RESISTANCE & PROPULSION

OF SHIPS

1
 Module 1
Introduction

The estimation of ship propulsive power is fundamental to the process of designing and
operating a ship. A knowledge of the propulsive power enables the size and mass of the
propulsion engines to be established and estimates made of the fuel consumption and
operating costs. The estimation of power entails the use of experimental techniques,
numerical methods, theoretical analysis, and software analysis for the various aspects of the
powering problem. The requirement for this stems from the need to determine the correct
match between the installed power and the ship hull form during the design process. An
understanding of ship resistance and propulsion derives from the fundamental behaviour of
fluid flow.

 History
Up to the early 1860s, little was really understood about ship resistance and many of the
ideas on powering at that time were erroneous. Propeller design was very much a question
of trial and error. The power installed in ships was often wrong and it was clear that there
was a need for a method of estimating the power to be installed in order to attain a certain
speed.
In 1870, W. Froude initiated an investigation into ship resistance with the use of models.
He propounded that the total resistance could be divided into skin friction resistance and
residuary, mainly wavemaking, resistance. His proposal was initially not well received, but
gained favour after full-scale tests had been carried out. HMS Greyhound (100 ft) was towed
by a larger vessel and the results showed a substantial level of agreement with the model
predictions. Model tests had been vindicated and the way opened for the realistic prediction
of ship power.
Before 1910, propeller design was hampered by a lack of understanding of negative, or
apparent, slip; naval architects were not fully aware of the effect of wake. In 1910, Luke [1.8]
published the first of important papers on wake, allowing more realistic estimates of wake
to be made for propeller design purposes. Cavitation was not known as such at this time,
although several investigators, including Reynolds, were attempting to describe its presence
in various ways. As a result, propeller blade area was based simply on thrust loading,
without a basic understanding of cavitation.
By the 1890s the full potential of model resistance tests had been realised. During these
years various tests were also being carried out on series of models. The next era saw a
steady stream of model resistance tests, including the study of the effects of changes in hull
parameters and the effects of shallow water and to challenge the suitability and correctness
of the Froude friction values. There was an increasing interest in the performance of ships in
rough water. Several investigations were carried out to determine the influence of waves on
motions and added resistance, both at model scale and from full-scale ship measurements.
From about the start of the 1980s, the potential future of computational fluid dynamics
(CFD) was fully realised. This would include the modelling of the flow around the hull and
the derivation of viscous resistance and free-surface waves.

 Powering: Overall Concept

The overall concept of the powering system may be seen as converting the energy of the
fuel into useful thrust (T) to match the ship resistance (R) at the required speed (V) (see the
figure below). The main components of powering may be summarized as the effective
power PE to tow the vessel in calm water, where PE = R x V and the propulsive efficiency ƞ,
leading to the propulsive (or delivered) power PD , defined as: PD = PE / ƞ.

2
Reference:
Molland, A. F., Turnock, S. R., Hudson, D. A. (2011), Ship Resistance and Propulsion,
Cambridge University Press.

3
 Module 2
Steps and Theories for Estimating the Propulsive Power

 Components of Propulsive Power

During the course of designing a ship it is necessary to estimate the power required to
propel the ship at a particular speed. This allows estimates to be made of:
(a) Machinery masses, which are a function of the installed power, and
(b) The expected fuel consumption and tank capacities.
The power estimate for a new design is obtained by comparison with an existing similar
vessel or from model tests. In either case it is necessary to derive a power estimate for one
size of craft from the power requirement of a different size of craft. That is, it is necessary to
be able to scale powering estimates.
One fundamental division in conventional powering methods is to distinguish between the
effective power required to drive the ship and the power delivered to the propulsion unit(s).
The power delivered to the propulsion unit exceeds the effective power by virtue of the
efficiency of the propulsion unit being less than 100%.
The main components considered when establishing the ship power comprise the ship
resistance to motion, the propeller open water efficiency and the hull–propeller interaction
efficiency, and these are summarised in Figure 1.

Fig. 1

Ship power predictions can be made either by:

(1) Model experiments and extrapolation, or
(2) Use of standard series data (hull resistance series and propeller series), or
(3) Theoretical (e.g. components of resistance and propeller design).
(4) A mixture of (1) and (2) or (1), (2) and (3).
(5) Use of softwares.

Reference:
Molland, A. F., Turnock, S. R., Hudson, D. A. (2011), Ship Resistance and Propulsion,
Cambridge University Press.

4
 Propulsion Systems
When making power estimates it is necessary to have an understanding of the performance
characteristics of the chosen propulsion system, as these determine the operation and
overall efficiency of the propulsion unit.
A fundamental requirement of any ship propulsion system is the efficient conversion of the
power (P) available from the main propulsion engine(s) [prime mover] into useful thrust (T)
to propel the ship at the required speed, Figure 2.

Fig. 2

There are several forms of main propulsion engines including:

- Diesel engine.
- Gas turbine.
- Steam turbine.
- Electric motor.
- (And variants / combinations of these).
and various propulsors (generally variants of a propeller) which convert the power into
useful thrust, including:
- Propeller, fixed pitch (FP).
- Propeller, controllable pitch (CP).
- Ducted propeller.
- Waterjet.
- Azimuthing podded units.
- (And variants of these).
Each type of propulsion engine and propulsor has its own advantages and disadvantages,
and applications and limitations, including such fundamental attributes as size, cost,
efficiency and maintenance requirements. All of the these propulsion options are in current
use and the choice of a particular propulsion engine and propulsor will depend on the ship
type and its design and operational requirements, as well as ship's level of quality.

Fig. 3

5
 The overall assessment of the marine propulsion system for a particular vessel will
therefore require:
- A knowledge of the required thrust (T) at a speed V, and its conversion into required
power (P).
- A knowledge and assessment of the physical properties and efficiencies of the
available propulsion engines.
- The assessment of the various propulsors.
- The assessment of the various engine-propulsor layouts.

References:
 Molland, A. F., Turnock, S. R., Hudson, D. A. (2011), Ship Resistance and Propulsion,
Cambridge University Press.
 Basic Principles of Ship Propulsion, likely to be compiled by a MAN B&W team.

