Marine propellers and propulsion 4th

Edition J. S. Carlton
Visit to download the full and correct content document:
https://ebookmass.com/product/marine-propellers-and-propulsion-4th-edition-j-s-carlt
on/
Marine Propellers and Propulsion
Marine Propellers
and Propulsion
Fourth Edition

J S Carlton FREng
Professor of Marine Engineering, City University
London
President of the Institute of Marine Engineering,
Science and Technology 2010/11
Butterworth-Heinemann is an imprint of Elsevier
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States

© 2019 Elsevier Ltd. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including
photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on
how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as
the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions.

This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be
noted herein).

Notices
Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes
in research methods, professional practices, or medical treatment may become necessary.

Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information,
methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own
safety and the safety of others, including parties for whom they have a professional responsibility.

To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury
and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation
of any methods, products, instructions, or ideas contained in the material herein.

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library

ISBN: 978-0-08-100366-4

For information on all Butterworth-Heinemann publications

visit our website at https://www.elsevier.com/books-and-journals

Publisher: Matthew Deans

Acquisition Editor: Carrie Bolger
Editorial Project Manager: Carrie Bolger
Production Project Manager: Anitha Sivaraj
Cover Designer: Christian J. Bilbow
Typeset by SPi Global, India
Dedication

To Jane and Caroline

Preface to the Fourth Edition
As with the previous editions, in the intervening time since
their publication several important advances in propeller
and propulsion technology have taken place. No less so with
this edition. Consequently, the text has generally been
updated to reflect the changes that have taken place during
the 6 years since the Third Edition of the book.
In addition to these continuing developments the chapter
on ship resistance and propulsion has been significantly
extended to embrace high speed craft: particularly in
relation to hydrodynamically supported vessels, aerostati-
cally supported vehicle propulsion, and aerodynamically
driven propulsion. Similarly, the discussion on high-speed
propellers has been considerably extended to reflect the
additions in Chapter 12. A new chapter on propeller-ice
interaction and operation in arctic climates has been intro-
duced. This new chapter includes information of ice types
and their mechanical properties; ship resistance and pro-
pulsion in ice; a discussion on models of propeller action;
model testing; ice interaction induced cavitation as well
as considering the results of ships operating in ice. With
regard to the layout and structure of the book some changes
have taken place and a consistent order has been introduced
into the referencing system.
As with previous editions thanks are once again due to
many colleagues around the world who have made very
valuable suggestions and comments as well as providing
me with further material for inclusion from their own
libraries and archives. I would like to particularly
acknowledge Mr. J.Th. Ligtelijn, for his continued support
and also Professor Tong Sun and Dr. Len Rogers in relation
to their many discussions about instrumentation methods. I
would also like to acknowledge the contributions of four of
my recent PhD students who have increased our under-
standing of various aspects of propulsion hydrodynamics:
Dr. Elena Hauerhof, Dr. Saeed Javdani, Dr. Ioannis Arma-
kolas, and Ms. Elham Derafshi. Finally, thanks are again
due to Jane, my wife, for her encouragement, support and
proof-reading activities in undertaking this new edition of
the book.
J.S. Carlton
Battle, United Kingdom
xiii
Preface to the Third Edition
Since the second edition was published in 2007, a number of
important changes have and are taking place, which relate to
ship propulsion. Among these are the developments arising
from the IMO’s initiatives on the Energy Efficiency Design
Index and the bringing to fruition of their earlier work on
hull coatings stemming from their resolution on the subject
in 2001. In addition to these legislative initiatives is a
greater understanding of the physics associated with cavi-
tation development and collapse and its subsequent erosive
effects on propeller materials. Moreover, some new types of
propulsor are being developed while in other technical
fields there is a greater awareness, for example, of the
effects of shipping activity on the behavior of marine
mammals. It is the intention of this third edition to capture
these and other developments that have occurred. In
addition, a certain amount of rearrangement of the subject
matter has also taken place.
As with previous editions, thanks are once again due to
many colleagues around the world who have made very
valuable suggestions and comments as well as providing
me with further material for inclusion from their own
libraries and archives. As in previous years, I would like
to particularly acknowledge Mr. P.A. Fitzsimmons, Mr. J.
Th. Ligtelijn, Dr. D. Radosavljevic, and Prof. Dr. T. van
Terwisga who have continued in their support as well as par-
ticular contributions for this addition from Mr. T. Veit-
komeno, Mr. J. Gonzalez-Adalid, Mr. P. van Terwisga,
and Dr. S. Whitworth. Finally, thanks are again due to Jane,
my wife, for her encouragement, support, and proof-reading
activities in undertaking this new edition of the book.
J.S. Carlton
Battle, United Kingdom
xv
Preface to the Second Edition
It is now rather over a decade since much of the material was
written for the first edition of this book. During that time,
advances have been made in the understanding of several
branches of the subject and it is now time to incorporate
much of that material into the text. These advances in under-
standing, together with the natural progression of the
subject, relate particularly to cavitation dynamics, theo-
retical methods including the growing development of com-
putational fluid dynamics in many parts of the subject, and
the use of carbon fiber materials for certain propeller types.
Moreover, podded propulsors have emerged in the inter-
vening years since the first edition was written and have
become a propulsion option for certain types of ship, partic-
ularly cruise ships and ice breakers but with a potential to
embrace other ship types in the future.
Some other aspects of the subject were not included in
the original publication for a number of reasons. In this
new edition, I have attempted to rectify some of these omis-
sions by the inclusion of material on high-speed propellers,
propeller-rudder interaction as well as a new chapter dealing
with azimuthing and podded propulsors and a substantial
revision to the chapter on cavitation. These additions,
together with a reasonably extensive updating of the
material and the removal of the inevitable typographical
errors, in the first edition form the basis of this new addition.
Furthermore, experience in using the book over the last
10 years or so has shown that the arrangement of some of
the material could be improved. As a consequence, it will
be seen that a certain amount of regrouping of the subject
matter has taken place in the hope that this will make the
text easier to use.
Finally, thanks are once again due to many colleagues
around the world who have made very valuable suggestions
and comments as well as providing me with further material
for inclusion from their own libraries and archives. Fur-
thermore, the normal day-to-day discussions that are held
on various aspects of the subject frequently trigger thought
processes, which have found their way into various parts of
the narrative. In particular, my thanks are due to Mrs. W.
Ball, Mr. P.A. Fitzsimmons, Mr. M. Johansen, Mr. J.Th.
Ligtelijn, Dr. D. Radosavljevic, Prof. Dr. T. van Terwisga,
and Mr. J. Wiltshire. Thanks are also due to Dr. P. Helmore
who, having read the book some 10 years ago, kindly sup-
plied me with a list of errata for this edition. Finally, thanks
are also due to Jane, my wife, for her encouragement and
support in undertaking this revision to the book in a rela-
tively short time frame.
J.S. Carlton
Hythe, United Kingdom
xvii
Preface to the First Edition
Although the propeller normally lies well submerged out of
sight and therefore, to some extent, also out of mind, it is a
deceptively complex component in both the hydrodynamic
and the structural sense. The subject of propulsion tech-
nology embraces many disciplines: for example, those of
mathematics, physics, metallurgy, naval architecture, and
mechanical and marine engineering. Clearly, the depen-
dence of the subject on such a wide set of basic disciplines
introduces the possibility of conflicting requirements within
the design process, necessitating some degree of com-
promise between opposing constraints. It is the attainment
of this compromise that typifies good propeller design.
The foundations of the subject were laid during the latter
part of the last century and the early years of this century.
Since that time much has been written and published in
the form of technical papers, but the number of books that
attempt to draw together all of these works on the subject
from around the world is small. A brief study of the bibliog-
raphy shows that, with the exception of Gerr’s recent book
dealing with the practical aspects of the design of small craft
propellers, little has been published dealing with the subject
as an entity since the early 1960s. Over the last 30 or so
years an immense amount of work, both theoretical and
empirical, has been undertaken and published, probably
more than in any preceding period. The principal aim,
therefore, of this book is to collect together the work that
has been done in the field of propeller technology up to
the present time in each of the areas of hydrodynamics,
strength, manufacture, and design, so as to present an
overall view of the subject and the current levels of
knowledge.
The book is mainly directed toward practicing marine
engineers and naval architects, principally within the marine
industry but also in academic and research institutions. In
particular, when writing this book, I have kept in mind
the range of questions about propeller technology that are
frequently posed by designers, ship operators, and surveyors
and I have attempted to provide answers to these questions.
Furthermore, the book is based on the currently accepted
body of knowledge of use to practical design and analysis;
current research issues are addressed in a less extensive
manner. For example, recent developments in surface panel
techniques and Navier-Stokes solutions are dealt with in less
detail than the currently more widely used lifting line, lifting
surface, and vortex lattice techniques of propeller analysis.
As a consequence knowledge of mathematics, fluid
mechanics, and engineering science is assumed commen-
surate with these premises. Notwithstanding this, it is to
be hoped that students at both undergraduate and post-
graduate levels will find the book of value to their studies.
The first two chapters of the book are essentially an
introduction to the subject: first, a brief history of the early
development of propellers and, second, an introduction to
the different propeller types that are either of topical interest
or, alternatively, will not be considered further in the book;
for example, paddle wheels or superconducting electric pro-
pulsion. Chapter 3 considers propeller geometry and, conse-
quently, this chapter can be viewed as a foundation upon
which the rest of the book is built. Without a thorough
knowledge of propeller geometry, the subject will not be
fully understood. Chapters 4 and 5 concern themselves with
the environment in which the propeller operates and the
wake field in particular. The wake field and its various
methods of prediction and transformation, particularly from
nominal to effective, are again fundamental to the under-
standing of the design and analysis of propellers.
Chapters 6–15 deal with propulsion hydrodynamics,
first in the context of model results and theoretical methods
relating to propellers fixed to line shafting, then moving on
to ship resistance and propulsion, including the important
subjects of propeller-hull interaction and thrust augmen-
tation devices, and finally to consideration of the specific
aspects of fixed and rotable thrusters and waterjets.
Chapter 17 addresses the all-important subject of sea trials
in terms of the conditions necessary for a valid trial, instru-
mentation, and analysis.
Chapters 18–20 deal with the mechanical aspects of pro-
pellers. Materials, manufacture, blade strength, and
vibration are the principal subjects of these four chapters,
and the techniques discussed are generally applicable to
all types of propulsors. The final five chapters, Chapters
21–25, discuss various practical aspects of propeller tech-
nology, starting with design, then continuing to operational
problems, service performance and, finally, to propeller
inspection, repair, and maintenance.
xix
xx Preface to the First Edition
In each of the chapters of the book, the attainment of a
fair balance between theoretical and practical consider-
ations has been attempted, so that the information pre-
sented will be of value to the practitioner in marine
science. For more advanced studies, particularly of a the-
oretical nature, the data presented here will act as a starting
point for further research: in the case of the theoretical
hydrodynamic aspects of the subjects, some of the refer-
ences contained in the bibliography will be found to be
of value.
This book, representing as it does a gathering together of
the subject of propulsion technology, is based upon the
research of many scientists and engineers throughout the
world. Indeed, it must be remembered that without these
people, many of whom have devoted considerable portions
of their lives to the development of the subject, this book
could not have been written and, indeed, the subject of pro-
peller technology could not have developed so far. I hope
that I have done justice to their efforts in this book. At
the end of each chapter, a series of references is given so
that, if necessary, the reader may refer to the original work,
which will contain full details of the specific research topic
under consideration. I am also considerably indebted to my
colleagues, both within Lloyd’s Register and in the marine
industry, for many discussions on various aspects of the
subject over the years, all of which have helped to provide
a greater insight into, and understanding of, the subject. Par-
ticularly, in this respect, thanks are given to Mr. C.M.R.
Wills, Mr. P.A. Fitzsimmons, and Mr. D.J. Howarth who,
as specialists in particular branches of the subject, have also
read several of the chapters and made many useful com-
ments concerning their content. I would also like to thank
Mr. A.W.O. Webb who, as a specialist in propeller materials
technology and colleague, has given much helpful advice
over the years in solving propeller problems and this
together with his many technical papers has influenced
much of the text of Chapters 17 and 25. Also, I am partic-
ularly grateful to Mr. J.Th. Ligtelijn of MARIN and to Dr.
G. Patience of Stone Manganese Marine Ltd, who have sup-
plied me with several photographs for inclusion in the text
and with whom many stimulating discussions on the subject
have been had over the years. Thanks are also due to the
many kind ladies who have so painstakingly typed the text
of this book over the years and without whom the book
would not have been produced.
J.S. Carlton
London, United Kingdom
Nomenclature

Upper Case Jp ship polar moment of inertia

K Prandtl or Goldstein factor
A cross-sectional area Kn Knapp’s similarity parameter
AC admiralty coefficient Kp pressure coefficient
AD developed area KQ propeller torque coefficient
AE expanded area KQS spindle torque coefficient
AM midship section area KT thrust coefficient
AO disc area KTN, KTD duct thrust coefficient
AP projected area KTP propeller thrust coefficient
AR aspect ratio KY side force coefficient
B molded breadth of ship L length of ship or duct
BP propeller power coefficient lift force
BAR blade area ratio section centrifugal bending moment arm
CA correlation factor LP sound pressure level
section area coefficient LPP length of ship between perpendiculars
Cb ship block coefficient LR length of run
CD drag coefficient LWL length of ship along waterline
CF frictional resistance coefficient M moment of force
CL lift coefficient Ma Mach number
CM moment coefficient N rotational speed (RPM)
section modulus coefficient number of cycles
CP pressure coefficient number of fatigue cycles
ship prismatic coefficient NS specific speed
propeller power coefficient P propeller pitch
CT thrust loading coefficient PB brake power
total resistance coefficient PD delivered power
CW wave-making resistance coefficient PE effective power
D drag force PG generator power
propeller diameter PS shaft power
Db behind diameter Q flow quantity
Do diameter of slipstream far upstream propeller torque
Ds shaft diameter QPC quasipropulsive coefficient
F force QS total spindle torque
fetch of the sea QSC centrifugal spindle torque
FB Bollard pull QSF frictional spindle torque
Fn Froude number QSH hydrodynamic spindle torque
G boundary layer unique shape function R radius of propeller, paddle wheel, or bubble
nondimensional circulation coefficient specific gas constant
H hydraulic head RAIR air resistance of ship
Hp pump head RAPP appendage resistance
I dry inertia Re real part
Ie polar entrained inertia RF frictional resistance
IVR inlet velocity ratio Rn Reynolds number
J advance coefficient

xxi
xxii Nomenclature

RT total resistance iG section generator line rake

RV viscous resistance iP propeller rake
RW wave-making resistance iS section skew-induced rake
S surface tension iT total rake of propeller section
ship wetted surface area j counter
SA additional load scale factor k counter
Sa apparent slip kc lifting surface camber correction factor
SBF solid boundary factor ks mean apparent amplitude of surface roughness
SC camber scale factor kt lifting surface thickness correction factor
T temperature kx lifting surface ideal angle of attack correction factor
draft of ship (1 + k) frictional form factor
propulsor thrust l counter
TA draft aft length
TF draft forward lcb longitudinal center of buoyancy
TN, TD duct thrust m mass counter
Tp propeller thrust m_ specific mass flow
UT propeller tip speed n rotational speed (rps)
V volume velocity p section pitch
Va speed of advance pressure
Vs ship speed pc cavity variation-induced pressure
X distance along coordinate axis pH propeller-induced pressure
Y distance along coordinate axis po reference pressure
W resultant velocity noncavitating pressure
width of channel pitch of reference section
We Weber number pv hull-induced vibratory pressure
Z blade number vapor pressure
distance along coordinate axis p1 apparent-induced pressure
Zm section modulus q dynamic flow pressure
r radius of a propeller section
rh hub or boss radius
Lower Case s length parameter
t time
a propeller axial inflow factor
thrust deduction factor
a1 propeller tangential inflow factor
section thickness
ac crack length
tF thickness fraction
ar resistance augmentation factor
tmax maximum thickness
b span of wing
to notional blade thickness at shaft center line
c wake contraction factor
u local velocity
section chord length
v local velocity
cd section drag coefficient
va axial velocity
cl section lift coefficient
vr radial velocity
cli ideal section lift coefficient
Vt tangential velocity
cm section moment coefficient
vT tide speed
cmax limiting chord length
w downwash velocity
f frequency
mean wake fraction
function of …
wF Froude wake fraction
g acceleration due to gravity
wmax maximum value of wake fraction in propeller disc
function of …
wn nominal wake fraction
h fluid enthalpy
wp potential wake fraction
height
wT Taylor wake fraction
hydraulic head
wv viscous wake fraction
hb height of bulbous bow centroid from base line in
ww wave-induced wake fraction
transverse plane
x distance along a coordinate axis
i counter
 Nomenclature xxiii

nondimensional radius (r/R) ho propeller open water efficiency

xc distance along chord hp pump efficiency
radial position of centroid hr relative rotative efficiency
xcp center of pressure measured along chord u pitch angle
xo reference section transformation parameter
y distance along coordinate axis momentum thickness of boundary layer
yc camber ordinate ufp face pitch angle
yL section lower surface ordinate uip propeller rake angle
yt thickness ordinate unt nose-tail pitch angle
yU section upper surface ordinate uo effective pitch angle
z distance along coordinate axis us section skew angle
usp propeller skew angle
uw angular position of transition wake roll-up point
Suffixes
L frequency reduction ratio
m model l wavelength
s ship source-sink strength
U upper ship-model scale factor
L lower m coefficient of dynamic viscosity
b bound, behind r density of water
F free ra density of air
O reference value rL leading edge radius
x reference radius rm density of blade material
s cavitation number
stress on section
Greek and Other Symbols
sa alternating stress
a angle of attack gas content sF corrosion fatigue strength
ad cavitation bucket width si inception cavitation number
ai ideal angle of attack sL local cavitation number
aK air content ratio sMD mean design stress
a0 zero lift angle sn cavitation number based on rotational speed
b advance angle relative shaft angle
b« hydrodynamic pitch in the ultimate wake so free steam cavitation number
bi hydrodynamic pitch angle sR residual stress
G circulation ss blade solidity factor
g local vortex strength sx blade stress at location on blade
length parameter τ shear stress
ratio of drag to lift coefficient (Cd/Ce) τC thrust loading coefficient
gg correction to angle of attack due to cascade effects y coefficient of kinematic viscosity
D change in parameter f angle of rotation in propeller plane
displacement of ship hull-form parameter
d boundary layer thickness velocity potential
linear displacement angular displacement
propeller speed coefficient flow coefficient
« thrust eccentricity shaft alignment angle relative to flow
transformation parameter c transformation parameter
z Bendemann static thrust factor gas content number
damping factor energy transfer coefficient
transformation parameter V angular velocity
hb propeller behind hull efficiency v angular velocity
hh hull efficiency — volumetric displacement
hi ideal efficiency
hm mechanical efficiency
xxiv Nomenclature

