The design of seakindly ships

AU6.

AkCF-IIEF

Numbers refer to Bibliography at the end of thó Pap&. VV

Lab

g$

_Luwkunde

.'a°1ior

NORTH-EAST COAST INSTITUTION OF ENGINEERS AND S i El

'BOLBEC. HALL, NEWCASTLE UPON TYNE

ADVANCE COPYSUBJECT TO REVISION

This paper is issued in advance on the understanding that neitherthe whole nor any portion of It shall be published until after it has been read at a g neral meeting of the

Institution. -

-It is to be read at a meeting of the Institution to be held in Newcastle upOn Tyne on 29th June 1950.

THE.. DESIGN OF $,

LY SHIPS

By J. L. KENT, C.B.E., Member 29th June, 1950

SYN0PSIS.The qualities of a seakindly ship are enumerated, and with data

obtained from experienced seamen, the effect of fullness upon the best direction for the ship to meet the weather to ensure easy handling in a seaway is deduded. The effect of the wëOther üpon steerihg is'discussed dnd the essential features of hull form for easy steering in rough weather are deduced from model experi-ments and öbservätion&made in ships at sea. The various motions of a vessel

in a seaway are considered and attention called to those portions of hull shape

and loading which produce uneasy motions of the ship in storms. The causes of shipping seas and -spray are examined, and the different features in hid! design

that addio.jhe quantity of water and spray taken inboard, are pointed out. The

reasons for eddies, back' draughts and .excessively high wind speèds over open

decks àre dircussed and suggestions made which would add to the safety and

comfort of passengers and crew traversing such decks in stormy weather.

-Introduction

APAPER

on "Sea-Kindliness añd ShipDesi," by Captain K.

Macdonald and Dr. E. -V-. Telfer' was read before this Institutjòn -in i 93:8 and tovoked considerable discussion, indicating a

wide-spread interest in this subject.

In the hope of adding a.litt]e more to

the knowledge of those qualities of ship-design which prOmote the

seakindliness of vessels in a seaway, this paper, has- been written from

informátion' gathered by the Author at sea

in various types of ship,

A seakindly ship is one which rides the seas in rough weather, with 'decks

free of seawater: that is, green seas are not shipped and little spray comes

matter in which direction the. wind and waves, meet the ship,

she will stay on her course with only an occasional uin of helm, she will respond quickly, to small rudder, angles and maintain :a 'fair speed without .sinmniing,

abnormal fluctuations in shaft torque, or periodic racing.of her engines. Open decks will be easy to traverse in ail weathers, without danger or discomfort to her passeñgers and crew, and hér behaviour iii a seawayi.e. her.rolling, pitching, yawing, heaving, surging and leeway driftwill 'be. smooth. and free from bauilcs orshocks. Expert. händliñg of the ship by her master and crew will álways bd required'.to pröduce a seakindly performance .by any vessel in rough weather; and in this paper (which is confined to .the influence of ship form upon hér seakihdly, qualities) expert handling is assumàd to be always.

COI?PlCEN?.

Fig. 1 is a useful rough guide to the direction in which to navigate a ship into the weather, to secure her best behaviour in a seaway.

So that the Author could judge for himself the differences in the ship's

behaviour, a practical demonstration was given to him in each ship, by bringing the vessel off her true course to the desired direction for a short time. During these demonstrations the changes in the behaviour of each of the different ships fully confirmed the master's experience; less water was shipped, the motion was easier, and the helmsman had less trouble in keeping the vessel on her course. In this paper an attempt has been made to account for these

factsat least in partand to draw attention to those portions of a ship's

design which affect her behaviour in rough weather.

ii auai

o

Ship Type Direction of weather

Q.s.s. Mauretania .. Liner point weather bow 595

Q.s.s. Berengaria

..

T.s.s. Oroya .. S.s. inkosi .. ..

S.s. London Mariner

Liner and cargo Liner and cargo

Fast cargo

i to 1 ,, ,, ,,

i to 1f ,, ,, ,,

1f points weather bow

689

73

T.s.s. Montcalm .. Liner and cargo 1f ,, ,, ,, .735

T.s.s. Oropesa .. Liner and cargo 1f ,,. ,, ,, 74i

S.s. San Alberto .. Tanker 1f ,, ,, ,, 738

S.s. San Tirso .. Tanker 2 ,, ,, ,, 778

S.s. San Gerardo .. Tanker 2f ,, ,, ,, 835

418 THE DESIGN OF SEAKThIDLY SHIPS

The seakindly qualities of a ship have been considered under four headings,

namely:-(i) Steering.

(ü) Ship movements in a seaway.

Shipping seas and spray.

Safety and comfort of passengers and crew on weather and other

open deck spaces.

An exhaustive consideration of all these qualities would be impossible within the limits of this paper, and only their more important bearings on the ship's seakindliness have been touched upon.

3. Direction of the Ship into the Weather for Easiest Steaming Conditions During stormy weather on the Atlantic passage, in vessels of different types,

the Author asked each ship's master in which direction experience had taught

him to navigate his particular

ship to ensure the easiest control

and smoothest ship movements in the weather prevalent at that time.

The directions of the wind and

waves, relative to the ship, at which

each shipmaster would have

pre-ferred to drive his vessel are given

in Table 1, and plotted to a base

of block

roughly on a straight line (Fig. 1). Fig.l Best Direction of Weather to Ship TABLE 1Best Direction of Weather to Ship

THE DESIGN OF SEAKINDLY SHIPS - '19

--. 4. Steering in Rough Weather

-Most ships steaming into wind and waves are in unstable equilibrium about a -vertical axis, and they are prevented from -yawing off their courses by the contmual use of the helm from moment to moment In many ships, corn paratively stable equilibrium is found by experience to be a course with the -wind and waves a pöiht br two oh-the bow when- oñly-an occisiónal change

of helm is needed to -keep- the ship on her couÎse. The principal forces

producmg a yawing couple on a ship m rough weather are

-(i) Wmd frce on the ship s superstructure and lateral water resistance On the vessel's i.mderwater body, due tO a small angle of yaw to

her direction of motion. - - -

-(u) Hull resistance and screw thrust during a roll, and also to a maI1er

extent during a pitch. - - .

(iii) Differences in thrust between port and starboard s5rews in twin and quadruple screw ship. -.

5. Wind and Water Forces which Produce a Yawing couple on the Ship's Hull.

When the wind is blowing at an angle to the bOw, the ship proceeds with its longitudinal middle line plane at a small angle of yaw to its direction of motion. Assuming that the rudder angle is zero, then the forces acting upon the. vessel are (Fig. 2) :the resultant relative wind force W, at an angle

- tô the ship, acting through the oentre of effort C; the rösultant water resistance

CREw 7h7,. . -.

--- --- Fzg.2

R at angle fi, acting through the centre -of lateral resistan L; and the

thrust-of the screw. along the centre-line plane thrust-of-the ship. Resolving the wind and water forces along and at right angles to the niiddle line plane of the vessel,

these forces reduce to (Fig.

3):-a couple R sin fi x (CL).

-the algebraic sum Qf -the screw thrust and -the resolved wind and

- water forces acting along the ship's longitudinal axis, namely, T - W cos

- R cosfi.

-A small force W sin - R sin fi such that the tangent of the angle

-W sin aR sin fi.

I

IC

.,. C'.

ó.

