps, 109
Ajax, 169
Captain, stability of, 242
Angle of maximum stability, 243
of vanishing stability, 243 Carlings, 44
Cast steel, tests of, 15
Annealing of steel plates, 13
Anti-corrosive paint, 140 Caulking sheathing, 144
steel plates, etc., 20
Anti-fouling paint, 145 wood
Area of curvilinear figures, 185 decks, 62
- of rectangle, etc., 185 Cement, 146
in bolt holes of sheathing, 144
Armour bars, 177 Centre of buoyancy, 206
bolts, 182 on
, influence stability, 213
, compound, 153 of flotation, 234
, figure of merit, 151 of gravity, 206
, Harveyed, 155 Centurion (new), 169
, Krupp, 168 Chambers, water, to diminish 257
rolling,
protection, 148
177
scuttles,
Change of trim, 231
Circle, area of, 185
wrought-iron, 150
Circles, turning, 268
Arrogant, section, 35 Clutches to steering gear, 106
stern, 86
,
Coal bunkers, ventilation of, 126, 129
, turning, 271 Coal stowage as affecting stability, 220
Aatrcea, turning, 271 Coefficient of fineness, 200
Audacious, 169 Cofferdam bulkhead, 65
Automatic doors, 78
Cofferdams, 179
valves, 77 Collision bulkhead, 65
Auxiliary steering gear, 108
Colossus, 169
Comparison, law of, 287
BACKING, teak, behind armour, 181 Composite system, 131
Balanced rudders, 90
Battens for observing rolling, 263 Compound armour, 153
Conditions of equilibrium, 207
Battle-ships, sections of, 21-24 of stable equilibrium, 209
framing of, 22-29
Conqueror, 169
,
plating of, 56
Copper sheathing, 141
,
Beams, construction of, 42 Cork cement, 147

strength of, 1
Corresponding speeds, 284
,
support to, 45
Corrosion, 138
,
Eellerophon (old), protection of, 150

Countersinking, 17
(new), protection of, 167
County class of cruisers, protection of, 173
Bilge keels, action of, 255 Crank ship, steadiest in a seaway, 220
construction of, 257
Cressy class, protection of, 171
,
Bilged compartment, sinkage due to, 204

, turning of, 272
change of trim due to, 239
Cruiser, first class, construction of, 30
,
Black Prince, protection of, 173

Cruiser, third class, construction of, 35
BM, longitudinal, 231 , watertight subdivision, 64
, approximation to, 231
,
Cruisers, protection of, 170
BM, transverse, 210 Curve of displacement, 199
, approximation to, 211
,
of horse-power, 279
Board margin, 192, 303 of stability, 242
Bolts to armour, 182
of tons per inch, 200
to sheathing, 144
Bolts to wood deck, 52
Bottom plating, 54 DAMAGE, change of trim due to, 239
Bracket frame, 26 , sinkage due to, 204
Breadth of ship, definition of, 192 Deck planking, 51
Bulkhead armour construction of, 160 plating, 47
Bulkheads, numbering of, 70 protection, 148
watertight, 65
,
Dee i> load condition, definition of 193
Bulwark class, protection of, 161 Design of war-ships, 298
Buoyancy, 196 Destroyer, framing of, 39
centre of, 206
, , plating of, 62
reserve of, 203
, Devastation, protection of, 160
335
336 INDEX
Diadem and Powerful, comparison of I.H.P., Greyhound, towing experiments on, 278
286 Grummets, 52, 144
Diadem, section of, 32 Gunmetal composition, 140
turning of, 27 '2
,
Difference of draught, salt and river water. HALF beams, 44

202 Harfield's steering gear, 100
Displacement, 197 Harvey process of manufacturing armour,
curve of, 199
,
155
Diverging waves, 283 Hatchways, 44
Docking, conditions of stability, 226 Heel caused by flooding wings, 240
Doors, watertight, automatic, 78 by turning, 266
, hinged, 72
, Hercules, 169
,
horizontal sliding, 73
, Hinged watertight doors, 72
, quick-closing, 73
, Hogging, 4
, vertical sliding, 73
, Horse-power, definition of, 281
Double bottom, access to, 58 , effective and indicated difference
flooding of, 114
, between, 281
pumping from, 114
,
:
, indicated, curve of, Drake, 280

, value of,
22 , Powerful and Diadem, 286
,
Dowel in deck planking, 52 Hyacinth and Juno comparison, 306