Abbreviations LES large eddy simulation

a.c. alternating current LNG liquid natural gas
AEW Admiralty Experiment Works, Haslar MARIN Maritime Research Institute of the Netherlands,
AP after perpendicular formerly NSMB
ATTC American Towing Tank Conference MEPC Marine Environment Protection Committee of IMO
BHP Brake horsepower MCR maximum continuous rating
BS British Standard mph miles per hour
CAD computer-aided design NACA National Advisory Council for Aeronautics
CAM computer-aided manufacture NC numerically controlled
cwt hundred weight NCR normal continuous rating
(1 cwt ¼ 12 lbf ¼ 50.8 kgf) OD oil distribution
DES design PIV particle image velocimetry
DHP delivered horsepower PHV propulsor hull vortex
DTNSRD David Taylor Naval Ship Research and Design Centre qrs quarters (4 qrs ¼ 1 cwt; 1 cwt ¼ 50.8 kgf)
EEDI energy efficiency design index RANS Reynolds averaged Navier Stokes
EHP effective horse power RH right handed
ft feet rpm revolutions per minute
HMS Her Majesty’s Ship shp shaft horsepower
hp horsepower SM Simpson’s multiplier
HSVA Hamburg Ship Model Basin SPA self polishing antifouling
IMO International Maritime Organization SSPA Statens Skeppsprovningsanstalt, G€ oteborg
ISO International Standards Organization TE trailing edge
ITTC International Towing Tank Conference THP thrust horse power
LDV laser Doppler velocimetry VLCC very large crude carrier
LE leading edge VTOL vertical take-off and landing
Chapter 1

The Early Development of the Screw

Propeller

Both Archimedes (c.250 BC) and Leonardo da Vinci (c.1500) mathematicians and scientists of Europe were attracted by
can be credited with having considered designs and ideas, this offer and names such as d’Alembert, Euler, and Ber-
which would subsequently be explored by ship propulsion noulli appear in the contributions. Bernoulli’s contribution,
engineers many years later. In the case of Archimedes, his for which he won a prize, introduced the propeller wheel,
thoughts centered on the application of the screw pump, shown in Fig. 1.2, which he intended to be driven by a New-
which bears his name and this provided inspiration to the comen steam engine. With this arrangement, he calculated
19th-century engineers involved in marine propulsion. Unfor- that a particular ship could be propelled at just under 2½
tunately, however, it also gave rise to several subsequent mis- knots by the application of some 20–25 hp. Opinion,
conceptions about the basis of propeller action by comparing it however, was still divided as to the most suitable propulsor
to that of a screw thread. In contrast Leonardo da Vinci, in configuration: indeed, as it was to be for many years to
his sketchbooks, which were produced some 1700 years after come. For example, the French mathematician Paucton,
Archimedes, shows an alternative form of screw propulsion working at about the same time as Bernoulli, suggested a
based on the idea of using fan blades, which had a similar different approach, illustrated in Fig. 1.3, which was based
appearance to those used for cooling purposes today. on the Archimedean screw principle.
The development of screw propulsion as we recognize it Thirty-three years after the Paris invitation Joseph
today can be traced back to the work of Robert Hooke who is Bramah in England proposed an arrangement for a screw
perhaps better remembered for his work on the elasticity of propeller located at the stern of a vessel, which, as is seen
materials. Hooke in his Philosophical Collections, pre- from Fig. 1.4, contains most of the features, which we
sented to the Royal Society in 1681, explained the design
of a horizontal watermill, which was remarkably similar
in its principle of operation to the Kirsten-Boeing vertical
axis propeller developed two and a half centuries later.
Hooke’s watermill comprised six wooden vanes, geared
to a central shaft and pinned vertically to a horizontal cir-
cular rotor. The gearing constrained the vanes to rotate
through 180 degrees about their own spindle axes for each
complete revolution of the rotor.
During his life, Hooke was also interested in the subject
of metrology and in the course of his work he developed an
air flow meter based on the principle of a windmill. He suc-
cessfully modified this instrument in 1683 to measure water
currents and then foresaw the potential of this invention to
propel ships if provided with a suitable means of motive
power. As seen from Fig. 1.1 the instrument comprised four,
flat rectangular blades located on radial arms with the blades
inclined to the plane of rotation.
Some years later in 1752, the Academie des Sciences in
Paris offered a series of prizes for research into theoretical
methods leading to significant developments in naval archi-
tecture. As might be expected, many of the famous FIG. 1.1 Hooke’s screw propeller (1683).

Marine Propellers and Propulsion. https://doi.org/10.1016/B978-0-08-100366-4.00001-8

© 2019 Elsevier Ltd. All rights reserved. 1
2 Marine Propellers and Propulsion
FIG. 1.2 Bernoulli’s propeller wheel (1752).
FIG. 1.3 Archimedean screw of Paucton.
FIG. 1.4 Bramah’s screw propeller design (1785). (Reproduced with
permission.)
FIG. 1.5 Shorter’s propulsion system (1802). (Reproduced with
permission.)
associate with screw propulsion today. It comprised a pro-
peller with a small number of blades driven by a horizontal
shaft, which passes into the ship’s hull below the waterline.
However, there appears to be no evidence of any trials of a
propeller of this kind being fitted to a ship and driven by a
steam engine. Nevertheless, in 1802, Edward Shorter used
a variation of Bramah’s idea to assist sailing vessels that were
becalmed to make some headway. In Shorter’s proposal,
Fig. 1.5, the shaft was designed to pass into the vessel’s hull
above the waterline, which eliminated the need for seals, and
the motive power for this propulsion arrangement was pro-
vided by eight men turning a capstan. Using this technique,
Shorter managed to propel the transport ship Doncaster in
Gibraltar and again at Malta at a speed of 1.5 mph in calm
conditions. Perhaps in view of the means of providing the
motive power no further application of Shorter’s propeller
was recorded. Nevertheless, it is interesting to note the enthu-
siasm with which this propeller was received by Admiral Sir
Richard Rickerton and his Captains as witnessed by the tran-
script of a letter dated the 4th July 1802; Fig. 1.6. However,
Edward Shorter recognized that this propulsion concept
could be driven by a steam engine.
Colonel John Stevens, who was a lawyer in the United
States and a man of substantial financial means, experi-
mented with screw propulsion in the year following Edward
Shorter’s proposal. As a basis for his work, he built a 25-ft
long boat into which he installed a rotary steam engine and
coupled this directly to a four-bladed propeller. The blades
of this propeller were flat iron plates riveted to forgings,
which formed a “spider-like” boss attachment to the shaft.
Stevens later replaced the rotary engine with a steam engine
of the Watt type and managed to attain a steady cruising
speed of 4 mph with some occasional surges of up to
8 mph. However, he was not impressed with the overall per-
formance of his craft and decided to turn his attention and
energies to other forms of marine propulsion.
In 1824 contrarotating propellers made their appearance
in France in a design produced by Monsieur Dollman. He
used a pair of two-bladed windmill type propellers rotating
in opposite directions on the same shaft axis to propel a
small craft. Following on from this French development,
 The Early Development of the Screw Propeller Chapter 1 3

FIG. 1.6 Certificate of performance for Mr. Shorter’s propeller arrangement. (Courtesy: Mr. J. Wiltshire, Qinetiq.)

the scene turned once again to England where John
Ericsson, a former Swedish army officer residing at that
time in London, designed and patented in 1836 a propulsion
system comprising two contrarotating propeller wheels.
His design is shown in Fig. 1.7, from which it can be seen
that the individual wheels were not dissimilar in outline to
Bernoulli’s earlier proposal. Each wheel comprised eight
short, wide blades of a helical configuration mounted on
a blade ring with the blades tied at their tips by a peripheral
strap. In this arrangement, the two wheels were allowed to
rotate at different speeds, probably to overcome the problem
of the different flow configurations induced in the for-
ward and after wheels and, perhaps, also blade passing
frequency issues. Ericsson conducted his early trials on a
3-ft model and the results proved successful enough to
encourage him to construct a 45-ft vessel, which he named
4 Marine Propellers and Propulsion
FIG. 1.7 Ericsson’s contrarotating screw propeller (1836).
the Francis B. Ogden. This vessel was fitted with his pro-
pulsion system, which had blade wheels having a diameter
of 5 ft 2 in. Trials were conducted on the River Thames in
the presence of representatives from the Admiralty and the
vessel was observed to be capable of a speed of some
10 mph. However, in his first design, Ericsson placed the
propeller astern of the rudder and this had an adverse effect
both on the steering characteristics of the ship and on the
water flow into the propeller. The Admiralty Board expressed
disappointment with the trial, although the propulsion results
were good when judged by the standards of the day. It was
said that one reason for their concern was over a vessel’s
ability to steer reliably when propelled from the stern. Fol-
lowing this rebuff, Ericsson left England for the United States
and in 1843 designed the US Navy’s first screw-propelled
vessel, the Princeton. It has subsequently been suggested that
by around this time the US merchant marine had some 41
screw-propelled vessels in operation.
The development of the screw propeller depended not
only on technical development but also upon the availability
of finance as well as being subjected to political influences
and the likely return on the initial investment that might
accrue to the inventor or his backers. Francis Petit Smith
was rather more successful in these respects than his con-
temporary Ericsson. Smith took out a patent in which a dif-
ferent form of propeller was used, more akin to an
Archimedean screw, but, more importantly, based on a dif-
ferent location of the propeller with respect to the rudder.
This happened just a few weeks prior to Ericsson estab-
lishing his patent and shortly after Ericsson’s trials the
British Admiralty modified their view of screw propulsion
due to Smith’s work.
Despite being frequently referred to as a farmer, Smith
had a sound classical education and explored the concepts
of marine propulsion by making model boats and testing
them on a pond. From one such model, which was propelled
by an Archimedean screw, he was sufficiently encouraged
to build a 6-ton prototype boat, the F P Smith. This was
powered by a 6-hp steam engine and to which he fitted a
wooden Archimedean screw of two turns and the vessel
underwent trials on the Paddington Canal in 1837. However,
by one of those fortunate accidents, which sometimes occur
in the history of science and technology, the propeller was
damaged during the trials and about half of it broke off,
whereupon the vessel immediately increased its speed.
Smith recognized the implications of this accident and mod-
ified the propeller accordingly. After completing calm water
trials, he took the vessel on a voyage down the River
Thames from Blackwall in a series of stages to Folkestone
and then eventually on to Hythe on the Kentish coast.
Between these last two ports the vessel averaged a speed
of some 7 mph. On the return voyage to London, Smith
encountered a storm in the Thames Estuary and the little
craft apparently performed excellently in these adverse
weather conditions. In March 1830 Smith and his backers,
Wright and the Rennie brothers, made an approach to the
Admiralty, who then requested a special trial for their
inspection. The Navy’s response to these trials was suffi-
ciently encouraging to motivate Smith and his backers to
construct a larger ship of 237 tons displacement, which he
called Archimedes. This vessel, which was laid down by
Henry Wilmshurst and engined by George Rennie, was
completed in 1839. It had a length of 125 ft and was rigged
as a three-masted schooner. The Archimedes was completed
just as the ill-fated Screw Propeller Company was incorpo-
rated as a joint stock company. The objectives of this
company were to purchase Smith’s patents; transfer the
financial interest to the company and sell licenses to use
the location of the propeller within the deadwood of a ship,
as suggested by Smith. However, the propeller design itself
was not included in the arrangement.
The Archimedes was powered by two 45-hp engines and
was finally fitted with a single-turn Archimedean screw,
which had a diameter of 5 ft 9 in.; a pitch of 10 ft and was
about 5 ft in length. This propeller was the last of a series
tried on the ship, the first having had a diameter of 7 ft with
a pitch of 8 ft and a helix making one complete turn. This
propeller was subsequently replaced by a modification in
which double-threaded screws, each of half a turn, were
employed in accordance with Smith’s amended patent of
1839: the propeller is shown in Fig. 1.8. After undergoing
a series of proving trials during which the speed achieved
was in excess of nine knots, the ship arrived at Dover in
1840 to undertake a series of races against the crosschannel
packets, which at that time were operated by the Royal
Navy. The Admiralty was duly impressed with the results
of these races and agreed to the adoption of screw
propulsion in the Navy. The Archimedes was then lent to
Brunel who fitted her with a series of propellers having
different forms.

FIG. 1.8 Propeller fitted to the Archimedes (1839).
Concurrent with these developments, other inventors
had also introduced novel features into their propeller
designs. In 1838, Lowes patented a propeller comprising
one or more blades where each blade was a portion of a
curve, which if continued would produce a screw. This
arrangement was equivalent to a pair of tandem propellers
located on a single shaft with each blade being mounted
on a separate boss. Subsequently, the SS Novelty was built
at Blackwall by Mr. Wilmshurst between the years 1839 and
1840 to test the principle of screw propulsion: indeed, this
ship can be considered as being the first screw-propelled
cargo ship. Additionally, in 1839, Rennie patented a
conoidal design of propeller in which he proposed increases
in pitch from forward to aft of the blade; three-bladed
helices and the use of skewback in the design. Taylor and
Napier, a year later, experimented with tandem propellers,
some of which were partially submerged. Subsequent to
these various developments taking place, by 1842 the
“windmill” propeller, as opposed to the Archimedean
screw, had developed to a fairly advanced state as witnessed
by Fig. 1.9.
This propeller, shown in Fig. 1.9, was fitted to the
Napoleon, a ship having a displacement of 376 tons, and
is particularly interesting in that it was developed into its
final form based on a series of model tests in which
diameter, pitch, blade area and blade number were all
FIG. 1.9 Propeller of the Napoleon (1842).
The Early Development of the Screw Propeller Chapter 1 5
varied. The first propeller in this series was designed with
three blades each having a length of a third of a turn of a
screw thread, thereby giving a high-blade area ratio. Never-
theless, as the design evolved, better results were achieved
with shorter-length blades of around 22% of a full thread
turn. The final propeller was manufactured from cast iron
and rotated at 126 rpm in association with a ship speed of
between 10 and 12 knots. The Napoleon was built by
Augustin Normand at Le Havre and the propellers were
designed and manufactured in Manchester by John Barnes
who also built the engines. Although the ship was originally
destined for postal service duties on the Mediterranean Sea,
she was later acquired by the French Navy and deployed as a
dispatch boat.
The result of Brunel’s trials with the Archimedes was
that the design of the Great Britain, now preserved in a
drydock at Bristol in England, was adapted for screw pro-
pulsion from the originally intended paddle propulsion. It
is, however, interesting to note that the general form of
the propeller adopted by Brunel for the Great Britain did
not follow the general design arrangement of the propellers
developed by Smith but was more like that proposed by
Ericsson: except that in the case of the Great Britain, the
propeller was not of the contrarotating type as seen in
Fig. 1.10. However, the original propeller as designed
by Brunel was subsequently modified since it tended to
break in service. While the original propeller was 16 ft in
diameter, had six blades, and was made from a single
FIG. 1.10 Replica of Brunel’s propeller for the Great Britain.
6 Marine Propellers and Propulsion
casting, the propeller that was finally adopted was a built-up
wrought iron propeller, also with six blades, having a
diameter of 15.5 ft and a pitch of 25 ft. It is also interesting,
and quite remarkable given the available design and
analysis capabilities of the time, to note that the pitch chosen
was not dissimilar, in effective pitch terms, from that which
would have been chosen today.
As a direct result of the Royal Navy’s commitment to
screw propulsion HMS Rattler, a sloop of approximately
800 tons and powered by a steam engine of about 200 hp,
was laid down in 1841 at Sheerness Dockyard. She then
underwent initial sea trials in the latter part of 1843 where
she achieved a speed of some 8¾ knots and subsequently ran
a race against her paddle half-sister, HMS Polyphemus. A
design study was then commissioned to study the various
facets of propeller design and also to optimize a propeller
design for Rattler and by January 1845 some 32 different
propeller designs had been tested. The best of these pro-
pellers was designed by Francis Petit Smith and propelled
the ship at a speed of about nine knots. This propeller
was a two-bladed design with a diameter of 10 ft 1 in., a pitch
of 11 ft, and weighed 26 cwt 2 qrs (1.68 tons). During the
spring of 1845, Rattler ran a series of competitive trials
against the paddle steamer Alecto. These trials embraced
both free-running and towing exercises as well as a series
of separate sail, steam, and combined sail and steam pro-
pulsion trials. By March 1845, the Admiralty was convinced
of the advantages of screw propulsion and had ordered
seven screw-propelled frigates together with a number of
lesser ships and in April of that year the famous “tug of
war” between the Rattler and the Alecto was held. However,
this event appears to have been more of a public relations
exercise than a scientific trial.
In 1846 Joseph Maudsley patented a two-bladed pro-
peller design in which the propeller could be lifted by a rope
and tackle connected to a crosshead and which permitted the
propeller to be raised to deck level. One year later HMS
Blenheim, which had been built in 1813, was fitted with a
similar arrangement to that proposed by Maudsley when
she was converted from sail to screw propulsion. The fol-
lowing year, 1848, he patented a further design in which
the blades of a two-bladed propeller, when not working,
could be turned into the plane of the shaft to reduce sailing
resistance. Indeed, this was an early example of a feathering
propeller, which is used in some hybrid propulsion designs
today. This theme of raising the propeller and, thereby,
reducing the resistance of the ship when under sail was con-
tinued by Seaward. In 1848, he developed a folding pro-
peller in which the blades were cut into five radial
segments, which could be folded to be contained within
the projection of the ship’s deadwood. Indeed, the configu-
ration of the propeller blades resembled, to some extent, a
lady’s hand-held fan in its form and operation. Later, in
1865, the Rev. P.A. Fothergill patented a self-feathering
propeller, which removed the need to raise the propeller
when under sail. In this design, the blades were so arranged
as to take up a position of least resistance when not being
rotated.
In 1853, John Fisher patented a two-bladed design with
perforated blades in which these perforations were in the
form of slots to disperse any air that may have been
entrained on the blades. A year later, Walduck patented a
design of propeller, which was intended to attenuate the cen-
trifugal motion of water over the blade surfaces by intro-
ducing a series of terraces, concentric with the shaft, but
each being greater in pitch than its inner neighbor. Indeed,
this theme was returned to many times during the subse-
quent development of the propeller, one of the later devel-
opments being in 1924 where chordal plates were
introduced into the blade design.
Peacock, in 1855, patented an auxiliary propeller in
which each blade was built from iron plate and supported
by a stay-rod projecting radially from the boss. Interest-
ingly, each blade was shaped to correspond to the general
form of a bee’s wing and the working surfaces of the blade
given a parabolic form.
Although accepted by the Navy, screw propulsion had
not been universally accepted for seagoing ships in pref-
erence to paddle propulsion as witnessed by the relatively
late general introduction of screw propulsion by the North
Atlantic Steamship companies. Nevertheless, the latter part
of the 19th century saw a considerable effort being
expended by many people to explore the effects of radial
pitch distribution, adjustable blades, blade arrangement
and outline and cavitation. For example, in 1860 Hirsch pat-
ented a propeller having both variable chordal pitch, which
we know today by the term camber, and variable radial
pitch. As an additional feature, this propeller also possessed
a considerable amount of forward skew on the blades.
A type of propeller known as the Common Screw
emerged and this was the most successful type of propeller
in use before 1860. The working surfaces of the blades were
portions of helices cut off by parallel lines, about an eighth
of the pitch apart, and located on a small cylindrical boss.
With these propellers, the blade chord lengths increased
from root to tip; however, Robert Griffiths modified a blade
of this type to have rounded tips and this was particularly
successful. Indeed, the Admiralty, which had several
Common Screws, reduced their broad tips by cutting away
the leading corners and this resulted in significant reduc-
tions in vibration.
During this period of rapid development, the compe-
tition between rival designers was great. In 1865 Hirsch
designed a four-bladed propeller for the SS P eriere, which
had originally been fitted with a Griffiths propeller design.
In this case, a one-knot improvement was recorded on
trial and similar results were noted when Hirsch’s pro-
pellers replaced other designs. At around the same time a