Fig. 4Positions of Centres of Wind Pressures

lk1

IlIUM

.IUUII_- I

I'll...

vn. w

Fig. 5Positions of Centres of Lateral Resistance of Hull

\

\

:

\

C-420 THE DESIGN OF SEAKINDLY SHIPS

For steady motion the angle of yaw ifr is always to windward of the ship's

direction of motion and the unbalanced couple R sin fi X (CL). must be

balanced by an equal and opposite couple applied by the rudder. When

C and L coincide, there is no turning moment on the ship and no rudder angle

is required to keep the vessel on her course. When L is forward of C, the

ship will tend to turn into the wind and the rudder angle will be to leeward. When L is aft of C, the ship will tend to fall off the wind and the rudder will

be to windward. For the ship to proceed steadily in the desired direction, the rudder should be used to give just sufficient turning moment on the vessel

to balance R sin fi X (CL). The helmsman endeavours to do this, by keeping

his compass bearing at a given course setting. But the compass is so arranged

that its zero reading is with the "lubbers' line" in the fore-and-aft line of the ship, which takes no account of the small yaw angle between the true course and the apparent direction of motion. In stormy weather this gives rise to a

small error to leeward in the ship's desired courseusually referred to as "drift "which is corrected daily in most vessels by a new course setting,

after her position has been checked by the sun or by other means. Except

00

Fig. 6Ship Turning Moments at Different Wind Speeds for Liner

-TT UUU

THE DESIGN OF SEAKINDLY SHIPS 421

steady for more than very short intervals of time, so that the rapidity of

movement along the ship of the centres of resultant wind pressure C, and

lateral water resistance L, with change in wind direction or angle of yaw, will be

a criterion of the rapidity with which the rudder must be moved to keep any particular ship on her course in rough weather.

6. Model Experiments

In a paper by Dr. Hughes2 the positions of C with change in wind direction,

and of L with change in yaw angle, are given for several ship models. For

three of these models (i) a liner (ii) a cargo ship and (iii) a tanker, the movements of C and L along the middle lines of the models are reproduced in Figs. 4 and 5. lt will be seen that the centre of wind pressure C moves steadily from forward

to aft, as the wind direction changes from dead ahead, through amidships, to dead astern. Also that the fore-and-aft position of C for any given wind

direction, does not differ greatly between ship and ship, in spite of the very great differences in the upper works of the three vessels. The fore-and-aft

positions of the centre of lateral resistance L with change in yaw angle (Fig. 5), for each ship model, are very different, however, the most noticeable being the

tanker model, where L first moves forward with increase in yaw angle, and then quite suddenly moves aft. From the data given in Dr. Hughes's paper2, the

couple R sin fi x (CL) (Fig. 3), has been calculated for each of the three ships, for relative wind speeds between 20 and 70 knots, and for wind directions from

forward to aft. The screw thrusts were obtained from data taken at sea on these three ships by the Author, and the angles of yaw (*) obtained for each wind direction and speed. The unbalanced couples so calculated are shown in Figs. 6, 7 and 8, as ratios of the lateral broadside water resistance of each

ship multiplied by its length between perpendiculars. These unbalanced

couples are the turning moments on the ship which must be supplied by the

rudder, if the ship is to proceed steadily on her course, with ail wind and water forces in equilibrium. (The actual application of the turning moment to the ship by the rudder has been assumed to leave the resultant wind and water

forces and their directions unchanged).

(i) Liner. (Fig. 6.) For all relative wind speeds up to 70 knots and relative

wind directions up to 20°, steady motion was obtained with a yaw of less than

one degree, by a very little turning moment applied by the rudder turned to

windward. This vessel can therefore steam close into the wind's eye with

w W0 SSOCTION U..*'Ifl 15 SNIP.

I

422 THE DESIGN OF SEAKINDLY SHIPS

very small helm angles, and rapid fluctuations in either wind velocity or

direction will not affect the ship's steering in all winds fine on the bow. In high wind speeds at wind angles greater than about 200, the ship turning moment increased steeply with increase in wind angle to the bow, so that greater rudder

movement would then be necessary to keep the ship on her course when the wind was fluctuating in force and direction. This was a model of the q.s.s. Mauretania, whose best performance in rough weather was obtained with the wind half a point (say 5°) on the bow (Table 1). During a winter voyage on this ship, not more than 3° of rudder angle was used ip winds up to B.N.8,

between 0° and 20° on the bow, and the helm angle was steady for long periods of time. With wind velocities up to B.N.12, with directions between 50° and 90° on the bow, about 5° of rudder angle was used, but the helm was then

changing between 0° and 5° fairly frequently.

(ü) Cargo Ship. Fig. 7 shows that this ship, left to herself, will always turn into the wind and a small angle of the rudder will always be required to keep her on her course. In general too, the rudder angle will always remain

on the same side,port or starboardand will be reasonably steady

relative wind directions up to about 10° on the bow. For greater wind angles,

continual helm movement will be necessary to keep the ship on her course.

This vessel was theLondonMariner, a l4fknot cargo ship, and on voyage it

was never found necessary to move the rudder from port to starboard tokeep her straight. She was easy to steer and her best point of sailing was with the wind about 10° on the bow (Table 1). The largest rudder angle used with the wind direction up to 45° on the bow and B.N.5 to 7, was only 5°. Even with the wind astern-30° on the starboard quarter and B.N.8,the largest helm angle

was only 9° with a fairly frequent rudder movement between 5° and 9°.

.,

..,

T

Fig. 7Ship Turning Moments at Different Wind Speeds for Cargo Ship

(iii) Tanker (Fig. 8). For high wind speeds the ship turning moment rises

rapidly as the wind changes from "head on" to between 15° and 20° on the

bow. It then drops abruptly to zero before again increasing. This is primarily

due to the erratic movement of the centre of lateral resistance with increase

__i

_U!.I1

-cOU

Fig. 8Ship Turning Moments at Different Wind Speeds for Tanker greater than 3°. This was a model of the tanker San Gerardo. Her easiest movements in a seaway occurred when the wind was 2 points on the bow, i.e. 30° àpproximately (Table 1), but throughout th&\royage change of helm was used continuously to keep her on her course. With the wind 30° on the port

bow, the rudder was to starboard, but as the angle of the wind to the bow became

less, the continual movement of the rudder increased. Thus the rudder angle varied between 0° and 8° to starboard with the wind 30° on the port bow, and from 50 to 25° to starboard with practically the same wind force acting between 5° and 10° ort the port bow. With the wind direction greater than

35° on the bow, the rudder was in continual movement from starboard to port.

7. Form of Ship which Causes IndWerent Steering

During winter voyages in each of these three ships, excellent steering ability

was shown by both liner and cargo ship in all kinds of rough weather, but the steering qualities of the tanker were not good. Figs. 4 and 5 suggest that to design a merchant ship which will steer easily in rough weather, it is the movement of the centre of lateral resistance with change in yaw angle, which

is of primary importance, as the movement of the centre of wind resistan

with change in wind direction is not very different for either of the three vessels. That is, the underwater shape of the ship affects her steering ability more than

the distribution of her upperworks. The principal differences in the hull shapes of the three vessels

were:-Liner hollow bow lines and fine lines aft.

Cargo Ship rather straight bow lines and fuller lines aft.

Tanker very full convex bow lines and extremely full afterlines.

When underway, the liner had no measureable dead water aft. Jn the cargo ship there was a little dead water between the water surface and the top of the

screw disc; and in the tanker a very large quantity of dead water aft was

dragged along with the ship. This dead water was very noticeable on voyage.

A

THE DESIGN OF SEAKINDLY SHIPS .423

in yaw angle from 2° to 3° (Fig. 5). With the ship running with 0° to 3° yaw

angle, she will be difficult to steer, as quite small changes in the force or direction of the wind will alter very considerably the ship turning moment to be

counter-acted by the rudder. Fig. 8 shows that in 60 to 70-knot winds, the easiest

424. THE DESiGN OF SEAXINDLY SHIPS

Blocks of wood dropped over the stern into the dead wateiV remained with the

ship for long periods of time in the tanker, and they were only left behind when the ship gave a rather more violent swerve than usuaL In the cargo ship, wooden blocks remained with the vessel for short periods of time, only if the blocks were dropped right aft and close to the side of the ship; but in

the liner the blocks would not stop with the ship, and in every attempt to

make them do so, they were immediately swept away aft.