 Definitions (see also Fig. 4)

A. Effective power (PE) = power required to tow the ship at the required speed
= total resistance × ship speed
= RT × VS

B. Thrust power (PT) = power delivered by the propeller to water

= propeller thrust × speed past propeller
= T × VA

C. Hull efficiency (ƞH) = PE / PT

ƞH = (1 - t) / (1 - w)
T -RT
Thrust deduction coefficient (t) = (Basic Principles of Ship Propulsion)
T
V - VA
Wake fraction coefficient (w) = S (Basic Principles of Ship Propulsion)
VS
D. Delivered power (PD) = power required to be delivered to the propulsion unit (at the
tailshaft).

E. Propeller efficiency, behind hull (ηB) = PT / PD

ηB = ηO x ηR

ηO : Propeller efficiency, open water

ηR : Relative rotative efficiency

F. Quasi-propulsive coefficient (QPC) (ηD) = PE / PD

The QPC is also called propulsive efficiency.

6
 The total installed power will exceed the delivered power by the amount of power lost in
the transmission system (shafting and gearing losses), and by a design power margin to
allow for roughness, fouling and weather, i.e. :
PD
G. Installed Power (PI )   margin (roughness , fouling and weather)
ηT
where ƞT is the transmission efficiency of the propulsion system.
The installed power is also often called brake power (PB).
The required power margin for fouling and weather will depend on the areas of operation
and likely sea conditions and will typically be between 15% and 30% of installed power.

Note that PS in Fig. 4 above refers to the shaft power, different with the service power
defined subsequently.

References:
 Molland, A. F., Turnock, S. R., Hudson, D. A. (2011), Ship Resistance and Propulsion,
Cambridge University Press.
 Basic Principles of Ship Propulsion, likely to be compiled by a MAN B&W team.

7
 Components of the Ship Power Estimate
The various components of the ship power estimate and the stages in the powering process
are summarized in Figure 5.
The total calm water resistance is made up of the calm water resistance and the air
resistance, where the calm water resistance itself consists of the bare hull resistance and the
resistance of appendages.
The effective power, i.e. the power required to propell the ship at the required speed, is
obtained by multiplying the total calm water resistance with the required speed.
The propeller quasi-propulsive coefficient (QPC), or the propulsive efficiency, ηD , is
made up of the hull efficiency ηH and the propeller efficiency behind hull ηB , where the
propeller efficiency behind hull ηB itself is made up of the open water propeller efficiency
ηO and the relative rotative efficiency ηR . Meanwhile the hull efficiency ηH is derived as
(1 - t)/(1 - w), where t is the thrust deduction factor and w is the wake fraction coefficient.
For clarity, the model-ship correlation allowance is included as a single-ship correlation
factor, SCF, applied to the overall delivered power. Current practice recommends more
detailed corrections to individual components of the resistance estimate and to the
components of propeller efficiency.
Transmission losses, ηT , between the engine and tailshaft/propeller are typically about
ηT = 0.98 for direct drive engines aft, and ηT = 0.95 for transmission via a gearbox.
The margins in stage 9 account for the increase in resistance, hence power, due to
roughness, fouling and weather. They are derived in a scientific manner for the purpose of
installing propulsion machinery with an adequate reserve of power. This stage should not be
seen as adding a margin to allow for uncertainty in the earlier stages of the power estimate.
The total installed power, PI, will typically relate to the MCR (maximum continuous
rating) or CSR (continuous service rating) of the main propulsion engine, depending on the
practice of the ship operator.

Reference:
Molland, A. F., Turnock, S. R., Hudson, D. A. (2011), Ship Resistance and Propulsion,
Cambridge University Press.

8
 Fig. 5

 Brief Review of Some Fluid Mechanics Concepts

 Classification of Fluid Flow
The flow of fluids can be classified as steady or unsteady, and uniform or non-uniform. In
steady flow the various parameters such as velocity and pressure at any point in the flow do
not change with time. Whereas an uniform flow over a region is developed when the
various parameters such as velocity and pressure do not change from point to point over
that region, at a particular instant.
For example, in a constant section pipe (and neglecting the region close to the walls) the
flow is steady and uniform. In a tapering pipe, the flow is steady and non-uniform. If the flow
is accelerating in the constant section pipe, then the flow will be unsteady and uniform, and
if the flow is accelerating in the tapering pipe, then it will be unsteady and non-uniform.

 Streamline

9
A streamline is an imaginary curve in the fluid across which, at that instant, no fluid is
flowing (see Fig. 7). At that instant, the velocity of every particle on the streamline is in a
direction tangential to the line, for example line a–a in Figure 7. This gives a good indication
of the flow, but only with steady flow is the pattern unchanging. The pattern should
therefore be considered as instantaneous. Boundaries are always streamlines as there is no
flow across them. If an indicator, such as a dye, is injected into the fluid, then in steady flow
the streamlines can be identified. A bundle of streamlines is termed a streamtube.

Fig. 7

 Continuity of Flow
Continuity exists on the basis that what flows in must flow out. For example, consider the
flow between (1) and (2) in Figure 8, in a streamtube (bundle of streamlines). For no flow
through the walls and constant flow rate and density, then for continuity,

Q = A1V1 = A2V2 = constant.

Where Q is the volume flow rate in m3/s.

Fig. 8

 Forces Due to Fluids in Motion

Forces occur on fluids due to accelerations in the flow. Applying Newton Second Law:

Force = mass × acceleration

or:
Force = Rate of change of momentum

A typical application is a propeller where thrust (T) is produced by accelerating the fluid from
velocity from V1 to V2 , and:

where m is the mass flow rate.

10
 General Description of Fluid and Viscosity
A fluid is a substance that deforms continuously when subjected to a shear stress, no matter
how small that shear stress may be. Meanwhile the average shear stress over a surface is
shear force divided by the area of the surface.
τ=F/A
A shear force F is the force component tangent to a surface (see Fig. 9). In this figure, a
substance is placed between two closely spaced parallel plates so large that conditions at
their edges may be neglected. The lower plate is fixed, and a force F is applied to the upper
plate, which exerts a shear stress F/A on any substance between the plates. Note that A is
the area of the upper plate in m2.