four-bladed, 22.8-ft diameter, 21.37-ft pitch, 11.7-ft-long
propeller was constructed for HMS Lord Warden. This pro-
peller was of a built-up design with the blades bolted
through slots, to permit adjustments to the blade pitch,
and with this propeller the ship attained a speed of 13.5
knots during trials. In the 1890s, Hirsch also introduced
the idea of bolted-on blades, thereby providing another early
example of built-up propellers. This concept also achieved
considerable popularity in the first half of the 20th century
when the design and manufacture of monoblock propellers
was difficult.
Thornycroft, in 1873, designed a propeller with
restricted camber in the mid-span regions of the blade and
combined this with a backward curvature of the blades in
an attempt to suppress tangential flow. Zeise carried the
ideas of the development of the radial pitch distribution a
stage further in 1886 when he increased the pitch of the
inner sections of the blade so as to try and make better
use of this region of the blades.
In parallel with the development of what might be
termed fixed pitch propeller designs, during the period
1844 through to about 1911, several inventors turned their
attention to the potential for controllable pitch propellers.
In reality, however, a number of these designs would be
better termed “adjustable” rather than “controllable” pitch
propellers. Bennett Woodcroft in 1844 patented a design
with adjustable blades and this had blades with increasing
pitch from forward to the after edge, which was in keeping
with his earlier patent of 1832. Later, in 1844, he patented a
further modification where short links to the blade stems
replaced his earlier idea of grooves on a collar to actuate
the blades. To fix the blades in the desired position, a similar
collar was provided aft with two wedge-shaped arms that
acted on small sliding pieces. In 1868, Mr. H.B. Young pat-
ented a method of altering pitch by which the shanks of two
blades are inserted in a hollow boss and extend through it.
These shanks were retained in position by arms projecting
from them, which were controlled by a nut on a screwed
rod. This rod extended through the main boss and was
turned by a key, which was manipulated from within the
ship. In the same year, Mr. R. Griffiths introduced his
concept of an adjustable pitch propeller. In this design,
within the propeller boss, the shank of a blade was provided
with an arm, connected to a link with a collar, and screwed
on a sleeve that was loose on the propeller shaft. The after
end of the sleeve turned in a groove, in which fitted a brake,
and could be tightened on a collar by actuation from on deck
by a tab on a screw. The movement for this pitch actuation
was achieved by slowly rotating the propeller shaft, thereby
screwing the collar along it and in the process setting the
blades to the desired position or, indeed, feathering them.
In a further arrangement, patented in 1858, Young designed
a system with a cotter, which passed through the stem of the
blades and rested in a sector-shaped recess in the boss. The
The Early Development of the Screw Propeller Chapter 1 7
alteration of pitch was effected by varying several packing
pieces. The Bevis-Gibson reversible propeller, patented in
1911, was a development of an earlier feathering propeller
patent in 1869 by Mr. R.R. Bevis, which had been used
extensively. This new patent provided a means of reversing
a small vessel driven by an internal combustion engine. In
this design, the roots of two blades were provided with
toothed pinions, which mated with a rack. The racks were
yoked together and were actuated parallel to the shaft by
means of a central rod which passed through a hollow pro-
peller shaft. This rod was then operated by a lever at the
control position in the craft to adjust the pitch into a forward,
astern or feathering position.
The contrarotating propeller concept received further
attention in 1876 when Mr. C.S. de Bay designed a pro-
pulsion system for the steam yacht Iolair; a 40.4-gross-
ton schooner rigged vessel having a length of 81.5 ft. His
design, a model of which is in the Science Museum in
London and is shown in Fig. 1.11, comprised two propellers
of equal and opposite pitch mounted on the same shaft but
revolving in opposite directions. The diameters of the pro-
pellers differed slightly with the larger having three blades
and the smaller four blades. As can be seen from the figure,
this propulsor utilized extremely complex blade shapes with
portions of the blades being cut out so that the remainder of
the blades could revolve in an interlocking manner. The idea
behind this blade shape complexity was to try and prevent
energy losses caused by the centrifugal and other motions
of the water. Comparative trials in 1879 were made between
this propeller and a Griffiths design, of a similar type to that
used on HMS Lord Warden, and it was stated that the de
Bay’s design achieved an efficiency at least 40% greater
than that of the competing design: this represented a speed
increase of around one knot.
FIG. 1.11 de Bay’s contrarotating propeller design.
8 Marine Propellers and Propulsion
In 1878, Col. W.H. Mallory in the United States of
America introduced the concept of the azimuthing pro-
peller. In this design, the propeller was carried in a frame,
which rather resembled a rudder and was rotated by a bevel
gear driven by an engine mounted on the deck.
Other developments of note in the context of the early
history of propeller design are those by Mangin, Zeise,
and Taylor. Mangin in, 1851, attempted to increase the
thrust of a propeller by dividing the blades radially into
two portions. Griffiths also used this idea in 1871, but
instead he used only a partial division of the blades in their
FIG. 1.12 Various early propeller developments. (Reproduced with permission from parts of references 2 and 3.)
center regions. Zeise in 1901 experimented with the idea of
flexible blades, in which the trailing part of the blade was
constructed from lamellae and Taylor, some 6 years later,
introduced air injection on the blade suction surface to try
and control the erosive effects of cavitation. Fig. 1.12 shows
a collage of some of these propeller designs, and those dis-
cussed previously, in which their novel features can be seen.
The latter part of the 19th century also saw the intro-
duction of some theoretical methods, which attempted to
explain the action of the screw propeller. Notable among
these were the works of Rankine and Froude. These,

together with subsequent developments, which occurred
during the 20th century, will be introduced in later chapters;
notably in Chapter 8.
The foregoing paragraphs have outlined some of the
activities and developments in the early years of moving
from sail and paddle propulsion to the application of pro-
pellers to ship propulsion. These innovations paved the
way for the further development of marine propeller tech-
nology during the 20th and latterly in the 21st centuries.
With perhaps the exception of ducted propellers in the early
years of the 20th century, propeller design after the turn of
the 19th century advanced principally through experiment
and theoretical methods with the aim of improving effi-
ciency, maneuverability, and controlling cavitation in the
context of either vibration or erosion. During the 20th
century, the systematic propeller series that were developed
in various experimental facilities around the world helped
both the understanding of propeller hydrodynamics as well
as forming the basis of design methods. In many cases, the
speed of development was particularly fast.
Within the context of development, in 1907, just 10 years
after Sir Charles Parsons had introduced the steam turbine
into marine practice at the fleet review on the 26th June
1897 with the 2000-hp Turbinia, the steam-turbine-driven
liner Mauritania was built and absorbed 70,000 shp through
four propellers rotating 180 rpm. These propellers, weighing
18.7 tons, had a diameter of 16.75 ft, a pitch of 15.5 ft, and a
blade area ratio of 0.467. The Mauritania achieved a speed of
26.3 knots and held the Blue Riband for the North Atlantic
from 1910 through to 1929. Some years later, the RMS Queen
Mary powered by four single-reduction geared turbine sets,
aggregating 160,000 shp over her four propellers, each
weighing around 35 tons, achieved speeds of between 30
and 32 knots. Due to the thrust density of her propellers,
the blade area ratios were increased significantly from those
of the Mauritania. Nevertheless, the Queen Mary initially
suffered badly from the effects of propeller cavitation, which
led to erosion of the blades and gave rise to significant cav-
itation excited vibration in the ship. Consequently, the first
series of propellers that had been fitted to the ship were rede-
signed, particularly with respect to blade shape and section
form, and this largely relieved the problems.
The Early Development of the Screw Propeller Chapter 1 9
Since then, there has been a progressive and sometimes
rapid growth in size of propellers. This has over the course
of time given rise to many hydrodynamic, stressing, met-
allurgical, and productions issues. Typical of these growth
rates was the rapid growth in the sizes of tankers during the
late 1960s and early 1970s. For example, a 380,000-dwt
single-screw VLCC (Very Large Crude Carrier) built in
the early 1970s was propelled by a six-bladed, fixed pitch
monoblock propeller. This propeller required the casting
of 93 tons of nickel-aluminum bronze to produce a pro-
peller of 70 tons finished weight. Today some of the pro-
pulsion challenges are to be found with the growth in
size of container and LNG ships, due mostly to port con-
straints. For example, in the case of a 12,500-teu container
ship designed to sail at 25 knots, this will absorb through a
single, six-bladed propeller some 67.3 MW at a rotational
speed of 90 rpm. Such a propeller will have a diameter of
9600 mm, a pitch ratio of 1.04, a blade area ratio of 0.85,
and will have a finished weight of around 130 tons. Apart
from increases in ship size or propeller thrust density, there
have been other challenges; for example, the introduction
of azimuthing and podded propulsors each of which have
introduced problems related to their design and operation.
Similarly, with transverse propulsion units and maneu-
vering issues such as dynamic positioning and the asso-
ciated control issues.
BIBLIOGRAPHY
Brown, D.K., 1983. A Century of Naval Construction. Conway, London.
Carlton, J.S., April 2006. The propulsion of a 12500 ton container ship.
Trans. I. Mar. EST.
Dirkzwager, J.M., October 1969. Some aspects on the development of
screw-propulsion in the 19th and early 20th century. 4th Lips Propeller
Symposium.
Ericsson, J., 1803–1899. RINA Affairs. 1836/43.
Geissler, R., 1921. Der Schraubenpropeller: Eine Darstellung Seiner
Entwicklung Nach dem Krayn. Inhalt der Deutschen, Amerikanischen
und Englischen (Patentliteratur). Krayn, Berlin.
Lambert, A.D., 1999. The Royal Navy and the introduction of the screw
propeller. Hist. Technol. 21.
Taggart, R., 1969. Marine Propulsion: Principles and Evolution. Gulf Pub-
lishing, Houston, TX.
Chapter 2
Propulsion Systems
Chapter Outline
2.1 Fixed Pitch Propellers
2.2 Ducted Propellers
2.3 Podded and Azimuthing Propulsors
2.4 Contrarotating Propellers
2.5 Overlapping Propellers
2.6 Tandem Propellers
2.7 Controllable Pitch Propellers
2.8 Surface Piercing Propellers
The early development of the propeller up to around 1900,
together with a few insights into its subsequent progress,
was considered in the previous chapter. Here we move
forward to the present day and discuss, again in outline,
the range of propulsion systems that are either currently
in use or have been under development. While many of
the concepts and systems considered in this chapter are
examined in greater depth in later chapters, it is useful first
to gain an overview of the various facets of the subject prior
to examining them in more detail.
2.1 FIXED PITCH PROPELLERS
The fixed pitch propeller has formed the basis of propeller
production over the years in either its monoblock or built-up
forms. While the monoblock propeller is the most common
type of propeller today, the built-up propeller, whose blades
are cast separately from the boss and then bolted or fixed in
some other way to it after machining, is now rarely used
except in a few niche markets. This, however, was not
always the case since in the early years of the last century
built-up propellers were very common: this was in part
due to the inability to achieve good quality large castings
at that time and due to difficulties in defining the correct
propeller blade pitch. Therefore, the bolting arrangements
in the blade palms gave some latitude to correct pitch errors
relatively easily. In this latter respect, the built-up propeller
has obvious advantages. Nevertheless, built-up propellers
generally have a larger diameter boss than their fixed pitch
counterparts, and this may cause difficulty with cavitation
problems in the blade root regions as well as adversely
affecting the propeller efficiency to a limited extent.
Marine Propellers and Propulsion. https://doi.org/10.1016/B978-0-08-100366-4.00002-X
© 2019 Elsevier Ltd. All rights reserved.
11 2.9 Waterjet Propulsion 21
13 2.10 Cycloidal Propellers 22
16 2.11 Paddle Wheels 22
17 2.12 Magnetohydrodynamic Propulsion 24
18 2.13 Rim-Driven Propulsors 26
18 2.14 Whale-Tail and Fish Propulsion 26
19 2.15 Hybrid Propulsion 27
21 Bibliography 27
The term monoblock propellers embrace a broad
spectrum of design types and sizes: typically ranging from
those propellers weighing only a few kilograms for use on
small power boats and some specialized craft to those, for
example, destined for large container ships, which may
weigh upwards of 100 tons and require the simultaneous
casting of significantly more metal than can be accommo-
dated in one ladle to produce the casting. Fig. 2.1 shows a
collage of various types of nonducted, fixed pitch propeller
in use today. These range from a large five-bladed propeller
fitted to a bulk carrier, seen in the figure in contrast to a man
standing on the dock bottom, through to highly skewed pro-
pellers for merchant and naval applications, to small high-
speed patrol craft propellers and a surface piercing propeller.
As might be expected, the materials used in manufacture
vary considerably over such a wide range of designs and
sizes. For the larger propellers, over 300 mm in diameter,
the nonferrous materials predominate: high-tensile brass
together with the manganese and nickel-aluminum bronzes
are the most favored types of materials with the nickel-
aluminum bronzes predominating. Nevertheless, stainless
steel has also gained limited use in niche markets and cast
iron, once a common material to produce spare propellers,
has now virtually disappeared from use. However, for the
smaller propellers use is frequently made of materials such
as the polymers, aluminum, nylon, and more recently carbon
fiber composites, which offer exciting potential in the field of
hydroelastic blade design to enhance propeller performance
both in terms of efficiency and cavitation performance.
In the case of fixed pitch propellers, the choice of blade
number, notwithstanding the important considerations of
blade-to-blade clearances at the blade root to boss interface,
11
12 Marine Propellers and Propulsion

FIG. 2.1 Typical fixed pitch propellers: (A) large five-bladed propeller fitted to a bulk carrier; (B) high-speed patrol craft propeller; (C) seven-bladed
balanced high-screw design; (D) surface piercing propeller; and (E) biased high-skew, low-blade-area ratio propeller.