It is possible that the initiâl movement forward of the centre of lateral

resistance L with increase in yaw angle frfor the tanker model shown in Fig. 5, was due in part to the full bow lines, and perhaps to a greater extent to the very

considerable portion of the length of the after body over which a break down in the stream flow occurred. Some support is given to the latter suggestion, in the results of wind-tunnel experiments with various air-ship models' & In these experiments the shape of the bow. was the same for models C, D, E,

R38 and Rbi, but their lengths and volumes varied. The movement of L

with increase in frwas steadily aft in the first three models, but in R38 and

RIOl the initial movement was forward. _{It would appear, therefore, that} it was the shape of the after ends of these air-ship models that affected the movement of L, and consequently their ability to steer a straight course.

Until more experiment work has been done on the redistribution ofpressure

along hull forms with change in yaw, it will not be possible to do more than

express an opinion on the best hull shape for good steering in all rough-weather conditions. In the Author's opinion, the presen of large quantities of dead water aft, is the principal cause of erratic. movements of L with increase in

yaw angle. As fris increased, a yaw angle is reached when quite suddenly.

a large quantity of dead water on the "upstream" side of the vessel, is swept away and replaced by stream-line motion, causing a sudden redistribution of the hydrodynamic pressures along that side of the hull, and producing an abiipt jump aft in L. If this is correct, then for gool steering in rough weather,

the slope of the aftei level liñes of the ship should not exceed 18021. As this i

not practicable in very full ships, or in short vessels with wide beams, the after

level lines should. be so drawn, that the: length of ship over which the slopes of the level lines aft exceed 18° are as short as possible, in order to reduce the vòlume of dead water to .a minimum

'.

8. Ship Conditions for Good Steering in Rough Weather

Fór göod sfeering in tough weather, the oentres of lateral resistance L and relative witd force C should be close together so that the ship turning moment due to the weather is always srnll As C always moves aft with increase in the angle of thé wind to th bow, L should move rapidly aft with increase in

yaw angle. At very small angles of yaw L.should be close to the stem, especially m ships with high forecastles and low poops where the centre of relative wind

force C is 'well forward for winds fine on the bow. In such vessels extreme

cut up of the bow profile and large stern trim will place L too far aft for small yaw angles and the steering will be erratic. In quite moderate winds, such erratic steering is common in tankers" flying light" with all t'nks émpty. The use of a jigger or trysail right aft, will sometimes bring C far enough aft.

to make reasonable steering possible in such vessels as drifters trimmed heavily

by the stern. While it is advisable for L t be as close to C as possible, it is inadvisable for C and L to cross one. another at large angles of yaw, as the

ship turning moment will then rise steeply each side of the" no turning moment"

point, and in weather with continually varying wind direction, the movement

of the rudder backwards and forwards to latge angles each side of the fore-and-aft pOsition will be needed to-keep the ship straight.

9 Wéàzher Conditions that Cause Dffìcu1t Steèring

Very large sind forces, with relative wind. directiòna little forward of the'

THE DESIGN OF SEAKINDLY SHIPS 425

the rudder is capable. The vessel cannot then sail on her true course. Reducing

the speed s usually useless, for although the relative wind force may be less, the rudder momept also drops as the square of the ship's speed. The ship

can then steam only with the wind direction closer to the bow or abaft the beam,

and the master usually. tacks to and fro each side of his true course, or lays

to, "head on" to the wind and waits for such weather to pass. If the ship

happens to be fitted with a staysail forward, it is sometimes possible to bring, C

sufficiently far forward by süch means to reduce the ship-turning mOment

enough to enable the vessel to proceed on her true course under the prevailing weather -conditions.

It is adyisable to give the rudder sufficient area for the helm to be able to cope with all probable wind forces likely to be met at sea. The wind usually

blows in a series of rapidly alternating guets and lulls during stormy weather on the North Atlantic. If the differences in ship turning moment are large between the lulls and the gustsas for example between 40 and 70 knots wind speeds with the wind amidships in the cargo ship (Fig. 7)the rapid change in wind force will require an equally rapid change in helm angle, if the ship is to remain on her coùrse. Such weather conditions requre a rudder that will give a large increase in ship turning moment for a small change in rudder

angle.

Rapid changes in ship turning moment with small changes in the relative wind direction, will also call fôr skilful steering. lîl thd taiiker (Fig 8) this

was, particularly noticeäble for all wind directions between "head on" and

30°, with relative wind speeds greater than 40 knots. With winds of 60 to 70 knots fluctuating in dirtiön between 150 änd 25° of the bow, good steering was impossible, for the rudder changed continually backwards and forwards from port to starboard. The liner (Fig. 6) is an éxample of a good steering ship in high winds varying m direction between 0° and 200 of the,.bow and at

speeds fluctuáting between zero and 70 knots or more.

- 2

-10. Effect of Rolling and Pitching upon Steering

A ship which is rolling heavily in a seaway, vill usually be subjected to

periodically changing yawing moments, due to the line of action of the resultant

water resistance of the hull, alternating from, side to side of the middle-line plane of the ship in which the screw thrusts. This was frequently observable

on the San Gerardo when she was rolling, the white wake of the screw race making 'a serpentine pattern on the ocean. If it were practicable to make

all the transverse sections of the ship circulär in Shape, each with their centres on the shaft line, there would be no yawing couple due to rolling. It has been claimed' that if the loci of the centroids of the immersed and emerged wedges of the transverse sections are always parallel to the middle-line plane of 'the

ship, there- will be -no yawing due to heel. This still requires 'demonstration

'by measured experiment data, however, to place the claim beyond all doubt. It is' advisable to reduce rolling as much as possible by keeping the transverse metacentric height at a sale minimum, with the added aid of bilge keels, fins,

or other devices. 'If this is borne in mind during the design stage, the ship's

steering will be-improved. ' ' .

Pitching affects a ship's steering, if -the angle of pitch is. enough to raise the stern sufficiently out of w ter for the rudder- to lose a, portion of its grip. This

frequently Occurs in vessels of the cross-channel- type when the ship's' length

is comparable with the fairly regular wave-lengths usually met in stormy weather

in confined sea areas. Such a ship on an oblique course to the waves, is periodically poised on a single wave crest, and during a pitch her rudder

momentarily comes partially out of the water. ,The tendency -of -the ship is

to set herself square to the waves if she is heading into the wind, and to" broach

to" if running before the- weather. As the '-bow 'wends, the. rudder is' fully

426 THE DESIGN OF SEAKINDLY SHIPS

or less violent corkscrew motion of the vessel, which can only be stopped by an àlteration of the course. It is advisable in such craft, to place the rudder in the 't e Of the screw if possible, and to keep most of its area low dwn.

In twin- and quadruple-screw vessels a minor cause of indifferent steering is a periodic alteration in the resultant line of action of the screw thrust from side tò side as the ship ròlh. This is less ñoticeable with turbine than with reciprocating machinery, and it IS due to the change in screw slip affecting its thrust as the propeller approaches the water surface Fast running screws of small pitch ratio are less affected than the slower turning propellers. Even

if the ship is not rolling it is important to keep the thrusts of the port and

starboard screws the same to avoid needless rudder movement.

