Polish Maritime Research

Address of Publisher
& Editor's Ofce :

GDASK UNIVERSITY
OF TECHNOLOGY
Faculty
of Ocean Engineering
& Ship Technology
ul. Narutowicza 11/12
80-952 Gdask, POLAND
tel.: +48 58 347 17 93
fax : +48 58 341 47 12
e-mail : [email protected]
Account number :
BANK ZACHODNI WBK S.A.
I Oddzia w Gdasku
41 1090 1098 0000 0000 0901 5569
Editorial Staff :
Witold Kirkor Editor in Chief
Przemysaw Wierzchowski Scientic Editor
Maciej Pawowski Editor for review matters
Tadeusz Borzcki Editor for international relations
Cezary Spigarski Computer Design
Domestic price :
single issue : 20 z
Prices for abroad :
single issue :
- in Europe EURO 15
- overseas US$ 20
ISSN 1233-2585
NAVAL ARCHITECTURE
3 TADEUSZ KORONOWICZ,
ZBIGNIEW KRZEMIANOWSKI
Investigations of inuence of screw
propeller operation on water ow
around stern part of ship hull
10 TOMASZ TABACZEK, JAN KULCZYK,
MACIEJ ZAWILAK
Analysis of hull resistance
of pushed barges in shallow water
OPERATION & ECONOMY
16 LECH MURAWSKI, MAREK SZMYT
Stiffness characteristics
and thermal deformations of the frame
of high power marine engine
23 ZYGMUNT GRSKI,
ROMUALD CWILEWICZ
Usefulness assessment of standard measuring
instruments installed on sea-going ships
to perform energy measurements
28 TADEUSZ SZELANGIEWICZ,
KATARZYNA ELAZNY
Calculation of the mean long-term service speed
of transport ship. Part II - Service speed
of ship sailing on regular shipping route
in real weather conditions
POLISH
MARITIME
RESEARCH
in internet
www.bg.pg.gda.pl/pmr.html
PUBLISHER :
CONTENTS
POLISH MARITIME RESEARCH
No 1(51) 2007 Vol 14
The papers published in this issue have been reviewed by :
Prof. A. Charchalis ; Prof. J. Kolenda
Assoc. Prof. M. Pawowski ; Prof. J. Szantyr
P
h
o
t
o

:

C
.

S
p
i
g
a
r
s
k
i
POLISH MARITIME RESEARCH is a scientic journal of worldwide circulation. The journal appears
as a quarterly four times a year. The rst issue of it was published in September 1994. Its main aim is to
present original, innovative scientic ideas and Research & Development achievements in the eld of :
Engineering, Computing & Technology, Mechanical Engineering,
which could nd applications in the broad domain of maritime economy. Hence there are published papers
which concern methods of the designing, manufacturing and operating processes of such technical objects
and devices as : ships, port equipment, ocean engineering units, underwater vehicles and equipment as
well as harbour facilities, with accounting for marine environment protection.
The Editors of POLISH MARITIME RESEARCH make also efforts to present problems dealing with
education of engineers and scientic and teaching personnel. As a rule, the basic papers are supplemented
by information on conferences , important scientic events as well as cooperation in carrying out interna-
tional scientic research projects.
Editorial
Scientic Board
Chairman : Prof. JERZY GIRTLER - Gdask University of Technology, Poland
Vice-chairman : Prof. ANTONI JANKOWSKI - Institute of Aeronautics, Poland
Vice-chairman : Prof. MIROSAW L. WYSZYSKI - University of Birmingham, United Kingdom
Dr POUL ANDERSEN
Technical University
of Denmark
Denmark
Dr MEHMET ATLAR
University of Newcastle
United Kingdom
Prof. GRAN BARK
Chalmers University
of Technology
Sweden
Prof. SERGEY BARSUKOV
Army Institute of Odessa
Ukraine
Prof. MUSTAFA BAYHAN
Sleyman Demirel University
Turkey
Prof. MAREK DZIDA
Gdask University
of Technology
Poland
Prof. ODD M. FALTINSEN
Norwegian University
of Science and Technology
Norway
Prof. PATRICK V. FARRELL
University of Wisconsin
Madison, WI
USA
Prof. WOLFGANG FRICKE
Hamburg-Harburg
Germany
Prof. STANISAW GUCMA
Maritime University of Szczecin
Poland
Prof. ANTONI ISKRA
Pozna University
of Technology
Poland
Prof. JAN KICISKI
Institute of Fluid-Flow Machinery
of PASci
Poland
Prof. ZYGMUNT KITOWSKI
Naval University
Poland
Prof. JAN KULCZYK
Wrocaw University of Technology
Poland
Prof. NICOS LADOMMATOS
University College London
United Kingdom

Prof. JZEF LISOWSKI
Gdynia Maritime University
Poland
Prof. JERZY MATUSIAK
Helsinki University
of Technology
Finland
Prof. EUGEN NEGRUS
University of Bucharest
Romania
Prof. YASUHIKO OHTA
Nagoya Institute of Technology
Japan
Prof. ANTONI K. OPPENHEIM
University of California
Berkeley, CA
USA
Prof. KRZYSZTOF ROSOCHOWICZ
Gdask University
of Technology
Poland
Dr YOSHIO SATO
National Trafc Safety
and Environment Laboratory
Japan
Prof. KLAUS SCHIER
University of Applied Sciences
Germany
Prof. FREDERICK STERN
University of Iowa,
IA, USA
Prof. JZEF SZALA
Bydgoszcz University
of Technology and Agriculture
Poland
Prof. TADEUSZ SZELANGIEWICZ
of Szczecin
Poland
Prof. WITALIJ SZCZAGIN
State Technical University
of Kaliningrad
Russia
Prof. BORIS TIKHOMIROV
State Marine University
of St. Petersburg
Russia
Prof. DRACOS VASSALOS
University of Glasgow
and Strathclyde
United Kingdom
3
POLISH MARITIME RESEARCH, No 1/2007
INTRODUCTION
In the 1990s in Ship Propeller Division, Institute of Fluid-
-Flow Machinery, Polish Academy of Sciences (IMP PAN), the
computer model basin PANSHIP was elaborated. It has been
aimed at simulation of ship hull model tests in ship model basin
as well as relevant calculations for full-scale ship hull.