is largely an independent variable. However, the major
influencing factor in this choice is the achievement of a mis-
match with the range of hull, superstructure, and machinery
vibration frequencies, which are considered likely to give
rise unwelcome vibration characteristics in the ship. Fur-
thermore, blade number can also be a useful parameter in
controlling unwelcome cavitation characteristics generated
on the blade surfaces during each revolution of the pro-
peller. The range of blade numbers for merchant ships is
generally from two to seven, recognizing that two- and
seven-bladed propellers are relatively rare, although in
some specialized naval applications, where considerations
of radiated noise become important, blade numbers greater
than this range have been researched and used to solve a
variety of propulsion problems. For most merchant vessels,
however, four, five, and six blades are the generally used
number, although many tugs and fishing vessels commonly
use three-bladed designs. In the case of small work or
pleasure power boats, two- and three-bladed propellers tend
to predominate.
The early propeller design philosophies centered on the
optimization of propeller efficiency. Today this aspect is no
 Propulsion Systems Chapter 2 13

less important and in some respects, associated with energy

conservation and atmospheric emissions reduction, has
assumed a greater importance as witnessed by the intro-
duction of the EEDI requirements and MARPOL Annex
VI. However, other constraints on design have emerged.
These are in response to calls for the reduction of vibration
excitation and radiated noise from the propeller. This latter
aspect has, of course, been a prime concern of naval ship
propeller designers for many years; however, pressure to
introduce these constraints, albeit in a generally less
stringent form, into merchant ship design practice has
grown over recent years. These considerations have been
brought about by the increases in power transmitted per
shaft; the use of after deckhouses; the maximization of
the cargo carrying capacity of the ship which imposes con-
straints on the hull lines and, in turn, the inflow into the pro-
peller; production constraints on hull form and international
legislation. Moreover, in recent years, there has been a
growing awareness of the effects of underwater radiated
noise from ships on marine mammals and fish.
For most vessels of over 100 tons displacement, it is
FIG. 2.2 Propeller operating regimes: (A) noncavitating; (B) partially
possible to design propellers on whose blades it is possible cavitating; and (C) supercavitating.
to control, although not eliminate, the effects of cavitation
in terms of its erosive effect on the blade material, its ability
to impair hydrodynamic performance as well as being the susceptible to cavitation, typified by a low cavitation
source of vibration excitation and noise. In this latter context, number, there comes a point when even supercavitating pro-
it must be recognized that very few propellers are free from pellers will not perform satisfactorily. For example, if the
cavitation during operation, since the majority will expe- propeller tip immersion becomes so small that the propeller
draws air from the surface via some convenient path such as
rience cavitation at some position in the propeller disc. Sub-
marine propellers when operating at depth are generally an along the hull surface or down a shaft bracket: this is termed
exception and the propellers, for example, of towed-array ventilation.
frigates and research vessels, when operating at reduced If the propeller immersion is reduced sufficiently either
speed, are also notable exceptions. Indeed, the propellers from design considerations or operational constraints, the
for these types of ship are normally designed to have high- propeller tips will break the surface. This condition is well
cavitation inception speeds to meet their stringent noise known on cargo vessels when operating in ballast conditions
emission requirements and to minimize the probability of and, indeed, may incur certain disadvantages from the
points of view of material fatigue and induced vibration.
either detection or interference with their own instruments.
For some small, high-speed vessels where both the However, the surface breaking concept can be an effective
propeller advance and rotational speeds are high and the means of propelling relatively small high-speed craft. Such
immersion low, a point is reached where it becomes dif- propellers are termed surface piercing propellers and their
ficult, if not impossible, to control the effects of cavitation design immersion, when measured from the free surface
acceptably within the propeller design requirements. To to the shaft center line, can be reduced to zero: that is, the
address this problem, all or some of the blade sections propeller operates half in and half out of the water. Under
may be permitted to fully cavitate in such a way that the these partially immersed conditions, the propeller blades
are commonly designed to operate such that the pressure
cavity developed over the blade sections extends beyond
the trailing edge and, consequently, collapses into the pro- face of the blade remains fully wetted and the suction side
peller slipstream. Such propellers are termed super- is fully ventilated, or dry, where the blade surface suction
cavitating propellers and frequently find application on pressure is at atmospheric conditions.
high-speed naval and pleasure craft. Fig. 2.2C illustrates
schematically this design philosophy in contrast to noncavi-
tating and partially cavitating propeller sections, shown in
2.2 DUCTED PROPELLERS
Fig. 2.2A and B, respectively. Ducted propellers, as their name implies, generally com-
When the design requirements dictate a specific hydro- prise two principal components. The first is an annular duct
dynamic loading in association with an environment having an aerofoil cross-section, which most commonly has
14 Marine Propellers and Propulsion
a uniform shape around the duct and, therefore, is sym-
metric with respect to the shaft center line. Alternatively,
the duct can have certain asymmetric features to try and
accommodate the wake field flow variations in an attempt
to optimize its performance; however, due to the cost of
manufacturing wake adapted axisymmetric ducts, the sym-
metric ducts predominate. The second component, the pro-
peller, is a special case of a nonducted propeller in which the
design of the blades has been modified. These modifications
relate to the radial loading distribution of the blades as well
as accounting for the flow interactions caused by the
presence of the duct in the flow field. The propellers for
these units can be of either the fixed or controllable pitch
type and in some special applications, such as torpedo pro-
pulsion, may take the form of a contrarotating pair of
propellers.
Ducted propellers, sometimes referred to as Kort nozzles
by way of recognition of the Kort Propulsion Company’s
initial patents and long association with this type of pro-
peller, have found application for many years where high
thrust at low or zero ship speed is required: typically, in
towing and trawling situations. In these situations, the duct
generally contributes some 50% of the propulsor’s total
thrust at zero ship speed: termed the bollard pull condition.
However, this relatively large contribution by the duct falls
to more modest proportions with increasing ship speed
toward the free running condition and, beyond this it is also
possible for a duct to give a negative contribution to the pro-
pulsor thrust at very high-advance speeds. This latter situ-
ation would nevertheless be a most unusual design
condition to encounter.
There are two nominal types of duct form, the acceler-
ating and decelerating duct and these are shown in
Fig. 2.3A and D, respectively. The underlying reason for
this somewhat artificial designation can be appreciated, in
global terms, by considering their general duct outline in
FIG. 2.3 Duct types: (A) accelerating duct; (B) “pull-push” duct; (C) Hannan slotted duct; and (D) decelerating duct.
relation to the continuity equation of fluid mechanics. This
can be expressed for incompressible flow in a closed conduit
between two stations a-a and b-b as,
rAa Va ¼ rAb Vb (2.1)
where Va is the velocity at station a-a; Vb is the velocity at
station b-b; Aa is the cross-section area at station a-a; Ab is
the cross-section area at station b-b; and r is the density of
the fluid.
In this context station b-b can be chosen in way of the
propeller disc, while a-a is some way forward, although
not necessarily at the leading edge. In the case of
Fig. 2.3A, which shows the accelerating duct, it can be seen
that Aa is greater than Ab since the internal diameter of the
duct is greater at station a-a. Hence, from Eq. (2.1) and since
water is incompressible, Va must be less than Vb, which
implies an acceleration of the water between stations a-a
and b-b; that is, up to the propeller location. The converse
situation is true in the case of the decelerating duct shown
in Fig. 2.3D. To determine which form the duct actually
is, if indeed this is important, the velocities induced by
the propeller also need to be taken into account together
with the velocity distributions that are developed throughout
the duct.
With the aid of hydrodynamic modeling and analysis, it
is possible to design complex duct forms, which are
intended for specific application and duties. Indeed,
attempts at producing nonsymmetric ducts to suit varying
wake field conditions have been made. These result in a duct
with both varying aerofoil section shape and incidence, rel-
ative to the shaft center line, around the duct circumference.
However, when choosing a duct form for an application, the
hydrodynamic desirability of a particular form must be
balanced against the practical manufacturing problems of
producing the desired shape if an economic, structurally
sound, and competitive duct is to result. This tenet is firmly

underlined by appreciating that ducts have been produced
for a range of propeller diameters from 0.5 m or less up
to around 8.0 m. For these larger sizes, fabrication problems
can be difficult, not least in maintaining the circularity of the
duct and providing reasonable engineering but hydrody-
namically acceptable clearances between the blade tips
and the duct: recognizing that from the hydrodynamic view-
point these clearances should be as small as possible.
Many standard and proprietary duct forms are in use
today, but those most commonly selected are shown in
Fig. 2.3. While the duct shown in Fig. 2.3A, the Wagen-
ingen 19A form, which has a production-friendly shape,
is probably the most widely used because it has a good
ahead performance, its astern performance is not so good
due to the aerofoil form of the duct having to work in
reverse: that is, when operating in the astern mode the
trailing edge effectively becomes the leading edge. This
is of relatively minor importance in, for example, a trawler
or tanker, since for most of their operating lives they are
essentially unidirectional ships. However, this is not true
for all vessels since some, such as harbor tugs, are expected
to have broadly equal capabilities in both directions. In
these cases where a bidirectional capability is required a
duct form of the type illustrated in Fig. 2.3B, the Wagen-
ingen No. 37 form, might be selected. This is because its
trailing edge represents a compromise between a conven-
tional duct trailing edge and the leading edge profile of, for
example, the 19A form. By using this geometric com-
promise, the astern performance is improved but at the
expense of the ahead performance, thereby introducing
an element of compromise in the design process. In
addition, several other methods of overcoming the disad-
vantages of the classical accelerating duct form in astern
operations have been patented over the years. One such
method is the “Hannan Slot,” shown in Fig. 2.3C. This
approach, while attempting to preserve the aerodynamic
form of the duct in the ahead condition, allows water when
backing to enter the duct both in the conventional manner
as well as through the slots at the trailing edge to improve
the astern efficiency of the unit.
When the control of cavitation and more particularly
the noise resulting from cavitation is of importance, use
might be made of the decelerating duct form. A duct form
of this type, Fig. 2.3D, effectively improves the local cav-
itation conditions by slowing the water before it passes
through the propeller as well as enhancing the pressure
environment surrounding the rotating propulsor. Most
applications of this type of duct are found in naval
situations: most commonly associated with submarines.
Nevertheless, some specialist research ships also have
needs, which can be partially satisfied using this type of
duct in the appropriate circumstances.
A further development of the classical ducted propeller
form is found in the pump jet: Fig. 2.4. In the 1960s there
was considerable interest in development of pump jets,
Propulsion Systems Chapter 2 15
FIG. 2.4 Outline of a pump jet with the rotor ahead of stator ring.
particularly for the propulsion of underwater vehicles. This
interest stemmed from the potential to reduce cavitation and
rotor-induced noise while maintaining an acceptable pro-
pulsion efficiency. Particularly in the case of submarines,
the propulsor is immersed in the wake field of the boat’s hull
and, therefore, to achieve the full potential of the propulsor,
as with surface ships, the characteristics of wake field must
be considered. Two basic types of pump jet exist. The first,
Fig. 2.4 comprises a duct supported at its leading and
trailing regions by guide vanes and a stator ring, respec-
tively, with a set of rotor blades mounted on a hub con-
centric with the propulsion shaft in a more central
position. The second type of pump jet dispenses with the
inlet guide vanes and is designed such that the duct is sup-
ported only by the stator ring of blades. In this arrangement,
the stator blades are located forward of the rotor ring. In
general, the rotor and stator blade rings have a high, but
unequal number, of blades to avoid any unwelcome effects
resulting from issues such as blade passing frequency: typ-
ically, rotor and stator blade numbers often lay between 12
and 20 and, for example, the blade numbers might be 9 and
11, respectively. The duct form will be dependent upon the
primary reasons for the use of a pump jet system. If it is to
suppress cavitation, then a decelerating duct form might be
employed so that within the duct flow, the local velocities
are reduced and pressures increased in keeping with
Bernoulli’s general flow equation. The rationale for
deploying such advanced and expensive propulsors is to
be found in the design constraints surrounding the need
for minimizing noise emissions from the unit and the sup-
pression of cavitation. Consequently, the pump jet in this
form, recognizing that certain commercial thruster units
also style themselves as pump jets, is largely restricted to
military applications. For these types of propulsors, the
analysis of the blades and duct can be achieved by a variety
16 Marine Propellers and Propulsion
of means. These range from blade element and momentum
approaches (Lewis, 1996) through to lifting surface, vortex
lattice, and boundary element methods. More recently, more
advanced computational analysis techniques have been
deployed in the analysis of pump jets.
The ducts of ducted propellers, in addition to being fixed
structures rigidly attached to the hull, are in some cases
designed to be steerable. The steerable duct, which dis-
penses with the need for a rudder, is mounted on a system
of pintles whose vertical axis coincides with that of the pro-
peller disc. This allows the duct to be rotated about the pintle
axes, by an inboard steering motor, and consequently the
thrust of the propeller can be directed toward a desired
direction for navigation purposes. Clearly, however, the
arc through which the thrust can be directed is limited by
geometric constraints dictated by interferences between
the duct, propeller, and the hull. Applications of this type
can range from small craft, such as harbor tugs, to compar-
atively large commercial vessels as shown by Fig. 2.5.
A further application of the steerable ducted propeller,
which has gained considerable popularity in recent years,
particularly in the offshore field, is the azimuthing thruster;
in many cases these latter types of units can be trained
through a full 360 degrees.
FIG. 2.5 Steerable ducted propeller.
2.3 PODDED AND AZIMUTHING
PROPULSORS
Azimuthing thrusters have been in common use for many
years and are configured in either nonducted or ducted pro-
peller arrangements. They are further classified into pusher
or tractor units as seen in Fig. 2.6. The essential difference
between azimuthing and podded propellers lies in where the
engine or motor driving the propeller is sited. If the engine
or motor is sited in the ship’s hull, then the system would
be termed an azimuthing propulsor and most commonly
the mechanical drive would be of a Z or L type to the
propeller shaft. Usually, the drive linking the vertical and
horizontal shafts is by means of a system of spiral bevel gears.
In the case of a podded propulsor, the drive system nor-
mally comprises an electric motor directly coupled to the
propeller shaft and which is supported on rolling element
bearing systems: one frequently being a radial bearing
closest to the propeller while the other is a spherical roller
bearing at the opposite end of the shaft line. This latter
bearing system is designed to accommodate both the thrust
reaction as well as maintaining radial position and reaction
loads. In some designs, however, this thrust-and-support
role is separated into two distinct bearing systems. The pro-
pellers associated with these propulsors have been of the
fixed pitch type and are commonly of a built-up design
although their size is not particularly large. Currently, the
largest size of podded propulsor units are around the 23-
MW capacity and the use of podded propulsors has been
mainly in the context of cruise ships and ice breakers where
their maneuvering potentials have been fully realized.
Clearly, however, there are several other ship types, which
might benefit, and have benefitted, from the application of
these propulsors. Fig. 2.7 shows a typical example of a large
podded propulsor unit being maintained in dry dock.
Tractor arrangements of podded and azimuthing pro-
pulsors in a twin screw arrangement generally have an
improved inflow velocity field. This is because they do
not have conventional shafting and A-bracket systems
ahead of them to cause a disturbance to the inflow. This
tends to help suppress the blade rate harmonic pressures
FIG. 2.6 Pusher and tractor thruster units.

FIG. 2.7 Typical podded propulsor unit.
since the relatively undisturbed wake field close to zero
azimuthing angles is more conducive to maintaining low
rates of growth and collapse of cavitation activity. However,
there can be a tendency for these propellers to exhibit
broadband excitation characteristics and during the design
process care should be exercised to endeavor to minimize
these effects. Nevertheless, at high-azimuthing angles, the
flow field can be more disturbed than with conventional
arrangements. In the alternative case of pusher units, where
the propulsor operates behind the vertical strut holding the
shafting pod, the propeller will operate in the wake field of
the strut and some propulsor-strut interaction can be
expected. Similarly, in the tractor arrangement, some back
pressure from the strut may be encountered by the flow field
emanating from the propeller.
Azimuthing or podded propulsors offer significant
maneuverability advantages; however, when used in com-
binations of two or more care should be exercised in
minimizing operation conditions where the propulsors can
mutually interfere with each other. If this occurs, large fluc-
tuating forces and moments can be induced on the shaft
system and significant vibration can be encountered.
2.4 CONTRAROTATING PROPELLERS
The contrarotating propeller principle comprising two
coaxial propellers, sited one behind the other and rotating
in opposite directions, has traditionally been associated with
the propulsion of aircraft. However, Ericsson’s original pro-
posal of 1836, Fig. 1.7, used this method as did de Bay’s
design, in 1876, for the steam yacht Iolair and shown in
Fig. 1.11.
Contrarotating propulsion systems have the hydrody-
namic advantage of recovering part of the slipstream rota-
tional energy, which would otherwise be lost when using a
conventional single-screw system. In marine applications
of contrarotating propulsion, it is normal for the aftermost
propeller to have a smaller diameter than the forward
Propulsion Systems Chapter 2 17
FIG. 2.8 Contrarotating torpedo propeller.
propeller and, in this way, accommodate the slipstream con-
traction effects when the flow passes through the forward
propeller. Similarly, the blade numbers of the forward and
aft propellers are usually different; typically, four and five
for the forward and aft propellers, respectively. Furthermore,
because of the two-propeller configuration, contrarotating
propellers possess a capability for balancing the torque
reaction from the propulsor: this being an important control
stability matter for torpedo and other similar propulsion
problems; Fig. 2.8.
Contrarotating propeller systems have throughout the
development of propulsion technology been the subject of
considerable theoretical and experimental research. While
there are many applications of contrarotating propellers,
particularly in relation to small high-speed outboard units,
operating for example at around 1500–2000 rpm, the
mechanical problems associated with coaxial contrarotating
shafts in the longer line shafting systems of larger com-
mercial vessels have generally precluded them from use
on merchant ships. Interest in the concept has over the years
had a cyclic nature: interest growing and then waning for a
variety of reasons. Nevertheless, an upsurge in interest in
1988 resulted in a system being fitted to a 37,000-dwt bulk
carrier (IHI, 1993) and subsequently to a 258,000-dwt
VLCC in 1993.
More recently, however, an interesting variant of the tra-
ditional contrarotating propulsor arrangement has been pro-
posed and successfully fitted to some ships. This comprises
a combination of a traditional propeller, driven from a
conventional line shaft, as the forward member of the con-
trarotating pair, together with a podded propulsor acting as
the astern component of the configuration: Fig. 2.9 shows
this propulsion concept. Such an arrangement also has the
further potential benefit of dispensing with the rudder since
the azimuthing podded propulsor of the pair provides this
feature. Clearly, for these arrangements when at high-
podded propulsor azimuth angles, it will be seen that the
after propulsor of the pair operates obliquely in the helical
flow field generated by the forward propulsor.
18 Marine Propellers and Propulsion
FIG. 2.9 Contrarotating pair comprising a conventional and podded pro-
pulsor. (Courtesy: ABB.)
2.5 OVERLAPPING PROPELLERS
This arrangement of propulsors is again a two-propeller
concept. However, in this case the propellers are not
mounted coaxially, but are each located on separate shaft
systems with the distance between the shaft center lines
being less than the diameter of the propellers. Fig. 2.10 illus-
trates a typical arrangement of such a system. Again, this is
not a recent idea and references may be found dating back
nearly 200 years: for example, Fig. 1.12 showing Taylor’s
design of 1830.
FIG. 2.10 Overlapping propellers.
As in the case of the contrarotating propeller principle,
recent work on this concept has largely been confined to
research and development exercises and the system has
rarely been used in practice. These research activities have
largely focused on the shaft spacing to propeller diameter
ratio and its effects on the overall propulsion efficiency in
the context of particular hull forms (Kerlen et al., 1970;
Restad et al., 1973). This is because the principal aim of this
type of propulsion arrangement is to gain as much pro-
pulsion benefit as possible from the low-velocity regions
of the wake field and, thereby, increase the ship’s overall
propulsion efficiency. Consequently, the benefits derived
from this propulsion concept are intimately related to the
propeller and hull propulsion coefficients.
Despite one propeller working partially in the wake of
the other, cavitation problems are not currently thought to
pose insurmountable design problems. However, significant
increases in the levels of fluctuating thrust and torque have
been identified when compared to single-screw appli-
cations. In comparison to the conventional twin-screw alter-
native, research has suggested that the overlapping propeller
arrangement may be associated with lower building costs
and this is portrayed as a further advantage for the concept.
When designing this type of propulsion system, several
additional variables are presented to the designer for reso-
lution. These are the direction of propeller rotation; the dis-
tance between the shafts; the longitudinal clearance between
the propellers and the stern shape. At present, there are only
partial answers to these questions; however, it has been
suggested that the best direction for rotation is outward rel-
ative to the top dead center position and that the optimum
distance between the shafts is less than 0.8D. Additionally,
there are indications that the principal effect of the longitu-
dinal spacing of the propellers is to be found in vibration
excitation and that propulsion efficiency is comparatively
insensitive to this variable.
2.6 TANDEM PROPELLERS
Tandem propeller arrangements are not a new propulsion
concept. Perhaps the best-known example is that of Parson’s
Turbinia where eventually three propellers were mounted
on each of the three propulsion shafts to overcome the
effects of cavitation-induced thrust breakdown: Fig. 2.11.
Indeed, the principal reason for the employment of tandem
propellers has been to ease difficult propeller loading situ-
ations; however, these occasions have been relatively few.
The disadvantage of the tandem propeller arrangement
when applied to conventional single- and twin-screw ships
is that the weights and axial spacing of the propellers along
the shafts may create large bending moments, which must
be reacted, principally by the stern tube or A and P bracket
bearings.