Íl' Rudder Design

For good control by the helmsmah of the ship's steering in trough seas, - a

large change in ship turningmòment should be produced, by a small alteration in rudder angle. _{The effect upon the ship turning moment of a change in shape}

of semi balanced and unbalanced rudders, the addition of a fin and the effect

of screw race etc, have been investigated for both single- and twin-screw ships

by the late Mr. G. H. Bottomley and, the results given in a series of papers'.

Briefly stated, hef6und that in single-screw ships, increased ship turning moment

was obtained, by fitting a comparatively short fin in front of an unbalanced rudder, and that a deep narrow rudder was better than a long rudder of the same area. kcreased rudder area increased the ship turning moment and

the outline shape of the rudder had little effect upon it. Semi-balanced

rudders did not give as goòd a ship turningmoment as an unbalanced rud4er' of the same area placed behind a fin. In twin-screw ships the ship turning moment was -30 per cent less than the same rudder ares on a single-screw

ship fitted with a 11h. LOñ tuddérs which moved into the screw race

gave larger ship turning moments than narrower rudders operatmg wholly

outside the screw races Iaracti the-uppeç limit of rulder area is governed by the horsepowe &f Th'e stèeíinThine, and tlie ermissibie torque on the rudder stock; and eiperiments' by E. M. Keäry7 with à ship. model in rough

water, showed that 'in comparatively small waves, the maximum torque on the rudder stock maybe increased by 50 per cent of that for the same rudder' angle

in smooth' wáter. -.

-It is of interest to cdnsider the rudder, areas necessary to .cope with the maximum ship turning moments given m Figs 6 7 and 8 For the cargo ship (Fig. 7) a ship turning mOment of about 0 0035 would be e,q,erienced in a SO4not- wind about 10° forward of the beam, with the ship steaming with a

- 4° yaw. In Mr. Bottomley's papers fot single-scÉew ships', he giies a value of

-- 1 42 for a ship not dissimilar to the cargo ship, (16)-at speed uj, to 12 knots, for 30° rudder angle. The rudder area required to cope with the max mum ship "turning m(;ment in a 50-knot gale would therefore be

X (0035 X 5 68)-

₌

the middle line plane of the vessel. It is usual in merchant vessels tO make the rudder área X length x draught, which' should 'therefore be sufficient to

cope with medt ship turniiij morneiiIke' td be' experienced -at sea- in rough

weather.

PT 2Ship MOvements 'in, á'Seawiy,

Al the motions.oLa seakindly ship ¡h ,rough water, will be moderate in

amplitude'and free from shôck. :The shi's structure. will' not-. be unduly

strained 'by: sudden, violent 'reversals r of motion and 'all accelerations' ánd

-Fig. 9Criterion for Maximum Rolling for a 10-Knot Ship

with a 10-Second Period

The extent to which a ship rolls depends also upon the maximum rolling lever, which in turn depends upon the maximum wave slope, multiplied by

the sine of the angle between the directions of the ship and the waves. As the

direction of the ship to the waves approaches "head on ", the rolling lever will diminish in length to zero and the ship's rolling amplitudes will also

diminish.

1!ihDAhIi

viArAi

Nl1UNÒi

1VAVAi%'4M1

,Ivirk.jpi

THE DESIGN OF SEAIIINDLY SHIPS 427

12. Rolling

Of the three rotational and three translational motions performed by a ship in a seaway, the most important affecting seakindliness is rolling and much has been written about this motion. The extent to which a ship rolls in waves

of uniform length, depends upon her maximum rothng lever and natural rolling

period TR both of which are governed by her dimensions, design and loading. With the weather on her beam, if this period TR coincides with or is close to her period of encounter with the waves TE, she will roll considerably. The Tgsin a&, criterion for maximum rolling conditions has been shown to be

_TETR

where is the angle between the directions of the ship and the waves and Fig. 9

shows the direction of waves of various lengths, to a knot ship with a

10-second natural rolling period at which there will be considerable rolling. This

figure indicates that heavy rolling can be expected for wave directions about 150 each side of that direction to the ship at which there is exact synchronism between T and TE. Fig. 10 shows when this synchronism occurs, for 10-knot ships with natural rolling periods between 6 and 15 seconds in waves of different lengths. For wave directions aft of "broadside on ", there are two wave

lengths with the same period of encounter TE :the longer waves overtaking

the ship, and shorter waves in which the ship overtakes the waves travelling in

the same direction, This figure also shows the wave directions to the ship at 12 knots, with 10 seconds natural rolling period and it will be noticed that 2 seconds change in TR had a much greater effect upon the wave direction for heavy rolling, than a 2-knot change in ship speed. Therefore, to stop heavy rolling, a change in ship direction will be more effective than any practical

428 THE DESIGN OF .SEAXINDLY SHIPS

The worst rolling in any rough sea, will be with the waves meeting the ship broadside on at her natural rolling penod _{The length of the rolling}

lever is controlled by the amount of buoyancy transferred from port to starboard

as a wave passes The amount of this buoyancy increases with mcreases in wave slope ship s beam and fullness of those water lines in the vicinity of the load draught The rollmg Iver will be still further increased with decrease

in ship s displament, ie with decreased draught Large beam vessels with

full load water lines and comparatively small draught will have the worst rolling

in steep waves of Such length that they'rneet the ship "broadsideon "in her

naturäl rolling period.

Although no rolling would teke place in regular waves meeting the ship "head on ", the: pitching motion might be severe, so that the best- direction in which to meet the weather is at smc angle to the bow or quarter, where neither roll nor pitch will be large. The period of encounter will be longer with the wave direction on the quarter, than, when the ship meets the'saine

waves on the bow; and the chahces of synchronism betweeñ TR and TE will

also 'be greater. It follows that 'for the easiest rolling and pitching, the best

'direction for theship will be to meet the weather on her bow.

During the height ofa storm, ocean wave lengths rarely exceed 600 feet,, and they usually lie between- 400 feet and 600 feet. For the worst rolling conditions within these lithits Fig. 10 show&thãt increase in wave-length tends to move the

critical weather direction towards .the bow, while an increase in the' 'ship's natural rolling period sends it aft. ThUs in 400-ft. to 600-ft. waves, a 10-knot

ship with a 10-second rolling period, has its worst rolling when the weather direction covers the range 300 carli side 'of "broadside on ". Increasing the ship's natural rolling period to 12 secOnds, shifts this range of weather direction

for worst rolling, 300 aft, i.e; from "broadside on" tO 150°c from the bow;'

Fig. .10 suggests that it is advisable to. make,the shin's natural rolling ¡eriod

- r

'jç'0) " ttW

as large 'as possible. '

o

Fig. 10Synchronism between PeriOds of Encountéi and Rolling

THE DESIGN OF SEAKINDLY SHIPS 429

/k2

This period. TR'= 2ir'V fl- :where k is the transverse radius, of gyration and M the transverse metacenttic height. In practice it is oiily.M which can be controlled by the designer and ship s officers as k is settled by the ship s

dimensions to a great extent. The longer range of stability possible with high

freeboard, enables the designer to use a smaller M while still ensiring a safe dynamic stabthty so that m this respect high freeboard will add to the ship s

seakindly qualities. Some increase in the ship's ñatural rolling perio can be secured by fitting bilge keels, which increases k by adding to the virtual màss of the ship.

'The ship's natural rolling period and her period of encountef with the waves

are usually' different, and the ship rolls through angles which vary more or less regularly from a maximum to a minimum. As the damping by water resistance

is not the same for large and small ,rolls, the periods of the larger rolls will not be quiteisochronous and periodically theship will be late in recovery and roll into an on-coming wave. Under such circumstances,, if-. the waves are high and the freeboard insufficient, water will be shipped and should flooding

become serious, a comparatively small change in the ship's course will speedily

effect a cure.