Fig. 9

When the force F causes the upper plate to move with a steady (nonzero) velocity, no
matter how small the magnitude of F, one may conclude that the substance between the
two plates is a fluid.
The fluid in immediate contact with a solid boundary has the same velocity as the
boundary; i.e., there is no slip at the boundary. The fluid in the area abcd flows to the new
position ab'c'd, each fluid particle moving parallel to the plate and the velocity u varying
uniformly from zero at the stationary plate to U at the upper plate.
Experiments show that, other quantities being held constant, F is directly proportional
to A and U and is inversely proportional to thickness t.

In which µ is the proportionality factor and includes the effect of the particular fluid. If τ =
F/A for the shear stress, then:

Keep in mind that t in the above equations does not represent time, yet it represents the
thickness or depth of the fluid.
The ratio U/t above may also be written as du/dy, since both U/t and du/dy express the
velocity change divided by the distance over which the change occurs. However, du/dy is
more general, as it holds for situations in which the shear stress in the fluid change with y, or
in other words, the curve of u(y) is not linear.

11
 The proportionality factor µ is called the VISCOSITY, COEFFICIENT OF VISCOSITY,
DYNAMIC VISCOSITY, or ABSOLUTE VISCOSITY of a fluid, and this equation is called the law
of viscosity. The viscosity of a fluid represents the internal resistance of a fluid to motion or
the "fluidity".
The unit of viscosity is kg/m.s, or equivalently, N.s/m2 (or Pa.s where Pa is the pressure
unit pascal). A common viscosity unit is POISE, which is equivalent to 0.1 Pa.s (or centipoise,
which is one-hundredth of a poise). One centipoise is the viscosity of water at 20°C, and thus
the unit centipoise serves as a useful reference.
From the equation of the shear force F required to move the upper plate in Fig. 9 at a
constant velocity of U,

We can calculate µ when the force F is measured. Therefore, the experimental setup just
described can be used to measure the viscosity of fluids. Note that under identical
conditions, the force F will be very different for different fluids, and so does with the
viscosity.
The force a fluid exerts on a body moving in it is called the DRAG FORCE, or RESISTANCE,
and the magnitude of this force depends, in part, on viscosity.
In fluid mechanics and heat transfer, the ratio of dynamic viscosity to density appears
frequently. For convenience, this ratio is given the name KINEMATIC VISCOSITY, v and is
expressed as:

Two common units of kinematic viscosity are m2/s and STOKE (1 stoke = 1 cm2/s = 0.0001
m2/s).
In general, the viscosity of a fluid depends on both temperature and pressure, although
the dependence on pressure is rather weak. For liquids, both the dynamic and kinematic
viscosities are practically independent of pressure, and any small variation with pressure is
usually disregarded, except at extremely high pressures. For gases, this is also the case for
dynamic viscosity (at low to moderate pressures), but not for kinematic viscosity since the
density of a gas is proportional to its pressure (see Fig. 10).

Fig. 10. Dynamic viscosity of

gases, in general, does not
depend on pressure, but
kinematic viscosity does.

12
 Pressure and Velocity Changes in a Moving Fluid

The changes are described by Bernoulli's equation as follows:

And so it can be written as:

Which is strictly valid when the flow is frictionless, termed inviscid, and steady. H
represents the total head, or total energy per unit weight (w=m.g) and, under these
conditions, is constant for any one fluid particle throughout its motion along any one
streamline. In this equation, P/ρg represents the pressure head, u2/2g represents the
velocity head (kinetic energy) and z represents the position or potential head (energy) due
to gravity.
An alternative presentation of Bernoulli's equation in terms of pressure is as follows:

where PT is total pressure.

As an example, consider the flow between two points on a streamline, Figure 11, then,

where P0 and u0 are in the undisturbed flow upstream and PL and uL are local to the
body.

Fig. 11

13
Similarly, from this Figure,

In the case of air, its density is small relative to other quantities. Hence, the ρgz term
becomes small and is often neglected.
Bernoulli's equation is strictly applicable to inviscid fluids. It can also be noted that
whilst, in reality, frictionless or inviscid fluids do not exist, it is a useful assumption that is
often made in the description of fluid flows, in particular, in the field of computational fluid
dynamics (CFD).
If, however, Bernoulli's equation is applied to real fluids (with viscosity) it does not
necessarily lead to significant errors, since the influence of viscosity in steady flow is usually
confined to the immediate vicinity of solid boundaries and wakes behind solid bodies (see
Fig. 12).

Fig. 12

The remainder of the flow, well clear of a solid body and termed the outer flow, behaves
effectively as if it were inviscid, even though it is not so. The outer flow is discussed in more
detail in the next section, Boundary Layer.

 Boundary Layer
 Origins
When a slightly viscous fluid flows past a body, shear stresses are large only within a thin
layer close to the body, called the boundary layer, and in the viscous wake formed by fluid
within the boundary layer being swept downstream of the body, Figure 12. The boundary
layer increases in thickness along the body length.

 Outer Flow
Outside the boundary layer, in the so-called outer flow in Figure 12, shear stresses are
negligibly small and the fluid behaves as if it were totally inviscid, that is, nonviscous or
frictionless. In an inviscid fluid, the fluid elements are moving under the influence of
pressure alone.
Consideration of a spherical element of fluid shows that such pressures act through the
centre of the sphere to produce a net force causing a translation motion. Thus, the outer
flow has no rotation and is termed irrotational.

 Flow Within the Boundary Layer

Flows within a boundary layer are unstable and a flow that is smooth and steady at the
forward end of the boundary layer will break up into a highly unsteady flow which can
extend over most of the boundary layer.

14
 The region at the forward end of the boundary layer is called laminar flow region (see
Fig. 13). In this region, the flow within the boundary layer is smooth, orderly and steady, or
varies only slowly with time.

Fig. 13

Behind the laminar flow region, the smooth flow breaks down. This is called transition
region. And then behind the transition region, which is called turbulent flow region, the flow
becomes erratic with a random motion and the boundary layer thickens. It should be noted
that flow outside the turbulent boundary layer can still be smooth and steady and turbulent
flow is not due to poor body streamlining as it can happen on a flat plate.
Figure 14 shows typical velocity distributions for laminar and turbulent boundary layers.
At the surface of the solid body, the fluid is at rest relative to the body. At the outer edge of
the boundary layer, distance δ, the fluid effectively has the full free-stream velocity relative
to the body.