FIG. 2.11 Tandem propeller arrangement on a shaft line of Turbinia.
Nevertheless, some azimuthing and podded propulsor
arrangements employ a tandem arrangement by having a
propeller located at each end of the propulsion shaft, located
either side of the pod body. In this way, the load is shared by
the tractor and pusher propellers and the weight-induced
shaft moments controlled within the bearings at either end
of the pod body.
2.7 CONTROLLABLE PITCH PROPELLERS
Unlike fixed pitch propellers whose only operational var-
iable is rotational speed, the controllable pitch propeller pro-
vides an extra degree of freedom in its ability to change
blade pitch. However, for some propulsion applications,
particularly those involving shaft-driven generators, the
shaft speed may be held constant, thus reducing the number
of operating variables again to one. While this latter
arrangement is very convenient for electrical power gener-
ation, it can cause difficulties in terms of the cavitation char-
acteristics of the propeller by inducing back and face
cavitation at different propulsion conditions.
The controllable pitch propeller has found application in
most of the propeller types and arrangements so far outlined
in this chapter, with the possible exception of the podded
propulsors, contrarotating and tandem propellers. However,
even in these extreme examples of mechanical complexity,
some development work has been undertaken for certain
specialist propulsion problems. Since the mid-1950s, the
controllable pitch propeller has grown in popularity from
representing a small proportion of the propellers produced
to its current position of having a substantial market share.
Currently, the controllable pitch propeller has about a 35%
market share when compared to fixed pitch propulsion
systems and tends to be most popular in the ferry, general
cargo, tug and trawling markets.
The controllable pitch propeller, although of neces-
sity possessing greater mechanical complexity than the
fixed pitch alternative, does possess several important
Propulsion Systems Chapter 2 19
advantages. Clearly, maneuvering is one such advantage
where fine thrust control can be achieved through pitch
adjustment and in cases where constant speed machinery
is being used and operational changes can be achieved
without the need to accelerate and decelerate the propulsion
machinery system. Furthermore, fine control of thrust is par-
ticularly important in certain other operational scenarios:
for example, in dynamic positioning situations or where
frequent berthing maneuvers are required, such as in short
sea route ferry operations.
A further additional feature of the basic controllable
pitch principle is that the propeller hub design may in some
instances be modified to accommodate the feathering of the
propeller blades. In this context, the feathering position is
the blade angular position where the blades are aligned
approximately fore and aft so that they present least resis-
tance to forward motion when not rotating. Such arrange-
ments find applications on double-ended ferries or in
small warships, particularly those which have a wide oper-
ational speed spectrum and may employ a hybrid propulsion
plant. In these latter types of application, the vessel could,
for example, have a triple screw arrangement: the two wing
screws being used when cruising with the center screw not
rotating, implying, therefore, that it would benefit from
being feathered to produce minimum resistance to forward
motion in this operating condition. Then, when a sprint
mode of operation is required, all three propellers can be
used at their appropriate pitch settings to develop
maximum speed.
The details and design of controllable pitch propeller
hub mechanisms are outside the scope of this book because
here we are primarily concerned with the hydrodynamic
aspects of ship propulsion. It will, therefore, suffice to say
that each manufacturer has an individual design of pitch
actuating mechanism and that these designs can be broadly
grouped into two principal types: those with inboard and
those with outboard hydraulic actuation. Fig. 2.12 shows
these principal types of arrangement in schematic form.
For further discussion and development of these matters,
reference can be made to the works of Smith (1983),
Plumb (1987), and Brownlie (1998), who provide introduc-
tions to the mechanical aspects of this subject. Alterna-
tively, the propeller manufacturers’ catalogs frequently
provide sources of outline information about the mechanical
features of their controllable pitch propeller designs.
The hub of the controllable pitch propeller, in addition to
providing a housing for the blade actuation mechanism,
must also be sufficiently strong to withstand the propulsive
forces supplied to and transmitted from the propeller blades.
In general, therefore, controllable pitch propellers tend to
have larger hub diameters than those for equivalent fixed
pitch propellers. Typically, the controllable pitch propeller
hub has a diameter in the range 0.24–0.32D: in contrast,
fixed pitch propeller boss diameters are generally within
20 Marine Propellers and Propulsion
FIG. 2.12 Controllable pitch propeller schematic operating systems: (A) pull-push rod system and (B) hub piston system.
the range 0.16–0.25D. However, for some applications,
such as ice class units, controllable pitch hub diameters
may rise to as high as 0.4 or even 0.5D. While large boss
diameters may give rise to some complex hydrodynamic
problems, often cavitation related, for the majority of appli-
cations the larger diameter of the controllable pitch pro-
peller hub does not pose problems that cannot be either
directly or indirectly solved by known design practices.
Certain specialist types of controllable pitch propeller
have been designed and patented in the past. Two examples
are the self-pitching propeller and the Pinnate propeller,
both of which are modern versions of much earlier designs.
Self-pitching propellers are a modern development of Grif-
fiths’ work in 1849 where the blades are sited on an external
crank, which is pinned to the hub and, thereby, the blades
are free to take up any pitch position. The actual blade pitch
position taken up in service depends on a balance of the
blade force and spindle torque loading components, which
are variables depending on, among other parameters, pro-
peller rotational speed: at zero shaft speed but with a finite
ship speed the blades are designed to feather. At the present
time, these propellers have only been used on relatively
small craft. In contrast, the Pinnate design is to some extent
a controllable pitch-fixed pitch propeller hybrid. It has a
blade activation mechanism, which allows the blades to
change pitch about a mean position by varying angular
amounts during one revolution of the propeller. The purpose
of this concept is to reduce both the magnitude of the blade
cyclical forces and the extent of cavitation by attempting to
adjust the angular positions of the blades to accommodate
the varying inflow velocity conditions that are encountered
around the propeller disc. Trials of these types of propeller
have been undertaken on small naval craft and Simonsson
(1983) describes these applications.

2.8 SURFACE PIERCING PROPELLERS
Surface piercing propellers, Fig. 2.1D, are sometimes
referred to as ventilated or partially submerged propellers
and are normally used in special and particularly chal-
lenging cases of high-speed propulsion. With these types
of propellers and with the ship at rest, the propeller is
usually, although not in all cases, fully submerged but close
to the free surface. However, as the vessel accelerates to a
high-cruising speed and the hull starts to plane, the propeller
takes up a partially submerged attitude. When this happens,
the amount of partial submergence is a variable depending
upon the design but is frequently in the region of 0.5D: that
is operating with the propeller half-in and half-out of
the water.
The blade chordal section forms of surface piercing pro-
pellers differ considerably from more conventional aerofoil
types used in conventional subsurface propellers: typically,
the design might adopt a form not dissimilar to that shown in
Fig. 2.2C. During operation in the fully ventilated design
condition, the backs, or suction surfaces, of the propeller
blades should be surrounded by an air film, which extends
to the free surface and only the pressure faces remain
wetted. As such, these types of propellers have specific
design and analysis methods, which are applicable to their
mode of operation to achieve the correct performance and
power absorption characteristics.
Surface piercing propellers, in this context, should be
distinguished from conventional merchant ship propellers,
which are propelling a ship in a light draft condition and,
in so doing, are not fully submerged: the two propellers
and their operating regimes are quite different and should
not be considered in the same context.
2.9 WATERJET PROPULSION
The origin of the waterjet principle can be traced back to
1661, when Toogood and Hayes produced a description
of a ship having a central water channel in which either a
plunger or centrifugal pump was installed to provide the
motive power. In more recent times, waterjet propulsion
has found considerable application on a wide variety of
small high-speed craft while the application of this mode
of propulsion to larger craft has been increasing with the
introduction of tunnel diameters of upwards of 2 m.
The operating principle of present-day waterjets is that
in which water is drawn through a ducting system by an
internal pump, which adds energy and after which the water
is expelled aft at high velocity. The unit’s thrust is primarily
generated because of the momentum increase imparted to
the water. Fig. 2.13 shows, in outline form, the main features
of the waterjet system and this method of propulsion is
further discussed in Chapter 16.
Propulsion Systems Chapter 2 21
FIG. 2.13 Waterjet configuration.
The pump configuration adopted for use with a waterjet
system depends on the specific speed of the pump; specific
speed NS being defined in normal hydraulic terms as,
NQ1=2
NS ¼ (2.2)
H 3=4
where Q is the quantity of fluid discharged, N is the rota-
tional speed, and H is the head.
For low values of specific speed centrifugal pumps are
usually adopted, whereas for intermediate and high values
of NS axial pumps and inducers are normally used, respec-
tively. The prime movers usually associated with these
various pumps are either gas turbines or high-speed diesel
engines.
Waterjet propulsion offers a further dimension to the
range of propulsion alternatives and tends to be used where
other propulsion forms are rejected for some reason: typ-
ically for reasons of efficiency, cavitation extent, noise
or immersion and draft. For example, in the case of a small
vessel traveling at say 45 knots, one might expect that a
conventional propeller would be fully cavitating, whereas
in the corresponding waterjet unit the pump should not
cavitate. However, waterjet propulsors are not always
cavitation free as there are operating conditions where cav-
itation problems may be experienced. Consequently, the
potential for waterjet application, neglecting any small
special purpose craft with particular requirements, is
where conventional, transcavitating and supercavitating
propeller performance is beginning to fall off. Indeed
surface-piercing propellers and waterjet systems are to
some extent competitors for some similar applications.
Waterjet units, however, tend to be heavier than conven-
tional propeller-based systems and, therefore, might be
expected to be deployed on larger craft; for example, in
the case of large wave-piercing catamaran or trimaran
ferries.
In terms of maneuverability, the waterjet system is
potentially very effective, since deflector units are normally
fitted to the jet outlet pipe, which then direct the water flow
and hence introduce turning forces by changing the
direction of the jet flow. Similarly, for stopping maneuvers,
flaps or a “bucket” can be introduced over the jet outlet to
redirect the flow forward and hence apply an effective
reactive retarding force to the vessel.
22 Marine Propellers and Propulsion

2.10 CYCLOIDAL PROPELLERS

Cycloidal propeller development began in the 1920s, ini-
tially with the Kirsten-Boeing and subsequently the
Voith-Schneider designs. As mentioned in Chapter 1, it is
interesting to note that the Kirsten-Boeing design was very
similar in its hydrodynamic action to the horizontal water-
wheel developed by Robert Hooke some two and half cen-
turies earlier in 1681.
The cycloidal or vertical axis propellers basically com-
prise a set of vertically mounted vanes, six or eight in
number, which rotate on a disc mounted in a horizontal or
near-horizontal plane. The vanes are constrained to move
FIG. 2.15 Voith Schneider propulsion unit and guard fitted to a small
about their spindle axis relative to the rotating disc in a pre- ferry.
determined way by a governing mechanical linkage.
Fig. 2.14A shows schematically the Kirsten-Boeing prin-
center point from the center of the disc. By controlling the
ciple. It can be seen from the figure that the vanes’ relative
eccentricity, which in turn governs the vane-pitch angles,
attitude to the circumference of the circle, which governs
both the thrust magnitude and direction can be controlled
their tracking path, is determined by referring the motion
independent of rotational speed. In the case of the Voith-
of the vanes to a particular point on that circumference.
Schneider design, in contrast to the Kirsten-Boeing pro-
As such, it can be deduced that each vane makes half a rev-
peller, the individual vanes make one complete revolution
olution about its own pintle axis during one revolution of the
about their pintle axes for each complete revolution of the
entire propeller disc. The magnitude of the thrust developed
propeller disc. In many cases, the units are provided with
by this propeller design is governed by rotational speed
guards to help protect the propulsor blades from damage
alone and the direction of the resulting thrust by the position
from external sources: Fig. 2.15.
of the reference point on the circumference of the vane-
Vertical axis propellers do have considerable advantages
tracking circle.
when maneuverability or station keeping is a high priority in
In contrast, the design of the Voith-Schneider propeller
the ship’s operational spectrum since the resultant thrust can
is rather more complex since it comprises a series of
be readily directed along any navigational bearing and have
linkages, which enable the individual vane motions to be
variable magnitude. Indeed, this type of propeller avoids the
controlled from points other than on the circumference of
necessity for a separate rudder installation on the ship.
the vane-tracking circle. Fig. 2.14B demonstrates this for
Despite the relative mechanical complexity of these units,
a particular value of the eccentricity (e) of the vane-control
vertical axis propellers have shown themselves to be
reliable in operation over many years of service.

2.11 PADDLE WHEELS

FIG. 2.14 Vertical axis propeller principle: (A) Kirsten-Boeing propeller
and (B) Voith-Schneider propeller.
Paddle propulsion, as is well known, predates screw pro-
pulsion. However, this form of propulsion has today almost
completely disappeared except for very few specialized
applications. These are to be found largely on lakes and
river services, either as tourist or nostalgic attractions as
well as where limited draughts are encountered. Never-
theless, until relatively recently, the Royal Navy also made
use of this type of propulsion on certain classes of harbor tug
since in the configurations used they were exceptionally
maneuverable. The last example of a seagoing paddle
steamer, the P.S. Waverley, which operates coastal and river
services is seen in Fig. 2.16.
The principal reason for the demise of the paddle wheel
as a means of ship propulsion was its intolerance of large
changes of draft and the complementary problem of variable
immersion in seaways. Once superseded by screw