-The worst type of roll is a "lurch ". This can

happen-- - (a) if there is negative transverse stability in the upright position, or

(b) if the centre of gravity is considerably above the instantaneous centre

of oscillation. -

-With deck cargoes (á) is frequently a ship condition' and the lilrch takes

place when the angle of roll becomes so large that the ship s natural heel

changes from port to starboard or vice versa In general this motion will not be periodi; and the ship will úsually settle with a heel to leewaïd; bût during

a lurch she will be completely out of phased with the seas and may roll dafigerously into'an oncohing 'wave4 With wdol deck cargoes -it-is usiiàlly possible to stow the deck load in such fashion that there is a virtuàl increase

in freeboard; and so long as none of 'this deck load breaks lôose during a

"larch ", the ship may be free of seriotis deck flooding, even in bad storms. Low-freeboard ships with negative initial transverse stability, carrying in-securely fastêned deck loads, wifi always be liable to dangerous "lurches ",

which may lead to disaster bycapsizing,the vessel.

(b) When the centre of gtavity ofihe ship is considerably above her transverse

centre of oscillation, "lurches" will -be frequent in a beam sea in stormy

weather. A. W. Johnslo has shown that very square bilges,'or very làrge bilge keels, have the effect of" anchoring ' the lower'part of the hull in space, when

the ship is rolling; and- then the instantaneous centre of oscifiation can be

considerably below the centre of gravity as 'the vessel rises from her toll to the horizontal. This has the effect of periodically thrusting the ship's side near

the load water-line, more or less squarely into the waves, and under bad weather

conditions the ship will "check" and "lurch ".'

In high short length waves, the orbital velOcity of surface water particles is

considerable A 'ship of deep draught with sharp bilge turns and efficient

bilge keels, encountering such Waves "broadside on ", will be rolled to leeward

as the crest passes and then to windward in the trough.' If the waves are

very steep, the roll to windward may become a "lurch ", more or less violently checked on the leeward slope of the- passing wave. Such rolling motions are

opposed to seakindliness and they can be improved by well rounded bilge turns

and diminished transverse metantric height. 13. Dathped 'Rollink

In many..merdiant vessels, with the low freèboard required in ordér to handle cargoes, the complete preventiön.of all' rolling would,be: dangerous in a seaway

430 Tur DEsIGN, or. SEAKINDLY S}UPS

with waves of considerable height; for seawater would swamp the ship 'as' wave

succeeded wave, and the decks and hull structure might be strained beyond their elastic limit. By allowing, the vessel to roll moderately to the waves, the stresses put upon her *ill also be moderate, although of course excessvè

rolling must be avoided. .

The effect o changes in ship form or loading, .upon the damping energy

d''g rolling, was investigated by Lieut. M. E. Serat U.S.N.", by môdel

experiments and he found dat, (other conditions being unaltered), increased

dampiñg energy was obtained when increases were made in

the midship section area, due principally to the sharper bilge turns;

the height of M orofG above the keel, due to an increase iiI ditance between the centres of oscillation and gravity

when the natural rolling period was decreased; due to increased water resistence 'occasioned by the more rapid movement of the

model.

The increased damping energy due to (3).however, is offset by the increased chances of synchronism when the period of the ship is decreased.. The aim in design to improve seäkindliness should be to prevent shock to the ship by evening out the accelerations and decelerations of her rolling motion, and this

can be done by bilge. keels 'or mechanical aids.. Anti-rolling tanks, the

mechnism of activated fips'2 and gyroscopes9 occupy valuable cargo space, and the last has a high initial cost on which the return may be negligible in

cargo ships. Bilge keels have been generally adopted in the naval and mercantile

marine, as they do not occupy cargo space and they are cheap to manufacture

and to fit. Care should be taken to see that the run of éach bilge keel is along the natural streém flow, and :(if necessary) to split them into two or more keels

along each side of the ship s bilges to avoid crossmg this stream flow The

action of such keels m damping the rolling has been described by the late Dr. G. S. Baker'3 from rolling experiments ôn. a ship modél, during which

observations were made of the water movements around the keels during a roil. Most of the rolling energy was taken frprn the model immediately after the end of each roll and dissipated in the 'form of large eddies.

If. the bilge keels are too wide, or placed too near the water surface, the

decelerations may become large enough to causea shock andthe ship to perform

a baulked roil. The ends Of the keels should be rounded off into the hull, as the water eddies thrown 'off by these ends are large, and thê stresses on them

of consideräble magnitude. '

During a winter voyage 'made by the. Author in a large liner, when the ship

rolled heavily, eddies, many square yards in area, broke surface forwardand aft at the positions of the ends of the bilge keels, about a second after the downward

roll had commenced and occasionally they were augmented by similar eddies along the whole lengths of the bilge keels. When this occurred, the ship

performed a baulked roll. The damping of a roll by bilge kéels is directly proportiOnal to their area, and model experiments have shown' that flanged edges or perfQrated surfaces do not add appreciably to the reduction of the rolling of the ship. Other model experiments with the bilge keels split up

into a series of small fins and gaps between them, have shown increased damping, but considerable are must be taken in their alignment, or the ship's speed. may suffer through increased resistance.

14. Pitching

As in rolling, if the natural pitching period of the ship coincides with, or is close to the period of encounter with the waves, the pitching of a vessel in

a 'storm will 'be severe,' especially if synchronism occurs in waves of' such length that the pitching lever is a maximum. It has been shown'4 that the'maximum

tiön inspeed,.or direction, or both,. will reduce:the dipping. Ñkiing ih ioad water plane forward and Increasing its area aft, has the effec of producing

THE DESiGN OF SEAXINDLY' SHiPS 431

pitching lever occurs when the ratio 'of the ship's 'length 'to the wave-length

08 or .2 2 for average sized merchaht ships. TheSefore the worst pitching

conditons will be' when synchronous pitching periods occur in waves f lengtha

andandperiods'494

The ntura1 pitching period of á ship Tp =

2ir/-

longi-tudihal radius of gyration, M thö longitudinal metacentric height For cargo

L L2 L

ships in load condition, K is approximately equal to to O75 -

and

-to 17, so th t.the natural pitching period for such vessels, with due allowance for

damping'° will be approximately T= 35 'IL. The worst conditions for pitching

exist when the. pitching lever is. a maximum, at the same time that synchronism exists between TEa.nd Tp. Thiswill occur when the ship is running "head on" to

waves of length equal to - at a speed given by = 775, which is usually a greater speed thini that at which cárgo vessels would face stormy weather.

Syn-chronism between ship and wáves can also exist at the same time that the pitching lever is a maximum, if the ship is running before waves of length

at a speed of V= 135'IL.

course or speed will immediately reduce the pitching. Vessels with füller forms 'forward than aft in the neighbourhood of the load water line will have greater 'scending moments änd Smaller pitching moments in à head sea and will therefore pitch less' Also ships with the centroid of the load water

plane aft of amidships, will tend to have greater pitching angles and to be less seakiny in' head Seas, other conditions of freeboard and loriitudinal positiOn of the centre of gravity remaining unaltered

d- 1 '--. k- .t .. _{,,. ,..} . .._

-15. Pitch Damping

The resistance to pitching depends largely upon the longitudihal metacntric height"which is always large in therchãnt ships---and the calculated pitching period is generally increased by damping by as much as 50 per cent'6 It is

not usual, therefore, to fit to the ship .any axtifiial aids to pitch damping

In

a case withiti the Aùthó±'s knowlödge, however, it was desired to prevent all pitching oscillations on a model of an aircraft camer of novel design, and a well-ininiersed horizontal fin placed transversely at each end of the model, proved most effective in damping out all pitching in regular waves urider synchronous conditions, and added very little. to the power required to drive

the model.

