FUNDAMENTALS OF MARINE VEHICLES Dr B S Lee PART 1 FUNDAMENTAL CONCEPTS

1.1. BASIC SHIP GEOMETRY Most ships have one plane of symmetry, the centreplane, giving port-and-starboard symmetry. Some vessels used in offshore industry may have two or more planes of symmetry (e.g. semi-submersibles). Since the length of most ships are bigger than breadth, it is natural that the direction parallel to the length is referred to as longitudinal direction, as in, for example, longitudinal stiffeners (or simply longitudinals). The athwartship direction is referred to as transverse direction, as in transverse girders. (a) Major Components of a Ship A ship is essentially a box, usually water-tight, designed to float carrying passengers and/or cargo and to propel itself. The main body of a ship is known as the hull. The hull is enclosed on the top by the upper (or main) deck (that is, if it is not an open ship). The hull will have the sideshell and the bottom (called collectively a shell). The part of the ship above the hull is called the superstructure. The spaces within the hull for carrying cargo are called holds, and the openings in the deck to allow access to them are called hatches. Hatches usually have lips around them to accommodate hatch covers, called hatch coaming. On ships carrying liquid or gaseous cargo, the holds are called tanks. (b) Principal Dimensions (see Figs 1.1 1.3) Principal dimensions (sometimes called principal particulars) show the major features of the vessel and include the following items: Length between perpendiculars (LBP, LBP or LPP) It is the distance between the aft perpendicular and the forward perpendicular. This is often the length of the underwater body at the loaded draft, and therefore is used in most calculations in naval architecture. Length overall (LOA or LOA) It is the distance from the extreme point at the forward end to a similar point at the aft end. In most ships it exceeds LBP by a considerable amount. The excess includes the overhang of the stern and stem. If there is a bulbous bow, LOA is measured to the extreme point of the bulb. Length on the waterline (LWL) It is longitudinal distance of the waterline at any given draft. For most ships LWL will be different for different draft. Breadth moulded (B or BMLD) This is the breadth of the ship at its broadest point, measured to the inside surface of the side shell plating. This is the breadth commonly used for most calculations, but is not the greatest breadth of the ship. The breadth extreme is

the breadth measured to the outside surface of the shell plating (equal to BMLD plus twice the thickness of shell plating), or any side appendages if any (such as fixed fenders). Breadth extreme will, therefore, indicate the clearance required for the ship to pass through. Depth moulded (D or DMLD) It is the vertical distance measured amidships from the baseline (top of the keel plating) to the underside of the deck plating at side. (Note that the depth at centreline is usually greater than that at side the difference is known as camber). Draft moulded (T) This is the vertical distance from the moulded baseline to the waterline at which the ship is floating. The extreme draft, of course, is the vertical distance from underside of the keel plating to the waterline. The draft moulded is used for most calculations, but draft extreme is important in determining the minimum depth of water the ship can sail in (should give sufficient clearance because of a phenomenon known as squatting).

Stern

Superstructure

Bow

Main Hull Keel

Port quarter

Port beam Centreline (centreplane)

Port bow

Starboard quarter

Starboard beam

Starboard bow

Fig. 1.1 Main parts of a ship

AP

Aft sheer

Fwd sheer

FP

Summer load waterlineline Midship Length between perpendiculars (LBP) Length on waterline (LWL) Length overall (LOA) Bulbous bow

Fig. 1.2 Longitudinal principal dimensions

Breadth extreme

Fender
Camber
Depth mld (D)

Breadth mld (B)

Baseline (top of keel)

Inside of side shell

Fig.1.3 Transverse principal dimensions

(c) Weights When a ship is completed and ready for sea, but as yet empty other than just enough fuel and lubricating oil and water to operate the engine, its mass is known as lightweight. Whatever mass additional to the lightweight that the ship may have on board at any time is called deadweight (DWT). The deadweight will include pottable water, sludge, fuel and lube oil, stores, provision, crew and their gear and ballast as well as the payload. At all times the sum of the lightweight and the deadweight is equal to the (mass) displacement (). The maximum deadweight that can be put on board is limited by the strength and the seaworthiness of the ship, and is controlled by law (Load Line Regulations). This limitation is shown clearly by markings of the maximum allowable draft on the side of the ship (usually amidships) for a variety of different conditions. These markings are known as Plimsoll lines. (d) Other Features There are some items which may be considered minor but sometimes are very important for certain characteristics. These include the following: Sheer This is the progressive rise of the deck as one moves from midships towards the ends. This feature is used to provide as much spare buoyancy as possible at the narrow points of the ship, and also to protect the most vulnerable parts of the ship to some extent. Small fishing boats often show high degree of sheer, and it is to render the working platform at the side as near to water as possible. In older ships the deck-at-side line was usually a parabola and the sheer is quoted as its value at the forward and aft perpendiculars. The sheer forward was usually twice as much as the sheer aft. In modern ships, however, the deck at side line can either be flat with zero sheer over some distance on either side of amidships and then rise as a straight line towards the ends, or be completely flat with no sheer over the entire length.

Rise of floor Some ships have a flat bottom (usually large ships), but others may have a bottom rising from the centreline or a little off it. The height of the intersection of a line drawn along this rising floor and the vertical line at the moulded breadth is called rise of floor. High speed vessels often have high rise of floor (deep V-section) at the bow region which becomes smaller (shallow V-section or U-section) towards the stern. Freeboard This is the vertical distance from the waterline to the deck. Usually it will be a minimum amidships and will increase towards the ends (due to sheer). Obviously this is related to the maximum allowable load line, and is governed by the Load Line Regulations.

1.2

FORM DEFINITIONS (a) Lines Plan Majority of ships has hull surfaces with three-dimensional curvature. This means that the hulls cannot be represented by simple common geometric elements. Indeed many techniques had to be devised to represent the hull surfaces in mathematical/numerical forms so as to accommodate computerisation in the last three decades. However, the traditional method of representing the hull form relies on section shapes in planes parallel to the three major planes. The drawing containing these sectional shapes is known as a lines plan and consists of Half breadth plan (intersection of horizontal planes with ships moulded surface) Showing a collection of waterlines at various drafts. Each line represents the shape of a waterplane. The load waterplane (LWP), or load waterline (LWL), represents the waterplane at which the ship is designed to float in loaded condition. A design waterline (DWL) is the waterplane for which the ship is designed. This does not mean the ship cannot operate at any other draft, but that the designer tried to optimise the performance of the ship for that draft. This, as you can imagine, is often the same as the LWL. Body plan (intersection of transverse vertical planes with ships moulded surface) A collection of transverse sections at regular, designated stations along the ships length. For normal purposes, only one half of the section shape is shown. The forward half of the ship is shown on the right hand side of the centreline and the aft half is shown on the left hand side. The stations, normally numbered 0 to 10, with stem and stern regions having more stations and around the parallel middle body at intervals of 1. For example, a basic design lines plan may have stations 0, , , , 1, 1, 2, 3, 4, 5, 6, 7, 8, 8, 9, 9, 9, 9, 10. Station 0 is known as the forward perpendicular, and is the plane perpendicular to the centreplane and the waterplanes passing through the intersection between LWL and the forward edge of the stem. Station 10 is the aft perpendicular and positioned at the aft side of rudder post or centreline of the rudder stock. Sometimes extra stations are placed outside these two perpendiculars. Station 5 corresponds to midships. Sheer plan or profile(intersection of longitudinal vertical planes with ships moulded surface)

Shows longitudinal sections of the hull on the planes parallel to the centreplane at a number of offsets. The forward lines are known as bowlines, while the aft lines are called buttock lines or buttocks. The line at 0 offset will show the profile outline of the vessel together with the deck at centre line. The plan also shows deck at side line. A lines plan is usually a very long drawing, and to reduce the inconvenience of having to deal with such long drawings, it is sometimes drawn with the forward half superimposed on the aft half. In order to ensure fair hull form some section shapes of longitudinal planes other than the two orthogonal planes are also drawn below the centreline of the halfbreadth plan. These are known as diagonals, and are of particular significance for sailing yacht forms, as they often sail with a static heeling angle, and, therefore, fairness of the lines at heel is crucial for their performance. (b) Table of Offsets Offsets are transverse distances from the centreplane to the inside of the shell (moulded offsets). These are arranged in a tabular form with columns for stations and rows for waterlines, or the other way around. In addition, other information, such as sheer, may also be included. (c) Form Coefficients There are a number of non-dimensional coefficients which are very often used to indicate the major geometric characteristics and to give an idea of the performance of the vessel. For geometrically similar vessels the form coefficients will b identical. Block coefficient (CB) This is fraction of the volume of displacement of the underwater form ( ) occupying the rectangular block whose sides are equal to LBP, B and T. Thus,
CB = L BP B T

The block coefficient gives an indication of the fullness of the form and its value may vary from 0.38 for some high speed vessels, yachts and military vessels through 0.60 for passenger ships and aircraft carriers to 0.88 for large crude oil carriers. Midship section coefficient (CM) This is the ratio of the immersed area of the midship section (AM) to the area of the enclosing rectangle whose sides are equal to B and T. Thus,
CM = AM B T

This coefficient expresses the fullness of the midship section and its value is usually around 0.75 to 0.98. For extreme forms it may be as low as 0.65 or as high as 1.0.

Prismatic coefficient (CP) This is the ratio of the volume of the underwater form to the volume of a prism having a constant cross section of the immersed midship section and length equal to LBP. Thus,
CP = AM L BP

This coefficient is sometimes called longitudinal prismatic coefficient to distinguish it from the vertical prismatic coefficient (see below). It is a very important parameter determining the forward motion resistance of the vessel. As can be expected, the ships with lesser resistance will be finer in form and therefore will have smaller CP. Its value may range from about 0.55 to 0.80. Vertical prismatic coefficient (CVP) This is the ratio of the volume of the underwater form to the volume of a prism having a depth equal to the vessels mean draft (T) and a constant horizontal cross section identical to the waterplane at that draft. Thus,
CVP = AW T

where AW is the waterplane area. Waterplane area coefficient (CW) This is the ratio of the waterplane area to its circumscribing rectangle. Thus, AW CW = LWL B Unless otherwise stated, CW is for the load waterline of the vessel, at which LWL is often the same as LBP. Table 1.1 gives sample principal dimensions and/or some form coefficients. Perhaps you would like to calculate the items left blank and compare them between various ship types. Table 1.1 Examples of principal dimensions Tug Ferry Cargo Cargo Passenger Tanker LPP (m) 31.7 47.2 126.5 161.0 201.0 125.0 B (m) 8.50 13.25 18.9 23.2 28.5 18.9 T (m) 3.50 2.45 8.0 9.0 9.9 7.3 (tonnes) 388 888 12,480 21,340 35,590 13,680 AM (m2) 20.50 30.25 144.7 206.2 275.0 TPC (tonnes/cm) 1.99 4.68 19.60 28.10 43.85 20.70 CB CP CVP CW CM 0.99 Tanker 250 34.25 14.1 101,500 77.50

0.99

1.3. FLOTATION (a) Archimedes Principle Most people have heard usually at school the funny but exhilarating story of Archimedes running through the street stark naked shouting Eureka! We all experience lightening of our body when we get into a bath and therefore it was not this phenomenon itself that he discovered. The significance of his discovery is in quantifying that experience precisely, thereby enabling the application of this principle in engineering. It is so fundamental to naval architecture, that it is necessary to examine his principle with the help of more modern understanding. When a body is immersed in a fluid, the body will experience an upward force equal and opposite to the weight of the fluid displaced by it. This is known as the Archimedes Principle and the upward force is called buoyancy. We now know that the buoyancy is the resultant of all the normal pressures exerted by the fluid on the entire immersed surface of the body. This pressure is known as hydrostatic pressure and is proportional to the depth of immersion, or

P = gd where = mass density of the fluid g = gravitational acceleration (=9.806 m/s2) d = depth of immersion of the point.
For an arbitrary body floating in static equilibrium in a fluid shown in Fig.1.4, the vertical force (or the buoyancy force) is the sum of the vertical component of the surface pressure, or

B = P cos dA = gh dA cos
A A

It is evident that h dA cos is in fact the immersed volume of the column which presents the surface dA to the fluid at an angle at its end. Therefore, h dA cos is the total immersed volume of the object. Let the volume be V,
A

then

B = gV provided g is constant, which is practically true for incompressible fluid and for the kind of depth we are concerned with. This proves the crucial Archimedes discovery.
One can also prove easily that the sum of horizontal component of the pressure is zero.