FIG. 2.16 P.S. Waverley: Example of a side wheel paddle steamer.
propulsion for ocean-going vessels, their use was largely
confined through the first half of the 20th century to river
steamers and tugs. Paddle wheels, however, also suffered
from damage caused by flotsam in rivers and were relatively
expensive to produce when compared to the equivalent
fixed pitch propeller.
Paddle wheel design progressed over the years from the
original simple fixed float concept to the feathering float
system, which then featured throughout much of its life.
Fig. 2.17 shows a typical feathering float paddle wheel
design from which the floats attitude relative to the water
surface is controlled from a point just slightly off-center
of the wheel axis. Feathering floats are essential for good
efficiency on relatively small diameter and deeply
immersed wheels. However, on the larger wheels, which
are not so deeply immersed, feathering floats are not
FIG. 2.17 Paddle wheel. Reproduced from Hamilton, F.C., 1948. Famous Paddle Steamers. Marshall, London.
Propulsion Systems Chapter 2 23
essential and fixed float designs were normally adopted.
This led to the practice of adopting feathered wheels in
side-mounted paddle wheel applications, such as were
found on the Rivers Clyde and Thames excursion steamers,
due to the restriction on wheel diameter imposed by the draft
of the vessel. In contrast, with the stern-wheel-propelled
vessels, such as those designed for the Mississippi services,
the use of fixed floats was preferred since wheel diameter
restriction did not apply.
The design of paddle wheels is considerably more
empirical than that of screw propellers today. Nevertheless,
high-propulsion efficiencies were achieved and were of
similar magnitudes to those of the equivalent screw-
propelled steamers. Ideally, each float of a paddle wheel
should enter the water “edgeways” and without shock,
having taken due account of the relative velocity of the float
to the water. The relative velocity in still water has two com-
ponents: the angular speed due to the rotation of the wheel
and the speed of the vessel Va. From Fig. 2.18 it can be seen
that at the point of entry A, a resultant vector a is produced
from the combination of advance speed Va and the rotational
vector oR. This resultant vector represents the absolute
velocity at the point of entry and to avoid shock at this con-
dition, that is a vertical thrusting action of the float, the float
should be aligned parallel to this vector along the line YY.
However, this is not possible practically and the best that
can be achieved is to align the floats to the point B, which
is achieved by the linkage EFG, which is introduced into
the system. Furthermore, from Fig. 2.18 it is obvious that
the less the immersion of the wheel (h), the less is the
advantage to be gained from adopting a feathering float
system. This explains why the fixed float principle is
adopted for large, shallowly immersed wheels.
24 Marine Propellers and Propulsion
FIG. 2.18 Paddle wheel float relative velocities.
With regard to the overall design parameters, based on
experience it was found that the number of fixed floats on
a wheel should be about one for every foot (0.3 m) of
diameter of the wheel and for feathering designs this
number should be reduced to around 60% or 70% of the
fixed float “rule.” The width of the floats used in a design
was of the order of 25%–40% of the float length for feath-
ering designs, but this could be reduced for the fixed float
paddle wheel to between 20% and 25%. A further constraint
on the immersion of the floats was that the peripheral speed
at the top of the floats should not exceed the ship speed and,
in general, feathering floats were immersed in the water up
to about half a float width. In the case of the sternwheelers,
the tops of the floats were never far from the water surface.
The empirical nature of paddle design was recognized as
being unsatisfactory and in the mid-1950s Volpich and
Bridge (Volpich and Bridge, 1954–55, 1955–56, 1956–57)
conducted systematic experiments on paddle wheel perfor-
mance at the Denny towing tank in Dumbarton. Unfortu-
nately, this work came at the end of the time when paddle
wheels were in use as a common form of propulsion and,
therefore, never achieved its full potential.
2.12 MAGNETOHYDRODYNAMIC
PROPULSION
Magnetohydrodynamic propulsion potentially provides a
means of ship propulsion without the aid of either propellers
or paddles. The laws governing magnetohydrodynamic pro-
pulsion were known in the 19th century and apart from a few
isolated experiments, such as those by Faraday when he
attempted to measure the voltage across the Thames induced
by its motion through the earth’s magnetic field and the work
of Hartmann on electromagnetic pumps in 1918, the subject
had largely to wait for engineering development until the 1960s.
The idea of electromagnetic thrusters was first patented
in the United States by Rice in 1961 (Rice, 1961). Following
J B
FIG. 2.19 Principle of magnetohydrodynamic propulsion.
this patent, the United States took a leading role in both the-
oretical and experimental studies, culminating in a report
from the Westinghouse Research Laboratory in 1966. This
report showed that greater magnetic field densities were
required before the idea could become practicable in terms
of providing a realistic alternative for ship propulsion. In the
1970s, superconducting coils enabled further progress to be
made with this concept.
The fundamental principle of electromagnetic pro-
pulsion is based upon the interaction of a magnetic field
B produced by a fixed coil placed inside the ship and an
electric current passed through the sea water from elec-
trodes in the bottom of the ship or across a duct, as shown
diagrammatically in Fig. 2.19. The Lorentz force equation
can be written as,
F ¼ e½J + ðv  BÞ
where J, B, and v are vectors defining, electric and magnetic
fields and the velocity v of the charge carriers, respectively.
It states that a force F is experienced by a charge of mag-
nitude e is equal to the vector sum of that due to the electric
field, eJ, and that due to the motion of the charge through
any additional magnetic field, e(v  B). Here v  B is the
vector cross product of v and B and creates a force vector
normal to the plane containing vectors v and B. In its sim-
plest configuration, the electric and magnetic fields are
arranged orthogonal to each other with the normal to their
plane pointing in the direction of the desired thrust. The
electric field J leads to the transport of charge across the
duct and hence creates the velocity field v necessary to
produce the axial thrust from the cross product of v and B.
The thrust of an MHD drive is proportional to sB2,
where s is the conductivity of the liquid being pumped;
in the case of a ship this being sea water. The necessary
magnetic fields can be large even by modern standards.
For torpedoes with a high-top speed it may be necessary
to create fields in the range 15–20 T; however, for ships
and submarines at typical speeds the magnetic field can
be lower, around 5–10 T. This, however, raises serious con-
cerns around magnetic stealth in relation to warships.
The electric field has certain, detrimental, consequences.
Unlike fresh water, in which hydrogen appears at the
cathode and oxygen at the anode, in seawater trace

FIG. 2.20 Internal magnetic field electromagnetic propulsion unit: (A) the dipole propulsion unit with internal magnetic field and (B) a cross-section
through a prototype propulsion unit.
elements, which typically are ions of sodium, chlorine, mag-
nesium, sulfur, potassium, and calcium gives rise, in
addition to oxygen, to the production of other chemical
species and to chlorine gas at the anode. The high pH at
the cathode leads to scale production, typically of calcium
hydroxide or magnesium hydroxide, which are electrically
insulating. Consequently, over a period of time, the current
can reduce and with it the effectiveness of the drive. Where
copper or aluminum anodes have been used, these can be
severely corroded as CuO2, AlO2, or their chlorides are
produced and hence careful selection of electrode material
is necessary. The electrolysis process produces gases at the
electrodes and these have the effect of blanketing the elec-
trodes, and consequently the cathode is best placed at the
bottom of the duct to encourage hydrogen to rise into the
induced flow and be swept out of the duct.
Iwata et al. (1983) and Iwata (1990) present an inter-
esting description of the state-of-the-art superconducting
propulsion. In theory, the electrical field can be generated
either internally or externally, in the latter case by posi-
tioning a system of electrodes in the bottom of the ship.
This, however, is a relatively inefficient method for ship
propulsion and the environmental impact of the internal
system is considerably reduced due to the containment of
FIG. 2.21 Yamato 1: Experimental magnetohydrodynamic propulsion craft.
Propulsion Systems Chapter 2 25
the electromagnetic fields. Most work, therefore, has con-
centrated on systems using internal magnetic fields and
the principle of this type of system is shown in
Fig. 2.20A in which a duct, through which sea water flows,
is surrounded by superconducting magnetic coils, which are
immersed in a cryostat. Two electrodes are placed inside the
duct, which create the electric field necessary to interact
with the magnetic field to create the Lorentz forces for pro-
pulsion. Nevertheless, the efficiency of a unit is low. The
efficiency, however, is proportional to the square of the
magnetic flux intensity and to the flow speed, which is a
function of ship speed. Consequently, to arrive at a rea-
sonable efficiency, it is necessary to create a strong mag-
netic flux intensity through the use of powerful magnets.
To investigate the potential of these systems at prototype
scale a small craft, Yamato 1, was built for trial purposes by
the Japanese. Fig. 2.20B shows a cross section through one
of the prototype propulsion units, indicating the arrangement
of the six dipole propulsion ducts within the unit. Fig. 2.21
shows the experimental craft, Yamato 1.
Magnetohydrodynamic propulsion does have certain
potential advantages in terms of providing a basis for noise
and vibration-free hydrodynamic propulsion. However, a
major obstacle to the development of this form of propulsion
26 Marine Propellers and Propulsion
until relatively recently was that of the design of the super-
conducting coil and its attendant refrigeration equipment
to maintain its zero-resistance property. However, develop-
ments in superconductivity have in the last few years shown
potential to produce marine propulsion motors using the
high-temperature superconductors. In the United States, a
5-MW demonstrator machine has proved satisfactory and
a 25-MW demonstrator was also constructed as a demon-
strator for marine propulsion purposes.
The relatively small diameter of these machines, if
adopted, may have beneficial implications for podded pro-
pulsors since the hub diameter may then be reduced, given
that this parameter is principally governed by the electric
motor size.
2.13 RIM-DRIVEN PROPULSORS
Although ring propellers, which are a close but very dif-
ferent relation, have been available for many years, rim-
driven propulsors are a relatively new concept. Rim-driven
propulsors combine aspects of electrical propulsion and
ducted propellers. From the electrical point of view, they
can be configured with either induction or dc machines,
however, due to having a greater power density the per-
manent magnet machines tend to be preferred. Regarding
their configuration, in outline, embedded within the duct,
or annular housing, is the motor stator comprising the
windings of the electrical machine. In this propulsion
concept, the propeller blades are rigidly attached at their tips
onto a ring, which rotates in the proximity of and concentri-
cally with the stator: the ring then has a series of magnets
fixed to it, thereby completing the motor configuration.
Yakovlev et al. (2011) have undertaken a numerical and
experimental study of rim-driven propulsors, which resulted
in a design procedure being developed for these types of
propulsor.
As such, the propeller blades do not require a hub,
although this may be beneficial in some high-power cases
to transmit the motive power to the propulsor. Clearly, if
no shaft line is required, then the inflow to the propulsor will
be less disturbed than in conventional ship propulsion. Cur-
rently, these propulsors have been applied in the smaller
power ranges, but development is taking place toward
accommodating higher powers.
2.14 WHALE-TAIL AND FISH
PROPULSION
Throughout the ages, designers have endeavored to mimic
marine mammals and fish in trying to develop enhanced
and more efficient solutions for ship propulsion. When com-
paring conventional ship propulsion to that, for example, of
whales, in the case of the ship the ratio of the propulsor
swept area to that of the ship’s midship section area is
low, while the converse is the case for the marine mammal.
Such an observation suggests relatively light loading of
the tail of the mammal, particularly when swimming at
high speed.
During the late 1980s, work at Glasgow and Memorial
Universities and elsewhere bore testament to this quest to
learn from the marine mammals and fish. Subsequently,
van Manen and van Terwisga (1996) developed a whale-tail
propulsion concept. This was based, in part, on earlier
studies of vertical axis and trichoidal propulsors where they
noted, albeit in a vertical plane, that the blade angular ori-
entations, as the propulsor passed through one complete
revolution, were not dissimilar to the tail movements of a
whale: Fig. 2.14. When the normal vertical axis of these pro-
pulsors was turned through 90 degrees to become a hori-
zontal axis, the motion of the blades became a reasonable
approximation of the action of the movements of a whale’s
tail. This led to the whale-tail propulsor concept, which
comprised a system in which the blades were supported at
both ends, unlike the Voith-Schneider or Kirsten-Boeing
propulsors, and typically was designed to accommodate
between four and seven blades located around the periphery
of the wheel.
Two advantages accrue from this concept. These are,
first, the possibility of large propulsor sizes and flow
regimes over the blades, which are almost two dimensional.
Secondly, since the blades are lightly loaded in terms of
thrust loading per unit envelope area, this implies the pos-
sibility of achieving high efficiency and the development
of a good cavitation environment. Indeed, simplified quasis-
teady computational studies supported this premise in that at
low propulsor loading coefficients and high-equivalent
open water efficiencies, approaching the ideal efficiency
of propulsors, could be a possibility. Subsequently, more
detailed estimates attenuated this potential to some extent,
but the concept was sufficiently attractive to be taken to a
full-scale demonstrator.
More generally, when considering the way in which fish
propel themselves through the water, Videler (1993) made
estimates of their propulsion efficiency. In the case of a
0.30-m cod, it was estimated that at a speed of 0.86 m/s,
equivalent to a Froude Number of 0.5, the propulsion effi-
ciency was about 0.73.
Returning to marine mammals, the humpback whale has
a series of tubercles on the leading edges of its flippers.
These protuberances, having a generally round cross
section, have been found to improve the lift-drag ratio as
well as delaying the stall point of aerofoil-type forms.
Research studies are being undertaken and suggest they
may enhance certain aspects of hydrodynamic performance
of blading, Weichao et al. (2017). However, further work is
required in an attempt to quantify any beneficial effects in
relation to propellers.

2.15 HYBRID PROPULSION
Hybrid propulsion is not new. Variants of the concept can be
traced back to Greek and Roman times and beyond with
various combinations of oar and sail propulsion. In the
19th century with the advent of steam propulsion, hybrid
propulsion manifested itself again in combinations of steam
and sail as seen in Fig. 2.22.
In the 20th century, various combinations within a ship
of diesel, gas, and steam turbines, electric motors, and fixed
and controllable pitch propellers were seen in warship
designs: the machinery fits gave rise to various acronyms
associated with the propulsion plant such as CODAG,
COGAS, COLAG, COGAG, and so on. Again, more
recently some merchant ships, particularly in the context
of cruise ships, have adopted combinations of diesel and
gas turbine machinery to develop economic electrical power
for propulsion, hotel and cargo servicing requirements.
While battery technology had not progressed signifi-
cantly in overall concept until the arrival of lithium-ion
technology, these batteries have permitted a further evo-
lution of the hybrid propulsion concept to evolve in the
marine industry. Two examples of this are a small inter-
island ferry, which uses a combination of diesel propulsion
and Li-ion batteries either in concert or separately for
propulsion. In this example, the batteries may be charged
from the diesel engines or, more commonly, from green
shore power when the ship is not on deployment. Using this
arrangement, significant efficiency and environmental
savings have accrued. A second example, again with a
rather larger ferry, the Prinses Benedikte, which maintains
its diesel engines running at their optimum power-
speed rating and when either less power is required the
FIG. 2.22 Early example of hybrid propulsion.
Propulsion Systems Chapter 2 27
engine recharges the battery banks, or in the converse case
draws power from the batteries. Further promising battery
technology in future years may derive from the Li-air
and Mg-air types, but further research in this area is
necessary.
In the context of reciprocating prime movers, it might be
argued that dual fuel systems fall under the category of
hybrid forms of propulsion. In such systems, typically com-
binations of diesel fuels and LNG are used: commonly
diesel for starting the engine and LNG for the remainder
of the voyage or when in areas of particular environmental
sensitivity. For these systems, the technology is now well
known and rules are in existence from most of the major
classification societies.
Notwithstanding the inboard options for hybrid
machinery combinations and fueling, there are many other
propulsion arrangements that may arise for various opera-
tional scenarios: most commonly for reasons of propulsion
efficiency in complex operational profiles. Typical among
these arrangements are:
l Combinations of fixed and controllable propellers:
sometimes with blade feathering capabilities.
l Arrangements of azimuthing and nonazimuthing
thrusters or podded propulsors.
l Combinations of propellers and retractable thrusters.
l Ship propulsion by means of some form of conventional
propulsors together with auxiliary propulsion assistance
from wind energy or, in certain cases, solar energy
devices.
There is an exceptionally large set of options from which the
designer can choose to meet the desired operational profile
of a vessel. The eventual choice being dependent upon the
propulsion efficiency, the environmental constraints;
internal and external vibration noise requirements, capital
costs, through life maintenance costs, the availability and
the likely reliability for the proposed vessel.
BIBLIOGRAPHY
Brownlie, K., 1998. Controllable Pitch Propellers. I. Mar. EST. ISBN:
1-902536-01-X.
Hamilton, F.C., 1948. Famous Paddle Steamers. Marshall, London.
IHI, 1993. CRP system for large merchant ships. Ship Technology Interna-
tional ‘93. SPG.
Iwata, A., 1990. Superconducting electromagnetic propulsion system. Bull.
Mar. Eng. Soc. 18 (1).
Iwata, A., Tada, E., Saji, Y., 1983. Experimental and theoretical study of
superconducting electromagnetic ship propulsion. Paper No. 2, 5th
Lips Propeller Symposium, May.
Kerlen, H., Esveldt, J., Wereldsman, R., 1970. Propulsion, Cavitation and
Vibration Characteristics of Overlapping Propellers for a Container
Ship. Schiffbautechnische Gesellschaft, Berlin.
Lewis, R.I., 1996. Turbomachinery Performance Analysis. Arnold.
Plumb, C.M., 1987. Warship Propulsion System Selection. I. Mar. EST.
28 Marine Propellers and Propulsion
Restad, K., Volcy, G.G., Garnier, H., Masson, J.C., 1973. Investigation on
free and forced vibrations of an LNG tanker with overlapping propeller
arrangement. Trans. SNAME.
Rice, W.A., 1961. US Patent 2997013 22.
Simonsson, P., 1983. The Pinnate Propeller. Department of Mechanics.
Royal Institute of Technology, Stockholm.
Smith, D.W., 1983. Marine Auxiliary Machinery, sixth ed. Butterworth-
Heinemann, Oxford.
van Manen, J.D., van Terwisga, T., 1996. A new way of simulating whale
tail propulsion. ONR Symposium.
Videler, J.J., 1993. Fish Swimming. Fish and Fisheries Series, vol. 10.
Chapman & Hall.
Volpich, H., Bridge, I.C., 1954–55. Preliminary model experiments. Trans.
IESS 98.
Volpich, H., Bridge, I.C., 1955–56. Systemic model experiments. Trans.
IESS 99.
Volpich, H., Bridge, I.C., 1956–57. Further model experiments and ship
model correlation. Trans. IESS 100.
Weichao, S., Atlar, M., Norman, R., 2017. Humpback whale inspired
design for tidal turbine blades. 5th International Symposium on Marine
Propellers, SMP’17, Espoo, Finland, June.
Yakovlev, A.Y., Sokolov, M.A., Marinich, N.V., 2011. Numerical design
and experimental verification of a RIM-driven thruster. 2nd Interna-
tional Symposium on Marine Propellers, SMP’11, Hamburg,
Germany, June.
Chapter 3
Propeller Geometry
Chapter Outline
3.1 Frames of Reference
3.2 Propeller Reference Lines
3.3 Pitch
3.4 Rake and Skew
3.5 Propeller Outlines and Area
3.6 Propeller Drawing Methods
3.7 Section Geometry and Definition
3.8 Blade Thickness Distribution and Thickness Fraction
To fully appreciate propeller hydrodynamic action from
either an empirical or theoretical standpoint, it is essential
to have a detailed understanding of propeller geometry and
the corresponding definitions used. While each propeller
manufacturer, consultant, or test tank has developed propri-
etary ways of detailing propeller geometric data on drawings
or in dimension books, these differences are most commonly
in matters of presentation rather than in fundamental changes
of definition. Consequently, we will generally not concern
ourselves with a detailed account of each of the different
ways of representing propeller geometric information, rather
we will consider propeller geometry in general terms, which
will then act as an adequate basis for any application with
which the reader will be concerned.
3.1 FRAMES OF REFERENCE
A prerequisite for the discussion of the geometric features of
any object or concept is the definition of a suitable reference
frame. In the case of both propeller geometry and hydrody-
namic analysis, many reference frames are encountered in
the literature, each, no doubt, chosen for some advantage
or preference of the author concerned. However, at the
10th International Towing Tank Conference (ITTC) in
1963, the preparation of a dictionary and nomenclature of
ship hydrodynamic terms was initiated; this work was com-
pleted in 1975 and the compiled version presented in 1978
(ITTC, 1978). The global reference frame proposed by the
ITTC is that shown in Fig. 3.1A, which is a right-handed,
rectangular Cartesian system. The X-axis is coincident with
the shaft axis and positive forward; the Y-axis is positive to
starboard and the Z-axis is positive in the vertically
Marine Propellers and Propulsion. https://doi.org/10.1016/B978-0-08-100366-4.00003-1
© 2019 Elsevier Ltd. All rights reserved.
29 3.9 Blade Interference Limits for Controllable Pitch
29 Propellers 45
30 3.10 Controllable Pitch Propeller Off-Design Section
33 Geometry 45
35 3.11 Miscellaneous Conventional Propeller Geometry
38 Terminology 46
39 Bibliography 46
44
downward direction. This system is adopted as the global
reference frame for this book since little other general
agreement exists in the field of propeller technology. For
propeller geometry, however, it is convenient to define a
local reference frame (x,y,z) having a common axis such that
OX and Ox are coincident, but allowing the mutually per-
pendicular axes Oy and Oz to rotate relative to the OY
and OZ fixed global frame as shown in Fig. 3.1B where f
denotes the angular displacement of Oz from OZ.
3.2 PROPELLER REFERENCE LINES
The propeller blade is defined about a line normal to the
shaft axis called either the “propeller reference line” or
the “directrix”: the word “directrix” being the older term
used for this line. In the case of controllable pitch propellers,
the term “spindle axis” is frequently synonymous with the
reference line or directrix. However, in a few special design
cases, the spindle axis has been defined to lie normally to the
surface of a shallow cone, which is coaxial with the shaft
axis and tapers toward the aft direction. For these arrange-
ments, the spindle axis is inclined to the reference line by a
few degrees. In the majority of cases, therefore, the terms
spindle axis, directrix, and reference line relate to the same
line as can be seen in Fig. 3.2. These lines are frequently, but
not necessarily, defined as being at the origin of the Car-
tesian reference frame identified in the previous section.
The aerofoil sections, which together comprise the blade
of a propeller, are defined on the surface of cylinders whose
axes are concentric with the shaft axis; hence, the term cylin-
drical sections, which is frequently encountered in propeller
technology. Fig. 3.3 shows this cylindrical definition of the
29
30 Marine Propellers and Propulsion

section from which it will be seen that the section lies

obliquely over the surface of the cylinder and thus its nose-
tail line, connecting the leading and trailing edges of the
section, forms a helix over the cylinder. The point A shown
in Fig. 3.2 where this helix intersects the plane defined by the
directrix and the x-axis is of particular interest since it forms
one point, at the radius r of the section considered, on the
“generator line.” Occasionally, however, in some designs,
the nose-tail line lies at a short distance from the point A
and care needs to be exercised in such cases by making the
correct offset transformation. Consequently, for the majority
of cases, the generator line is the locus of all such points
between the tip and root of the blade as can be inferred from
Fig. 3.2. Occasionally the term stacking line is encountered;
this is most frequently used as a synonym for the generator
line; however, there have been instances when the term has
been used by designers to mean the directrix. Therefore, care
is needed for all cases except the special case when the gen-
erator line is the same as the directrix.