Downton hand pump, 114
Drainage, 109 INCLINING experiment, 216
Drake curves of I.H.P. and revolutions, 280 Indefatigable, 175
, protection of,
171 Indiarubber washers to armour bolts, 184
Draught, change of, due to different density to watertight doors, 72
of water, 202 Indicated horse-power, definition of, 281
, light, normal, deep,
193 Drake, curve of, 280
,
Dreadnought, 165 estimates of, 288

,
turning, 272
, Powerful and Diadem compared,
,
Dry dock flood to magazines, etc., 117 286

Duke of Edinburgh, protection of, 173 Indomitable, 173, 334
turning of, 272
, Inflexible (old), protection, 151
Duncan class, protection of, 161 (new), 173, 334
section of, 22
, stability, 221
Dynamical stability, 246 Initial stability, 209
Inner bottom, construction of, 56
EDDY-MAKING resistance, 282 corrosion of, 146
,
Edgar, effect of shallow water on speed, 293 , manholes in, 58,

,section, 31 Inspection of watertight doors, 78
Edge laps, 18 Insulation of copper sheathing from hull, 144
strips, 19 Intercostal framing, 25
Effective horse-power, 281 Invincible (new), 173, 334
Elastic cup washer to armour bolts, 182 Iron Duke, 170
Equilibrium, neutral, 208 Isochronous rolling, 254
,stable, 208
, unstable, 208 JOINTS, lap and butt, 19
Eurydice, stability of, 24G Juno and Hyacinth, comparison, 306
Exfoliation of copper sheathing, 141
KEEL, flat plate, 56
FIGURE of merit of armour, 151 , vertical, etc., of battle-ship and cruiser
Fire main, 118 compared, 4
First class battle-ship, construction of, 22 , vertical, 23
, shaft brackets, 96 King Edicard VII., armour, etc., 163
- 80 stem, section, 23
-,
stern, 83 shape of stern, 86
,
cruiser, construction of, 32 King George V., 169

rudder and sternpost, 93
, Kingston valve, 110
Floating dock, information for, 241 Krupp armour, 158
Flood valve, 116, 123 Krupp's formula, 159
Flooding of double bottom, 117
of magazines, 117-122 LAP joints, 19
of wing compartments, 117 Latitude in weight of steel, 12
Formidable class, protection of, 161 Leak, rate of inflow of water, 109
Fouling, 141 Length of ship, definition, 192
Framing of battle-ship, 22-29 Length, influence of, as regards economy of
of destroyer, 39 propulsion, 284
of 32
first class cruiser, Light condition, definition of, 193
second class cruiser, 33
of stability, 219, 248
,
of
sloop, 38 Linoleum, 53
third class cruiser, 35
of Lion, 175
Freeboard, influence on stability, 245 Load, deep, condition, definition of, 194
of merchant ships, 204 stability, 219
,
Free water, influence of, 227 Load, normal condition, definition of, l'J4
French war-ship, section of, 158 stability, 219
,
Frictional resistance, 282 Local strains, 8

to rolling, 255 Log, stability of, 212
Froude's experiment with Greyhound, 278 Longitudinal bulkheads, 68
law of comparison, 287 framing, 23
metacentre, 231
GALVANIC action, a cause of corrosion, 140 metacentric height. 231
Galvanizing, 139 strength of ships, 3
INDEX 337
Longitudinal system of construction. 6 Rectangle, area of, 185

Lord Nelson, 22, 165 Reserve of buoyancy. 203
Resistance of ships, 278
MAGAZINES, flooding of, 117, 122 to rolling, 255
, ventilation of, 133 Resisted rolling in still water, 255
Magenta, stability of, 247 Rivet steel and rivets, tests, 14
Main drain, 110 Riveting in outer bottom, 55
Majestic class, armour, 156 Rivets and rivetting, 17
through Suez Canal, 235 Rolling among waves, 258
Mallock's rolling indicator, 263 in still water, 252
Manhole through bulkhead, 60 "
Rosbonite
" for
fresh-water tanks, 147
to double bottom, 58 Roijal Arthur, stability of, 247
Manual power for steering, 108 Royal Sovereign, armour protection of, 154
Measured-mile trials, 291 midship section, 21
,
Mensuration, rules of, 185 stability of, 247