L
h

ds

(a) Hydrostatic pressure on the immersed surface

(b) An elemental volume

Fig.1.4 Mechanism of buoyancy force

(b) Notation The volume of the ship underwater is known as volume displacement and is represented by the symbol . The normal unit for volume displacement is m3. The mass of the water displaced is called mass displacement and is written . Thus, = . The unit of mass displacement is usually tonnes (the mass of 1 m3 of fresh water is 1 tonne, that of 1 litre is 1 kg). Note that is the same as the mass of the ship and g (sometimes called force displacement) will be the same as the weight of the ship. Indeed one hardly ever sees any special symbol for the ships weight, since it can be represented by the displacement. The unit of force displacement is N, but since mass is in tonnes, it is more convenient to use kN (weight of 1 tonne is approximately 9.806 kN). (c) Static Equilibrium Although the buoyancy force is distributed all over the wetted surface, it is often more convenient to consider that the total buoyancy force acts upwards from a single point. This is the point about which the moment of the hydrostatic pressure, when summed over the entire wetted surface, becomes zero. It is called the centre of buoyancy. In a similar manner the gravitational force (i.e. the weight of the object) can also be regarded to act from one point: this is the centre of gravity. In general, a state of equilibrium exists if and only if (a) the sum of all the forces is zero; and (b) the sum of all the moments is zero. An object freely floating in still water, therefore, will have just the right volume immersed to provide exactly the same buoyancy force as its weight, and will take an attitude which puts the centre of gravity vertically in line with the centre of

buoyancy (regardless of vertical relative position). This state is known as static equilibrium in contrast to dynamic equilibrium which will be discussed later. Of course, if its total volume cannot provide sufficient buoyancy, it will simply sink. If the total volume can produce buoyancy exactly the same as the weight, it will be neutrally buoyant, as submarines can be in certain conditions. If the first of the two equilibrium conditions is not met, then the object will either sink further or rise until that condition is fulfilled. If only the second condition is not met, then the object will rotate itself until the equilibrium is reached. In general, however, these two processes occur at the same time. Unlike objects of simple geometric shapes, the volume displacement of a ship for any given draft is not simple to calculate. These days, all that can be done by a computer at a blink of an eye, and thus presents no problem. However, in the days when every single computation had to be done by hand, it was immensely important to devise methods of instantly reckoning the expected draft for a given loading condition or the mass displacement for a given draft. Naval architects developed many different curves and other hydrostatic items for this purpose, and we shall examine some of the more important ones here, as, despite the modern technology, these are still used extensively. (d) Curve of Displacement The volume displacement calculated for a range of draft can be plotted against draft, and the curve is smoothed (naval architects are, if nothing else, past masters of generating smooth curves). From this curve, the volume displacement at any mean draft can be readily obtained and vice-versa. This curve presupposes that the ship will have a fixed pre-defined longitudinal attitude (trim). Trim, however, is the subject of a later section. (e) Tonnes Per cm Immersion (TPC) The loading condition of a ship changes often and it is important to keep track of what draft the ship will float at the end of a minor loading/unloading operation. For this the curve of displacement may be used, but for obtaining a quick indication of the change of draft for a given change in mass displacement, a ratio known as tonnes per cm immersion is very useful. The TPC of a ship at any given draft is the mass required to increase the draft by one cm (parallel to the existing waterline, and hence the terms parallel sinkage and parallel rise). Consider a vessel floating at a draft T in water of density (by the way, 1.025 or 1.026 tonnes/m3 is often taken for the normal mass density of sea water). Its waterplane area is AW at this draft. Some cargo is now placed on the vessel so that it sinks by 1 cm. This increases the underwater volume by approximately

d = AW 1 cm = AW 0.01 (in m 3 , if AW is in m 2 ) Therefore, the increase in mass displacement is

d = d = 0.01AW

Note that this increase in displacement will be identical as the mass of the cargo loaded which caused the parallel sinkage.

Conversely, we can find the parallel sinkage or rise caused when a mass m is loaded or unloaded as follows parallel sinkage = m / TPC A moments reflection will show that TPC is in fact equivalent to the slope of the curve of displacement. Note also that in the above discussion we have assumed the layer of the body which is immersed by increased mass of the object has a constant waterplane area. Strictly speaking, this is not true, but most ships are virtually wall-sided, and it is accurate enough for most engineering purposes for small changes in draft. Indeed this type of approximation, simplification and linearisation used to be essential before the advent of computers. Remember that computers began their serious inroads into ship design only in 1970s!
(f) Effect of Water Density

Ships travel, and the seas of different areas have different mass density, due to changes in salinity and temperature. A more pronounced change in mass density of water is experienced by ships on moving from fresh water into sea water (as in the Great Lakes and St Laurence Seaways). There is about 2.5% difference in the density of fresh water and sea water, and obviously this will cause sinkage or rise of the vessel during the transit. In all these cases, only one principle should be remembered: the mass displacement of the ship should be the same as the mass of the ship at all times, whatever the density of water. This means change in the underwater volume which can only be achieved by altering the draft. Consider a vessel of displacement 15,000 tonnes floating in sea water of density 1.025 tonnes/m3 at a draft of 7m . The volume displacement in sea water is
S = 15000 / 1.025 = 14634 m 3

When this ship moves into fresh water of density 1.0 tonne/m3, its volume displacement will be
F = 15000 / 1.0 = 15000 m 3

The increase in volume displacement, 366 m3, can only be achieved by the (parallel) increase of draft. If the waterplane area of the ship is 2000 m2 at 7 m draft, then the parallel sinkage is
dT = 366 / 2000 = 0.183 m or 18.3 cm

Note that in general, the new waterline will not necessarily be parallel to the original waterline, but it is convenient to consider the parallel rise or sinkage first and then the change in trim separately.

10

TUTORIAL No.1
1. A hollow steel cylinder is 1.83m in diameter, 7.62m long and weighs 16.26 tonnes. At what draft will it float in sea water ( = 1.025 tonnes/m 3 ), when its axis is vertical? What will then be the hydrostatic pressure at any point on the bottom? 2. When a ship, without changing weight, passes from sea water to fresh water, she will sink further in the water, because fresh water is less dense than sea water. Show that the increase in draft, say d (in cm), is given by d= s f

f TPC
= displacement of the ship in tonnes s = density of sea water in tonnes/m3 f = density of fresh water in tonnes/m3 TPC is for sea water.

where

3. Show that for any ship

C CP = W CVP C M
4. A ship, 122m long, has a beam of 15.25m and floats in sea water with an even keel draft of 5.50m. If the block coefficient, CB, is 0.695, what is the displacement? The immersed midship section area is 82.50m2. Calculate the values of CP and CM. 5. A ship has the following particulars: L = 128m, B = 19.20m, T = 8.85m, CB = 0.713, CM = 0.945. Calculate - Moulded (mass) displacement in sea water - Area of immersed midship section - Prismatic coefficient. 6. A ship of displacement 1747.0 tonnes has a beam-to-draft ratio equal to 3.53 when floating in sea water. If the immersed midship section is 30m2, CB = 0.537 and CM = 0.834, calculate the length, beam and draft of the ship.

11

1.4. FUNDAMENTAL MOMENT THEORIES AND THEIR APPLICATIONS

One of the most important concepts in engineering is moments, of various order, of lines, areas, volumes, masses, forces and, even, moments. Indeed, the area of a 2-D shape can be called its 0th moment (the lever is brought up to the power of 0). Of these, the first and second moments are encountered the most often in naval architecture. Here we shall confine our discussion to the first and second moments of area, volume, force and mass.
(a) Area

The area of an arbitrary 2-D shape in the xy-plane shown in Fig.1.5 is calculated by first considering a thin strip of width dx. The length of this strip is = y1 y2, and thus its area is dx. Therefore, A = dx
x1 x2

When it is not possible to express in a mathematical form, this integration can be performed numerically (using e.g. Simpsons rules).

y1

dx

y2 x
Fig.1.5 Calculation of area

x
Fig.1.6 Calculation of first moments

(b) First Moment of Area

The first moment of an infinitesimal area about a given axis is the area multiplied by the distance between that axis and the area (lever). The word infinitesimal is used here to signify that the area is so small that the lever from the reference axis to that area can easily be determined. This is analogous to the moment of force, often known simply as moment, where the force is multiplied by the lever. Consider an arbitrary, enclosed 2-D shape of area A shown in Fig.1.6. If we take a very small part of it, and let the area dA and the lever to y-axis x. Then the moment of this element about y-axis is xdA. Therefore, the moment of A about y-axis is

12

M y = x dA
A

The symbol

means integration over the whole area A. Similarly

M x = y dA
A

The lever can be positive or negative. It follows, therefore, that a first moment of anything can be positive, zero or negative. Indeed it can be seen at once that the first moment of an area about the axis of symmetry (if it exists) is zero. Furthermore, in Fig.7, My is obviously positive, since all the levers are positive; similarly Mx is also positive. Consider now another extreme case where the y-axis is entirely on the right hand side of the area and the x-axis is well above the area. Then, both moments will be negative, since the levers are always negative. This means that as the y-axis moves from the left hand side of the area to the right, it will pass a point at which My is zero keep the y-axis there. Similarly as the x-axis moves upwards from below the area, it will pass a point at which Mx is zero, and we keep the x-axis there. The intersection of these two axes is known as the centroid of the area. The definition of the centroid of an area, therefore, is that it is the point of intersection of any two axes, the first moments of the area about which are zero. Consider an arbitrary 2-D area A and two parallel axes distance tg apart, say y and y1. Then the first moments of the area about these axes are

M y = x dA and
A

M y1 = x1 dA = x + t g dA = x dA + t g dA
A A A A

= M y + t g dA = M y + t g A
A

If we now assume that the axis y passes through the centroid, then My =0 Therefore M y1 = t g A In other words the first moment of an area can be obtained by multiplying the area with the distance from the centroid to the axis. Conversely, the position of the centroid relative to any reference axis can be found by dividing the first moment of the area about that axis by the area. Actually these two principles are applicable to any kind of first moments, and are used in determining centre of gravity, centre of buoyancy, centre of flotation, and so on.

13

Centroid of Waterplane

Identifying the centroid of waterplane for a given draft is very important, as it is required to estimate the longitudinal stability and trim. Normal ships have portand-starboard symmetry. Therefore, it is obvious that the centroid of the waterplane will be on the centreline. It remains to identify the longitudinal location of the centroid, and this can be done for any reference point, but normally it is done for the midship. The longitudinal distance from the midship to the centroid of waterplane is known as LCF (standing for Longitudinal Centre of Flotation), the reason for which will be discussed later on.

Fig. 1.7 A waterplane with the axis system

Consider a waterplane shown in Fig.1.7. The origin of the axis system is on the waterline amidships, so that LCF can be estimated relative to the midship directly without any adjustment. The first moment of the WP about y-axis is
Lf

My =2
where

La

xydx
La and Lf are the length of aft body and fore body respectively y is the half breadth of the waterplane (a function of x).

The area of the WP is

Lf

AW = 2

La

ydx

Therefore, midship to centroid is

Lf

x=

My AW

La LF

xydx ydx

La

14

Note that, in this notation, a positive LCF indicates the centroid is forward of the midship. Most often longitudinal positions of all kinds are given as aft or fwd of midship rather than as signed numbers. Nevertheless, having a sign convention is invaluable in many calculations.
(c) Second Moment of Area

The second moment of an infinitesimal area about a given axis is the area multiplied by the distance between the axis and the area (lever) squared. A second moment of area is not as easy to visualise as the first moment, but it is analogous to a second moment of mass, which is also known as mass moment of inertia. The second moment of an area about Ox-axis is often represented by Ix or Ixx, and it is in fact a moment about x-axis of the first moment of the area about x-axis. Note that one can have a moment about y-axis of the first moment of the area about xaxis, which is known as product of area and represented by a symbol Ixy. Consider an arbitrary area of Fig.1.8. In this case

I x = y 2 dA and
A

I y = x 2 dA
A

For example, for the rectangle shown in Fig.1.9, second moment of an elemental strip about y-axis is x 2 bdx . Therefore

1 1 I y = x bdx = x 3 b = a 3 b 3 0 3 0
2

Similarly, it can be shown that

1 I x = ab 3 3

Fig. 1.8 an arbitrary area on x-y plane plane

Fig. 1.9 A rectangle on x-y

Parallel Axis Theorem

Consider an arbitrary area shown in Fig.1.10. y-axis passes through the centroid of the area, and an -axis is parallel to it at a distance l away.