16. Heaving and Dipping

The amplitude of dip of à ship in regular waves, depends upon the amount of the change in static buoyancy as a wave passes,. and to the degree of syn-chromsm between the ship's natural dipping period (Te) and her period of encounter (TE) with the waves The former is governed by the area of the

load water, plane, which is settled by stability and speed considerations. For

ordinary merchant vessels, the amphtude of the dipping lever, will be a maximum

in waves 0 58 of the ship s length meeting her "head on ' Dr Havelock

has shown'6 that the natural periods of "pitch ", and "dip" are nearly' the

same, i.e. T1, = 035 /72 approximately for cargo ships of moderate dimensions. It follows that the worst dipping will occur in following waves of OE58 L length when the ship's speed and direction are given by

033. An

altera-H

large differences in dipping móment, for very small changes in wave-length, at

certain ratios of ship length/wave-length's.. When meeting head seas, quite small changes in wave-length or small alterations in course, may cause large

increases in the heaving motion of ships with centroids of the load water plane

considerably aft of the midship section.

-As the heavingdipping motion of a ship acts as a large artificial

wave-maker, creating waves which travel away from the ship the natural damping is large and no artificial aids are needed to reduce the amplitude of heave and dip. - The most important effect of this motion upon the seakindiiness of a

vessel is its action in producing slamming 17 _{In addition to possible} damage or excessive straining of the hull structure slamming has an

un-pleasant psychological effect upon the -passengers. and crew if the" slams" are frequent and of considerable magnitude, and give the ship the undesirable

reputation of an unseakindly vessel To eliminate or to reduce slamming

a reduction in speed is always effectivé and the intensity of the " slams" can be reduced until they -become negligible, by giving the prOfile of the stem a sufficient cut-up to avoid hollow lines near the keel, and the transverse sections

aft of the cut-up a pronounced rise of floor and well-rounded bilge turn, for

at least one-eighth of the ship's length aft of the fore perpendicular..

17. Surging

The -principal causes producing yaw and drift in a seaway, have been discussed

in Part I of this paper and the measures which can be taken to reduce these motiOns have been cnsidered. The remaining ship motion is "surging ", iie.

a periodic acceleration and deceleration of the ship's speed. Owing to the large

inertia of a ship underway, variations in screw thrust due to changes in slip caused by pitching and dipping rarely result in noticeable surging In long

high waves the orbital velocities of the surface water particles are considerable

and if theship's length is small in compar on with that of the waves and her alternate1peed up on a wave ttough and slov doWhronr a wave' crest, when running intO-the weather more' or less ''hed on". _Duel'

to the large inertia of a ship,, she will not keep in time with the waves, and her

period of surge will lag behind her period of encounter; so that she wilJ- be alternately accelerated and retarded through the watér out of phase with the

waves. If the ship's bow lines are very full and she has a sharply rising

curveas in most low-speed coastersand if the waves are steep 'sided, as

those found in cotifined and shallow waters, then the surging may become very noticeable, and the ship suffer periodic shocks due to a large retardation

of her speed, when her nose is thrust into on-coming wave crests. It is

advisable in stich craft to rake the stem, to ease away thé turn of the bow lines and also the lower portions of the buttock lines, to give the ship a better chance

of rising to the Seas, and also reducing fluctuatiOns in propeller thrust by an

easier flOw of water to the screw.

-PÂliT 3. Shipping Green Seas and Spray

Nd ship can earti a reputatiön for seakindliness if her weather decks are àontinually washed by ocean waves or covered with sea Spray in moderately

rough- weathet While it is probably not possible to design a cargo vessel that will not ship seawater -under any weather conditions, and at the same time will be an economical and efficient ship to handle for her trade yet some improvement is possible in this respect by attention to details of design, while leaving her freeböardánd trim as cargo handling and other practical working conditions demand.

-- 18. Shipping solid water

In rough weather, whed a ship meets the waves "head on" or on either

bow, green seas are sometimes shipped forward and amidships which flood

ww

LgI* O? C11. I.

- to 1% 0 -*

'

Fig. I 1Movement of L.C.B. as Wave Gest - Fig. 12change 'in Buoyancy as Wave - Passes 'alOng- Hull (see §19) Crest Passes àlong Hull (see § 19)

a range of wave-lengths. It was shown that ;when -the model speed was

constant, she pitched in her period - of encounter with 'the waves (TE) so 'long.

as TE was approximately equal to or greater than her natural resisted pitching

period Tp añd dipped. in the same period when - TE was -z or> T0. The

positions of:the wave crests- along the hull at lowest pitch and. dip were then approximately the same for each consecutive wave. These positions Were - between sectiOns 9 and 9k for lowest pitch, and between sections 6k and 7 for lowest dip, moving slightly forward.with increase in wave-length. In shorter waves, however, when TE was -less than Tp. or T0, pitching and dipping

were tiot isochronous, and the instantaneous -position of the wave crest along

the hull at löwest pitch or dip, changed with each succesive wave and oscillated

rj ç

ìÏiIh:

'I'll".

---i I T---iI---iÏ

THE DESIGN OF SEAICINDLY SHIPS 43,3

as th bow uses --ok the ship rolls. A vessel ',vill be most likely to shij, süch-séàs fórward, when the- bow is- at- the lowest, point of its pitch, the centre -of gravity ät the lowest point of its dip, and an ooean wave óretis at the stem or break of the forecastle; and all three conditions occût at the sàme iitàxit Most water will be shipped when the pitch and dip are large and the amount

f water takeú- ihbdard will be-increased if the vessel's form is such that -her bow breaker in smooth water is high and the ship's freeboard and sheer smalL The practical steps which can be taken in design to lessen the amplitudes of

-pitch, dip and roll, have been considered in Part 2.

-19. Position oJ Ocean Wave Crest along the Ship at Lowest Pitch and Pip During experimentsm rough water with ship rnodels,iß& 20 the positions of the wáve crest along the' mödel hulls at the instant that they were at their

lowest' pitches änd dips, - were obtained experimentally and in Figs. 11-and .12

the loci of these positions are shown for model 607. at constant speed over

It will be seen that the movement of the wave crest at lowest pitch, to and fro about a mean position on the hull, occurred in this ship as in the model

experiments.

TIME IN MMP ruNurfl.

I

IIIIIIIIIIIIIIj!IIIlIIIJf I

Wave-length: 152 ft. Wave height: 11 ft.

Wave direction: 10° Starboard bow Wind speed: 30 knots

Wind direction: head on Ship speed: 1352 knots Period of encounter: 30 seconds

Fig. 13Pitching Diagram taken at Sea 1.4.38

The longitudinal positions of the l.c.b. each side of its still water position, (9 3 feet forward of amidships), were calculated for model 60718. when an

ocean-wave crest was at each transverse section of the model, and they are shown

diagramatically in Fig. Il. Similarly, changes in buoyancy are shown in

Fig. 12. The assumed waves were of trochoidal form and represented ocean

waves 8 feet high from crest to trough and from 160 to 480 feet in length passing

a 400-foot ship "head on ". The movement of l.c.b. was nil for waves of approximately 230 feet, i.e. for a ship-length/wave-length = 174; and there was zero change in buoyance in 320 feet waves; or LIA = I These are almost identical with the ratios obtained from a different set of ship lines in waves of sinusoidal form'4, and may be taken to apply to all modem cargo

ships. When the wave crest was at the stem for waves greater than 380 feet, and also between 190 and 230 feet in length, the buoyancy moment tended to lift the bow; but for wave-lengths from 230 to 380 feet it tended to tip the No. of

Pitch 1 2

345678910.1112131415

Wave cres!

position in

feet 92 140 77 61 92 61 61 15

46 46 Nil

434 tHE DESIGN OF SEAKINDLY SH1P1

backwards and forwards within well-defined limits. A typical pitching diagram

taken at sea on a single-screw cargo liner is shown in Fig. 13. By pressing a button, an observer on the forecastle head recorded on the pitching record, the exact instant when each ocean wave crest passed the stem. From the

information on this diagram, the consecutive positions of the wave crest along

the hull at the moment of lowest pitch were determined, and are given in Table 2. TABLE 2. Wave Crest Aft of Stem at Consecutive Pitches

i-...