Fig. 14

The onset of the transition from laminar to turbulent flow will depend on the fluid
velocity (v), the distance (l) it has travelled along the body and the fluid kinematic viscosity
(ν). This is characterised by the Reynolds number (Re) of the flow, defined as:

It is found that when Re exceeds about 0.5 x 106 then, even for a smooth body, the flow
will become turbulent. At the same time, the surface finish of the body, for example, its level
of roughness, will influence transition from laminar to turbulent flow.
Transition will also depend on the amount of turbulence already in the fluid through
which the body travels. Due to the actions of ocean waves, currents, shallow water and
other local disruptions, ships will be operating mainly in water with relatively high levels of
turbulence. Consequently, their boundary layer will normally be turbulent.

 Flow Separation
For flow along a flat surface, with constant pressure in the direction of flow, the boundary
layer grows in thickness with distance, but the flow will not separate from the surface. If the
pressure is falling in the direction of the flow, termed a favourable pressure gradient, then

15
the flow is not likely to separate. If, however, the pressure is increasing along the direction
of flow, known as an adverse pressure gradient, then there is a relative loss of speed within
the boundary layer (see the bernoulli equation).
This process can reduce the velocity in the inner layers of the boundary layer to zero at
some point along the body length, such as point S, Figure 15. At such a point, the
characteristic mean flow within the boundary layer changes dramatically and the boundary
layer starts to become much thicker. This process can reduce the velocity in the inner layers
of the boundary layer to zero at some point along the body length, such as point S, Figure 15.
At such a point, the characteristic mean flow within the boundary layer changes dramatically
and the boundary layer starts to become much thicker.

Fig. 15

This also explains why golf balls with dimples that promote turbulent flow have less
drag and travel further than the original smooth golf balls, Figure 16. It is also worth noting
that a thick wake following separation should not be confused with the thickening of the
boundary layer following transition from laminar to turbulent flow, described earlier.

Fig. 16

 Wave Properties
Winds create natural waves on the oceans and ships create waves during their passage
through water. The water particles move in orbital paths which are approximately circular,
Figure 17. These orbital paths decrease exponentially with increasing depth.
A small floating object will simply rise and fall with the passage of a wave beneath it.
The orbital motion of wave is generally not of concern for large displacement ships. There
may be some influence on the wake of twin-screw ships, depending on whether the
propellers are near a crest or trough when the wake will be increased or decreased by the
orbital motion. Smaller craft may experience problems with control with the effects of the
orbital motion of a following sea.
The wave contour is given by a trochoid function, which is a path traced out by a point
on the radius of a rolling circle, Figure 17. Other applications tend to use a sine wave which
can include the orbital motion of the water particles and is easier to manipulate

16
mathematically, Figure 18. The actual difference between a trochoid and a sine wave is
small.

Fig. 17

Fig. 18

17
Wave theory yields the wave velocity c as follows:

where h is the water depth from the still water level and λ is the wavelength, crest to crest.

Main References:
 Molland, A. F., Turnock, S. R., Hudson, D. A. (2011), Ship Resistance and Propulsion,
Cambridge University Press.
 Streeter, V. L., Wylie, E. B. (1981), Fluid Mechanics, 7th edition, McGraw-Hill Kogakusha,
Ltd.
 Yunus A. Cengel, John M. Cimbala (2006), Fluid Mechanics Fundamentals and
Applications, The McGraw-Hill Companies, Inc.
 Giancoli, D. C. (2014), Physics Principles with Applications, 7th edition, Pearson
Education, Inc.

 Ship Total Resistance and its Components

 Classification of Ship Total Resistance Components
The first step for obtaining a ship's propulsive power is the determination of its total calm
water resistance RT (see Fig. 5). This resistance of a ship at a given speed is the force
required to tow the ship at that speed in smooth water, assuming no interference from the
towing ship.
Observation of a ship moving through water indicates two features of the flow, namely
that there is a wave pattern moving with the hull (see Fig. 19 & 20) and there is a region of
turbulent flow building up along the length of the hull and extending as a wake behind the
hull (see Fig. 12 & Fig. 21).

Fig. 19

18
 Fig. 20

Fig. 21. Waves and wake

Both of these features of the flow absorb energy from the hull and, hence, constitute a
resistance force on the hull. This resistance force is transmitted to the hull as a distribution
of pressure and shear forces over the hull (see Fig. 22). The shear stress arises because of
the viscous property of the water (see Fig. 23).

Fig. 22. Frictional and

pressure forces.

19
 Fig. 23. The shear force (F) is acting on a liquid. It can be formulated as F = µ.A.U / t

This leads to the first possible physical breakdown of resistance which considers the forces
acting:
a) Pressure resistance
The fore and aft components of the pressure force P acting on each element of hull surface,
Figure 22 & 24, can be summed over the hull to produce a total pressure resistance.

b) Frictional resistance
The fore and aft components of the tangential shear forces τ acting on each element of the
hull surface, Figure 22, can be summed over the hull to produce the total shear resistance or
frictional resistance.

Fig. 24. Pressure forces and their components.

The frictional resistance arises purely because of the viscosity, but the pressure resistance is
due in part to viscous effects and to hull wavemaking.

An alternative physical breakdown of resistance considers energy dissipation.

a) Total viscous resistance
Bernoulli’s theorem states that (P/ρg) + (V2/2g) + h = H and, in the absence of viscous
forces, H is constant throughout the flow. Surely this is not the case for fluid flow around a
ship, which has varying values of H throughout its length. By means of a Pitot tube, or, more
precisely, Pitot-static probe, local total head can be measured (See Fig. 25).
Since losses in total head are due to viscous forces, it is possible to measure the total
viscous resistance by measuring the total head loss in the wake behind the hull, Figure 26.
This resistance will include the skin frictional resistance and part of the pressure resistance
force, since the total head losses in the flow along the hull due to viscous forces result in a
pressure loss over the afterbody which gives rise to a resistance due to pressure forces.

20
 Fig. 25. The Pitot-static probe

Flow Velocity (V):

Fig. 26. Measurement of total viscous resistance.