15

Fig.1.10 Parallel axis theorem

Then

I y = x 2 dA and
A

I = ( x + l )2 dA = x 2 + 2 xl + l 2 dA = x 2 dA + l 2 dA + 2 xldA
A A A A A

Since l is a constant,

I = x 2 dA + l 2 dA + 2l xdA = I y + Al 2 + 2lM y
A A A

where A is the area of the shape and My is first moment of the area about y-axis. However, since y-axis passes through the centroid, My = 0. In other words,

I = I y + Al 2
This means that the second moment of an area about any axis is equal to the second moment of that area about an axis parallel to that axis passing through the centroid of the area plus the area multiplied by the distance squared. This applies equally to second moment of mass and volume as well. It is a very important principle in many branches of engineering. In particular, the second moment of a waterplane area about the transverse axis passing through its centroid (i.e. centre of flotation) is known as the longitudinal second moment of waterplane area, or IL, and that about the centreline is referred to as the transverse second moment of waterplane area, or IT. We shall deal with practical applications of these items later on.
Example 1

Calculate the second moment of area of a rectangle (b x h) about an axis passing through the centroid and parallel to the b-side. Then calculate the second moment of this rectangle about the b-side. Put a Cartesian co-ordinate system with the origin at the centroid of the rectangle and the x-axis parallel to the b-side and the y-axis parallel to the h-side. Consider a thin strip of width dy, lengthwise parallel to the x-axis and at a distance y away from the x-axis. Its area is bdy and its second moment about the x-axis is

16

dI x = y 2 bdy

Therefore the second moment of the whole rectangle about the x-axis is
Ix = bh 3 1 y 2 bdy = y 3 b = 3 h / 2 12 h 2
h2 h/2

The distance between the x-axis and the b-side is h/2. Using the parallel axis theorem,

bh 3 bh 3 bh 3 h I b = I x + bh = + = 12 4 3 2
Example 2

A right angle triangle has two shorter sides of length b and h. Find the second moment of this triangle about the side of length b. Using this result, calculate its second moment about an axis parallel to that side, but passing through the centroid of the triangle. Put the x-axis along the b side, and the y-axis along the h side. Then the hypotenuse can be expressed as

y=

h b x + h or x = b y b h

There are two ways of going about it. The first method takes a narrow strip of width dy, lengthwise parallel to and at a distance y from the x-axis. Its area is

b xdy = b y dy h
Its second moment about x-axis is

y3 b dy dI x = y 2 b y dy = b y 2 h h
Thus the second moment of the area is
4 3 3 y3 dy = b y y = bh I x = b y 2 4h h 3 0 12 0
h h

We may also take the strips of width dx, parallel to the y-axis. In this case the strip may be considered to be a rectangle. Its area is ydx and its second moment about the x-axis is (from Example 1)

17

dI x =

y 3 dx 1 h = x + h dx 3 3 b

When this is integrated along the x-axis from 0 to b, the same result as above is obtained. Now, the second moment about the centroidal axis can be found by using the parallel axis theorem (remembering that the least second moment of an area is about its centroidal axis)

I xx = I x

1 h bh 3 bh 3 bh 3 bh = = 2 3 12 18 36

Example 3

A twin pontoon type semi-submersible has four columns of diameter 12m. The transverse distance between the centres of the columns is 70m. Calculate the transverse second moment of the waterplane area in its operational condition. Ignore all bracings. The second moment of area of a circle of diameter d about d 4 . its diametrical axis is 64 Semi-submersibles are designed to operate so that only its columns and associated structures (such as bracings) pierce the water surface. The area of the circle is d 2 / 4 . The distance from the centre of the circle to the longitudinal axis of the semi-submersible is 70/2 = 35m. The second moment of the circle about the longitudinal axis (centreline) can be found by using parallel axis theorem
IT =

d 4
64

d 2
4

35 2 = 1017.88 + 138,544.24 = 139,562.12 m 4

Since all four columns are of identical size and at the same distance away from the centreline,
I T = 4 139,562.12 = 558,248.48 m 4

It can be seen that the contribution of the second moment about its own diametrical axis is very small (less than 1%) indeed. Nevertheless, the second moment is comparable to a monohull vessel despite its small waterplane area thanks entirely to the separation. As will be seen later, the transverse second moment of the waterplane is one of the main factors which influence the stability of a vessel. This explains why catamarans and semi-submersibles have very good initial stability characteristics.

18

TUTORIAL No 2

1. Find the first moments of area about x and y axes for the various shapes shown below, and determine the location of their centroid.
y

y (6, 10)
(2, 5) (4.5, 5)

(2, 4)
(1, 1) (6, 1) x

(9, -1)
y

(-2, 5)

(6, 5)
8

30

(2, 1)
30

x (1, -1.5) (6, -1) (4, -2)

(-2, -2)

2. For the waterplane shown below calculate first moment of area about midship and hence LCF from midship; longitudinal and transverse second moments of area (about axes through the centroid).

20

45

35

30

20

All dimensions in m.

19

3. Calculate the longitudinal and transverse second moment of area of a catamaran shape below.

15m

5m

7m 2m

4. An old pentagon-type semi-submersible has 5 columns of 15m diameter equally spaced around a circle of radius 30m. Calculate the second moments of the waterplane of this semi-submersible.

20

1.5. PROPERTIES OF 3-D SHAPES

The first and second moments of area are usually about an axis. The first moment of volume and mass, on the other hand, is calculated about a plane. Their second moment, however, is calculated about an axis, as it is the inertia of the object in rotational motion about that axis (hence the alternative name, mass moment of inertia). Despite these little differences, the properties of volume and mass are really the same as those of 2-D shapes. For instance, with an arbitrary object in space, the vertical mass moment about x-y plane is

M xv = zdm
M

where z is the vertical distance from the reference plane (x-y plane in this case) to the elemental mass dm. M denotes the whole mass. Therefore the centroid of the object is at a point

z=

M xy M

zdm
M

above the reference plane. The same principle applies in the other directions. Note that volume can be treated in exactly the same way as mass. This is how the centres of gravity and buoyancy are calculated. For example, the centre of buoyancy can be found in the following manner. Consider a Cartesian co-ordinate system where the origin is fixed at a point amidships on the centreline on the water plane as shown in Fig. 12. Positive x is forward, y is positive starboard and z is positive upward, maintaining the right-hand rule. Then the moments of the underwater volume about the x-y and y-z planes are

M xy = zdv and M yz = xdv

V V

where V is the whole underwater volume and dv is the elemental volume. The longitudinal position of the centre of buoyancy (LCB) is and its vertical position (VCB) is convention. The mass moment of inertia of an object is calculated about an axis, as mentioned before. For example the moment of inertia about x-axis is

M yz

from the midship

M yz

from the waterline, remembering the sign

I xx =

(y

+ z 2 dm

21

Note that y 2 + z 2 is the squared distance of dm from the x-axis.

Movement of Centroid Due to Addition or Subtraction of Mass or Volume

Consider an object, with mass m1 and the centroid l1 from a reference plane. Some mass m2 is added to it and its centroid is at l2 from the plane. Then the distance of the new centroid from the reference plane, l, is

l=

m1l1 + m2 l 2 m1 + m2

This is quite obvious from the principles that - The distance to the centroid can be obtained by dividing the first moment by the mass; and - The moment of an object about an axis can be found by summing the moments of its component parts about the axis. Of course, when m2 is subtracted from m1, then the sign of m2 will be negative. Sometimes it is more convenient to know how much the centre of gravity moves in this situation. The movement of the centroid (from l1, that is), dl is

dl = l l1 =

m1l1 + m2 l 2 m l + m2 l 2 m1l1 m 2 l1 m 2 l 2 m2 l1 m2 (l 2 l1 ) = l1 = 1 1 = m1 + m 2 m1 + m 2 m1 + m 2 m1 + m2

The numerator of the last is the mass moment of m2 about the plane passing through the centre of gravity of m1 and parallel to the original reference plane. This idea can be applied to the situation where parts of a given object are moved to new locations. This case can be regarded as some parts removed and the same parts added elsewhere. Some worked examples will illustrate this very much more clearly.
Example 4

A ship of mass displacement 10,000 tonnes has its centre of gravity at 7m above baseline. Some cargo of mass 1000 tonnes is loaded onto the ship so that its centre of gravity is at 10m above baseline. Calculate the new vertical position of the ships centre of gravity (VCG). The new KG =

10000 7 + 1000 10 = 7.273m 10000 + 1000

(We may get the same result by taking the moment about the original centre of gravity, thus the new CG relative to the original CG =

10000 0 + 1000 (10 7 ) = 0.273m ) 10000 + 1000

22

Example 5

A ship of mass displacement 8,000 tonnes has its centre of gravity at 4m aft of midship. Some cargo of mass 500 tonnes is moved forward by 20m. Calculate the position of the longitudinal centre of gravity (LCG). We may regard the movement as a combination of two operations: unload 500 tonnes from a location x, say, and load 500 tonnes at a location x + 20m. Taking fwd of midship as positive, The new LCG = 8000 ( 4) 500 x + 500 ( x + 20) ) 8000 ( 4) + 500 20 = = 2.75m 8000 500 + 500 8000 I.e. 2.75m aft of midship. The important thing to note here is that when a mass is simply moved, we do not need to do this in two operations. All we have to do is add the moment of the movement. It is important to note in this case, however, that the total mass has not changed.
Example 6

A ship of mass displacement 10,000 tonnes has its centre of gravity at 8m above baseline. The following operations are carried out: Load Discharge Load Move 500 tonnes 3m AB 300 tonnes 5m AB 200 tonnes 12m AB 1000 tonnes downward by 2m

Calculate new VCG. The new VCG =

10000 8 + 500 3 300 5 + 200 12 1000 2 = 7.731m 10000 + 500 300 + 200

In fact when a number of operations are carried out like this, it is best to compile a table, something like this

Operation Original ship Load Discharge Load Move

Mass (tonnes) 10000 500 300 200 1000

v.c.g. (m) 8 3 5 12 by -2

d 10000 +500 -300 +200 0

Moment 80000 1500 -1500 2400 -2000 80400

New VCG =

10400

80400 = 7.731m 10400

23

PART 2 NUMERICAL INTEGRATION TECHNIQUES

2.1. INTRODUCTION

Not surprisingly integration in various forms keeps appearing in hydrostatics, since area is calculated by integrating line lengths and volume by integrating sectional areas and so on. It is hardly an exaggeration to say that most of hydrostatic calculations were based on moment theories and integration of one sort or another. When we are dealing with geometrical shapes which can be readily expressed in a mathematical form (particularly polynomials), the integration in most cases can be accomplished by manipulating the mathematical expression. Ship hulls are difficult to represent in a convenient mathematical expression (the piecewise polynomials often used to express hull forms are in general more difficult to integrate). In any case, naval architects of bygone era before computers arrived had to devise some methods to do the integration. They developed a number of numerical integration techniques which are used extensively even now. All these numerical integration produces approximate answers, although sufficiently accurate for engineering purposes. Indeed ingenious, albeit somewhat clumsy to use requiring infinite care and patience, devices called integrators were invented to carry out accurate integration, but thankfully these have disappeared from ship design offices, as computers can produce more accurate results in a much shorter time. In this part, we will explore some of the more popular numerical integration techniques. Before doing so, however, it will be useful to summarise various ways of finding some hydrostatic quantities.
To get WP area Section area Integrate breadths breadths WP area section area (SA) Direction longitudinally vertically vertically horizontally

volume displacement vertical moment of displacement longl moment of displacement IT of WP IL of WP

vertical moment of SA vertical moment of WPA longl moment of SA longl moment of WPA transverse second moment of breadths longitudinal second moment of breadths

longitudinally vertically longitudinally vertically longitudinally longitudinally

2.2. TRAPEZOIDAL RULE

Consider a curve A-D as shown in Fig.2.1(a) with the known points of (xi, yi), i = 0, 1, 2, 3 ... n, and we would like to calculate the area ABCD. Note that point (x0, y0) is at A and (xn, yn) is at D. The curve A-D is assumed to be made of n straight line segments.

24

Then an approximate area of ABCD can be found by simply summing the areas of the trapezoids thus formed as shown in Fig.2.1(b).

y0

y1

y2 y3

yn-2 y

n-1

D
yn

x0

x1 x2 x3 .. x n-2

xn

Fig.2.1(a) Area to be calculated

Fig.2.1(b) Simplification to trapezoids

Thus area ABCD =

1 [(x1 x0 )( y 0 + y1 ) + (x 2 x1 )( y1 + y 2 ) + ...... + (x n xn1 )( y n1 + y n )] 2

The accuracy of this technique depends largely on how closely the curve can be represented as a collection of straight lines. It is obvious therefore the more points on the curve are known, the more accurate the result will be. The method is extremely simple to understand and use, and despite its somewhat crude approximation can often produce sufficiently accurate results. Indeed some very sophisticated computer programs sometimes rely on this method. Trapezoidal rule is one of the Newton-Cotes formulae for numerical integration which cover polynomials of any order, trapezoidal being for linear polynomial approximation. For any curve other than straight line, approximating it with a higher order curve will produce more accurate area, and for this we turn to Simpsons rules which are again special cases of Newton-Cotes formulae.