-e I_-.,

THE DESIGN OF SEAKINDLY SHIPS 435 nose of the ship downwards into the waves (Fig. 1.1). - This latter tendency

persisted over a dumnjshmg range of wave lengths as the crest passed sections

9+ and 9. Comparison between calculation and experiment (Fig. 11) shows

that for a 400 foot ship at 12 4 knots in regular waves, the vessel s pitch down

wards persisted from to i second after the buoyance moment had passed thtough zero and .was tending to lift the bow. A similarcomparison between

calculation arid experiment for dipping, gives approximately the same time lag between the calculated change m buoyancy and the response of the ship (Fig 12) This time lag was probably due to inertia of the ship It will be

noticed that model 607 was never at her lowest pitch and dip at the same

instant, when pitching in her period of encounter (TE), and that her lowest, pitch occurted about i seconds befdre her lowest dip. The calculated change

in displacement from dip to husve, also lags beuind the calculated change in buoyance moment from pitch to 'scend by about the same time. It fóllòws

from these calculations and experiments

that ifwithout alteration in the

position of the l.c.b.those level lines just above and below the load watet plane, to a depth of an average ocean-wave height, are so drawn, that the

centroids of their water planes are well forward of amidships, then the

instan-taneoùs position of the wave crest at. the moment of lowest pitch, can be placed

well forward of the f.p., so that the stern is always rising to the slope of an oncoming wave, when the ship is pitching in her period of encounter An example of a dry ship model with the ocean wave crest well forward of the stem at the position of lowest pitch, is shown in Fig. 13 of a paper read in 1941

before the Scottish Institution2o.

Dr. Havelock's work'6 lends some support to this theory, in that he shows.

that for a symmêtricäl form, with the centroids of all the water planes amidships, the instantaneoUs position of the wave trough at lowest pitch is only one quarter

of a wave-length forward of amidships. Hence, to ensure that the stem of

the ship will always rise to the forward slope of an approaching ocean wave

in very rough weather, the ntroid of the load water plane should be well

forward of amidships, even inships with their Lc.b's aTt of the centre of length.

20. Height of Bow Wave

1f the bow bieaker of a ship is large in volume and of coñsiderable height,. the chances of shipping water in a seaway are increased, as the vertical move-ment of the surface water in an ocean-wave crest is increased by at least the height of the bow wave, as its crest passes the stem. Raked stems tend to reduce the height of the bow waves. The height of the water above the still

water level along the side of a floating body of parabolic form when underway

in a smooth sea has been investigated mathematically by. W. C. S. Wigley'9, with certain assumptions, and compared. with the wave proffles obtained by experiments on modela. The comparison showed good agreement at the forward end, and an empirical, formula for the height of the bow breakers of merchant slaps based on this work has been deduced, the constant being

obtained from wave profiles given by experiments on a number of ship models,

run over a range of speeds in the Teddington Tankc

This formula is H = 0083 x - x V2

where H is the height of the bow breaker infeet. B'the ship's beam infeet.

LE the length of the entrance in feet V the ship's speed in 'knots . -- .

and gives good average values for thè height of the bów Wave for ordinai-$r

r

436 THE DESIGN OF SEAKIÑDLY smps

-I 21., Water shipped when Rolling

-.Water may. also be shipped amidships where the freeboard is,low, when the vessel is foiling. If the centre Of gravity of the ship is high, the vessel may "lUrch" intö ah on-coming wave of a beam sea and ship water; but even with seas on the bow green water is sometimes shipped on decks just forward or just aft of amidships. When a ship is underway.a seres of.ftansverse waves is'fonhed by the bow, and the first of these wäves can be of appreciableheight

in vessels with short entrances and full bow coefficients. An ocean-wave

rest, travelling in the opposite direction to the ship, grows rapidly in height

when passing through this transverse wave and may break green water coming

inboard as the ship rolls to windward. In the early stages öf design, when the

reqUired ship speed is known, the positions along the hull of the crests of the bow-wave system, can he obtained by a simple calculation22, andthe distribu-tion 'of displacement alöng the entrance adjusted, to avoid placing the first

trañsverse wave crest near a deck with low freeboard.

'kö-reduce the height of'this transverse wave and to ensure that its shape

döse to the hull will be mOré sinusoidal than trochoidal, and not have steep slopes, the bow 'linés near the extreme beam should have long gentlC slopes fòrward with very easy turns at the bilge. This can be secured by filling out the' fore-deck lines and using fairly large bilge' radii at the forward shoulders of the ship. '

To secure low wave-making resistance, the placiuig f the wave crests. of the

two main transverse wave-making 'systems of the vessel due to her entrance

and run,.should be so arranged that they do.not coincide at normai ship speeds.22

If other more important coi siderations make this impbssible, then following ocean waves on either, quarter may cause «large upheavals of water s their

àrests' pass through the combined bow- and steri -wave systems, and the after de&orpoop to 'be periodically flooded.: Such a periodic.growth of

theocean-wavefcrësts rattherafterrend;of theuhip

the' Author in a largO' twin-screwiiner running before a strong south-westerly

gale.

22. Spray over Ship's Decks

When an ocean wave breaks lose to the weather side. of ,a ship, spray-is

thrown ùp into 'the air and may be driven inboard by the wind, wetting the ship's decks. Ocean waves that are not breaking,' may do so as they pass through

the ship waves, especially if the latter are steep-sided. It is commOn knowledge

that ships, with high thin bow divergent waves have their fore decks drenched

with spray, even 'when the'waves in the ocean 'swell are smaU. If the directiOn

of the ocean waves towards the sliip is approximately the, same as that.of the ship. divergent waves away from the vessel, and of the same length, then high

oscillating waves wifi be formed on 'the weather side and these waves generally

break, the broken water being thrown up to a considerable height above the

sea. Should the ship's form be such that the divergent wave crests are high

close to the hull, the broken water of the oscillating waves will be' driven inboard

asspráy. "1

-Observations were made by the Author of the spray thrown up in all

weathers, by a single-screw cargo liner, with a forecastle 43 feet, long ,and 23

feet above still-water level, when steaming at speeds between 13 and 14 knots.