The computer model basin is a computer software system
consisted of a dozen or so mutually cooperating programs
[16]. Crucial elements of the system are the programs capable
of taking into account the inuence of screwpropeller operation
on ow around ship hull. The software contains the programs
with the use of which a change of hull resistance resulting
from propeller suction action can be determined, and those by
which the inuence of propeller operation on velocity eld in
behind-the-hull ow can be taken into account.
The initial calculations performed by using the PANSHIP
software have yielded generally correct results with the excep-
tion of one element : changes of hull resistance resulting from
propeller operation. In ship theory such change is expressed
in the form of the so-called thrust deduction t :
t = (R
T
R
o
)/T = (T R
o
)/T
where :
R
o
resistance of hull without propeller
T propeller thrust
R
T
resistance of hull with operating propeller (identied
with propeller thrust)
The quantity t is usually determined during every ship model
propulsion tests in model basin.
As such tests have been performed every year for many
ship models, a very rich collection of experimental data in this
Investigations of inuence
of screw propeller operation on water ow
around stern part of ship hull
Tadeusz Koronowicz
Zbigniew Krzemianowski
Institute of Fluid-Flow Machinery,
Polish Academy of Sciences
in Gdask
ABSTRACT
This paper presents results of measurements of velocity eld in before- the - propeller ow in presence of
a ship model hull of two congurations, as well as comparative calculations of velocity eld on a full-scale
ship. Analysis of the research results showed that input data to Biot-Savart formula should be modied in
the case of calculations of propeller-induced velocities on ship hull surface.
Keywords : ship hydromechanics, propeller-induced velocities, Biot-Savart equation.
domain has been gathered. Basing on them one can unambigu-
ously state : the more full form of a ship the greater value of
its thrust deduction t.
In the preliminary version of the computer model basin
in question, for hulls of more full forms, greater and greater
differences between calculated values of thrust deduction and
those experimentally determined for the same hulls, were
obtained (Fig.1). Due to prior research on owaround propel-
ler it was possible to diagnose that the velocities induced by
whirls representing the propeller, determined by means of
the Biot-Savart formula, obtained erroneous values on the
hull surface. In Fig.2 it can be observed that the more full form
of a hull the smaller values of the induced velocities calculated
from the original Biot-Savart formula, therefore the calculated
pressures (under-pressures) on the hull surface take also smaller
values.
Fig. 1. A simplied diagram of the relation between the thrust deduction
and hull block coefcient (in reality the relation is more complex
as it depends rst of all on fullness of stern part of hull) .
By analyzing the diagrams presented in Fig.2 and 3
it can be explained why such results have been obtained.
0
0.05
0.1
0.15
0.2
0.25
0.3
0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9
Test
Original calculation
Calculation with a modified
Biot-Savart equation
t
4 POLISH MARITIME RESEARCH, No 1/2007

Fig. 2. Distribution of axial component of velocity in before-the-propeller
ow, obtained from the original Biot-Savart formula, shown on the backgro-
und of frame sections of hulls having various block coefcients .
If the velocities induced by the whirl system which repre-
sents propeller blades and propeller race, are calculated from
the original Biot-Savart formula then the velocity distribution
starts at the hull plane of symmetry (the point C in Fig.3).
The similar velocity distribution is presented in Fig.2, where
simultaneously the frame sections of 3 ship hulls of different
values of the block coefcient are shown. It can be observed
that the greater fullness of the hull the smaller obtained values
of velocities induced on its surface.
Fig. 3. Schematic presentation of the modication
of data input to the Biot-Savart formula.
In the up-to-date version of the PANSHIP, was implemented
a new method of calculation of induced velocities by means
of Biot-Savart formula, (called the engineering method). The
propeller-induced velocities were calculated in the point C
(Fig.3), but they were considered as the velocities calculated
on the hull (the point C on the hull). It means that zero-value of
the coordinate perpendicular to the hull plane of symmetry was
put in the Biot-Savart formula. The calculation results appeared
signicantly better. Values of the thrust deduction obtained from
calculations and those from experiments became more and
more similar to each other. Obviously the described method of
determination of induced velocities is only approximate, how-
ever, as revealed from practice, it yielded satisfactory results
in engineering applications without any special modications
of the software.
For many years the so-modied computer software PAN-
SHIP has been in use, and the hypothesis associated with
the modication of input data for Biot-Savart formula was
conrmed by comparing calculation results with experimental
ones. However it was necessary to test the hypothesis by means
of direct measurements of the eld of the propeller - induced
velocities around the hull. Such a verication is the subject of
the presented work.
The model tests were performed at the Ship Hydromecha-
nics Centre of CTO [10]. They consisted in measuring the velo-
city eld around stern part of ship both without any propeller
and with operating propeller.