2.3. SIMPSONS RULES

Simpsons rules are the most widely used numerical integration techniques because of their simplicity and wide applicability. Consider a curve A-D as shown in Fig.2.2(a) with three known points the x values of which are such that x2 x1 = x3 x2 (in other words evenly spaced). For ease of calculation we will put the y axis at x = x2, thus making x2 = 0. Let x1 = h, then x3 = h. We now assume that the curve A-D can be expressed as a cubic polynomial of x, thus

y = a 0 + a1 x + a 2 x 2 + a 3 x 3

(2.1)

25

y
y2

D
y3

y1

A
y1 y2

F E
y3 y4 y5

x1

O B
x2

x3

x1

x2

x3

x4

x5

Fig.2.2(a) Simpsons First Rule

Fig.2.2(b) Five equally spaced ordinates

The area ABCD, say A, can be found as

1 1 1 2 A = ydx = a 0 x + a1 x 2 + a 2 x 3 + a3 x 4 = 2a 0 h + a 2 h 3 h 2 3 4 3 h
h

(2.2)

Since

y = y 2 when x = 0 , a 0 = y 2 , and y = y1 when x = h , y1 = a 0 a1 h + a 2 h a 3 h

2 3

(2.3) (2.4) (2.5)

2 3

y = y 3 when x = h , y1 = a 0 + a1 h + a 2 h + a 3 h

Adding (2.4) to (2.5) and solving the result for a2 gives

1 ( y1 + y3 ) y 2 2 a2 = h2
Substituting this into (2.2) gives

1 ( y + y3 ) y 2 2 3 2 1 area ABCD = 2 y 2 h + h 3 h2

h = ( y1 + 4 y 2 + y 3 ) 3

This is known as Simpsons First Rule and a moments reflection will show that as long as three points on the curve, the x-values of which are evenly spaced, are known, this rule can be applied regardless of what those x-values are (remember we derived the above for x2 = 0, but x-values do not appear in the rule except the common interval). This rule is sometimes known as Simpsons 1/3 rule or 1-4-1 rule for obvious reasons.

26

Note that this rule produces a precise value for any polynomial up to cubic. Therefore, the accuracy of this method depends largely on how closely the curve can be approximated as a cubic polynomial. If, on the other hand, the curve contains segments which can only be expressed in polynomials of different orders, for example a cubic curve adjoining a straight line segment, the accuracy will decrease. If we have to deal with 5 points on the curve with the x-values equally distributed as shown in Fig.2.2(b), then we can divide the area into two segments: ABCE and ECFD and apply Simpsons first rule to each portion, thus area ABFD =

h ( y1 + 4 y 2 + y 3 ) + h ( y 3 + 4 y 4 + y5 ) = h ( y1 + 4 y 2 + 2 y 3 + 4 y 4 + y 5 ) 3 3 3
The numbers 1, 4, 2, 4 and 1 are known as Simpson multipliers. If there is a half station at the beginning for example, 0, , 1, 2, 3, 4, 5, the integration can be performed by applying the rule separately to the segments 0, , 1 and 1, 2, 3, 4, 5. Of course the common interval for the first segment will be half of that for the latter. Therefore, the Simpson multipliers in this case will be 0.5, 2, 1.5, 4, 2, 4, 1. One often encounters Simpson integration in a tabular form and a sample of this is given in Table 2.1. Simpsons First Rule, as we have seen, applies to three equally spaced ordinates. There is another Simpsons rule which applies to four equally spaced ordinates, and it is known as Simpsons Second Rule (sometimes known as Simpsons 3/8 rule or 1-3-3-1 rule). With the four ordinates y1, y2, y3 and y4 at x1, x2, x3 and x4, and the common interval h, area =

3h ( y1 + 3 y 2 + 3 y 3 + y 4 ) 8

You can work out how this may be so and prove that it produces accurate results for up to cubic polynomials quite easily.

Table 2.1 An example calculation of area using Simpsons first rule

Station 0 1 1 2 3 4 5 Ordinates 10.00 10.61 11.30 12.07 12.94 14.92 17.25 22.94 30.00 30.00 SM 0.25 1 0.5 1 0.75 2 1.5 4 2 4 f(A) 2.50 10.61 5.65 12.07 9.70 29.84 25.88 91.75 60.00 120.00

27

6 7 8 8 9 9 9 9 10

30.00 19.13 10.50 7.03 4.13 2.88 1.78 0.82 0.00

2 4 1.5 2 0.75 1 0.5 1 0.25

60.00 76.50 15.75 14.06 3.09 2.88 0.89 0.82 0.00

f(A) =

542.00

If the length of the base is 150m, the common interval h = 150/10 = 15m (since there are ten equally spaced stations). Thus, the area =

15 542 = 2710 m2 3

2.4. OTHER NUMERICAL INTEGRATION METHODS

Although Simpsons rules are enough in most cases encountered in naval architecture, there are a few other numerical integration methods which can be useful in certain situations. We shall have a look at Tchebycheffs rules here.
Tchebycheffs two-ordinate rule

As with the Simpsons rules, we start with a function which is expressed as a cubic polynomial of x as in (2.1) for which the definite integral from x = h to x = h is given by (2.2). We now assume that there is a value x = x1 and x2 = x1 such that Area ABCD can be expressed as M ( y1 + y 2 ) , where

y1 = f ( x1 ) = a 0 + a1 x1 + a 2 x12 + a 3 x13 and y 2 = f (x 2 ) = a 0 a1 x1 + a 2 x12 a 3 x13

Substituting these into the expression of area, Area = M 2a 0 + 2a 2 x12 . Comparing this with (2.2) gives

M =h 1 Mx12 = h 3 3
Therefore, x1 =

h 3

28

In other words the area under the curve from x = h to h can be obtained by multiplying the sum of two ordinates at x = 0.57735h and x = 0.57735h with the common interval h. This is for a special case where the base is divided equally about the y-axis. In general, the first ordinate can be taken at x1 which is 0.42265h from the left boundary of the base and the second ordinate similarly from the other end.
Tchebycheffs three-ordinate rule

Again we start with (2.1) and (2.2), but this time we set up three ordinates at x = x1, 0 and x, with their respective values of y1, y2 and y3. Again we express the area ABCD as M ( y1 + y 2 + y 3 ) , and we know

y1 = f ( x1 ) = a 0 + a1 x1 + a 2 x12 + a 3 x13

y 2 = f (0) = a 0

y 3 = f ( x1 ) = a 0 a1 x1 + a 2 x12 a 3 x13

Substituting these into the expression for the area and comparing it with (2.2) gives Area = M 3a 0 + 2a 2 x12 = 2a 0 h + Therefore M =

2 a2 h 3 3

1 2 h h and x1 = 3 2

In other words the first and the third ordinates are at x = 0.707107h . The area is then

2h ( y1 + y 2 + y 3 ) . 3

We have seen two of Tchebycheffs rules, but in fact there is a series of them for various numbers of ordinates and corresponding order of polynomials. For us, however, what we have examined here suffices.

2.5. POLAR INTEGRATION

We have discussed some integration methods with the Cartesian co-ordinate system as the starting point for derivation. However, essentially all these numerical methods can be used for

t2

t1

f (t ) dt . In some cases it is much easier to use polar co-ordinate system

and all these methods can, therefore, equally apply to polar integration. We shall first have a look at how polar integration (sometimes known as radial integration) can be used in certain situations. Consider an area represented by two lines OA and OB and a curve segment AB as shown in Fig.2.3. The angle of the line OA to the x-axis is 1 and OB is 2 away from the x-axis. We construct an elemental area dA of angle d as shown at an angle from the x-axis. Since this area can be regarded as a triangle with the height of rd and base of r,

29

x
1
Fig.2.3 Polar integration

dA =

1 1 rrd = r 2 d 2 2

Therefore, the area is

A=

1 2 r d 2 1 1 2 2 r d r sin = r 3 sin d 3 2 3 1 1 2 2 r d r cos = r 3 cos d 3 2 3

Also, the first moments of the elemental area about the x-axis and y-axis are

dM x = dM y =
Therefore,

and

Mx =

1 3 r sin d 3 2 1 M y = r 3 cos d 1 3
2 1

You can work out how these integrals can be executed with the numerical integration methods discussed in this Part. This type of integration can be useful in the calculation of rotational stability, as will be seen in the study of transverse stability.

30

NUMERICAL INTEGRATION TUORIAL 1. A curve has the following ordinates spaced 1.68m apart: 10.86, 13.53, 14.58, 15.05, 15.24, 15.28, 15.22. Calculate the area under this curve using (a) Simpsons first rule (b) Simpsons second rule (c) the trapezoidal rule. 2. A curvilinear figure has the following ordinates at equidistant intervals: 12.4, 27.6, 43.8, 52.8, 44.7, 29.4 and 14.7. Calculate the percentage difference from the area found by Simpsons first rule when finding the area by (a) Simpsons second rule and (b) the trapezoidal rule. Explain this discrepancy. 3. The effective girths of the outer bottom plating of a ship 27.5m between perpendiculars are given below, together with the mean thickness of plating at each ordinate. Calculate the volume of the plating. If the plating is steel of mass density 7700 kg/m3, calculate the weight in meganewtons.
Ord No Girth (m) t (mm) AP(0) 14.4 10.2 1 22.8 10.4 2 29.4 10.6 3 34.2 11.4 4 37.0 13.6 5 37.4 13.6 6 36.8 12.8 7 28.6 10.4 8 24.2 10.1 9 2.6 10.1 10(FP) 23.2 14.2

4. The half-ordinates of a vessel 144 ft between perpendiculars are given below:

Ord No ord (ft) AP(0) 17.0 1 20.8 2 22.4 3 22.6 4 21.6 5 18.6 6 12.8 7 5.6 FP(8) 0

In addition there is an appendage 21.6ft long abaft the AP whose half-ordinates at equally spaced intervals are respectively, 0.0, 9.6, 14.0, 17.0 (AP). Find the area and position of the centre of the area of the complete waterplane. 5. The cross sectional area of a sip 72m LBP, 11.5m beam and 4.3m draught are as follows: Station CSA (m2) Station CSA (m2) 0 0 7 45 17.1 8 31 1 28 8 24 1 35 9 17 2 40 9 14.2 3 45 9 8 4 46 9 4.9 5 46 10 0 6 46

Determine the volume of displacement, mass of displacement in salt water, LCB, CB, CP, and CM. Compute LCF and longitudinal second moment of the area.

31

PART 3

TRIM AND LONGITUDINAL STABILITY

Usual ships have sufficient longitudinal stability and, therefore, it is of little interest to us in normal circumstances. However, it affects the trim of the vessel directly, and trim is a very important factor in ensuring operational efficiency of ships. We shall briefly examine here the longitudinal stability with emphasis on trim. There are three conditions of trim:
even or level keel trim by the stern trim by the head or bow 3.1 Longitudinal Centre of Floatation (LCF)

Consider the following figure which represents the forward wedge which submerges when the vessel changes trim without alteration in its displacement.

For the elemental strip of half breadth y and length x, the volume is 2yx tanx. Therefore, the total volume of the wedge is

2 tan xydx.
0

Lf

But 2

Lf

xydx is the first moment of the original waterplane about XOX .

Similarly for the aft portion of the vessel, the volume is 2 tan

La

xydx.

(Note that La and Lf are length aft and forward respectively from XOX.) Since the vessel has not changed its displacement, the volumes of these two wedges should be the same, i.e.

2 tan

Lf

xydx = 2 tan xydx or

La

Lf

La

xydx = 0

In other words, the first moments of the original waterplane (about the axis XOX) fwd and aft are the same in magnitude although opposite in sign. This shows that in order for the vessel to change trim only, the trim axis XOX should be such that the first moment

32

of area of the original waterplane about this axis is zero. Put in another way, XOX must pass through the centroid of the waterplane, and this point is called the centre of floatation. The longitudinal position of CF is often referred to as LCF.
3.2 Moment to Change Trim

Consider the following figure which shows a vessel floating at waterline WL. A piece of cargo of mass w tonnes is then moved horizontally through a distance h as shown. This will cause the vessel to change trim about LCF and finally settle at waterline W1L1. When the cargo is moved, the CG of the vessel moves from G to G1. After trimming the CB will be vertically below G1 (since that is the new equilibrium).

The moment causing the trim is the same as the righting moment of the vessel at W1L1, which is GMLtan. If the change in trim is t and the length of the vessel is L, then tan = t/L. We can see that the moment to alter trim by t is GML t/L. Therefore, the moment to change trim by 1cm (or MCT) is

MCT =

GM L tonne-m/cm. 100L

3.3 Changes in Loading Condition

Consider a vessel initially floating in equilibrium. Several masses (m1, m2, , mn) are loaded/discharged at distances (x1, x2, , xn) from the original LCG. We wish to find the final draft and trim when loading/discharging is complete. A table is then prepared as follows: MASSES + m1 - m2 - m3 + m4 ______ LEVER ( from LCG) + x1 + x2 - x3 - x4 MOMENT + m1x1 - m2x2 + m3x3 - m4x4 _______

mx

33

Note:

+ for masses added, fwd of LCG, trim by the bow - for masses discharged, aft of LCG, trim by the stern.