The fore deck from the break of the forecastle to the bridge structure_was

81 feet long and about 15 feet above the water surface, and the bridge deck

35 feet above still water. Meeting waves only 2 feet high, spray broke over the

forecastle and in 4-ft. waves, spray was thrOwn over the fore deck. In 5-ft.

waves spray reached the bridge deck and in 8-ft. waves, green'water in addition

to spray was shipped on the fore deck. In 18-ft. waves dlouds of spray were thrown ovei the forecastle and bridgh and even over the top of 'the funneL This spray was wholly due to the shape of the forward end of- the ship. The

'i

THE DESIGN OF -SEAKINDLY SHIPS 43'? bow waterlines were hollow, but this hollow had been formed by grafting

hollow ends on to a convex-shaped vessel. The bow breaker was considerably Smaller than the first divergent waves, which were in fact the trca bow breakers,

añd which in smooth water were breaking waves, commencing a little aft of

the stein, extending to about 10 feet aft of the break of the forecastle and

hugging the ship's side. Practically all the spray thrown oVet the fOrecaStle and fote döck came froin this first divergent wave on the weather side. The

convéx portion pf the ship's lines had too great a slope for the ship's speed,

and when even a small ocean wave passed through this ship wave broken water was thrown violently against the ship's side, which was practically

vertical in the neighbourhood of the fore deck and not flared outwards. This

broken water burst into spray-which wai blown upwards and inboard: In a much iàrgdi and faster liner with hollow bow lines, spray was thrown up over the bridge well over 100 feet above the water surface This spray came from both the bow breaker at the stem and the first divergent wave on the weather side, which started some distance forward of the bridge structure and hugged the ship's side. As an ocean wave passed through these ship waves,, they grew rapidly m height and broke all along their lengths This broken water

was blown mboard aft of the break of the forecastle and caught by an

up-draught of win'd caused by. the stepped back forward ends of the decks in the bridge structure, spray was carried upwards to wet the bridge deck leavmg only twO small areas òf it dry. This sortS of defedt in a vessel's performance, can be avoided by a careful placing by calculation of the bow divergent waves,

before the lines are drawn, ensuring thatthe breaking crests 'of such waves are well away from the ship's 'side and invisible to the naked eye where they meet

'thehull. These examples of "wet" ships pOint to the extreme' care which

shouki be given to the drawing of hollow bow lines: To graft a hollow forward

end on to a convex bow ship will mevitably lead to an unseakmdly vessel

When-the-ship's speed demands h011ow bow lines, they should be drawn without

ferencó,anystpf.convex bowjinesand the utmost care taken by calculation

_,IØ

venient positions with respect to the upper deck and bridge. Above these maximum slopes and for some distan aft of them, the transverse lines should

flare outwards.

The steeper sides of bow divergent waves are normally outboard and these waves usually break away from the ship; but occasionally, as in- the

single-screw caigo liner referred to above, the steeper sides of these waves are inboard, with the waves breaking towards the ship's side. When an ocean wavò passes through 'such divergeht waves, broken water is thrown with increased violence

against the ship's side and much mote spray rèaches the decks than with

divergent waves of the normal pattern. More research is needed on the hull shapes that produce such waves, before reliable theories can be fOrmed as to

the typin of bow lines to be avoided. At present they can only be conjectured.

23. Conditions fòr Breaking Waves iñ an Ocean Swell

The probability of shipping spray is greater in a sea of breaking waves, than when the ocean swell is composed of smooth non-breaking rollers. From a

mass of data taken at sea of the heights and lengths of the waves in ocean swells, it was noticed that the maximum wave height h to wave leñgth A did not exceed

i _{Assuming an ocean wave to be of .trochoidhl form, the maximum slope}

will be' ir

'x4

standing wävessimilar to those thrown off by ¿bstacles in riverS or tideways will- not have wave slopes greater than 18. This is about the same angle as the hmitmg tangent to the slope of the lines of the after body of a ship, above which stream lines leave the ship form and dead water is produced2'

438 THE DESIGN OF SEAKINDLY SHIPS

For progressive waves oftrochoidal form, the velocity (y) of a water particle

'g7rh2 V2 h

...:..

-in feet/sec. = -u-- or

h I y2

For limiting conditions

₌

accelera-tion of a surface water. particle exceeds lO- 1 ft. /sec.2, the wave breäks. PAkT 4

24 Safety and cOmfort of Passengers and Çrew on Weathèr Decks

In a seakindly ship, passengers and crew will be able to move about in

stormy weather without danger or discomfort. This means that aU ship mOve-ments will be easy and free from sudden violent checks or baulks, no seas will

be shipped on open deck spaces accessible to passengers and crew, and these

weather decks will also be free from sudden unexpected high wind gusts, eddies,

or back draughts, which might make walking dangerous or impossible. The steps which can be taken during the design stage to ensure easy ship movements

n a seaway and to avoid shipping seas, have been discussed in Parts 2 and 3. of this paper, and the causes of sudden high wind speeds along open decks

remain to be. considered.

That very high wind velocities do occur along open decks, even when the actual wind blowing at the time has a low speed, is shown in Fig 14 which was drawn from measured wind speeds taken simultaneously by pitot tubes along the windward sides of the boat and sun decks of the qs.s. Mauretania. The vertical distance between these decks was about II. feet, and the side of the sun deck was about 15 feet inboard from that of the boat. deck. It will be noticed that the wind velocities at the boat-deck level were from two to three times those measured ti "suh!abók. '- .

-Il

GAS MAURITANIA OHIKRVATI0N.S

fl

THUS NUIS Intl ASS UNITI.

,.LATNI !NH HISACTIHTI 2 POUT POW. SHIP SPItS ¿455 NUITS

RIN HICK.

PITO? TAIl .0111,0Ml ALONG DOCK

Fig. 14'- Wind Speeds over Boat and Sun Decks 25. Streamlined Forms of Superstructure .

"Stresmiiriing" the form of the front Of a superstructure to lessen the

wind pressure and to satisfy an aesthetic taste, is an attractive idea to a designer.

But by Bernoulli s law any decrease in wind pressure due to streamline form

is accompanied by an increase in wind velocity So that a streamlined

super-structure nsmg from an open deck will inevitably induce high wind speeds

Fig. 15Contour of Sun Deck Rising from Boat Deck 26. Bridge Screens.

To protect navigating officers from the effect of high wind speed on open

bridges, it is the custom to fit sailcloth " dodger curtains " or permanent screens

of light plating from 3 ft. 6 in. to 4 ft. 6 in. in height, along the front of the

transverse bridge deck. These screens deflect the wind at a man's height from

the bridge deck and allow the navigating officers to perform their duties

without being troubled by violent wind gusts. The action of these bridge

screens in deflecting the wind, was examined by pitot tube and direction

indicator on a low-speed liner in a head wind. The bridge of this vessel spanned

the ship from port to starboard beam, and the superstructure between the

bridge and the open sun deck was vertical and plated all the way up to the

full width of the ship.

Ail along the upper horizontal edge of the screen, a wind eddy formed with a horizontal core. Below this eddy and close to the screen, the wind divided at about the centre of the ship and moved outwards and slightly downwards

At each vertical edge of the screen at the beam of the ship, a wind eddy formed

with a vertical core and a little forward of the screen the wind moved strongly upwards and passed aft at a distance above a man's height from the bridge

deck level, leaving a space above the screen in comparative calm.

27. Other Critical Portions of the Superstructure

Stepped back deck erections induce greater upward wind velocities than a vertical wall of such erections. If the bridge tops the stepped back decks,

and if it is itself covered by an erection such as the standard compass platform,

the upward draught of air may be caught and turned into a back draught if

the platform has a fair length longitudinally. Such a back draught was

experienced athwart the uppermost promenade deck of a twin-screw liner, where

a rather deep compass platform caught the uprising wind and turned it back

ÌHE DESKN OF SEAXTNDLY srnP 439

shown during some observations of wind velocities and directions made on a liner running at 24 knots into a true wind speed of just over 30 knots meeting the ship at 18° on the port bow. At the boat deck level the shape of the super-structure rising from it was as drawn in Fig. 15. Measurements by pitot tube of the wind velocities around the starboard (i.e. the leeward) side of the ship on the boat deck near the bridge structure, showed increases in relative wind speeds from over 70 knots to something greater than 175 knots, which was the limit of the pitot tube gauges used in the observations. From two to three

days, this portion of the promenade deck had to be forbidden to passengers, even though the weather was fair. Had the bridge superstructure not been "streamlined ", but shaped somewhat as shown dotted in Fig. 15, the wind speeds would have been quite low and the promenade deck always available