3.3 PITCH
Consider a point P lying on the surface of a cylinder of radius
r, which is at some initial point P0 and moves so as to form a
helix over the surface of a cylinder. The equations governing
the motion of the point P over the surface of the cylinder
(points P0, P1, P2, …, Pn) in Fig. 3.4A are as follows:
9
x ¼ f ðf Þ =
y ¼ r sin ðfÞ (3.1)
FIG. 3.1 Reference frames: (A) global reference frame and (B) local ;
reference frame. z ¼ r cos ðfÞ

FIG. 3.2 Blade reference lines.

 Propeller Geometry Chapter 3 31

FIG. 3.3 Cylindrical blade section definition.
where f is the angle of rotation in the Y-Z-plane of a radius
arm r relative to the OZ-axis in the global reference frame.
When the angle f ¼ 360 degrees, or 2p radians, then the
helix, defined by the locus of the points Pn, has completed
one complete revolution of the cylinder and again intersects
the X-Z-plane but at a distance p measured along the OX-axis
from the origin. If the cylinder is now opened out into a two-
dimensional plane, as shown in Fig. 3.4B, we see that the
locus of the point P, as it was rotated through 2p radians
on the surface of the cylinder, lies on a straight line. In the
projection one revolution of the helix around the cylinder,
measured normal to the OX direction, is equal to a distance
2pr. The distance moved forward by the helical line during
this revolution is p and hence the helix angle (y) is given by
 p 
y ¼ tan 1 (3.2)
2pr
The angle y is termed the pitch angle and the distance p
is the pitch. Hence Eq. (3.1), which defines a point on a
helix, can be written as follows:
9
x ¼ rf tan y =
y ¼ r sin ðfÞ (3.1a)
;
z ¼ r cos ðfÞ
There are several pitch definitions that are of importance
in propeller analysis and distinguishing between them is

FIG. 3.4 Definition of pitch: (A) helix definition on a cylinder of radius r and (B) development of helix on the cylinder.
32 Marine Propellers and Propulsion
important if serious analytical mistakes are to be avoided. In
all cases, however, the term pitch when used in propeller
technology refers to the helical progress along a cylindrical
surface rather than, for example, as in mechanical gear
design where pitch refers to the distance between teeth.
The important pitch terms with which the analyst needs to
be thoroughly conversant are as follows:
1. Nose-tail pitch.
2. Face pitch.
3. Effective or “no-lift” pitch.
4. Hydrodynamic pitch.
Fig. 3.5 shows these pitch lines in association with an arbi-
trary aerofoil section profile. The nose-tail pitch line is
today the most commonly used reference line about which
to define blade sections and it is normally defined by a pitch
angle ynt to the thwart-ship direction as seen in Fig. 3.5. The
nose-tail line has a hydrodynamic significance too, since the
section angles of attack are defined relative to it in the con-
ventional aerodynamic sense.
Face pitch is now relatively rarely used by the large pro-
peller manufacturers, but it will frequently be seen on older
propeller drawings and is still used by some smaller manu-
facturers. Furthermore, many of the older model test series,
for example the Wageningen B series, use this pitch ref-
erence as a standard to present the model open water char-
acteristics. However, face pitch has no hydrodynamic
significance: it was a device invented by the manufacturers
to simplify the propeller production process by obviating the
need to hollow-out the surfaces of the propeller mold to
accommodate that part of the section between the nose-tail
and face pitch lines. Consequently, the face pitch line is
basically a tangent to the section’s pressure side surface
FIG. 3.5 Pitch lines.
and, therefore, has a degree of arbitrariness about its defi-
nition since many tangents can be drawn to the aerofoil
pressure surface.
The effective pitch line of the section corresponds to the
conventional aerodynamic no-lift line and is the line which
if the incident water flowed along it, zero lift would result
from the aerofoil section. The effective pitch angle (y0) is
greater than the nose-tail pitch angle by an amount corre-
sponding to the three-dimensional zero lift angle of the
section. As such, this could be considered as the funda-
mental pitch angle since it is the base about which the hydro-
dynamic forces associated with the propeller section are
calculated in classical analysis procedures. Finally, the
hydrodynamic pitch angle (bi) is the angle at which the
incident flow encounters the blade section and is a hydrody-
namic inflow rather than a geometric property of the pro-
peller. Neither this angle nor the effective pitch angle
would be expected to be found on the propeller drawing
under normal circumstances.
From this discussion and by reference to Fig. 3.5, it can
be seen that the three pitch angles, effective, nose-tail, and
hydrodynamic pitch, are related by the equations:
Effective pitch angle ¼ nose-tail pitch angle
+ 3D zero lift angle
¼ hydrodynamic pitch angle
+ angle of attack of section
+ 3D zero lift angle:
The fuller discussion of the effective pitch, hydrodynamic
pitch, and zero lift angles will be left until Chapters 7 and 8.
These terms have been included here to underline the differ-
ences between them and, thereby, prevent confusion.
Another random document with
no related content on Scribd:
bridegroom part of the way to the bride’s apartment. So, lifting the
heavy Mohammed, I carried him a few paces. He was evidently
pleased at my doing him this friendly service, and, the form having
been gone through, sprang quickly down, and, taking me and one of
his other friends each by a hand, began to run. Before us sped a
young man; the rest followed. We were breathless when we reached
the caves.
All was in order. A crowd of spectators began to gather
immediately, and we slipped in through the gates and down the
passage, rapidly crossed into the first court, thence through the
underground passage and out into the other court. This was half-
dark, but from one of the caves shone a light. Here we entered. The
vaulted oblong room with its whitewashed walls was brilliantly lit up.
At the far end a carpet hung right across the room, concealing
something on the ground; in front was spread another carpet. Here
Mohammed seated himself, facing the door. There was no other
furniture visible.
On the bridegroom’s left his friend took a seat, pointing to me to
take my place on the right. There was not the slightest sign on the
features of the former expressive of any emotion, either of gladness
or gravity. To the looker-on he appeared merely phlegmatic, and sat,
wrapped in his cloak, staring into vacancy. His friend, who was also
clothed in red, sat, like himself, in silence.
In the open doorway I saw the faces of Belkassim and Mansur,
also some children, Jews, and the men who had followed us. No
women were present.
When we had been seated thus for a while, there appeared, from
the part of the room divided from us by the hangings, a large dish of
kus-kus and, soon after, a pitcher of water. These were placed
before Mohammed, who took a mouthful of the food—the first meal
prepared for him by his bride.
We sat silent a moment longer, then Belkassim dismissed the
spectators from the door, and I rose, shook hands with my friend the
bridegroom, and left. In the doorway I looked back. There sat the
bridegroom, dumb and stiff, but behind him I saw the carpet being
drawn a little aside, and in the dim light beyond it fancied I caught
sight of a woman’s face. Whether it were pretty, young, or smiling, I
know not. I only know that it must have been the bride’s.
In the open air the festival was in full swing. Closely packed in
front of the gate, and all along the approach, sat veiled women. The
banks and hollows were white with spectators. The negroes danced,
played, and drummed. There a mulatto sang a droll ballad; here two
men danced a stick dance, and so on.
I had been requested to take my place amongst the bridegroom’s
friends, who held themselves in a group apart, prepared to show him
this last day’s homage.
About an hour elapsed, then from the bridal cave a muffled gun-
shot was heard; it was scarcely noticeable, as the shrieking and
booming of the music overpowered all sounds. Ali hastily handed me
a gun, which I discharged, and several shots were fired from our
group. Every one of those present knew what this meant, and
rejoiced, but none more so than the family of the bride. They, who
had waited anxiously, were reassured, for she would not now, under
cover of the silent dark night, and wrapped in a grey blanket, be
hunted at a given word out of the village, and driven home to
sorrowing and disgraced parents and relatives.
The festival was not interrupted, but continued as before.
With the Khalifa by my side, and surrounded by his sons, the
sheikhs, and the principal guests, I remained seated all the evening
and far into the night, watching the entertainment, that in course of
time became very monotonous.
Now and again some men stepped forward, either singly or two
together. Over their shoulders hung red cloaks, and they posed in
graceful attitudes, with their heads held high, one foot forward, and
the left arm hidden beneath the burnous and the red cloak, whilst the
right hand was extended. On each side of them crouched a negro,
with the flaps of his burnous spread out before him to catch the coins
shortly to be thrown to him. Round these figures danced other
negroes, whilst the drums played.
 Now one of the red figures raised an arm and threw a coin into the
negro’s lap, then again, slipping his hands into the folds over his
breast, pulled out another coin. This went on incessantly, that all
might witness how much money was distributed.
First it was the turn of the representatives from Beni Sultan, then
from Zaraua, Tamezred, or other villages, who in this way paid the
tribute expected of them on such festive occasions for the benefit of
the negro musicians.
When at last the men ceased, and the chink of coin was no longer
heard, one of the negroes advanced towards the group of women,
and, half-singing, half-declaiming, told them that the men of such
and such villages had given so much, at the same time praising not
only their generosity, but also their other virtues. Now and then his
song was interrupted by the “Yu, yu” of the women, which this
evening, owing to the number of voices, sounded quite imposing.
When he concluded, the applause was deafening.
Now and then a solo was sung, two or three voices joining in the
chorus that followed, the singers sticking their noses as close
together as they could during the performance. These songs are
always sung in a nasal tone, without any modulation, and the time
never varies.
 CHAPTER XI