The investigations were conducted on the ship model
having its main particulars as follows :
The hull frame sections are shown in Fig.4.
Fig. 4. Image of the panels projected on the model frame sections .
The applied measuring instrument (the measuring sounder
tted with the single ve-hole spherical head PKN(5+4)/8/1)
made it possible to measure the velocity components Vx, Vy,
Vz in the hull-xed rectangular coordinate frame :

The measurement space was located
at the port side of the hull.
Distance of propeller working plane from
aft perpendicular Xp = 124 mm
Distance of propeller axis from
plane of symmetry Yp = 0.0 mm
Distance of propeller axis from
base plane Zp = 109.1 mm.
The measurement plane was located X = 157 mm
fore from the propeller working plane.
EXPERIMENTAL TESTS
The measurements were performed at one value of the ship
model velocity V
M
= 1.75 m/s and four values of rotational
speed of the propeller model. The rst value of rotational
speed was determined for zero-value of propeller thrust. It
was assumed that this was the rotational speed at which pro-
peller-induced velocities on the hull were of negligibly small
values, hence the velocity measurements could be considered
equivalent to the tests on the hull without propeller. The value
of n
o
= 7.3 rps resulted from the tests (during the tests values
of both propeller - induced thrust and torque as well as hull
resistance were measured).
The next three values of rotational speed were so selected
as to obtain only signicantly large values of propeller - model-
-induced velocities. With taking into consideration the working
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Vx
= 0.44 = 0.60 = 0.85
Length b.p. 6.515 m
Breadth 0.977 m
Draught 0.376 m
Model scale = 33
Stern Bow
x component : along ship axis of sym-
metry and hull motion direction
y component : perpendicular to the
hull plane of symmetry
z component : perpendicular to the
hull water plane.
5
range of the measuring dynamometer the following three values
of rotational speed were selected :
At the obtained values of rotational speeds the values of
propeller thrust were many times greater than that of hull re-
sistance at the speed V
M
= 1.75 m/s.
The measurements were conducted along two measurement
lines perpendicular to the longitudinal plane of symmetry of
the hull, XZ, placed by X = 157 mm apart from the propel-
ler working plane. The rst line is placed at the height of
propeller rotation axis, the other - 50 mm above the men-
tioned axis.

The rst measurement point was selected as close to the
hull surface as possible, and the successive points were placed
at every 20 mm up to the distance assumed negligible from the
point of view of propeller induced velocities.

The selected measurement results are presented in Fig. 58
whereas the complete set of them - in CTOs report [10], and
their graphical representation - in the IMP PAN report [11].
Values of the velocity components Vx, Vy, Vz and of the total
velocity Vc can be found there. During all the tests the ship
model speed V
M
was kept equal to 1.75 m/s.
In Fig.5 are presented the measurement results at the
rotational speed n = 7.3 1/s corresponding with zero-value
of propeller model thrust. Hence it can be assumed that the
velocities shown in Fig. 5 correspond with those around the
hull without propeller. They have been taken as the reference
point for determining the velocities induced by working pro-
peller model.
Fig. 5. Velocity components along the measurement line located at the
height of the propeller axis, for n = 7.3 1/s (thrust of zero-value) .
In Fig.6, 7 and 8 are presented results of the measurements at
higher rotational speeds of propeller model, for which induced
velocities should already show signicant values. In Fig.9 it
can be observed in which way values of the axial component
(marked x) change along with rotational speed changing.
Fig. 6. Velocity components along the measurement line located
at the height of the propeller axis, for n = 25 1/s .
Fig. 9. Modules of velocities along the measurement line located
at the height of the propeller axis, for various values
of propeller rotational speed .
In Fig.10 are presented the differences between the veloci-
ties obtained at high values of rotational speed and the velocity
corresponding with the thrust of zero-value. They should cor-
respond with the propeller-induced velocities but the character
of the changes indicates that the inuence of viscosity on the
velocity distribution is signicant (induced velocities make
velocity distribution in the boundary layer changing).
Fig. 10. Axial component and module of propeller-induced velocity along
the measurement line located at the height of the propeller shaft axis .
n
1
= 25 1/s ; n
2
= 30 1/s
n
3
= 35 1/s.
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 5 10 15 20 25
V [m/s]
y [cm]
Vy
Vz
Vx
Vc
n = 7.3 1/s
Vm = 1.75 m/s
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 5 10 15 20 25
V [m/s]
y [cm]
Vy
Vz
Vx
Vc
n = 25.0 1/s
Vm = 1.75 m/s
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 5 10 15 20 25
V [m/s]
y [cm]
Vy
Vz
Vx
Vc
n = 30.0 1/s
Vm = 1.75 m/s
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 5 10 15 20 25
V [m/s]
y [cm]
Vy
Vz
Vx
Vc
n = 35.0 1/s
Vm = 1.75 m/s
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 5 10 15 20 25
V [m/s]
c
y [cm]
Vm = 1.75 m/s
n = 30.0 1/s
n = 25.0 1/s
n = 7.3 1/s
n = 35.0 1/s
0 5 10 15 20 25
Vi [m/s]
y [cm]
Vm = 1.75 m/s
n = 30.0 1/s
n = 25.0 1/s
n = 35.0 1/s
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Vix
Vic
Vic
Vic
Vix
Vix
Similar measurements were performed at the measurement
line located in the same plane but at the height above the pro-
peller rotation axis by 50 mm. For the measurements only the
nal diagram of the propeller induced velocities is presented.