,
Merchant steamers, stability of, 221 supports behind armour, 180

,
Metacentre, longitudinal, 231 Rudders, 86, 87, 89

, transverse, 209 Rust, 138
Metacentric diagram, 213
shapes of, for various ships, 219
, SAGGING 4 ,
Metacentric height, longitudinal, 231 Screw steeringgear, 102, 106

transverse, 209, 221
, Screw-down, non-return, and flood valve, 116
Metacentric heights for different ships, 221 Screw-down valve, 116
Miantonomah, stability of, 247 Sections of steel, 16
Mild steel, tests of, etc., 10 Sentinel and Topaze, comparison, 307
Mill scale, removal of, 138 Shallow water, influence on speed, 293
Minotaur (old), 150 Shaft brackets, 96
(new), 173 Sheathed ships, planking of, 144
Moment of inertia of area about an axis, 210 Sheathing, cost of, 145
of weight about an axis, 253 Shift of butts, 54
Moment to change trim 1 inch, 233 Side lights, 76
approximation to, 233
, Simpson's rule for areas, 186
Monarch, stability of, 243 Sinkage of vessel due to damage, 204
(new), 169 due to passing from sea to
Monmouth, protection, 173 river water, 202
Sluice valves, 77
NARROW belt, disadvantages of. 157 Solid plate frames, 26
Naval brass, composition of, 143 Spacing of rivets, 20
Navy List displacement, 192 Spare gear to steering gear, 105
Neptune, 169 Speed of ships on measured mile, 291
Neutral axis, definition, 6 Splinter nettings, 179
equilibrium, definition of, 208 Stability, curves of, for various ships, 247
Non-return drain valve, 117 , dynamical, 246
and flood valve, 116 , initial, 209
Normal load draught, 193 -,statement, 251
,
statical curve of, 242
OBSERVATIONS of rolling, 261 Stable equilibrium, definition of, 209
Ollis's steering gear, 102 Steadiness of ship at sea, 253
Orion, 169 Steam steering gear, 103
Orlando, turning of, 271 Steel castings, tests of, 14
Oscillations of ships, 252 ,high tensile, 11
Outer bottom plating, 54 ,
sections of, 16
,tests of, 11
PANTING, 8 Steering by manual power, 104
Period of rolling, 253 by steam, 103
different ships, 255
, Steering gears, 98
Phosphor-bronze castings, tests, 15 Stems, 80
composition of, 15
, Sternposts, 83
Pickling of steel plates, 13 Sterns, shapes of, 86, 87
Pillars, fixed, 45 Still water, rolling in, 252
,portable, 47 Strength of ships, 1
Pitch of rivets, 20 Structural strains. 1
Pitching, 264 Submarine boat, stability of, 236
Powerful and Diadem, comparison of J.H.P.. Suez Canal, passage through, of large ships.
286 235
Preservation of ships, 139 , tonnage, 195
Princess Royal, 175
Progressive speed trials, 291 TESTS, for rivets, etc., 14
Propulsive coefficient, 282 for steel plates, etc., 11
Protection of ships, 148 for strength of bulkheads, 70
Protective decks, construction of, 50 for watertightness, 70
Pumping and flooding, 109 Temeraire (old), protection of, 150
capacity of large ship, 109 Third class cruiser, construction of, 35
Queen Elizabeth, 175 Thunderer, 169
Tiger, 175
Queen Mary, 175 Tonnage, 194
Questions, 315 Tons per inch immersion, 199
approximation, 200
,
RADIUS of gyration, 253 Topaze and Sentinel, comparison, 307

Range of stability, 243 Transverse bulkheads, 05
Rapson's slide steering gear, 98 framing, amidships, 25
338 INDEX
Transverse framing, ends, 29 Ventilation with motor fans, 132
metacentric height, 209 steam, 130
for various classes, 221 Victoria, loss of, 309
strength, 8
system of construction, 35 Warrior (old), armour of, 148
Trapezoid, area of, 185
(new), 173
Trapezoidal rule, 185
Treatment of mild 12 Washer, elastic cup, to armour bolts, 184
steel,
Trials of H.M. 194 Water-chambers, 257
ships,
on measured mile, 291 Water, density of, 198
difference of draught, salt and river,
Triangle, area of, 185
,
202
Trim, change of, 232
moment to change 1 inch, 233 Water-testing, 70
,
Watertight bulkheads, 65
Triumph, H.M.S., 163
Tumble home, effect on stability, 249 doors, hinged, 72
, horizontal sliding, 73
Turning of ships, 265
trials in Royal Navy, 274 , vertical sliding, 73
flats, 50
Twin screw ship, turning of, 274
frames, 26
UNRESISTED 252 hatches, 53
rolling in still water,
inner bottom, 58
Unstable equilibrium, 209
subdivision, 65
Upper deck of a cruiser, 47 Watertightness, 70
VALVE, screw-down, 116
Wave-making resistance, 283
and Wave, ship on, 4
, screw-down, non-return, flood,
Waves, rolling among, 258
116
Wood backing, 181
Ventilation, artificial, 126
126 decks. 51
, natural,
of coal bunkers, 1 26, 129 sheathing, 144