(This sign convention is by no means universal!) If the original displacement was 0,

= 0 + m and
new LCG = (LCG)0 + mx/ LCB can be obtained from the table of hydrostatics at the displacement . This will also give the even keel draft and BML. (See an alternative approach of approximate calculation below.) Trimming moment = * (distance between LCB and LCG) and Trim = (trimming moment)/(100*MCT)

(metres).

3.4 Approximate Method of Calculating Trimming Moment

Consider a vessel originally floating at the waterline WL. At this condition the centres of gravity and buoyancy of the vessel (G0 and B0) are vertically in line, since the vessel is in equilibrium. Let xG the distance of G0 from the midship. (Note that the distance from B0 to midship will also be xG.) A piece of cargo of mass m is loaded at a distance xm from midship as shown in the diagram below.

Let the longitudinal distance of the new centre of gravity (G1) be xG. Then
' xG =

xG + m xm +m

The parallel-immersed strip should displace water of mass m, and its centroid can be said to be approximately at the LCF of WL. If the LCF is at distance xF from the midship, then the new centre of buoyancy B1 will be away from the midship by xB, where

xB =

xG + m xF +m

As we have seen above, the trimming moment = * (distance between LCB and LCG). The distance between LCB and LCG is
' xG xB =

xG + m xm ( xG + m xF ) +m

m( x m x F ) +m

34

The trimming moment, therefore, is

( + m)

m( x m x F ) +m

= m( x m x F )

This approximate method is useful when the approximation used in the derivation (the centroid of the parallel-sinkage strip can be said to be at the LCF of WL) is valid. This usually happens for a conventional ship-shape when the change in displacement is fairly small (for example, draft change of a few cm rather than a few tens of cm). The method allows the trimming moment to be calculated without having to find new LCG and LCB first. With this method, the table given above will have levers calculated from LCF of the original waterline and the moment sum is the trimming moment. This can then be divided by the MCT for the new draft to produce the change in trim.
Example

A vessel 100m L x 30m B is operating at an even keel draft of 6m. The following loading/unloading operations have been carried out at a port
Mass(tonnes) Loading/Unloading xg (m from midships)

100 50 200 50

L U L U

10 fwd 40 aft 30 aft 20 fwd

Calculate the trim and consequently the drafts fwd and aft.

Solution
Mass Lever Moment

+ 100 - 50 + 200 - 50

+ 10 - 40 - 30 + 20

+ 1000 + 2000 - 6000 - 1000

m = 200

mx = - 4000

0 = 100 30 6 = 18,000 m 3

0 = 18000 1025 = 18,450 tonnes .

T1 = 181951 / (100 30) = 6.065 m, trimming moment = - 4000 tonne - m . MCT = 18,650 137.4 / (100 100) = 256.25 tonne - m / cm trim = -4000 / 256.25 = 16 cm by aft. I L = 30 100 3 / 12 = 2,500,000 m 4 BM L = I L / 1 = 137.4 m

1 = 18,450 + 200 = 18,650 tonnes 1 = 18650 / 1025 = 18,1951 m 3 . .

Since LCF is amidships, trim forward and aft will be identical. Thus,

35

Taft = T1 + 016 / 2 = 6145 m, . .

and T fwd = T1 016 / 2 = 5,985 m .

3.5 Derivation of BML

We have already discussed how the waterline appears to swing about the transverse axis through LCF. From the second figure of the notes, it can be seen that BB1 = BM L tan . The centre of buoyancy moves to B1, because some volume has been transferred from aft to forward. Referring to the first diagram in the notes, the longitudinal moment of volume transfer can be calculated by
Lf

moment =

La

( elementary volme ) x = ( 2 yx tan dx ) x = 2 tan

La

Lf

Lf

yx 2 dx

La

The last item in the above expression is actually the longitudinal second moment of the waterplane, IL, multiplied by tan. Also the shift in centre of buoyancy, BB1, can be obtained by dividing the moment by the total volume. Therefore, BB1 = I L tan = BM L tan BM L = I L As stated earlier, some people prefer to use BML instead of GML, but, wherever possible, GML should be used.

36

Part 4 STATICAL TRANSVERSE STABILITY 4.1 INTRODUCTION

All ships should have adequate stability in all their operating conditions. This statement sounds almost too obvious and simple to deserve a serious consideration, but it raises two important questions: firstly, what is stability; and secondly how do we define adequate? We cannot even begin to answer the second until we have answered the former. The word stability can mean different things to different people: for example, mathematicians talk about the stability of solutions, dynamicists discuss the stability of a system, and chemists worry about the stability of a compound they are working on, while many people know roughly what is meant by economic, political or social stability. Although stability is used in Naval Architecture in a variety of contexts as well, for example in directional stability, the unqualified word stability in normal usage has a specific meaning, reflecting our preoccupation in keeping a ship upright, implying that achieving this aim cannot be taken for granted. A more precise definition of ship stability may be the ability of the ship to return to the normal upright condition, when disturbed from this attitude, without endangering itself or the cargo and human life it carries. The keyword here is return, and this means that the disturbance from the normal upright condition (equilibrium), whatever caused it, will automatically generate an opposing force or moment to return it to the equilibrium. The nature of these restoring forces and moments will become clearer if we examine the cases of disturbances in different directions. First of all, it is easy to see that the disturbances on the horizontal plane, i.e. displacements along x- and y-axes and rotation about z-axis (surge, sway and yaw respectively) will not generate any restoring force or moment, as the system in these cases is in complete neutral equilibrium. Displacement along z-axis (heave) can be relied on to produce restoring force, provided the waterplane area is non-zero. In the case of normal ships this, therefore, is no problem, except that there should be sufficient free board so that the ship faces less chance of being swamped. In any case, the disturbance in heave direction is straight forward and does not require complicated theory to explain the phenomenon. On the other hand rotational disturbances about x- and y-axes (roll and pitch, or their static equivalent of heel and trim) are very different indeed. One knows, instinctively, that the disturbance about the y-axis (i.e. trim) is no problem, because longitudinal stability is so high for conventional ships. Nevertheless this is an important factor in determining the performance of ship propulsion and has been discussed in Part 3. This leaves the stability in the transverse direction as the really important issue as we cannot be sure that the ship will generate sufficient moment to return to upright condition when heeled and how much heel the ship can suffer without completely capsizing.

We shall, therefore, examine the transverse stability carefully, with particular emphasis for the cases where the watertight integrity of the ship is intact (intact stability). When we are completely at ease with this we shall examine the approaches normally used for studying the stability when the ship is damaged and the normally watertight skin of the ship is broken.

37

4.2 BASIC PRINCIPLES OF STABILITY

Before we start examining the nature of the restoring moment (or more commonly known as righting moment), we need to make some fundamental assumptions as follows: (a) the ship is floating in calm water (b) the ship is stationary (c) the ship is rigid, i.e. it maintains its original shape whatever it might assume. Some of these assumptions can be removed either partially or wholly, and we shall deal with that later.
Righting Moment

We shall now have a look at the mechanism through which the righting moment is generated when an arbitrary object is inclined from an equilibrium position. Consider an object floating in water at equilibrium. The weight of the object is W and its centroid of mass is at G. Since the object is in equilibrium the underwater W volume of the object will be , where is the mass density of the water and g g is the gravitational acceleration. Furthermore the zero sum of moments dictates that the centroid of the underwater volume, B, be vertically under G as shown in Fig.4.1(a).

L1

L
W1

G B1

Fig.4.1(a) Object in equilibrium

Fig.4.1(b) After

heeling If this object is then inclined by an angle , as shown in Fig.4.1(b), without altering the displacement, the centroid of the underwater volume will move to a new location B1, while the mass centroid of the object remains at G. It is easy to see that except in very special circumstances, B1 will in general move out of the vertical line passing through G. The sum of the forces is zero but a moment is created. Therefore, the object is no longer in an equilibrium, and the resulting moment may be in either direction. We shall define this moment as the righting moment, and it will be positive if it tends to put the object back to the original upright position, and negative if it tends to increase the angle of inclination. The

38

magnitude of this moment is simply the product of the weight W and the distance between the two vertical lines through G and B1. Note that a heeling moment tends to increase the angle of inclination. Therefore, a positive heeling moment will be in the same direction as a negative righting moment. If we divide the righting moment by the weight of the object, we obtain righting lever or righting arm. Of course we can define heeling lever and heeling arm in a similar manner.

Symbols and Terminology

When stability is discussed in naval architecture we need to define the meaning of certain terminology and symbols. While some authors use their own systems, the most commonly used symbols and terminology are illustrated in Fig.4.2.
M L1 W W1 B N K G O B1 Z L

WL: upright waterline W1L1: heeled waterline G: centre of gravity of the vessel B: centre of buoyancy (upright) B1: centre of buoyancy (heeled) K: keel (or the point at which the baseline crosses the centreline) M: metacentre : angle of heel KB: height of V.C.B. KG: height of V.C.G. GM: metacentric height BM: metacentric radius GZ: righting lever (or righting arm)
Fig.4.2 Definition of terms

39

Metacentric Height

It is convenient to assign a positive or negative sign to the angle of heel, and throughout this course we shall use the convention that heel to starboard is positive and heel to port is negative (viewed from the stern). For example, a heel angle of 10 means 10 heel to starboard and -10 denotes 10 to port. A vessel floating in equilibrium at WL is heeled to starboard by an angle so that the new waterline can be represented by W1L1. Note that at this waterline the displacement should be the same as the original one, and for a small angle of heel W1L1 can be assumed to pass through point O of Fig.4.2. The CG of the vessel is still at G (assumed to be on the centreline of the vessel), while CB will move to B1. A righting moment is generated as indicated by the two forces equal in magnitude, but opposite in direction acting on two vertical lines separated by a distance GZ. Thus,

M R = GZ
GZ is the righting lever. For small , it can also be shown that GZ = GM sin GM Remember that is in radians in the above expression. Now that we have the stability quantified, albeit in a very rudimentary way, we need to examine what the point M signifies and how we can calculate GM. Point M, or the metacentre, is formally defined as the point of intersection of the two vertical lines passing through the centres of buoyancy at two very close angles of inclination. It is easy from this definition to see why the length BM is called metacentric radius. The metacentre at the heel angle of zero is called the initial metacentre, or more often simply as the metacentre. The metacentres at non-zero angles of heel are known as pro-metacentres. It is easy to observe that

GM = BM + KB KG
Therefore, if we can calculate BM and KB, and if we know KG, we can obtain GM. The method of calculating KB and KG has already been discussed in Part 1, and thus the problem is reduced to determining BM. Consider Fig.4.3, which shows the whole ship and we attempt now to calculate GM (called metacentric height, or more correctly initial metacentric height). The displacement in the heeled state is the same as the initial displacement. Therefore, the volume of the emerged wedge (out wedge), v1, should be the same as that of the immersed wedge (in wedge), v2. In other words, v1 = v 2 = v , say.

40

dx

L/2

L/2

dx

L1 2y/3 G 2y/3 g2

W g1 W1 y

'out' wedge

'in' wedge

Fig.4.3 Derivation of GM
2 y 3 away from the centreline in transverse direction. The moment of transfer of buoyancy for this thin strip of thickness dx is

For very small angles of heel, the local centre of gravity of the wedges is

2 1 2 y y y tan dx = y 3 tan dx 3 2 3

If we now integrate this moment along the whole length of the ship (i.e., from L L x = to x = ), we obtain the total moment of transfer for the whole ship. 2 2

L 2 L 2

2 2 3 y tan dx = tan 2L y 3 dx 3 3 2

We can relate this moment to the transverse movement of the centre of buoyancy which we will call h, as shown in the enlarged figure below. From the moment theory we know that

41

h=

2 tan 2L y 3 dx 3 2

On the other hand, for small

h BM tan Therefore

BM =

2 2 3 L y dx / 3 2

Note that y is half breadth of the waterline, and therefore is a function of x.