Fig.11. Axial component and module of propeller-induced velocity
along the measurement line located at the height
above the propeller shaft axis by 50 mm .
Analyzing the above presented results of the investigations,
especially those of Fig. 10 and 11, one can state that the share
of propeller-induced velocities in the total velocity is signicant
(intentionally the values of propeller rotational speed conside-
rably exceeded the own propulsion point of the model, which,
for the velocity V=1.75 m/s, approximately corresponded with
the rotational speed of 13 rps).
To conrmhull inuence on calculation results of propeller-
-induced velocities it should be necessary to perform measure-
ments in two frame cross-sections located nearby to each other
but having signicantly different transverse offsets (breadth).
Unfortunately, for many years the ship models designed and
tested have been characterized by slender stern forms. For
this reason the second cross-section was chosen beyond the
stern. In order to maintain the distance between the working
plane and measurement cross-section the way of fastening the
propeller shaft was changed. The stern part of the hull was
modied by extending the stern tube in such a way as to get
the measurement cross-section placed beyond the stern and the
propeller model placed at the same distance as in the case of
basic tests (Fig.12). The tests on such hull version were called
the tests on the Ship 2.
Fig. 12. Location of measurement lines on the ship model
with the modied stern. PP - Base plane, PR - Aft perpendicular .

The measurements were performed at the same ship model
speed V
M
= 1.75 m/s and four values of rotational speed of
propeller model, in the same way as in the rst cycle of inves-
tigations. Their results for the sounder located at the height of
the shaft axis are presented in Fig.13. (The comprehensive set
0 5 10 15 20 25
y [cm]
Vm = 1.75 m/s
n = 30.0 1/s
n = 25.0 1/s
n = 35.0 1/s
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Vix
Vic
Vic
Vic
Vix
Vix
30 35
Vi [m/s]
of the results from the measurements and calculations can be
found in the CTO report [10] and IMP PAN report [11]).
Fig. 13. Axial component and module of propeller-induced velocity
measured at the height of the propeller shaft axis of the ship model
with the modied stern, at the distance of 157 mm before the propeller .
Comparing, with each other, the measurement results for
both versions of propeller fastening (Fig.14) one can state that
the curves are mutually shifted. The difference is approximately
equal to the difference of hull breadth and propeller shaft in the
places where the measurements have been performed in both
versions of the tests.
Fig. 14. Comparison of induced velocities obtained from the tests
on the models with the original stern and modied one .
Therefore the experimental tests fully conrmed
the put hypothesis as follows :
Calculations of the velocities around ship hull, induced by
whirl systems representing the propeller itself and propeller
race, make it necessary to modify input data to Biot-Savart
formula, and as a result of the tests in question the proposed
modication of the input data has been proved correct.
TESTS ON FULL-SCALE SHIP
The scientic aim of the presented investigations is to
improve the algorithm applied in the software for calculating
3D velocity eld in the stern part of full-scale ship with taking
into account propeller operation [7]. Therefore an important
element of the investigations is to verify such eld on a full-
-scale ship.
It is very hard to achieve reliable results from full-scale
measurements of such eld. In the subject-matter literature are
known results of the measurements performed, both in model
- and full-scale, on the hull of HSVA tanker ship, realized under
the auspices of the model basin in Hamburg.
The measurements of the velocity eld before the propel-
ler working on the full-scale ship were carried out through
a window panel tted in the stern part of the hull. They were
conducted with the use of a laser anemometer but only respec-
tive to the axial component of total velocity (i.e. with taking
into account propeller-induced velocities)
.
The measurement
0 5 10 15 20 25
y [cm]
Vm = 1.75 m/s
n = 30.0 1/s
n = 25.0 1/s
n = 35.0 1/s
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Vix
Vic
Vic
Vic
Vix
Vix
30
Vic [m/s]
0 5 10 15 20 25
y [cm]
Vm = 1.75 m/s
n = 30.5 1/s
n = 25.5 1/s
n = 35.5 1/s
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
30
Vic [m/s]
tests on the level
of the shaft axis
tests on the original hull tests with a rebuilt stern
n = 30.0 1/s
n = 25.0 1/s
n = 35.0 1/s
7
results are

presented in the form of diagrams of isotachs, and
only in the range covered by laser beam (Fig.15). The measu-
rements were performed at the distance X = 0.21D from aft
perpendicular. The broken line denotes the propeller circle of
the diameter D = 6.1 m.
Fig.15. Results of the measurements of axial velocity component,
performed on the ship with operating propeller .
And, in Fig.16 are presented the calculation results obta-
ined from the modied PANSHIP software, also concerning
only the axial component of velocity in the same cross-section
before the propeller.
Fig. 16. Results of the measurements of axial velocity component,
performed on the ship with operating propeller .
On the basis of analysis of the achieved results of calcu-
lations, performed on the background of measurement results
(compare Fig.15 and Fig.16), a qualitative similarity of both
elds can be stated. Quantitative comparison can be more
clearly presented in another form. In Fig.17, 18 and 19 the
same results are shown in the form of diagrams of velocity at
a given radius. To this end three radiuses : r/R = 1.0, 0.7 and
0.5 were selected.
If only accuracy of the measurements on the full-scale
ship are taken into consideration (the curves presented in the
gures should be symmetrical respective to the ship plane of
symmetry) then the so-presented results are found astonishin-
gly similar for the radiuses r/R = 1.0, and especially r/R = 0.7).