This can be explained using the figure 24, which illustrates that if the fluid has been
assumed to be without viscosity, the pressure forces around a moving ship will everywhere
be normal to the hull. Over the forward part of the hull, these will have components acting
towards the stern and therefore resisting the motion. Over the after part, the reverse is the
case, and these components are assisting the motion. It is shown in the figure that the
resultant total forces on the fore and after bodies are equal, and the body therefore
experiences no resistance.
In a real fluid the boundary layer alters the virtual shape and length of the fluid flow at
the stern, which result that the pressure distribution there is changed and its forward
component is reduced (see Fig. 27). There is therefore a net force on the body acting against
the motion, giving rise to a resistance which is variously referred to as pressure resistance
force or viscous pressure force or form drag or viscous pressure drag.
As the result, the pressure resistance force is actually a difference between the total
backward component of the pressure forces and the total forward component of them. This
difference is caused by the viscous forces between the ship's skin and the water it moves
within.

21
 Fig. 27. Flow of real fluids around moving bodies.

b) Total wave resistance

The wave pattern created by the hull, Fig. 19 & 20, can be measured and analysed into its
component waves. The energy required to sustain each wave component can be estimated
and, hence, the total wave resistance component obtained.

A summary of these basic hydrodynamic components of ship resistance is shown in

Figure 28. When considering the forces acting, the total resistance is made up of the sum of
the tangential shear and normal pressure forces acting on the wetted surface of the vessel,
as shown in Figure 22 and at the top of Figure 28.
When considering energy dissipation, the total resistance is made up of the sum of the
energy dissipated in the wake and the energy used in the creation of waves, as shown in
Figure 21 and at the bottom of Figure 28.

Fig. 28

22
 Figure 29 shows a more detailed breakdown of the basic resistance components
together with other contributing components, including wave breaking, spray, transom and
induced resistance. The total skin friction in the figure has been divided into
two-dimensional flat plate friction and three-dimensional effects.
Wave breaking and spray can be important in high-speed craft and, in the case of the
catamaran, significant wave breaking may occur between the hulls at particular speeds in
the aft part of the catamaran. Meanwhile in the case of warships, an example of spray
developed at their bows is shown in Fig. 30. Wave breaking and spray should form part of
the total wavemaking resistance, but, in practice, this energy will normally be lost in the
wake; the dotted line in Figure 29 illustrates this effect.

Fig. 29

Fig. 30

23
 The transom stern, used on most high-speed vessels, is included as a pressure drag
component. It is likely that the large low-pressure area directly behind the transom, which
causes the transom to be at atmospheric pressure rather than stagnation pressure (see also
Fig. 25), causes waves and wave breaking and spray which are not fully transmitted to the
far field. Again, this energy is likely to be lost in the wake, as illustrated by the dotted line in
Figure 29.
Induced drag will be generated in the case of yachts, resulting from the lift produced by
keels and rudders, Fig. 31 - 34. Catamarans, Fig. 35, can also create induced drag because of
the asymmetric nature of the flow between and over their hulls and the resulting production
of lift or sideforce on the individual hulls. An investigation indicates that the influence of
induced drag for catamarans is likely to be very small.

Fig. 31 Fig. 32

Fig. 33 Fig. 34

Fig. 35

24
 Definition of Ship Source Resistance R and Resistance Coefficient C
To move a ship, it is first necessary to overcome resistance, i.e. the force working against its
propulsion. The calculation of this resistance R plays a significant role in the selection of the
correct propeller and in the subsequent choice of main engine.
The total resistance RT, consists of many source resistances R, which can be classified as
water and air resistances. Mean-while the water resistance itself is classified as:
a) The skin frictional resistance, or simply frictional resistance.
b) The residuary resistance.
Water with a speed of V and a density of ρ has a dynamic pressure of:

1/2 x ρ x V 2
This formula is deduced from the Bernoulli equation, Fig. 36.

Fig. 36

Thus, if water is being completely stopped by a body, the water will react on the surface of
the body with the dynamic pressure, resulting in a dynamic force on the body. This
relationship is used as a basis when calculating or measuring the source-resistances R of a
ship’s hull, by means of dimensionless resistance coefficients C.
Thus, C is related to the reference force F, defined as the force which the dynamic
pressure of water with the ship’s speed V exerts on a surface which is equal to the hull’s
wetted area AS. The rudder’s surface is also included in the wetted area. The general data for
resistance calculations is thus:
Reference Force (F): F = 1/2 x ρ x V 2 x AS
and Source Resistances (R): R = C x F
(Ref: Basic Principles of Ship Propulsion)
From the equations above, the resistance coefficients C can be formulated as follows
(Harvald, 1983):

Note that S in the last equation is similar with AS in the preceding equation, which
represents the wetted surface area of a ship.

 Another Graphical Representation of Ship Total Resistance Components

As we know the definition of ship resistance coefficients C, we can represent the ship total
resistance components in graphical form by using another term called Froude number, Fn.
The Froude number is formulated as follows:

25
Where V and L are ship's speed and length, respectively, and g is the acceleration of gravity,
and so Fn is a dimensionless quantity. When we make a graph with C as the abscissa and Fn
as the ordinate, we can represent the ship resistance coefficient curves like to be shown in
Fig. 37.

Fig. 37. Resistance Coefficient Curves

Graph of similar coordinate axes can also be used to represent the ship total resistance
components, like to be shown in Fig. 38. We can compare this figure with Fig. 39, which is
similar with Fig. 29, to enhance our understanding of the ship total resistance components
and their respective values at various speeds.
From Fig. 38 above, we can see the various ship source resistances or ship total
resistance components, where the description of each are as follows (Harvald, 1983):
a) Frictional Resistance, RF
The frictional resistance is the component of resistance obtained by integrating the
tangential stresses τ over the wetted surface of the ship in the direction of motion.
b) Residuary Resistance, RR
The residuary resistance is a quantity obtained by subtracting from the total
resistance of a hull, a calculated friction resistance obtained by any specific
formulation.