2 The integral 2L y 3 dx happens to be the second moment of the waterplane area 3 2 about the centreline, or IT (transverse second moment of waterplane).
L

Implications of GM

Having found a way of calculating GM, we now need to explore its implications a little before we go further into stability.
(i) GM as a stability parameter

As we have seen, the righting moment of a vessel when heeled to a small angle is Righting moment = GM sin GM In other words GM is a direct measure of stability at small angles of heel. The reason why people prefer GM as a parameter to GZ, for example, is that GM does not vary whilst GZ is a function of . Thus, GM is a parameter showing the initial stability characteristics of the vessel. For reasons which will become apparent later on, at larger angles GM often gives misleading information about stability and consequently it is unwise to use it alone to indicate stability.
(ii) GM as a roll motion parameter

Although the full implication of GM in terms of roll motion characteristics have to be explored later, it is worth noting at this stage that the natural period of roll motion of a vessel is a function of GM, i.e. Tn = 2k GM g

42

where

Tn = natural roll period k = radius of gyration (often about 0.42 B) g = gravitational acceleration
It can be seen, therefore, that a vessel with high initial GM (i.e. a stiff ship) will have short natural roll period, and a soft ship (low GM) will have long natural period. A stiff vessel will roll relatively rapidly in a jerky fashion which may be detrimental to the comfort of passengers and crew. Note the use of terms analogous to the characteristics of a spring.
(iii) Methods of influencing GM

The easiest way of changing GM of an existing vessel is by using ballast, often in solid form. This often is an expensive, albeit necessary, way of increasing GM, as it reduces the carrying capacity of the ship. At the design stage, however, any perceived shortcomings in intact transverse stability can be remedied by increasing the second moment of the area, either by increasing the breadth of the ship, where it is possible, or changing the shape of the waterplane. It must be remembered that the waterplane is often governed by the requirements of speed or resistance.
4.3 FACTORS AFFECTING GM

The GM calculated as above may not be the effective GM for the ship depending on its conditions. There are a number of factors which need to be taken into account when trying to arrive at an effective GM and the two typical cases are suspended weight and free surface effects.
Suspended Weight

Consider a vessel with a weight of mass m suspended from point P as shown in Fig.4.4(a). First, if we assume that the weight is tied to the post so that it does not move, then the righting moment of the vessel for a small angle of inclination is RM = GM sin (Note that the righting moment in this expression is mass moment, and not force moment. This is the usual convention but there are occasions when one needs to use force moment. So watch out.)

43

h m

Fig.4.4(a) Suspended weight in upright condition weight heeled

Fig.4.4(b) Suspended

If the weight is gently released in this heeled condition, the weight will move to a position vertically below the point P as shown in Fig.4.4(b). Then the movement of the c.g. of the weight is

gg1 = h sin
we can now approach the problem in two ways. (i) The heeling moment due to the shift of the weight c.g. is m h sin Therefore the net righting moment = (GM mh ) sin = GM mh sin In other words, GM is effectively reduced by mh / . (ii) The shif of the vessel C.G. due to the shift of the weight c.g. is
GG1 = mh sin

Therefore, the net righting moment mh = (GM sin GG1 cos ) GM sin , since cos 1 for << 1.

44

This is the same expression as obtained in (i). The apparent reduction in GM of

mh is equivalent to an increase in VCG by the same amount and this signifies that, when evaluating the VCG of a vessel, any suspended weight free to swing should be considered to be situated at the points of suspension.

Free Surface

The free surface effects are by far the most important factor which influences the magnitude of initial GM. These effects occur when there is any container with liquid content having free surface, i.e. in the case of an enclosed tank the tank is partially filled so that the surface of the liquid is free to maintain horizontal position whatever the ships attitude (termed a slack tank). Consider a vessel with a tank partially filled with liquid of density T . The density of the water the ship is floating on is W . The righting moment of the vessel if the surface is not allowed to move (e.g. by assuming a rigid surface film) at a small angle of heel is GM sin . This righting moment is often called solid righting moment, and GM is called solid GM and sometimes denoted as GMs.

L W W L

m g g1

2b

Fig.4.5 Vessel with a slack tank

Now the thin film on the surface is removed and the liquid is allowed to move. For the sake of clarity we shall consider a rectangular tank of breadth 2b and length l as shown in Fig4.5. The point m denotes the intersection of the vertical

45

lines through g and g1, where g is the original centre of gravity and g1 is the centre of gravity of the liquid at the heeled condition. Then The moment of volume transfer = Therefore,
gg1 = gm tan = i 2 3 lb tan v = T tan . 3 v

2 2 b b b tan l = b3l tan 3 3

Again there are two approaches. (i) The c.g. of the liquid will move from g to g1 and therefore the C.G. of the vessel moves to G1.

GG1 =
where

i v T i w gg 1 w gm tan w iT = tan = T tan = T T tan = v v

w is the liquid mass = v T v is the liquid volume iT is the local transverse second moment of the free surface area.

Thus, the net righting moment i = (GZ GG1 cos ) = GM T T sin = (GM GG2 ) sin , say. This item GG2, which is equal to correction. The term GM GMf. It can be seen that the effect of the free surface is to increase the VCG effectively by this amount. In other words, recalling a similar situation with suspended weights, the c.g. of the liquid appears to be at the point m and not at g. (ii) Heeling moment due to the shift of the liquid c.g. is

iT T is known as the GM fluid and denoted

iT T , is called the free surface

w gg1 cos = w gm tan cos = w gm sin

Therefore, the net righting moment is

46

w gm GM sin w gm sin = GM sin i = GM T T sin as obtained in (i). It is not difficult to see a certain similarity between free surface effect and suspended weight effect. For suspended weight mh net righting moment = GM sin For free surface w gm net righting moment = GM sin
4.4 INCLNING EXPERIMENT

It can be seen from the above that in order to get an accurate idea of stability we need to know the accurate position of the VCG of the vessel. The VCG is estimated at the design stage, but this estimation is not sufficient. Further more, the VCG can change after being in service for a while. Therefore, all ships have to undergo a measurement process known as inclining experiment or test after launch and at regular intervals thereafter. Since directly measuring KG is not possible, GM is estimated by imposing a known heeling moment and measuring the resulting heel angle. Since KM can be calculated from the hydrostatics, KG can be estimated from this. A clear windless day and a site with minimum current is selected for this experiment. All mooring lines are slackened off and all non-essential personnel are taken off the ship. There should be no suspended weight or slack tank, but, if it is not possible to ensure this, the details of these should be taken so that they can be taken into account. The draught marks around the ship are read off so that the displacement, VCB and LCB can be calculated. Any items which are not a normal part of the ship should also be removed or if it is not possible taken a careful note of. A number of similar weights, w, are then moved across the deck through a known distance, h, so as to incline the vessel to both port and starboard by a small angle. The angles of inclination are carefully measured at each stage by either an inclinometer or a long pendulum with a plumb bob in a bucket of oil to damp its movement. If identical inclining weights are used, then an average change in inclination after each movement of the weight is obtained, and then

47

GG1 =

wh but GG1 = GM T tan wh tan

Therefore GM T =

If a pendulum is used, tan can be obtained by dividing the horizontal movement of the pendulum by its length. Having obtained KG ( KG = KB + BM GM T ), it is relatively simple to undertake corrections for removing non-lightship items, as we need to keep a careful track of KG for the light ship condition.
4.5 LARGE ANGLE STABILITY

The stability characteristics of a vessel at large angles of heel are in general very much different from those of initial stability or stability at very small angles of heel. It may be recalled that stability of a vessel can be quantified in terms of righting moment or, for a given displacement, righting arm. Moreover, the righting arm could be represented by using the initial metacentric height in the form

GZ = GM sin
The most essential assumption in the derivation of this expression was that << 1. In other words, at very small angles of heel we can take the initial metacentric height as the stability parameter. However, as increases, this relationship becomes less valid and beyond the heel angle of about 10 GM loses its meaning as the stability parameter entirely. Therefore, it is now necessary to go back to the righting arm for a meaningful indication of stability. This section examines various ways of calculating the righting levers at large angles of heel, but first we shall discuss what happens to the by now familiar terms of metacentres etc at large angles.

Metacentres at Large Angles

When dealing with initial stability, BM =

IT

was developed with the assumption that

the centre of buoyancy moves horizontally, i.e. that the transfer of wedges was horizontal. Obviously this cannot be true at large angles and the centre of buoyancy moves out round a curve called the isoval or metacentric involute.

48

When the ship is upright, this curve is perpendicular to the centreline of the ship and the radius of curvature is B0 M = I T / . The verticals through the heeled centres of buoyancy do not normally intersect at M and any two verticals very small angles apart will have a corresponding metacentre - the pro-metacentre for that angle. If a curve is drawn from M through these prometacentres, this is the metacentric evolute.
Attwood's Formula (1796)

Consider a vessel heeled to a large angle without any change of displacement. Then the volume of the immersed wedge LSL1 must equal that of the emerged wedge WSW1, but the inclined waterline WL will not necessarily pass through the point O. Let the volume of the wedges be v with centroids g1 and g2, and the volume displacement be .

W G W1 g1 h1 v Z v h2 g2 B L1 L R B1

Fig.4.6 Attwoods formula

Horizontal moment of buoyancy transfer = v h1 h2 = BR Therefore, BR = v h1 h2 Also from the figure BR = GZ + BG sin Therefore, GZ = v h1 h2 / BG sin This is known as Attwood's formula, but in a way it is incomplete, because it does not say how to obtain v h1 h2 . This is done by Barnes for example (see below).

49

Barnes's Method

h2 W b1 W1 W2 h1 b'1 h' 1 O F2 b2

h2 b2
'

'

L1 L2 L

Fig.4.7 Barness method

W2L2 is the waterline we are looking for. v1 = buoyancy of emerged wedge between WL and W1L1 v2 = buoyancy of immersed wedge between WL and W1L1 v = buoyancy of emerged and immersed wedges between WL and W2L2 buoyancy of the layer between W1L1 and W2L2 = v2 v1 b1' , b2' : centroids of v on each side b1 , b2 : centroids of v1 and v2 respectively h1 , h2 : projections of b1 and b2 onto W1L1 respectively h1' , h2' : projections of b1' and b2' onto W1L1 respectively F2 is the projection of centroid of the layer (v2 v1) onto W1L1, but since the layer is very small, we can assume it to be identical to the t.c.f. of W1L1. The moment of volume transfer is
' v h1' h2 = v1 Oh1 + v2 Oh2 ( v2 v1 ) OF2

If we evaluate the RHS of this relation, then we can use Attwood's formula. By using radial integration and introducing a dummy variable as defined in the Fig.4.8,

2r 1 2r 1 v1 Oh1 + v2 Oh2 = r12 d 1 + r22 d 2 cos ( ) dx 2 3 2 3 0 0

L r 3 + r23 = 1 cos ( ) d dx 3 0 0

where r1 and r2 are distance between the point O and side hull port and starboard respectively. v1 and v2 can be calculated in a similar manner as follows:

1 v1 = r12 d dx 2 0 0

50

1 v2 = r22 d dx 2 0 0
Waterplane area =

( r
0

+ r 2 )dx

where r 1 and r 2 are r1 and r2 at = respectively. First moment of area about the longitudinal axis through O is

2 r
0

1 + r 2 dx 2
(note: positive to sta'b'd)

Then OF2 = (first moment)/(area)

This method can produce the righting arm value at any heeling angle without having to find the true heeled waterline W2L2 first.

L1

O L

Fig.4.8 Radial integration used for stability calculation

Wall-Sided Formula (Scribanti)

If the vessel is wall-sided at all points in the length, irrespective of the waterline shape, the areas of immersion and emersion at each section will be equal and opposite right angled triangles and the heeled waterlines for the same displacement will all intersect at the same point on the centreline. Then by geometry distance between g1 and g2 perpendicular to WL = area of each triangle = y2 tan 4 distance between g1 and g2 parallel to WL = 3y Therefore, moment of transfer of buoyancy parallel to WL =

2 y tan 3

(2 y
0

4 tan y )dx 3

51

2 tan y 3 dx 3 0

Moment of transfer of buoyancy perpendicular to WL =

2 y
0

2 tan y tan dx 3
L

1 2 3 = tan y dx 3 0
L

2 3 But y dx = IT , i.e. the transverse moment of inertia of the waterplane. 30

Therefore

X = IT tan or X = BM tan Y = 1 1 IT tan 2 or Y = BM tan 2 2 2

If the transfer of buoyancy had been purely horizontal (i.e. from B to P), the righting lever would have been

GH = GQ sin
But X = BQ tan and also BM tan . Therefore, Q is the initial metacentre M.

y L1 ytan L

Q (M) Z H

W ytan W1 g1

g2

B1 B X Y B X

B1 Y P

Fig.4.9(a) Wall-sided formula

Fig.4.9(b) Geometry of GZ in wall-sided formula

52

The righting lever then is

GZ = GH + HZ = GM sin + Y sin 1 = GM + BM tan 2 sin 2

This formula gives a close approximation of GZ for ships with normal form below the angle of deck edge immersion.
Example

Consider a log of a constant square section of a homogeneous material with the specific gravity of 0.5 floating in fresh water. Its length is L and the side of the square is a.

V = La I=

a 1 = L a2 2 2

1 L a3 12 2L a3 a BM = = 12 L a 2 6 a KB = 4 GM = KB + BM KG = a 12

I.e. the log is unstable when floating on one of its sides. The equilibrium point is when GZ = 0. Based on the wall-sided formula,

tan E =

2GM 2a 12 = =1 BM a6

E = 45
In other words the log floats with a corner down.