It means that the PANSHIP software correctly determines the
velocity eld in the behind- the- hull owand correctly expres-
ses the velocity eld induced by the propeller. In the diagrams
are presented total velocity values which are formed along
a considerable length of hull stern part at a signicant share
of propeller-induced velocities. It conrms that the PANSHIP
can be successfully applied to the scaling of velocity elds on
full-scale ship [7].
Fig. 17. Comparison of the axial component of before-the-propeller
velocity at the radius r/R = 1.0, obtained from measurements
and calculations for the full-scale ship, respectively .
Fig. 18. Comparison of the axial component of before- the - propeller
Fig. 19. Comparison of the axial component of before- the - propeller
FINAL REMARKS
O It can be concluded that the obtained experimental tests
fully conrmed the proposed hypothesis :
Calculations of velocities around the ship, induced by
whirl systems representing the propeller itself and pro-
peller race require input data to Biot-Savart formula
to be modied.
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
.35
0.3
D
/
Z
-0.5 0 0.5
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
-Vx/Vs
0.95
0.9
0.85
0.8
0.75
0.7
0.65
0.6
0.55
0.5
0.45
0.4
0
Y/D (starboard )
(Hamburg Test Case) : full scale total axial
velocity distribution; Vs = 9.26 m/s
EXPERIMENT
x - measurement
location
X-Xo/D = 0.210
-Vx/Vs
Axial component of effective velocity eld behind the hull, V
X
/V
S
Effective wake fraction w
n
= 0.2249[-]
Effective wake fraction w
n1
= 0.2273[-]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-200 -150 -100 -50 0 50 100 150 200
[grad]
Test
Calculation
r/R = 1.0
Vx/Vs
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-200 -150 -100 -50 0 50 100 150 200
[grad]
Calculation
r/R = 0.7
Vx/Vs
Test
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-200 -150 -100 -50 0 50 100 150 200
[grad]
Test Calculation
r/R = 0.5
Vx/Vs
O The proved hypothesis can be considered as a kind of disco-
very in uid mechanics (in handbooks on hydromechanics
and subject-matter literature no mention on that theme can
be found)
O However further theoretical research aimed at building
a correct form of Biot-Savart formula in multiply connected
space, is necessary
O The proposed modication of the way of calculations
of velocities induced around ship hull can be tentatively
implemented in engineering practice as an effective appro-
ximation.
NOMENCLATURE
B hull breadth
D propeller diameter
L hull length
n number of propeller revolutions per second
r radius of cylindrical cross-section around propeller axis
R propeller radius
R
o
hull resistance
R
T
resistance of hull with operating propeller
t thrust deduction
T propeller thrust, also hull draught
V
M
model speed
V
S
ship speed
V
c
total velocity
2 2 2
V V ) V V
(
+ + +
M x y z
V
i
induced velocity
V
ic
total induced velocity
2 2 2
V V V + +
x y z
Vx, Vy, Vz

velocity components
X
p
, y
p
, z
p
coordinates of propeller axis location
y distance of measurement points from hull plane of symmetry
hull block coefcient
hull buoyancy
V
i
velocity induced by whirl laments element
model scale
BIBLIOGRAPHY
1. Bugalski T., Koronowicz T., Szantyr J., Waberska G.: Computer
software for determining ow around the ship together with
wave system (in Polish). Proceedings of 10
th
Symposium on ship
hydromechanics, Gdask. 1993
2. Bugalski T., Koronowicz T., Szantyr J., Waberska G.: Computer
system for calculation of ow, resistance and propulsion of
a ship; Paper presented at CADMO94. Southampton UK, 1994
3. Bugalski T., Koronowicz T., Szantyr J., Waberska G.: A metod
for calculation of ow around the hull of a ship moving in calm
water with constant velocity; Marine Technology Transactions,
vol. 5. 1994
4. Koronowicz T., Tuszkowska T., Bugalski T., Grabowska K.: The
inuence of propeller operation on the pressure eld on the hull
and around it; Proceedings of 12
th
International Conference on
Hydrodynamics in Ship Design, HYDRONAV97, Szklarska
Porba, September 1997
5. Koronowicz T., Szantyr J., Bugalski T.: Theoretical model
for determining the pressure eld resulting from hull ow and
operation of the marine propeller; Polish Maritime Research,
September 1997
6. Koronowicz T., Tuszkowska T., Waberska G.: Computer
software system for determining the pressure eld resulting from
hull ow and operation of the marine propeller; Polish Maritime
Research, December 1997
7. Koronowicz T.: A computer method for prediction of the velocity
eld behind a full-scale ship hull, Polish Maritime Research No
1/2003, vol. 10
8. Koronowicz T., Tuszkowska T., Waberska G.: Modernization
of the calculation model applied to the computer software
PANSHIP (in Polish). IMP PAN Report
9. Koronowicz T., Tuszkowska T., Waberska G., Kaniecki M.,
Krzemianowski Z.: Analysis of preliminary results of tests
(in Polish). Report no. 4643/04. IMP PAN
10. Jaworski S.: Results of measurements of velocity eld before the
propeller which operates on ship model and axially symmetrical
bodies (in Polish). Technical report no. RH-2004/T-149, Ship
Design & Research Centre
11. Koronowicz T., Tuszkowska T., Waberska G., Kaniecki M.,
Krzemianowski Z. : Analysis of results of velocity eld tests on
the model 1 (in Polish). Report no. 4864/04, IMP PAN (Institute
of Fluid-Flow Machinery, Polish Academy of Sciences in
Gdask)
12. Koronowicz T., Koronowicz J., Niewiadomski J., Bunikowski J.,
Huk G.: Measurements of propeller-induced velocity eld
around an axially symmetrical body in cavitation tunnel
(in Polish). Report no. 5065/05, IMP PAN
13. Koronowicz T., Tuszkowska T., Wawerska G., Kaniecki M.,
Krzemianowski Z., Koronowicz J., Chaja P., Bednarek A.:
Analysis of results of tests of an axially symmetrical body
in model basin (in Polish). Report no. 5104/05, IMP PAN
14. Koronowicz T., Bugalski T., : Analysis of results of tests of
a ship (in Polish), Report no. 5220/05, IMP PAN
15. Koronowicz T., Krzemianowski Z.: Experimental research
on propeller inuence on ow around ship hull (in Polish).