of magazines, 133
of spirit room, 126 ZINC protectors, 141
IN* GREAT J51UTAIN BY WILLIAM CLOWES AND SONS, LIMITFP, BFCrT.F..S.

UBgASÛSe'
^tfttff^
JU
UNIVERSITY OF CALIFORNIA LIBRARY
"L fc

Warships - A Text Book - Atwood (1917)

Uploaded by

Copyright:

Available Formats

Warships - A Text Book - Atwood (1917)

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Warships - A Text Book - Atwood (1917)

Uploaded by

Copyright:

Available Formats

WAR-SHIPS

THE CONSTRUCTION, PROTECTION, STABILITY,

EDWARD L. ATTWOOD, M.lNsr.N.A.

WITH NUMEROUS DIAGRAMS

LONGMANS, GREEN AND CO

Goytiay Gordia of the Spanish Navy in .

Sixth Edition January,

Approved for issue to the Officers' Libraries of II. M. ships

officers' standpoint, and certain parts of the subject have been

requirements. I am indebted to several officers for valuable

suggestions as to the subject-matter likely to be of most use.

space to note particulars and details peculiar to the ship on

carefully drawn diagrams. A number of questions have been

I am indebted to the Controller of the Navy, Rear- Admiral

F.E.S., LL.D., for his kindly interest in the undertaking. The

Although prepared primarily for naval officers, yet it is

believed that the work will prove a useful introduction to the

subject of Naval Architecture for apprentices and students at

the Iloyal Dockyards and elsewhere.

ROYAL NAVAL COLLEGE,

it has been found necessary to embody a considerable number

of modifications in the present edition. The important 1906

Admiralty memorandum concerning the design of H.M.S.

Dreadnought has been retained as an Appendix.

IT. TESTS OF STEEL, ETC., SECTIONS, RIVETS, JOINTS, ETC. . . 10

ITT. FRAMING OF VARIOUS TYPES OF SHIPS 21

TV. BEAMS, PILLARS, AND DECKS 42

VIT. STEMS, STERNPOSTS, RUDDERS, AND SHAFT BRACKETS ... 80

XX. THE TURNING OF SHIPS 265

XXI. THE RESISTANCE AND PROPULSION OF SHIPS 278

XXIII NOTES ON THE Loss OF H.M.S. VICTORIA 309

APPENDIX A. QUESTIONS 315

APPENDIX B. ADMIRALTY MEMORANDUM ON H.M.S. DREAD-

THE STRENGTH OF SHIPS

BEFORE considering in some detail the structural arrangements of

huge beam or girder. In order to bring out the various points

arrangement of the material to make the ship strong enough, we

If, now, we placed six such planks as in Fig. I,

further, we should havea log 12 in. wide and 12 in. deep of

homogeneous material, and we should find that such a beam

compressed and the lower layers are being stretched, and it is

of the beam standing thegreatest weight at the centre, without

of lattice-work. (Charing Cross Eailway Bridge is an instance of

Longitudinal Strains. In arranging the structure of a ship

In attempting to make these strains the subject of calculation

KEEL OF BATTLESHIP KEEL or CRUISER

accurately determine the maximum strains the ship may reasonably

unsupported, or, again, if the ends only were supported. It would,

ACROSS WAVE TROUGH

In a well-constructed ship these are only tendencies, and the

assumptions give a certain stress on the material, and the ship is

adoption of special steel of high tensile strength. Instead of the

inch, this special steel has to stand a tensile test of between 34

contribute anything like its fair share to the structural strength,

design of more recent large cruisers there was the adoption of a

tinuous, but the transverse framing is made the continuous part

either side of the keel. The longitudinal strength is not so

in the latter case. There are several reasons which contribute to

assistvery materially in the structural strength. Most merchant