53

4.6 CURVE OF STATICAL STABILITY

We can learn a great deal about the stability of the vessel by looking at its intact still water stability curve, and here we shall discuss some of the key features of stability curves which give us the clue. A typical stability curve is shown in Fig.4.10.

Righting moment

Angle of heel
Fig.4.10 A typical intact stability curve

(a) At small angles of , GZ = GM sin . Therefore,

dGZ d = GM cos = GM at = 0 dGZ d is the slope of the stability curve and it can be concluded therefore that the slope of the righting lever curve at = 0 is GM. See Fig.4.11.
GZ

GZ max

GM 1 rad (57.3)

54

Fig.4.11 GM and maximum GZ on the curve

(b) GZ max is the maximum heeling moment the vessel can withstand without capsizing. See Fig.4.11. (c) If C.G. of the vessel is not on the centreline, as static heel can occur. (d) If C.G. moves say from G to G1 vertically, the new righting lever at heel angle is

G1Z1 = GZ GA = GZ GG1 sin for all angles. See Fig.4.12.

(e) If C.G. moves transversely from G to G1, the new righting lever at angle is

G1Z1 = GZ GG1 cos for all angles. See Fig.4.13.

L1

G1
W

Z1
L
W

L1 L

G
W1

A B

Z
W1

G B

Z G1 Z 1 B1

B1

Fig.4.12 Vertical movement of C.G. C.G.

Fig.4.13 Horizontal movement of

(f) The area under the righting moment curve represents work or ability of the vessel to absorb energy imparted to it by external forces. Work required to heel from 2 to 1
2 1

is

g GZ d .

See Fig.4.14 (in this diagram the term g is omitted, as both

items are constant). If the vessel moves from angle 2 to 1, the same amount of work will be done on the vessel in the form of increased kinetic energy in the form of increased roll velocity. It is also important to note that indeed the area under the curve is a kind of potential energy (remember different forms of energy can be mutually converted).

55

GZ

Fig.4.14 Work done to change heel from 1 to 2

(g) Range of (positive) stability is the range of angles within which the vessel has positive stability. The final point at which the stability becomes negative is known as the vanishing point. See Fig.4.15.

GZ

GZ

vanishing angle vanishing angle range of stability range of stability

Fig.4.15 Illustration of vanishing angle (v) and range of stability

(h) Some curves have concave and some others convex shape at the beginning. Think about which one is preferable for the same GM. (i) A negative GM does not necessarily mean the vessel will capsize. It may have a range of positive stability. Note here that the point of positive equilibrium is at the point where the curve crosses the angle-axis upwards for positive angles. (j) For a symmetric vessel with the C.G. at the centreline, the curve can be expressed as 3 5 an odd function of heel angle, for example, GZ ( ) = a1 + a3 + a5 + ...

4.7 ISOCLINE METHOD

We have seen that one of major concerns of the foregoing methods is that the displacement at every heel angle should be identical to the one of upright position. This

56

approach may be called constant displacement method. Clearly the major issue here is having to find the correct heeled waterline (even though Barnes's method gets around this problem by approximation). There is another approach which makes no attempt to keep the displacement constant whatever, and it is called isocline method. The isocline method is an indirect way of obtaining a statical stability curve, and indeed it has an additional advantage that it can cater for any alteration in the loading condition. The result of this method when presented in a graphical form is called cross curves of stability, an example of which is shown in Fig.4.16. It is a convenient and effective method for normal ships but may not be as effective for vessels such as semisubmersibles because the cross curves will then have many discontinuities. A conventional hull form may require only about 10 or so drafts and the results are then faired into smooth curves. The essential procedure of this method can be summarised as follows: (a) Decide on the angles of heel (often these are 15, 30, 45, 60, 75 and 90). (b) For each angle of heel draw several waterlines at least covering the range of displacement of interest. (c) Compute the displacement and righting moment about the vertical line through an assumed centre of gravity (assumed KG is necessary because KG will change according to the loading condition and hence displacement). (d) Plot the righting moments or righting levers on the base of displacement, one curve for each heel angle. The resulting curves can also be presented in the form of solid of stability which is in fact cross curves drawn upon two bases, viz displacement and angle of heel, forming a surface. Often KG of 0 m is assumed and in this case the cross curves are sometimes called KN curves. Whatever value of KG is assumed, it is essential that it is clearly shown in the curves. When using the curves to generate a statical stability curve for a given displacement, it is then necessary to be corrected for the actual KG for that loading condition. This correction is called sine correction (see 4.6 (d) above), since it is of the form

GZ = ( GZ )0 G0G sin
where (GZ)o = GZ values obtained from the cross curves Go is the assumed centre of gravity GoG is distance between the assumed and real centre of gravity (positive if G is above Go and negative if below).

57

GZ (m)
3.5 45

CROSS CURVES OF STABILITY

30

Assumed KG = 3m

60 75

3.0

15 90 90 10 75 60 45

2.5

2.0 5

30

1.5

1.0 5

15 10

0.5

1000

2000

3000

4000

5000

6000

7000

8000

Displacement (tonnes)

9000

10,000

11,000

Fig.4.16 An example of cross curves of stability

58

STATICAL STABILITY TUTORIAL 1

1.

A box-shaped barge of 100mL x 10mB x 3mT is floating in calm sea. (a) Assume a heel angle of 5 and plot the section on a graph paper. Find the new centre of buoyancy and graphically locate the metacentre. Compare the BMT thus obtained with calculated value. Comment on the result. (b) If the VCG of the barge is 4m above the baseline, find the initial metacentric height. (c) How high the centre of gravity can be raised without causing GM to become negative. (d) With KG = 4m, increase length, breadth and draught by 10% one at a time and compute GM for each case. Comment on the results.

2.

A log of square section with each side 20cm long is floating in calm fresh water. The specific gravity of the log is 0.5 and the log is homogeneous. (a) Find the initial GM (b) Find the attitude of stable equilibrium.

3.

A box-like vessel of 100mL x 20mB x 10mD is operating in sea water at a draught of 6m. In this condition KG = 7m. It has a box-shaped oil tank measuring 10mL x 10mB x 3mD. The centreline of the tank is at midship and the tank is initially full. Assume that the tank rests on the bottom of the vessel. (a) Find the initial GM. (b) If half of the oil (specific gravity 0.9) is pumped out, what is GMf. (c) Instead of pumping the oil out as in (b), one third of the original full tank content is pumped into a settling tank measuring 10mL x 5mB x 3mD and situated right above the oil tank. Find GMf for this situation.

4.

A box-like vessel, 180mL x 31.5mB, floats in sea water at a level keel draught of 12m and in this condition the value of GM is 0.891m. A tween deck space, 30mL x 31.5mB, is accidentally flooded with sea water to a depth of 1.35m. The deck on which the water is lying is level and is 12.6m above the baseline of the vessel. Estimate the virtual GM of the vessel. (Note: GMf is sometimes called virtual GM.) A vessel 100m long and 6m deep is of a constant rectangular section, 12m broad and symmetrical about the centreline. It is floating in sea water initially at a draught of 3m even keel. Weight is added so as to trim the vessel in such a way that the deck at one end is just immersed while the draught at the other end remains at 3m. Calculate the magnitude of the weight required and the position of its centre of gravity. A rectoidal vessel, 50mL x 12mB, is operating in sea water at a level keel draught of 3m. It goes aground on a narrow ledge 15m from the aft end on the centreline of the vessel. The tide then falls by 0.5m. Find its final attitude.

5.

6.

59

STATICAL STABILITY TUTORIAL 2

1.

A box-shaped vessel, 36m long x 6.4m beam, floats in sea water at a level keel draught of 2.44m. In this condition the value of GM is 0.789m. Cargo is now loaded, part with c.g. at 1.22m above the bottom of keel and the remainder with c.g. at 2.5m. The final draught is 2.75m even keel and the final KG is 1.804m. Find the weight of cargo loaded at each position.

2.

The following data, which may be assumed to be constant except for Taft, are given for a vessel as follows: TPC = 18.0 tonnes/cm, MTC = 160 tonne-m/cm LCF is amidships, original Taft = 10.0m After ballast is discharged from a tank with c.g. at 52m aft of midship, the vessel is found to float at a mean draught of 9.5m and to have a trim of 50cm by the stern. Calculate the amount of ballast discharged and the original draught forward.

3.

A rectangular vessel 100mL x 20mB x 8mT goes aground at one end. VCG of the vessel is 10m and LCG is amidships. The tide then falls by 1m. In order to effect refloating of the vessel several pieces of cargo weighing total 1000 tonnes with their centroid at 30m aft of midship are to be moved. Find the minimum horizontal distance by which they have to be moved. A box-like barge, 80mL x 15mB x 10mD, is operating in sea water at an even keel draught of 3m. When some cargo of mass 200tonnes was moved horizontally in the transverse direction by 3.2m, the barge heeled by 3. - Estimate GMT and BMT, and hence the position of the VCG. - If the same cargo is moved aft by 10m instead of transversely, what will be the trim and hence the draughts forward and aft? - Having returned the moved cargo to its original position, it is now decided to load the barge with some containerised cargo of mass 2000tonnes on the deck. The centre of gravity of the additional cargo when loaded will be at 11.5m above keel. Estimate the transverse metacentric height. - What is the allowable cargo mass which can be loaded on the deck, if the minimum allowable GMT is 1m? (Ignore trim and assume the c.g. of the cargo at 11.5m above keel.)

4.

5.

A semi-submersible comprises two rectangular cross-section pontoons, each 90m long, 11m wide and 7m deep. The deck is supported by four cylindrical columns each 9.5m in diameter, the centrelines being spaced 70m apart longitudinally and 60m apart transversely. The vessel floats at a draught of 23m in sea water and has a vertical centre of gravity 18.5m above the keel. Calculate the transverse metacentric height. A mass of 800tonnes is added on board at a position 35m above the keel. Calculate the new values of draught and vertical centre of gravity.

60

PART 5 DAMAGE STABILITY AND WATERTIGHT SUBDIVISION

5.1

INTRODUCTION

A damage to a vessel which compromises the watertight integrity of the hull will lead to ingress of water into the compartment(s) of the vessel. The flooding consequent to the broaching of the watertight skin of the ship will affect the attitude of the vessel, i.e. trim, draught and heel, and the stability characteristics will be affected, usually to the worse. The remaining stability after sustaining damage is known as residual or damage stability. All types of vessels are subject to risk of being lost if they are damaged whether by collision, grounding or internal mishaps such as fire and explosion. Such accidents are frequent enough in practice that some degree of protection against the eventualities of flooding should be given. For example, sufficient residual stability is provided so that the survival of the passengers/crew and ultimately the vessel and cargo. On of the difficulties in doing this is the fact that damages are not planned (as distinct from the design activities which work on planned state of affairs), and thus are unpredictable. It is not known where the damage will occur and what the extent will be indeed it is unknown whether the vessel will sustain any damage during its lifetime at all. This means that we have to consider the effects of possible damage scenarios and their probability of occurrence. The probability of occurrence has been traditionally incorporated into the damage stability regulations as a factorial system in the practice of watertight subdivision, but with the industry having taken a step towards goalsetting rather than prescriptive regulatory regime, the so-called probabilistic assessment of survivability of ships and passengers/crew is in vogue at present time. Whatever the stability regulatory regime, we need to have some methods of dealing with various damaged situations. Clearly we need to understand first, however, in what way the ship is affected after a damage. We can consider here only the outwardly visible effects which can be readily measured directly as follows: Draught The draught will change so that the displacement of the remaining unflooded part of the ship is equal to the displacement of the ship before damage less the weight of any liquids or other cargo which were in the space opened to the sea. Trim The underwater body shape changes and the flooded water also causes the centre of gravity to change. It is obvious that a new state of equilibrium will involve an alteration in trim. Heel If the flooding is asymmetrical, a static heel will be inevitable. A static heel poses a danger of capsizal and, therefore, regulations demand that in such cases crossflooding arrangement be made so that port and starboard balance can be achieved within a short time. Nevertheless, it is important to note that until such a balance can be achieved, the ship may go through interim state of static heel. Stability It is reasonably obvious why the stability characteristics will alter, and this will be discussed in more detail.

61

Freeboard One of the unpleasant consequence of the increase in draught is the reduction in the amount of freeboard available. Freeboard is a kind of reserve buoyancy and stability, and its reduction can be serious. For example, the so-called margin line is used to define the minimum freeboard allowed in any damaged states. Loss of ship Where changes to any one or more of the above factors are excessive beyond certain limits, loss of ship can occur, either through capsizal, plunging, sinking, structural breaking up or any combination of these processes. It can be seen that the hydrostatics of a damaged ship is somewhat more complicated. Essentially, however, it is still a floating body, and, as such, exactly the same principles as used in intact hydrostatics can be applied. The subject of damaged stability and the safety or survivability of damaged ships is a complex one and can easily form a separate subject. The discussion in this part, therefore, is necessarily brief and only deals with the most basic ideas.