Materials of MWK-2005, vol. III, Waplewo 17-20 May 2005
16. Koronowicz T., Krzemianowski Z.: The numerical and
experimental tests of the work of a screw propeller on the
ow around the hull on a ship, Materials of the International
Conference HYDRONAV05, Gdask-Ostrda
CONTACT WITH THE AUTHORS
Prof. Tadeusz Koronowicz
Zbigniew Krzemianowski, D.Sc., Eng.
Institute of Fluid-Flow Machinery,
Polish Academy of Sciences
Fiszera 14
REGIONAL GROUP
of the Section
on Exploitation
Foundations
On 25 May 2006 the Regional Group of the Section on
Exploitation Foundations, Machine Building Committee,
Polish Academy of Sciences (PAS), held its successive
scientic seminar organized by Faculty of Engineering
Sciences , Warmia - Mazury University in Olsztyn.
Scientic workers of the Faculty
presented the following papers :
A method for improving operation processes of track
engine by B. Kolator
Application of Exsys Covrid to maintenance of machi-
nes by K. Ligier and A. Rychlik
A system for maintaining the machines in tero-techno-
logical approach by P. Mikoajczak
After discussion and replies from the side of the authors
to questions directed to them, the organizers presented
scientic laboratories of the Faculty.
9
The books Ship Turbine Power Plants: Fundamentals of Thermodynamical Cycles and Introduction to the Theory of
Marine Turbines by Krzysztof Kosowski form the rst two volumes of a series of four monographs on marine turbine
power plants. They are meant for mechanical engineers and for graduate students of technical universities, as well as
marine and naval academies. They were initially elaborated as part of the EuroMTEC program for the module Advanced
Ship Propulsion and Equipment, and have now been remarkably developed and noticeably extended. When writing them
the author made use of some of the most outstanding works on the subject and to take into account his experiences from
work at university. Fundamentals of power plant cycles and principles of turbine operation were laid out using renowned
books ranging from the rst works of A. Stodola published 100 years ago to the latest scientic papers and information
given by major turbine producers.
Ship Turbine Power Plants: Fundamentals of Thermodynamical Cycles
This book deals with thermodynamical cycles of steam and gas turbines, and turbine power plant
arrangements.
Chapter 1: Fundamental Principles of Thermodynamics
Chapter 2: Steam Turbine Cycles
Chapter 3: Gas Turbine Cycles
Chapter 4: Combined Turbine Cycles
(ISBN: 83-922007-2-1, Published by: Foundation for the Promotion of Maritime Industry, Gdask,
2005, 280 pages, hardback, full colour)
Introduction to the Theory of Marine Turbines
This book deals with the fundamental aspects of axial turbine theory.
Chapter 1: Review of Gas Dynamics
Chapter 2: Axial Stage Theory
Chapter 3: Stage Internal Losses
Chapter 4: Efciency Characteristics
Chapter 5: Calculations of Flow in Turbine Stages
Chapter 6: Multi-stage Turbines
(ISBN: 83-922007-3-X, Published by: Foundation for the Promotion of Maritime Industry, Gdask,
2005, 261 pages, hardback, full colour)
The books have earned high appreciation of the reviewers:
The author deals not only with turbine theory, but also with a wide range of aspects connected with strength, opera-
tion, technology and dynamics. [...] The discussed problems are presented in a clear and concise way [...] and the book
meets all standards of an academic handbook.
The handbook contains not only a classical approach to the principles of turbomachinery, but also presents state-of-
-the-art design methods and results of recent research.
The form of presentation of the material and the collected examples deserve admiration [...], while the text and
gures are excellently chosen.
About the author
Krzysztof Kosowski is an Associate Professor at the Chair of Ship Automation and Turbine
Propulsion, Gdask University of Technology. His scope of interest includes selected problems in
the theory, design and construction of steam and gas turbines. He has delivered lectures on turbine
theory, turbines and compressors, power plants, uid ow mechanics, thermodynamics, turbine
design, nuclear turbines and rotating machinery.
His scientic activity is focused on theoretical and experimental research into turbine stages, dyna-
mics of rotor systems, active control of ows and mechanical vibrations, as well as three-dimensional
ow calculations in turbomachinery, optimisation of turbine ow parts and thermodynamical cycles.
INTRODUCTION
The barges operating in push-train mode are characterized
by great values of the hull block coefcients (C
B
> 0.85), that
ensures achieving large values of their displacement at assumed
main dimensions. On the other hand, to decrease their building
costs, usually is applied a simplied bow form consisted of
practically developable surfaces divided by chine lines, thus
relatively simple in building. Such approach is a rational and
economical compromise since service speed of ships on inland
waterways is of the order of 10 15 km/h.