26
 Residuary Resistance = “Total Resistance” - Frictional Resistance (Flat Plate) (see Fig.
38)
Or
Residuary Resistance = Pressure Resistance + Additional Frictional Resistance Due to
Curvature of Body (See Fig. 38 & 39)
Residuary Resistance = (Wave Resistance + Viscous Pressure Resistance) + Additional
Frictional Resistance Due to Curvature of Body (See Fig. 38)

Residuary Resistance = (Wave Pattern Resistance + Wavebreaking Resistance) +

Viscous Pressure Resistance + Additional Frictional Resistance
Due to Curvature of Body (See Fig. 38)

In general, the greater part of the residuary resistance of merchant ships will be
wavemaking resistance.

Fig. 38. Components of Specific Resistance of Ships (Harvald, 1983)

c) Viscous Resistance, RV
The viscous resistance is the component of resistance associated with the energy
expended due viscous effects.
Viscous Resistance = Frictional Resistance + Viscous Pressure Resistance (see Fig. 38
& 39)
d) Pressure Resistance, RP
The pressure resistance is the component of resistance obtained by integrating the
normal stresses over the surface of a body in the direction of motion (see Fig. 22).

27
 Pressure Resistance = Wave Resistance + Viscous Pressure Resistance (see Fig. 38)
Or
Pressure Resistance = “Total Resistance” - Frictional Resistance (see Fig. 38 & 39)

Fig. 39

e) Viscous Pressure Resistance, RPV

The viscous pressure resistance is the component of resistance obtained by
integrating the components of the normal stresses due to viscosity and turbulence.
This quantiy can not be directly measured except for a fully submerged body, where it
is equal to the pressure resistance (see Fig. 28).

Viscous Pressure Resistance = Total Viscous Resistance - Frictional Resistance

(see Fig. 38 & 39)

f) Wavemaking Resistance, RW
The wavemaking resistance is the component of resistance associated with the
energy expended generating gravity waves.

28
 g) Wave pattern Resistance, RWP
The resistance component deduced from measurements of wave elevations remote
from the ship or model, where it is assumed that the subsurface velocity field and,
hence, the momentum of the fluid can be related to the wave pattern by means of a
so called linearized theory. The resistance so deduced does not include wavebreaking
resistance.

h) Wavebreaking Resistance, RWB

The wavebreaking resistance is a resistance component associated with the
breakdown of the ship bow wave.

i) Spray Resistance, RS
The spray resistance is the component of resistance associated with the energy
expended generating spray.

To these resistance components some additional resistances, RA , should be added.

j) Appendage Resistance
This is the resistance of shafts, shaft bossing, shaft brackets, rudders, bilge keels, etc.
(see Fig. 40 & 41) When using physical models, the appendages are often fitted to the
models and the appendage resistance are then included in the measured resistance.
Normally bilge keels are not fitted. If the hull has no appendages fitted, the resistance
is called the bare hull resistance.

k) Roughness Resistance
This is the resistance due to the roughness, for instance, owing to corrosion and
fouling on the ship hull.

l) Air Resistance
This is experienced by the abovewater part of the main hull and the superstructures
owing to the motion of the ship through the air.

m) Steering Resistance
To maintain a straightline path it is in general necessary to use a rudder for
corrections. The use of the rudder when correcting the ship course, results in an extra
resistance component called the steering resistance.

 More About Frictional Resistance RF

The frictional resistance RF of the hull depends on the size of the hull’s wetted area AS, and
on the specific frictional resistance coefficient CF. The friction increases with fouling of the
hull, i.e. by the growth of algae, barnacles, etc. An attempt to avoid fouling is made by the
use of anti fouling hull paints to reduce the possibility of the hull becoming fouled by living
organisms.
The paints containing TBT (tributyl tin) as their principal biocide, which is very toxic,
have dominated the market for decades, but the IMO ban of TBT for new applications from 1
January, 2003, and a full ban from 1 January, 2008, may involve the use of new (and maybe
not as effective) alternatives, probably copper based anti fouling paints.

29
 When the ship is propelled through the water, the frictional resistance increases at a
rate that is virtually equal to the square of the vessel’s speed. Frictional resistance
represents a considerable part of the ship’s resistance, often some 70-90% of the ship’s
total resistance for low speed ships (bulk carriers and tankers), and sometimes less than 40%
for high speed ships (cruise liners and passenger ships). The frictional resistance is found as
follows:
RF = CF x F
(Ref: Basic Principles of Ship Propulsion)

Fig. 41. General layout of propulsion system

for the ship shown in Fig. 40

30
 More About Residual Resistance RR
Residual resistance RR comprises wave resistance and eddy resistance. Wave resistance
refers to the energy loss caused by waves created by the vessel during its propulsion
through the water, while eddy resistance refers to the loss caused by flow separation which
creates eddies, particularly at the aft end of the ship.
Wave resistance at low speeds is proportional to the square of the speed, but increases
much faster at higher speeds, Fig. 38. In principle, this means that a speed barrier is imposed,
so that a further increase of the ship’s propulsion power will not result in a higher speed as
all the power will be converted into wave energy.
The residual resistance normally represents 8-25% of the total resistance for low speed
ships, and up to 40-60% for high speed ships (see Fig. 38). Incidentally, shallow waters can
also have great influence on the residual resistance, as the displaced water under the ship
will have greater difficulty in moving aftwards. The value of the residual resistance RR can be
found by using the following formula:
RR = CR x F
(Ref: Basic Principles of Ship Propulsion)

 More About Air Resistance RA

In calm weather, air resistance is, in principle, proportional to the square of the ship’s
speed, and proportional to the cross sectional area of the ship above the waterline. Air
resistance normally represents about 2% of the total resistance. For container ships in head
wind, the air resistance can be as much as 10%.
The air resistance can, similar to the foregoing resistances, be expressed as RA = CA x F ,
but is sometimes based on 90% of the dynamic pressure of air with a speed of V, i.e.:

(Ref.: Basic Principles of Ship Propulsion)

The air speed V is similar in magnitude with the ship speed, where to convert the ship speed
in knots to m/s, we can multiply the value with 0,5144. Meanwhile ρair is the density of air in
kg/m3, and Aair is the cross-sectional area of the vessel above the water in m2. The density
values of air at 1 atm pressure is shown in the table below.