5.2

INTERMEDIATE STAGES OF FLOODING

During the intermediate stages of flooding (i.e. from the time of damage to the time when a final new equilibrium has been reached) water is continually flooding into the damaged compartment(s). Flooding may not be continuous, as the water flow is determined by the difference in head of the surface inside and out. On the outside the surface is continuously changing due to wave action, while inside the water already flooded in will slosh about due to the vessel motion. Therefore, the water flow may not even be one-way all of the time. Nevertheless, short of tackling this with a time-domain simulation, some simplification is necessary and usually the following assumptions are made: The vessel is assumed to be in static equilibrium at every stage of flooding, with water surface in the flooded compartment parallel to (but at a lower level) the surface of the sea. In the case of asymmetrical flooding involving spaces, which are cross-connected by pipes, ducts, etc., it is normally assumed that the flooding water in the damaged wing spaces reaches sea-level. During the intermediate stages of flooding, heel may occur either as a result of negative residual GM or from asymmetrical flooding. Since we are dealing with a transient state, some heel is acceptable provided it is not excessive and long-lasting and also the range of stability and maximum righting arm include sufficient margin of safety.

5.3

EQUILIBRIUM, DRAUGHT AND STABILITY AFTER FLOODING

In order to calculate the ships attitude in equilibrium, broadly speaking two methods are used representing perhaps the two fundamental ways of looking at the flooded water.

62

Lost Buoyancy Method In this approach the damaged compartment is treated to be completely open to sea and therefore no longer contributes to the ships buoyancy. Just imagine the normally watertight hull being suddenly changed to sieves with large holes! The compartments flooded are consequently removed completely from the hydrostatic calculations. To compensate for this lost buoyancy, the ship has to sink further to an increased draught until the displacement becomes identical to that of before the damage, as the ships mass is assumed to be unchanged. Te centre of gravity of the ship remains unchanged. It is easy to see that this is a physical over simplification, but the method is simpler to use than the added weight approach. Added Weight Method In this approach the damaged compartment is treated to be still intact, and the flooded water is treated in exactly the same manner as ballast water taken on board. In other words the mass displacement of the ship has been increased by the amount of water ingress. An increase in draught occurs to provide the extra buoyancy for this added weight of flood water. This of course will alter the centre of gravity of the vessel as well. The method is a little more complex than the LBM, but represents the reality a little more closely. It is useful for dealing with the intermediate stages of flooding and also when we wish to work out the length and location of the compartment which can be flooded without the waterline going over the margin line (floodable length calculation). It is important to note that both methods should produce identical values of the physically measurable quantities. For example, draught and righting moment are some of the items which can be measured, while the location of centre of gravity cannot be measured as it is essentially a conceptual point. The main points of the difference between the two approaches are summarised below:
LBM AWM

Displacement VCG Draught VCB Righting moment Righting arm Free surface effects

( KG )0
T + T

( KG )1 ( KB )2
R.M.

( KB )1
R.M .
n.a.

R.M . ( + )
yes

It is crucial, therefore, to understand that GM value alone is insufficient information unless accompanied by a displacement used (or more commonly the method used to calculate it, i.e. LBM or AWM).

63

Before we examine the two methods using some examples, we need to determine how much water a damaged compartment can admit into it. This point is dealt with by a factor called permeability, which is defined as the ratio of the volume which can be occupied by flood water to the total nominal volume of the compartment. Typical values of permeability are: accommodation spaces 95% machinery compartments 85% coal bunkers, stores, cargo holds tanks 0 95%

60%

Of course, cases can be made for differing values of permeability depending on circumstances. Example 1 To illustrate the general effects and principles, consider a box-shaped vessel as being the only one in which the numerical work is not unduly laborious. Take a box-shaped vessel, 120mL x 21mB x 8m even keel draught with KG = 6.0m. A compartment of length 30m and spanning the whole breadth of the vessel, situated amidships is flooded due to a damage to the hull. Assume 100% permeability. Calculate the residual GM. We will do this with both methods, but with LBM in the first instance.
LBM

original WPA = 120 x 21 = 2520m2 lost WPA = 30 x 21 = 630m2 Therefore, intact WPA = 2520 630 = 1890m2 intact TPC = 1890 x 1.025/100 = 19.37 tonnes/cm lost buoyancy = 30 x 21 x 8 x 1.025 = 5166 tonnes sinkage = 5166/1937 = 2.67m new draught = 10.67m thus, KB = 5.33m new BM =

(120 30 ) 213
12 120 21 8

= 3.45m

new KM = 5.33 + 3.45 = 8.78m KG = 6.00m Therefore, new GM = 8.78 6.00 = 2.78m Remember, the displacement is still 120 x 21 x 8 = 20,664 tonnes.
AWM

Let new draught be T. Since original displacement plus the added weight is the new displacement,

120 21 8 1.025 + 30 21 T 1.025 = 120 21 T 1.025

Solving it for T we get T = 10.67m and consequently KB = 5.33m

64

Now we must obtain the new KG new KG =

20, 664 6 + 6890 5.33 = 5.83m ( 20, 664 + 6890 )

30 213 = 3.45m 12 120 21 10.67

free surface correction =

Therefore residual GM = 5.33 + 3.45 5.83 0.86 = 2.09m Remember the new displacement = 27,554 tonnes. As can be seen from this example the residual GM values calculated by the two methods are not the same. However, the righting moment which is directly proportional to GM at small angles of inclination should be the same. Therefore we should get identical values of (displacement x GM). From the LBM 20,664 x 2.78 = 57,500 From the AWM 27,554 x 2.09 = 57,500. When the flooding is not symmetrical, the change in GM is accompanied by a heeling moment. The resulting heel makes the added weight method quite clumsy to use as will be illustrated by a simple case here. Assume a compartment is flooded as shown. With the LBM, it is quite easy to calculate the lost buoyancy, and thus the sinkage to W1L1. Then, of course due to the imbalance in buoyancy a static heel occurs which will be about the new transverse centre of flotation, F. With the AWM, however, we cannot determine W1L1 directly because we do not know the amount of flooded water until that waterline is determined. Even when that is done, as indeed it can be done through a graphic method, problem does not go away, because the heeling alters added weight yet again and this in turn changes the heeling moment and so on. This time, it is more difficult to decide the heel angle. To overcome this problem a hybrid method, called the quasi-added weight method, is sometimes used. Example 2 Consider the previous box-shaped vessel, 120mL x 21mB x 8m even keel draught with KG = 6.0m. A compartment of length 30m and spanning the entire starboard half-breadth of the vessel, situated amidships is flooded due to a damage to the hull. Although it is a quite meaningless exercise, we wish to find the original KG of the vessel before flooding when the resulting angle of heel is such that tan = 0.25 or 14 . This simplifies the problem so that we can examine it here. Assume 100% permeability.
LBM

original TPC = 25.83 lost TPC = 3.23 intact TPC = 22.60 tonnes/cm lost buoyancy = 1.025 x 30 x 10.5 x 8 = 2583 tonnes

65

parallel sinkage = 1.14m new draught = 9.14m KB = 4.57m For finding new BM Area original WP lost WP Residual 2520 -315 2205 lever from CL 0 5.25 1st mt 0 -1653.75 1653.75 2nd mt 0 8682 IT own CG 92610 2894 IT 92610 -11574 81034

CF = 1653.75/2205 = 0.75m from the centreline. Therefore the residual IT about the new TCF is IT = 81,034 2205 * 0.752 = 79,794m4 Displacement = 20,160m3 BM = 79794/20160 = 3.96m In this particular example, since the compartment is flooded to the bottom shell and the vessel is of box-like shape, the transverse shift of the centre of buoyancy is the same as the transverse shift of the centre of flotation, viz. 0.75m. The centre of gravity of the vessel is still on the centreline, and, therefore, the heeling moment is 0.75 cos . Using the wall-sided formula and equating the heeling moment to the righting moment,

0.75 = GM tan14 +
giving GM = 2.88m

3.96 tan 3 14 2

since KM = 8.53m, KG = 5.65m.

AWM

Let the draught on the centreline after flooding be T. Original displacement + added weight = new displacement Thus, Therefore, T = 9.33m. KB = 4.66m and new displacement = 24,108 tonnes. added weight = 3444 tonnes. CG of the flooded water

120 21 8 1.025 + 30 10.5 (T + 5.25 tan14 ) 1.025 = 120 21 T 1.025

a + 2b h 9.33 + 2 11.96 10.5 = = 5.47 m from the C.L. = 9.33 + 11.96 3 a+b 3
heeling lever of the flooded water = 3444 x 5.47 / 24,108 = 0.781m BM = (21 x 21)/(12 x 9.33) = 3.94m

66

Using wall-sided formula,

0.781 = GM tan14 +

3.94 tan 3 14 2

resulting in GM = 3.00m KM = 8.60m Therefore, KG = 5.60m in flooded condition. The v.c.g. of the flooded water is the vertical position of the trapezoidal shape and it is 5.35m. Tonnes v.c.g. Moment 24108 5.60 135,005 -3444 5.35 -18425 _____________________________________ 20,664 116,580 Therefore original KG = 5.65m. Please note that no free surface correction was necessary, as we already had the waterline at the final equilibrium and that the true centre of gravity of the flood water was used for calculating the heeling moment. The alternative was to take the CG of the flood water at the centre of the tank and make a free surface correction in the righting moment, or moment of added water = 3444 x 5.25 = 18,081 heeling lever = 0.750m giving GM = 2.88m f.s.e. = 2894/2418 = 0.12m corrected GM = 3.00m.

In the above two examples the flooding was limited to midship compartments or at worst symmetrical flooding about midship was assumed. In general flooding due to damage, however, is more likeley to be asymmetrical longitudinally. We need, therefore, to include the effects of flooding on trim. This is nearly always done with the LBM and the process is roughly summarised as follows: (a) calculate permeable volume of compartment up to original WL; (b) calculate TPC, longl and transverse positions of CF with the damaged area removed; (c) calculate revised second moments of waterplane area about the CF in the two directions and, hence new BMT and BML. (d) calculate parallel sinkage and the rise of CB due to vertical transfer of buoyancy from the flooded compartment to the parallel layer; (e) calculate new GMs (f) calculate angles of rotation due to the eccentricity of the loss of buoyancy from the new CFs. Example 3

67

A compartment having a plan area at the waterline of 100m2 and centoid 70m fwd of midships, 13m to starboard is bilged. Up to the waterline obtaining before bilging, the compartment volume was 1000m3 with the centres of volume 68.5m fwd of midships, 12m to starboard and 5m above keel. The permeability was 0.70. Before the incident the ship was floating on an even keel draught of 10m at which the following particulars are given

= 30, 000 tonnes KG = 9.40m KMT = 11.40m KB = 5.25m

KML = 170m WPA = 4540m2 LCF = 1m fwd of midships LBP = 220m

Calculate the heel and trim when the compartment is bilged. Use lost buoyancy method. permeable volume = 0.7 x 1000 = 700m3 damaged WPA = 4440m2 movement LCF =

100 ( 70 1) 4440

= 1.55 , i.e. LCF moves by 1.55m aft.

new LCF = -0.55m, or 0.55m aft of midships. movement TCF =

100 13 = 0.29m or 0.29m to port. 4440

new TCF = -0.29m.

4 original IT = (11.40 5.25 ) 0.975 30, 000 = 179.889m

damaged IT = 179,889 100 132 4440 ( 0.29 ) = 162, 620m 4

2

(we have ignored I of the damaged part about own axis for speed) damaged BM T =

162, 620 = 5.56m 0.975 30, 000

6 4 original I L = (170.0 5.25 ) 0.975 30, 000 = 4.819 10 m

damaged I L = 4.819 106 100 ( 69 ) 4440 (1.55 ) = 4.332 106 m 4

2 2

damaged BM L =

4.332 106 = 148.1m 0.975 30, 000

parallel sinkage = 700/4440 = 0.16m rise of CB =

700 (10 + 0.08 5 ) 0.975 30, 000

= 0.12m

damaged GM T = 5.25 + 0.12 + 5.56 9.40 = 1.53m damaged GM L = 5.25 + 0.12 + 148.1 9.4 = 144.1m

68

Angle of heel =

700 12.29 180 = 11.0 0.975 30, 000 1.53

Angle of trim

700 ( 68.5 1 + 1.55 ) 0.975 30, 000 1.53

= 0.01147radians

Change of trim = 0.01147 220 = 2.52m between perpendiculars.

69