An inspiration to undertake the research on hull resistance
of inland navigation cargo ships has been given by the informa-
tion coming from an inland navigation ship owner that the fuel
consumption on a given shipping route differs signicantly in
the case of pushed barges differing to each other rst of all by
their bow forms. These authors decided to investigate which
bow forms of pushed barges ensure obtaining the smallest hull
resistance values. To this end several characteristic bow forms
were selected [1]. Each of the selected characteristic forms
has been adjusted to a barge having the main dimensions :
L
C
B T = 48.75 9.0 1.7 m, under the assumption that
the bow length from the end of the cylindrical midship body up
to the bow transom plane is equal to L
E
= 8.0 m. Next, series
of calculations of the ow around the push-trains consisted of
two barges connected to each other by their stern parts, were
performed. The calculations were executed by means of the
FLUENT commercial computer software which makes it pos-
sible to take into consideration all factors of crucial inuence
on ship resistance, i.e. viscosity of water, turbulence of ow, as
well as wave system on water free-surface around the ship.
Quality of the calculation results of free-surface water ow
around inland navigation ships in shallow water, has been as-
sessed during the previous research investigations carried out
by these authors [2 , 3]. In view of a limited performance of
the computers being at the authors disposal most of the com-
Analysis of hull resistance
of pushed barges in shallow water
Tomasz Tabaczek
Jan Kulczyk
Maciej Zawilak
ABSTRACT
These authors performed a set of numerical calculations of water ow around pushed barges differing to
each other by bow forms. The calculations were executed by means of FLUENT computer software. Tur-
bulent free-surface ow of viscous liquid was considered. In this paper the calculated values of barge hull
resistance split into bow, cylindrical and stern part components, have been compared and presented.
Keywords : inland waterways ship, hull resistance
putations was performed for the hulls in a reduced scale. As
a rule the same scale has been applied as in the case of model
testing in a towing tank. A direct comparison of the results of
the calculations with those from model tests has conrmed that
the applied software is useful in calculating hull resistance and
determining wave prole on the ship side.
HULL FORMS
OF THE CONSIDERED BARGES
The calculations of water ow around hulls of the barges
were performed for 11 trains of barges tted with bows of the
following forms (Fig.1) : EIIB, EIIBM, EIIBV2, EIIBH, ELI,
ELIM, WALE, WALC, B, B3 and HEL. The rst four constitute
a group of similar forms. They have been elaborated on the basis
of the hull form of the EUROPA IIb pushed barge popular on
the West - European waterways. They differ to each other by
the shape of longitudinal cross-section contour in the plane of
symmetry. For the four barges similar results were achieved.
The hull form of ELI barge has been elaborated on the basis of
an elliptical bow (Ellipsenbug) proposed by Nussbaum [4]. The
ELIM form is a simplied version of the ELI form. Rounded
segments of frame sections have been replaced by straight-line
segments inclined by the angle of 45. As a result, the surface
between the bottom and side of hull has become a developable
surface. The WALE and WALC forms have vertical sides and
are of the simplest geometry. They differ to each other only by
a shape of water-planes which are elliptical in the rst case,
and in the other circular segments tangent to ship sides.
The B and B3 bow forms have been designed by the team
working on the project. The form B ensures obtaining a high
block coefcient value of the bow. Owing to the at bottom
it is possible to make the barge cubicoid hold much longer.
The other bow form is more ne it has a more inclined stem
and higher elevated line of the side chine. The HEL bow form
has been designed by these authors. Side surface of the bow
11
is a fragment of a regular helicoid with its axis located in the
plane of hull side. This shape was assumed to jostle water aside
like the WALC form and simultaneously to integrate a broad
deck and bow transom like in the case of the EIIB bow form.
In the below presented table, are given values of the block
coefcient of hull and that of bow which has been dened as
follows : C
BE
= V
E
/(L
E
B T).
The authors have intended to check if any unambiguous
relation between values of the above mentioned coefcients
and hull resistance, takes place.
Bow C
B
C
BE
EIIB 0.952 0.705
EIIBM 0.952 0.709
EIIBV2 0.956 0.742
EIIBH 0.951 0.675
ELI 0.950 0.695
ELIM 0.949 0.699
WALE 0.962 0.782
WALC 0.949 0.700
B 0.974 0.823
B3 0.952 0.663
HEL 0.939 0.589
The calculations were performed for the train of two barges
connected to each other by their stern parts. To elaborate grids
for numerical calculations the assumption was made that the
stern form of a single barge inuences trains resistance to the
same degree, irrespective of an applied bow version. For this
reason the aft bottom undercut was not modelled and the
cylindrical parts of both barges were made longer and joined
together in the aft transom plane.
The identical at bilge form of 200 mm in height, was ap-
plied to all the barges, except of those having ELI bow form,
where the cylindrical bilge form of 200 mm radius was used.
CALCULATION CONDITIONS

The ow calculations were performed for two values of
water depth: 2.0 m and 3.4 m. The rst of them models the
conditions of very shallow water (h/T = 1.18). In the case of
canalised rivers such conditions appear only in certain places
along short sections of a waterway. The other water depth
(h = 3.4 m) models shallow water conditions, which is more real-
istic for average service conditions on the domestic waterways.