Over the Mountains and across the Plain

from Hadeij to Metamer

During the night most of the guests wended their way homewards,
but a few still remained next morning; some of whom desired to
accompany me to Beni Sultan.
The bridegroom was expected to emerge from his cave at any
moment, so I lingered awhile, partly in hopes of bidding him farewell,
and also because I had been told he would be received with
rejoicings, and would distribute sweetmeats amongst the village
children. But the time fixed for my departure came, and I had to ride
off without witnessing this concluding scene of the festival.
Mansur’s mule was brought me. The Khalifa himself arranged my
saddle and lengthened the stirrups, thus showing me the final marks
of courtesy. He then gave the guide his instructions, and I took my
leave with warm expressions of thanks to him and to his sons, and
also to the assembled men. I rewarded little Ali for the services he
had rendered me, bowed respectfully to the Khalifa, and rode off with
my heart full of gratitude towards him and his people for their great
hospitality, and with the pleasant impression that my stay in the
Matmata mountains had given me the opportunity of seeing manners
and customs which, to my knowledge, no European had yet
witnessed in these regions. I thankfully recognised my good fortune
in having had the goodwill and assistance of the authorities; and
was, above all, grateful for the great hospitality of the people from
whom I had then just parted, and for their friendship which I flattered
myself I had gained.
 A mule saddle is very broad, and resembles somewhat a pack. Its
peculiarity is that the stirrup-leathers are not secured to it. A leather
strap with a stirrup hung at each end is slung over the saddle, so
that, to mount, one must either vault into the saddle without setting
foot in the stirrup or be lifted into it. To anyone accustomed to the
ordinary English saddle it is an extremely uncomfortable seat, as it is
necessary to bear equally on both stirrups, or one risks losing one’s
balance and falling off; but I must say my mule proved to be
altogether a success on the difficult mountain road.
For nearly an hour we rode along the mountain top, whence we
had a lovely view; then we descended into a long valley in which
were many half-dead olive trees and green palms. Just as we began
the descent, we met a couple of men on their way to Hadeij from
Beni Sultan to complain to the Khalifa that their sheep had been
stolen.
In the valley was a deep, broad river bed, then dry, and the
mountain sides were furrowed with deep watercourses leading
thereto. In these furrows stood a number of palms surrounded by
embankments.
We halted in a lovely grove of olives, amongst which sprung a few
palms. Here some of our guides awaited us. They had crossed the
mountain by a shorter but precipitous path, whilst we had circled
round by a less steep and fatiguing route.
From an eminence some way down the valley we observed a
village looking like an eagle’s nest. This was the ancient Beni Sultan,
now deserted and in ruins, the present village lying on the incline on
the farther side of the mountain. A few of the houses in this deserted
village were excavated in the ground.
We travelled directly across the valley, and by a very dangerous
and slippery path reached the lowest point of the mountain ridge.
From thence we looked down on the valley on the other side. Facing
us were the ruins of the old village, standing picturesquely against
the sky. We rode down in a zigzag line past the farms and houses
scattered on the mountain side; the dwellings were crowded with
domestic animals, with men and women, and especially with
children.
Not till we reached the valley did we halt, close to the descent to a
cluster of cave dwellings belonging to the Sheikh, in absence of
whom I was most cordially received by one of his nearest relatives.
The passage to the cave was not covered, and was cut into steps
where it sloped down into the ground towards the gate. On one side
of the wall by the steps was dug out a vaulted and somewhat
decorated cave; this was the guest-room where I dined. Afterwards I
visited the nearer of the Sheikh’s houses, with permission to
examine them from top to bottom.
In the main these dwellings were on the same plan as those of
Hadeij, but I found several cisterns in both the farms and the ruins.
Water flowed from the mountains into these through canals and
primitive pipes.
The caves were not all dug down and around a courtyard, but
were often high up on a perpendicular wall, and were reached by
steps.
The women offered me dates and showed me their looms. I saw
where they slept, generally on benches like low tables, called by
them “mokera.”
In one of the underground vaults, to which the access was
through a very heavy gate, was an oil-mill, and in another a granary.
After spending a couple of hours in the shelter of the caves, we
again started riding through the valley in a southerly direction, and
passing through large palm and olive groves. Nowhere in the
mountains had I seen such rich vegetation.
Close to the village were some ten women clad in dark blue,
drawing water from one of the few wells on this mountain. Two large
columns, formed of hewn palm stems, were inserted on either side of
the well, so as to slope inwards. These supported another palm stem
placed horizontally on the top of them; this again sustained a
wooden disk by means of which the water was hauled up. This
system of drawing water is rather comical, for the women, instead of
hauling up the bucket by moving their hands on the rope, seize the
latter and take a quick run, the distance covered being equal to the
depth of the well. When they have thus drawn a pitcherful of water
they return to the well to take another run.
We constantly passed spots in the valley planted more or less
largely with olive trees, but some of these were in an unhealthy
condition, showing grey or yellow instead of a deep fresh green. If
rain were not soon to fall these would die, and it would be many
years before others could be grown and bear fruit.
It cut me to the heart to see all this wealth on the verge of
destruction, and the more so when I learnt that the Khalifa owned
many of the trees. Rain had fallen in many other districts, but none in
this.
Quitting the valley we turned to the right, and rode in a westerly
direction amongst colossal cliffs and into a wild ravine, where we
were surrounded to the north, west, and south by towering rock
pinnacles. Only the very centre of this chasm was reached by the
sun, which, hidden behind the mountain, streamed in glorious
radiance through a rift in the wall of the cliff. On either side of this rift,
with the light playing on their roughly piled grey masses, were the
two villages of Tujan, clinging to the precipitous sides like swallows’
nests to a wall. On one side, high up the mountain, I caught a
glimpse of what appeared to be an eagle’s nest as the sunlight
glanced on it. On inquiring what it was I received the reply that in old,
very old, days the village people resided there, before they moved
lower down the slope.
When we arrived at a difficult pass, my guide, “Erzib ben Hamed,”
who had his home in the village, asked me to dismount. So, leading
our animals, we walked slowly up, our feet slipping, and the stones
rattling down behind us. Beneath some olive trees we again
mounted to make our entry.
We were now near enough to discern that the rift was a deep
ravine; on either side was an irregular mass of dreary, grey houses
piled one over the other, above which the nearly perpendicular cliffs
rose steeply to almost the very top of the mountain, broken only in
one place by a flat surface. On the side nearest to us stood the ruins
of the village of bygone days, perched like a mediæval castle on the
summit of the cliff.
I sat in silent contemplation of this imposing sight, till interrupted
by Hamed, the faithful Hamed, who came up dragging his horse
behind him. He told me to turn round and look at the view of the
Mediterranean.
Great heavens! how glorious was the sight when I raised myself
and looked back. It was so beautiful that even Hamed and the Arabs
were awed by its splendour.
Looking down directly over the slopes, the valley, and the
mountains on either side, we saw the blue sea far away beyond the
plains. In the evening light all the tints of blue, violet, brown, yellow,
and green were softly blended and intermingled as into a veil which
spread over the whole landscape, and imagination divined more than
was actually visible, thus adding to the fascination of the scene.
By the first house, the Sheikh’s, we halted. Some people came
out, one of whom, I suppose, acted as his representative, since he
invited me in; but Hamed was already off his horse and had gone in
to look at the quarters. He returned and announced that they were
very bad, upon which I inquired whether notice had been given of my
coming, and whether the Sheikh had not directed that I should be
given decent accommodation. The spokesman insisted that there
was no other room available. I suspected this to be false, and
ordered Hamed and Erzib to mount their horses at once, and we
rode up a narrow lane and alighted outside Erzib’s dwelling, where
he had already told me I should be welcome. I heard the man
following us, and saw that a number of people had gradually
assembled.
Erzib’s dwelling lay high up on the side of the cliff, but there were
others that were higher still, and yet others below. In front of these
dwellings ran a narrow path, that, starting from the highest farms, led
in a zigzag course down to those below. The outside of the path was
on the edge of a steep declivity, down which all refuse was thrown,
and was therefore dirty; looking, as did all the other banks when
seen from a distance, as though scored with black stripes. On the
slope below us was a house with a courtyard between it and the cliff
wall. This yard, in which a woman was working, was completely
open to view. Thence the ground fell gradually away till it ended in
palm-grown gorges and valleys; beyond these were low hills, then a
plain, and, last of all, the sea.
Below us, and a little to one side, was the principal edifice of the
village—a little mosque, or Marabout’s tomb, outside which a crowd
of men had gathered to perform their evening devotions. Kneeling
almost simultaneously, they kissed the earth and rose again. A few
of them presently disappeared through the open door of the
Marabout’s tomb, but the remainder stayed outside.
Looking upwards, the eyrie on the summit was visible above
Erzib’s house, that is, one could discern it by stretching one’s neck.
On the opposite side, at the end to the left, lay the other village.
In the evening the women sallied forth in numbers to fetch water
from the cisterns in the valley, and the village dogs barked,
answering each other from every side. Below us, at the foot of the
slope, a crowd of men gathered. I could hear their shouts and see
their gesticulations, as, with extended arms, they pointed to one
figure. Some of them turned and called up the bank to us, one of
them being the man who had met me on the Sheikh’s behalf. They
shouted that I should come down and live wherever I pleased with
the other men, and when I replied that I was well installed, they
informed Erzib that fowls, eggs, and bread would shortly be sent,
that the stranger guest might have a really good meal prepared for
him.
Through a very broad gateway I descended into a court. Opposite
was a long house with its own entrance, to the right another
resembling it; and between the two was a passage leading to a third
dwelling that was situated at the back. To the left was a wall.
On the flat roof of the nearest of these houses stood some
enormous rush-bins for corn, and in the courtyard was another.
There also were two fireplaces, one on either side, screened off with
branches. Behind the screen to the left sat a woman laying small
faggots on the fire to warm her hands, for it was cold since the sun
had set. Some children came out of the door, but fled when they
caught sight of me, wrapped as I was in the folds of Erzib’s burnous.
From the door on the left peeped out an elderly and rather nice-
looking woman.
These two were Erzib’s wives: each had her own house; the
children belonged to the woman I saw seated by the hearth.
Erzib told his wives to come forward. This they did quite naturally
and willingly, retiring again after I had shaken hands with them.
Soon after, I saw people arriving with screaming fowls and a
basket of eggs and bread. Erzib at once drew his knife and vanished
with the hens—his intentions were easy to divine.
In the meanwhile carpets had been spread on the floor of the
house, and a couch arranged for me. I lit a candle which had been
placed in a small square recess in the wall. The room was very
irregularly shaped. The floor was of beaten clay, and the walls not
whitewashed. In the background a door led into another room
containing a loom, and where gala dresses hung on a cord, and
household goods on the wall. Through yet another door in the wall to
the right was a room with a bed in it raised on four slight stumps: this
was made of twigs, and had no coverings.
This dwelling was inhabited by the younger wife and her children
—two boys and a little girl. The wife was pretty and not old.
In the house in the courtyard the elder wife resided. In this the
anteroom was larger, and contained household goods and
implements; behind it was the sleeping apartment.
A grown-up married son, then absent, occupied a house tucked
away at the back, and designed on the same plan as the others. His
wife was at home.
Whilst the pile of wood burnt and crackled in the yard and the
women were busy preparing food, I sat on a bank outside the house
in company of my host and several other men.
 The moon had risen and shone clearly over mountain and vale. I
could see down into a courtyard at the foot of the slope, where a fire
burnt brightly on a hearth. Over it hung a cauldron watched by the
housewife. She was young and pretty, and as she moved to and fro
a couple of little children trotted after her. Now and then she stood
still, shading her eyes with her hand, and gazing up in our direction;
possibly in the stillness of the night our voices reached her, for it was
not likely that she could see us. By the hearth a white dog lay and
growled, and when the woman paused and looked up he moved
restlessly, for he also was watching the stranger.
Erzib’s first wife came out and stood leaning against the doorway.
She did not speak, but was evidently interested in our conversation.
Her husband glanced at her and said abruptly—
“She has a great sorrow, and has grieved and wept for many
years. Ali, her only son, who was in service at Gabés, was sent to
prison, accused of having stolen money from the tradesman he
served. But he was innocent—that we know; he was a good boy, and
his mother loved him. It is now four years and four months since we
heard from him, and eight months more must pass before we can
have him home again.”
“Do you not even know if he lives?”
“Yes, we have learnt through strangers that he is alive, and
supposed to be imprisoned at Bona in Algeria.”
The old woman drew herself along the wall till she was close to
me when she heard of what we were talking.
“Are you from Bona?” she asked, whimpering.
“No,” I replied, “I come from a much more distant place, and have
never been in Bona.”
“Ah! then you do not know Ali,” she said, with a sob.
“No, poor woman,” I replied; “that I do not; but now you will soon
see your son alive. You have waited so long for him that the
remaining time will soon pass ere he return to you and be happy with
you again, for you love him. He will have thought so often of you,
and he will be so good to you that both of you will rejoice.”
“Ah! it was a great misfortune, for he was innocent—I am sure of
that; another must have been the culprit, for he was so young.”
“How old was he?”
“That I do not remember.”
“Don’t you know what year he was born?”
“No, I cannot recollect; we never know anything of that.”
“Don’t you know either, Erzib?”
“No, Sidi; but it was before the strangers came to this country.”[2]
The poor woman sobbed audibly, and Erzib pushed her inside the
door that her weeping might not trouble me, saying, “She is very
unhappy, Sidi.”
“Oh yes, Erzib. Would I were able to help you to get back your
son sooner, or at least to procure you tidings of him. But this I can
promise—I will speak to the Khalifa of Gabés on the subject, and, if
possible, send you greeting from your son.”
To my regret, however, I must confess that I was unable later to
do anything for these poor folk. Whether the boy is still in prison I
know not, and whether innocent or no, I know less. My sincere hope
is that he may be worthy of his parents’ touching affection.
The repast was now brought and set out in the house, on the clay
floor, where I enjoyed it; the father, surrounded by his children whom
he caressed, sat aside with Hamed and the younger wife.
When I had finished, and Hamed and Erzib had also eaten, we
remained seated. I talked with the wife about her children. The eldest
may have been about ten years old; he was a lively boy, who nodded
continually to me, and was indefatigable in showing me all the
treasures of his home, from an old musket to his father’s agricultural
implements. When I showed surprise at a very primitive and curious
harrow used to break up the soil, his father gave it to me.
 Next in age to the boy was a very pretty little girl about six years
old. Unfortunately she had lost one eye; her father told me that it was
in consequence of a severe attack of inflammation when she was
quite little, and that the eye had fallen out of itself. Here in the south
one meets with an alarming number of people who are blind or
suffering from eye complaints. A doctor told me that many are born
thus; with others it is the result of dust, heat, and uncleanliness.
The youngest child was a bright little fellow of two, who clung to
his father, whose neck he clasped tightly in his arms.
Feeling disposed to take a stroll before retiring to rest, I bade
Erzib follow me. As we crossed the court, he inquired whether I
would not like to see all the dwellings. Accordingly we went first to
visit the elder lady. When we entered with a light we found her
crouching in a corner, her face buried in her hands; beside her lay a
large dog which growled at me.
Thence we went into the son’s house. Asleep on the bed, quite
dressed,—for the natives never undress at night,—was a woman
wrapped in blue clothing; she was evidently the son’s wife.
We walked on and up amongst other houses till we were nearly at
the top of the village. Beneath, we saw the lights and fires in the
courts, and heard the incessant barking of dogs. Shortly after, we
climbed a difficult ascent just over the village, to a ledge or terrace of
some width cut in the side of the cliff, which from thence rose, quite
straight and steep, to the old deserted village that lay in darkness on
the very summit. According to Erzib, we could not reach it from the
side we were on.
I contented myself with examining some real cliff caves, which I lit
up by means of matches. They were excavated from the terrace,
and, according to tradition, had once been inhabited; they were
irregular in form, and not very large.
After an hour’s enjoyment of the beautiful evening, we descended
from this high point.
Wrapping myself in my burnous I lay down on my couch on the
ground; in the same room lay both Hamed and Erzib. In the side
chamber, of which the door remained open, slept the children and
their mother. Just as I was falling asleep a woman came and spread
a covering over me; it warmed me well, and I slept till daybreak, and
was only once disturbed by a little kid coming in through the open
door leading from the courtyard and tripping over me. I heard then
the children, who with their mother were sleeping in the next room,
Hamed and Erzib moving on their beds, and, out of doors, the distant
and continuous barking of dogs. I slept again, and when I awoke saw
that what had been spread over me was a brand new festal garment
that evidently was considered none too good for the guest.
From the doorway overlooking the courtyard I saw through the
gate and down into the valley, where grew a solitary palm, and at the
same time had a view of the flat roofs of several houses, and of the
path where the horses and mules stood ready saddled. From a side
chamber the head of a cow came peering in at the gate, and above
the gateway a white dog lay on the wall watching me.
I gave some money to the children, ate a couple of dates with a
sup of water, and, having thanked the women for their hospitality,
mounted, with Erzib in front and Hamed behind me. As we left, the
women came out to throw refuse down the slope, and vanished
again behind the wall.
From the hearths rose a light blue smoke that was wafted over the
valley beneath us.
We had a view over the mountains of the valley, the plains, and
the Mediterranean Sea, as we followed the route along the western
declivity of the Matmata range, which commands the low-lying land
that extends right away to Tripoli.
For a while we were accompanied by two women who were on
their way to the mountains. They tripped along beside our horses,
and stared at me in astonishment through carelessly drawn veils.
The mountain tops, where lie the villages of Shenini and Sguimi,
are a continuation of the southern range. As I was aware that the
inhabitants of these villages were absent sowing their crops, and
having been told that the dwellings were similar to those I had
already seen, I decided not to visit them. We therefore left them on
one side and rode down the mountain and across a small plain
encircled by hills, behind which lie the great steppes. Towards the
east this plain is bounded by low hills, where water springs are
found, and where we could descry herds grazing. It was here that,
when passing through a little thicket, we spied a covey of partridges
running amongst the bushes. Erzib tried to fire at them from his
horse, but it would not stand long enough, and when he got off it was
too late—the birds had flown.
Before traversing the last of these hills, we halted and partook of
dates, bread, and water, as many hours would elapse before we
could arrive at any place of habitation.
The ride on the mule had tired me, so I preceded the others on
foot, and reached the farther side of the acclivity. There lies an
interminable flat plain stretching as far as the eye can reach from the
east to the north-west; whilst towards the south the mountains fade
away in long undulations. In the midst of the plain I distinguished a
hill, and on its summit what appeared to be a tower or fortress. This
was the signal station near Metamer. It corresponds to the one we
saw near Gabés, and also to another farther south.
I wandered down the gentle slope, through bushes and among
stones, and crossed the bed of the river, that, coming from the
mountains, winds out into the plain. There were many paths, all
leading in an easterly direction. I followed one of these, crossed yet
another stony torrent bed, and continued steadily towards the east,
making the signal station my point of direction; until, looking round, I
discovered the two riders in their white burnouses far away towards
the south. They beckoned to me, as we were compelled to make a
détour to avoid a rough and uneven river bed.
Joining once more my party, we rode farther and farther over the
plain, which becomes dismally desolate and monotonous; with the
exception of the hill and its signal station, nothing breaks the long
line of the horizon.
At last we viewed in the distance a couple of palm trees, and
concluded that the Ksar of Metamer was probably near them, but we
could not see it at all, as it lay in a hollow.
For long, naught but these trees showed on the level horizon.
Then at last the tops of other palm trees appeared, and a little later
some huts; the number of these increased, and proved to be the
outskirts of the town. The huts—of straw and branches—were round,
as a rule, with a pointed thatch. But it was easy to infer that the
inhabitants were absent, as the network which usually encloses the
verandah that runs round each hut had been removed, and only the
centre of the huts remained, their thatched eaves sticking out all
round, so that they resembled thick mushrooms on short stalks.
As the day advanced, the heat became stifling, so that I took off
my gaiters and bared my legs. But after a couple of hours they were
so scorched by the sun that, on arrival at Medinin, I had to ask a
doctor to dress them for me, to ease the pain of the sun-scorch, and
it was eight days before they recovered.
 CHAPTER XII

Metamer and Medinin

Arriving at the palm grove in the hollow we had seen from the
distance, we found that it lay by a river bed. The trees were not
particularly well cared for, as could be seen at a glance; they were
far apart, and there were few ditches for irrigation.
On a slope to the east of the valley and above it, there is a village
of peculiar construction, with whitewashed buildings that are dazzling
in the sunlight. This is the “Ksar” Metamer. The ground plan of the
houses is oblong and rectangular, and their raised roofs are vaulted.
They lie lengthwise, as the houses do at home in towns dating from
the Middle Ages—the gable ends turning towards the streets. In
general they are erected round an open square. The fronts of those
facing the plain are without any aperture, except some loop-holes
here and there. In other words, every quarter, and also the town as a
whole, forms a little fortress. This is the style of building adopted
here in the plains; it is, in fact, the same plan as that employed in
cave construction, but in this case carried out aboveground; since
the natives have found it impossible to reach the inaccessible
mountain peaks, or to dig down into the rocky ground. The houses
are very often seven storeys high. On every storey there is a well-
barred door to the inner gable. This is reached by steps or by stones
projecting from the walls. The effect is most peculiar and
picturesque. Each inhabitant carries in his hand a key that he takes
with him everywhere. This locks his rooms, which are mostly used as
corn stores.
Not far from the “Ksar” are barracks for the little garrison, and
shops that supply the needs of the soldiers, not only of the place, but
also of those quartered in the neighbouring town of “Medinin.” I did
not wish to visit the camp just then, so dismounted outside an Arab
dwelling, and was invited to enter and partake of stewed kid.
After a hurried visit to the town, and having taken leave of Erzib,
who desired to ride a long way towards his home that evening, I
procured a new guide and rode eastwards over the plain, so as to
arrive before nightfall at the Ksar of Medinin. As we approached its
neighbourhood we turned into the highroad from Gabés.
Before us and to our left lay the “Ksar” of Medinin, illuminated by
the evening sun. The ends of the houses were turned outwards,
producing the effect of a circular wall scalloped at the top. Above
these vaulted gable ends I caught a glimpse of higher buildings, and
amongst them, in the centre of the town, a large square block. This
was the Kasba. Through a narrow opening in the row of houses I
saw the inner gable ends of dwellings, and doors disposed one
above the other, the whole calling to mind the pictures one sees of
Mexican “pueblos.”
 A STREET IN BENI BARKA.

Parts lay in deep shadow, parts blinding white in the sunshine.

These lights and shadows were mingled in such dazzling contrast
that the eye could scarcely discriminate what it beheld.
We rode along the exterior wall till we came to some palms;
farther on grew others. These plantations are to the south of the
Ksar and between it and the European quarter, which showed up
gradually on the right, and consisted of barracks for the cavalry and
infantry, quarters for the officers, and those occupied by the “Bureau
de Renseignement.” The soldiers work amongst the palms, and have
enclosed a plot of ground as a garden. In the beds I saw tender
young green plants sprouting, which proved to be cress. In the open
square in front of headquarters, and before the other houses, holes
were being dug for plants by soldiers in light linen clothing.
In the future the whole military quarter will be surrounded by a
beautiful palm grove, affording shade to the dwellings now
completely exposed to the glare of the sun.
I rode up to headquarters—a large building—where the flag was
hoisted half-mast high on account of the death of Marshal
MacMahon.
Lieutenant Henry, who was at the Bureau, came out to welcome
me. He told me that I was expected, and added that I should meet
the officers of the 4th Light Brigade, whom I had known well at
Gabés, they having arrived to relieve the southern station. I was
quickly conducted to real bachelor’s quarters, consisting of a couple
of rooms. All over the walls hung weapons and curiosities collected
in these regions. The furniture, though camp-like, was very
comfortable. At last I was able to indulge in the luxury of a bath and
change.
In the meantime Hamed arrived to say farewell. He wished to ride
back to Metamer on his donkey and accompany Erzib as far as
Tujan, whence he hoped to take the donkey back to Hadeij, and
return later to Gabés.
When I was dressed I called on the Commander-in-Chief of the
district, Commandant Billet, a young man, who invited me to be his
guest.
When I told him that I was most anxious to meet some Tuareg if
possible, he replied, to my great joy, that by riding some thirty-two
miles farther south I should probably have my wish gratified, as a
telegram had just arrived from the signal station that two of these