Temp.
0 5 10 15 20 25 30 35 40
(°C)
Density
1.292 1.269 1.246 1.225 1.204 1.184 1.164 1.145 1.127
(kg/m3)

(Ref.: Fluid Mechanics Fundamentals & Applications,

Yunus A. Cengel & John M. Cimbala )

 Towing Resistance RT

The ship’s total towing resistance RT is thus found as :

RT = RW + RA = (RF + RR) + RA = RF + (RW + RE) + RA

RT = RW + RA = (RF + RR) + RA = RF + RW + RE + RA

31
 RT = RW + RA = (RF + RR) + RA = (CF x F ) + (CW x F ) + (CE x F ) + (CA x F )
RT = RW + RA = (RF + RR) + RA = (CF + CW + CE + CA) F = CT x F

On the basis of many experimental tank tests, and with the help of pertaining dimensionless
hull parameters (CB, CP, etc.), methods have been established for calculating all the
necessary resistance coefficients C and, thus, the pertaining source resistances R. In practice,
the calculation of a particular ship’s resistance can be verified by testing a model of the
relevant ship in a towing tank.
The percentage of the four main components of ship total towing resistance could also,
as a guideline, be stated as shown in the table below. The right column of the table is valid
for low speed ships like bulk carriers and tankers, and the left column is valid for very high
speed ships like cruise liners and ferries. Container ships may be placed in between the two
columns.

The main reason for the difference between the two columns is, as earlier mentioned,
the wave resistance. Thus, in general all the resistances are proportional to the square of the
speed, but for higher speeds the wave resistance increases much faster, involving a higher
part of the total resistance.
This tendency is also shown in Fig. 42 for a 600 TEU container ship, originally designed
for the ship speed of 15 knots. Without any change to the hull design, the ship speed for a
sister ship was requested to be increased to about 17.6 knots. However, this would lead to a
relatively high wave resistance, requiring a doubling of the necessary propulsion power.
A further increase of the propulsion power may only result in a minor ship speed
increase, as most of the extra power will be converted into wave energy, i.e. a ship speed
barrier valid for the given hull design is imposed by what we could call a “wave wall”, Fig.
42. A modification of the hull lines, suiting the higher ship speed, is necessary.

 Increase of ship resistance in service

During the operation of the ship, the paint film on the hull will break down. Erosion will start,
and marine plants and barnacles, etc. will grow on the surface of the hull. Bad weather,
perhaps in connection with an inappropriate distribution of the cargo, can be a reason for
buckled bottom plates.
The hull has been fouled and will no longer have a “technically smooth” surface,
which means that the frictional resistance will be greater. It must also be considered that the
propeller surface can become rough and fouled. Therefore, the sources of fouling can be
depicted like to be shown in Fig. 43.

32
 Fig. 42. The “wave wall” ship speed barrier

The total resistance, caused by fouling, may increase by 25 - 50% throughout the
lifetime of a ship. Experience shows that hull fouling with barnacles and tube worms
theirself may cause an increase in drag (ship resistance) of up to 40%, with a drastical
reduction of the ship speed as the consequence.
Resistance will also increase because of wind and current, and heavy waves of the sea,
as shown in the table below for different main routes of ships. The resistance when
navigating in head on sea could, in general, increase by as much as 50-100% of the total ship
resistance in calm weather. However, analysis of trading conditions for a typical 140,000 dwt
bulk carrier shows that on some routes, especially Japan-Canada when loaded, the increased
resistance (sea margin) can reach extreme values up to 220%, with an average of about

33
100%. On the North Atlantic routes, the first percentage corresponds to summer navigation
and the second percentage to winter navigation.

In principle, the increased resistance caused by heavy weather could be related to:
a) wind and current against, and
b) heavy waves.
Among these factors, the latter may have great influence. Thus, if the wave size is relatively
high, the ship speed will be somewhat reduced even when sailing in fair seas. In practice,
however, it will be difficult to distinguish between these two factors.
Unfortunately, no data have been published on increased resistance as a function of
type and size of vessel. The larger the ship, the less the relative increase of resistance due to
the FORCES OF SEA, which is caused by wind and current against, and heavy waves. On the
other hand, the frictional resistance of the large, full bodied ships will very easily be changed
in the course of time because of FOULING.
As a rule of thumb, the average increase in resistance due to the forces of sea is shown
in the last table, with in certain condition it can reach 100%. Whereas the maximum value,
i.e. when navigating in head on sea could, in general, achieve as much as 50-100%, with in
certain condition it can reach 220%. Meanwhile, the total increase in resistance caused by
fouling, may reach 25 - 50% throughout the lifetime of a ship.

References:

Molland, A. F., Turnock, S. R., Hudson, D. A. (2011), Ship Resistance and Propulsion,
Cambridge University Press.
Basic Principles of Ship Propulsion
Cengel, Y. A, Cimbala, J M. (2006), Fluid Mechanics Fundamentals & Applications, The
McGraw-Hill Companies, Inc.
Harvald, Sv. Aa. (1983), Resistance and Propulsion of Ships, John Wiley & Sons, Inc.

34
 Some of the Ship Propulsive Power Determination Methods

 The Classical Treatment of Resistance

In the classical treatment of resistance, this is divided into two components, which are
governed by different laws:
a) The skin frictional resistance, which is governed by the Reynold’s number.
b) The residuary resistance, taken mainly to be wavemaking, which is governed by the
Froude number.
In the following paragraphs use will be made of the resistance coefficient C. This is related to
the wetted surface S, the speed V, and the mass density ρ by the following equation:

R = 1/2 . C . ρ . S . V 2 (Eq. 1)

The coefficient C is given two types of subscripts. The first of these refers to the subdivision
of resistance with t = total; f = frictional; r = residuary; w = wavemaking. Whereas the second
subscript distinguishes between model resistance = m; and ship resistance = s.
In the classical treatment, the skin frictional resistance coefficient of the model is
calculated based on the coefficient of friction applicable to a plank (flat plane) of model
length and having the same wetted area as the model. This is then deducted from the total
model resistance coefficient to establish the model residuary resistance coefficient (see also
Fig. 38):
Cr = Ctm - Cfm Eq. (2)

At a constant Froude number the residuary resistance coefficient remains the same for the
ship as it is for the model, so there is no need for a suffix to indicate “model” or “ship” in
this case.
The ship frictional resistance coefficient Cfs , is again calculated using the coefficient of
friction applicable to a plank, this time one of the same wetted area and the same length as
the ship. This is then added to Cr , to arrive at the total resistance coefficient of the ship, Cts .

Cts = Cfs + Cr Eq. (3)

It should be noted that the use of friction coefficients based on a plank for both model and
ship implies that the skin friction is independent of the shape of the lines.
A change in the treatment of resistance was made in 1957 when the basis for calculating
the friction coefficients was altered from the Froude line which had been used for many
years to the 1957 ITTC (International Towing Tank Conference) line. The 1957 ITTC line is
expressed by the formula:

Cf = 0.075/(log Rn - 2)2 Eq. (4)

where Rn is the Reynold number, which can be formulated as follows:

Rn = V.L/v Eq. (5)

where V is the ship speed in m/s, L is the ship length in m, and v is the kinematic viscosity in
m2/s.

35
 The Present Day (ITTC'78) Treatment of Resistance
The present day treatment recognises that the frictional resistances of both model and ship
differ from those of flat plates of the same length and area (see Fig. 38). The viscous
resistance coefficient (Cv), as the frictional resistance of a shaped body is now called, is
increased over the frictional resistance coefficient of the corresponding flat plate by a form
factor k so that:
Cvm = Cfm ( 1 + k ) Eq. (6)

where Cfm is formulated as follows:

Cfm = 0.075/(log Rn - 2)2 Eq. (7)

which is similar with the formula of Cf in Eq. (4).

The form factor in Eq. (6) can either be deduced from model experiments at very slow
speeds when Cr is reduced to nearly zero, or in some tanks by the direct measurement of Cr
from the energy delivered to the wave systems. Cr for the model is now calculated from the
formula:

Cr = Ctm - Cvm
Cr = Ctm - Cfm ( 1 + k ) Eq. (8)

As k tends to have a value of between 0.25 and 0.35 for most ship forms, a value of Cr
calculated from Eq. (8) is much smaller than one calculated from Eq. (2).
The total resistance coefficient of a ship is now considered as made up of

Cts = Cvs + Cr + Capp + ΔC + Cair

Cts = Cfs (1 + k) + Cr + Capp + ΔC + Cair Eq. (9)

where:
Capp is the resistance of appendages.
ΔC is the roughness allowance which is discussed later.
Cair is the air resistance coefficient, for which there is the following approximate formula:

Cair = 0.001 . At / S

where At is the projected cross sectional area of the ship above the waterline.

It should be noted that in association with Cts derived in this way the wetted surface S in Eq.
(1) is the total wetted surface inclusive of the surface area of the bilge keels, if fitted.

S (wetted) = S (bare hull) + S (bilge keels)

36
 Hull Finish and the Importance of Skin Friction Resistance
An understanding of the importance of hull finish requires a knowledge of the proportion of
the total ship resistance which is frictional. In ITTC’78 practice the proportion of the viscous
component is:

C vs C fs (1  k )  C

Cts C fs (1  k )  C  C r  C air
This may be compared with the ITTC’57 practice in which the frictional component
proportion was:
C fs C fs

Cts C water  C air
C fs C fs

Cts C fs  C r  C air

It is important to note that the Cr values in these two equations differ from one another,
indeed the biggest part of the difference between the two formulae occurs when the viscous
or frictional components respectively of the model resistance are subtracted from the model
total resistance to establish the respective residuary resistance coefficients, which results
the residuary resistance of Cr = Ctm - Cfm in the ITTC’57 method and Cr = Ctm - Cvm = Ctm -
Cfm (1+k) in the ITTC’78 method. The further change caused by the multiplication of the ship
friction coefficient by the form factor has lesser significance.
An example with some figures may help to make the difference clear. A ship of:

was tested using a model with the following particulars:

37
Apart from the change in the proportion of frictional/viscous resistance, i.e. from 56% to
95%, the very large reduction in Cts, i.e. from 2.646 x 10-3 to 2.049 x 10-3 or 29% should be
noted. This change in value is of course tied to the k value of 0.33 used in this example, a
figure which appears to agree with the Holtrop and Mennen’s formula. As the consequence,
the value of Cr will decrease considerably, which result in the significant reduction in Cts .

Although most tank authorities appear to have adopted the new method, others are
sticking to the use of ITTC’57. Designers can only hope that there will shortly be an end to
the succession of changes and variety of methods used by tanks which have caused them so
much difficulty in the last two decades.
This hope may, however, be a little premature as a 1993 R.I.N.A. paper by C.W.B.
Grigson “An accurate smooth friction line for use in performance prediction” questions the
accuracy of the 1957 ITTC line (see Fig. 44 for the drawing of ITTC line). He is almost certainly
right in doing this if this line is to be used as a base for a form factor as the ITTC line was
never claimed to be a friction line having originally been introduced as a ship-model
correlation line.
Manen & Oossanen (1988) mentioned that the 1957 conference adopted this as the
"ITTC 1957 model-ship correlation line," and was careful to label it as "only an interim
solution to this problem for practical engineering purposes," (ITTC 1957). Equation (4) was
called a model-ship correlation line, and not a frictional resistance line; it was not meant to
represent the frictional resistance of plane or curved surfaces, nor was it intended to be
used for such a purpose.
The Grigson’s paper shows that at Reynolds’ numbers in the model area (4 x 106 to 2
x107), Cf ‘57 values are up to 6% higher than what he suggests are the “correct” values,
whilst in the ship area which is of the order of (4 x 108 to 2 x 109), Cf ‘57 values are about 5%
below the “correct” values. This would mean that both Cr and Cfs are being under-estimated,
and as the paper also revises the (1 + k) values upward the overall effect is to increase Pd by
about 7% and propeller RPM by about 1.5%.

38
 model area ship area

Fig. 44. Skin friction lines (Manen & Oossanen (1988))

References:

Watson, D. G. M. (1998), Practical Ship Design, Elsevier Science, Ltd.

Manen, J. D. V., Oossanen, P. V. (1988), “Resistance”, in Principles of Naval Architecture,
Volume 2, ed. Lewis, E. V., The Society of Naval Architecs and Marine Engineers.

39