In both the cases the calculations were carried out for the
ship speed equal to 2.48 m/s (8.93 km/h). i.e. at the Froude
number Fn
h
= 0.56 in a more shallower water, and Fn
h
= 0.43
in a deeper water. At such speed a signicant sagging of the
ship should be taken into account, especially at the water depth
equal to 2.0 m. However the taking of sagging into account in
calculations is associated with a change of location of bound-
Fig. 1. Hull forms of the considered barges. PS - Plane of symmetry, PP - Base Plane, KLW - Design waterline .
EIIB
KLW
P
S
PP
EIIBM
KLW
PP
P
S
EIIBV2
PP
KLW
P
S
EIIBH
PP
KLW
P
S
P
S
KLW
PP
ELI
P
S
KLW
PP
ELIM
P
S
KLW
PP
WALE
P
S
P
S
KLW
PP
WALC
P
S
KLW
PP
B
P
S
KLW
PP
B3
P
S
KLW
PP
HEL
P
S
2
5
0
0
1
7
0
0
possible to easily determine forces acting on various parts of
the hull. In order to analyse a contribution of particular hull
segments in total hull resistance the authors split the entire hull
surface into three parts :
+ the bow (from the bow transom plane of fore barge to the
cylindrical midship body)
+ the cylindrical midship body (precisely joined midship
bodies of both fore and aft barges), and
+ the stern (from the cylindrical midship body of aft barge to
the bow transom plane of aft barge) (Fig.2).
Fig. 2. The train of two barges split into three parts
for resistance analysis purposes
The below presented values of hull resistance, calculated
by means of the FLUENT software for free-surface ow con-
ditions, take into account hydrostatic pressure.
In further considerations it was assumed that hull resistance
is a force acting in the direction opposite to ship speed vector
(i.e. aft). A negative value of stern resistance means that the
resultant force acting onto the stern is directed fore. The greater
the force the smaller the total hull resistance. In Fig.3 and 4 the
bow forms are ranked in a sequence resulting from increasing
value of total resistance.
In design and service practice, quality of a pushed barge hull
form is assessed by using the unit resistance values, i. e. those
taken per unit buoyancy or volume of underwater part of ships
hull. The unit resistance values are compared in Fig. 5.
aries of computation area and a signicant increase of time of
computations. The authors have assumed that the neglecting of
sagging introduces the same errors to resistance values of all
the considered hull forms. Hence the differences in calculated
resistance values would maintain the same, and to compare
directly the bow forms would be possible.
For the calculations performed within the frame of this
research work the authors made use of the same principles of
building the computational grids and controlling calculation
runs as those used in the previous research work [3].
All the calculations were performed in the model-scale
of 1:14. The computational grid covered the rectangular area
extending up to 41.25 m ahead the bow and behind the stern,
and 45.5 m overboard. The grid mesh was so designed as to
ensure precise modelling the hull form and the owaround hull
surface. As a rule a regular grid consisted of cubicoid elements
was applied, but irregular one only locally. For the reason of
a limited computer performance the number of elements did
not exceed 200 000.
In the FLUENT software the problem in question was dened
as non-stationary one. The equations were integrated till reach-
ing a stationary state. The applied time-step of 0.01s ensured
reaching the convergence of calculations after 30 000 steps.
To model the turbulence phenomenon the model RNG k- was
selected. A single run of calculations took 48 h on average.
RESULTS
In contrast to the experimental methods (towing tank model
tests) the application of the numerical computation method to
uid mechanics (CFD) makes it possible to split hull resistance
into the components resulting from normal stresses (pressu-
re) and tangential stresses (liquid viscosity). It also makes it
81.5 m
V
8.0 m 8.0 m
Cylindrical midship body
Bow Stern
S
Total resistance
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
Bow
E
L
I
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
R
T
M
[
N
]
Bow resistance
46.0
47.0
48.0
49.0
50.0
51.0
52.0
53.0
54.0
Bow
E
L
I
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
R
B
M
[
N
]
Pressure resistance
h = 2.0 m
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
Bow
E
L
I
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
R
P
M
[
N
]
Midship body resistance
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
Bow
E
L
I
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
R
M
M
[
N
]
Viscosity resistance
h = 2.0 m
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
Bow
E
L
I
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
R
V
M
[
N
]
Stern resistance
-46.0
-45.0
-44.0
-43.0
-42.0
-41.0
-40.0
-39.0
-38.0
Bow
E
L
I
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
R
S
M
[
N
]
Fig. 3. Resistance of the two-barge-train model at the water depth of 2.0m .
13
SUMMARY
O Onto the hulls jostling water aside (HEL, B, WALC, WALE)
is exerted a greater pressure resistance and smaller viscosity
resistance than onto the remaining hulls (Fig. 3 and 4).
However this is the pressure resistance which decides on
the value of total resistance and ranking sequence of the
bow forms. Also, onto those forms a greater aft pressure
force and simultaneously a smaller fore resistance acts
as a rule. These conclusions are also valid for full-scale
ships since in this scale the share of pressure resistance in
total resistance is greater than in the case of model-scale.
O At the water depth h = 3.4 m greater differences in hull
resistance values occur than at the depth of 2.0 m (Fig.7),
which means that though the resistance is smaller in the
deeper water, this is the bow form which more inuences
the hull resistance.
O In general, the hull and bow block coefcients constitute
a rough index of quality of pushed barge hull resistance,
but no unambiguous relation between those indices and hull
resistance has been revealed (Fig. 6, 7, 8).
Fig. 5. The unit resistance values of the two-barge-train model .
0.0
5.0
10.0
15.0
20.0
25.0
30.0
Bow
Bow
E
L
I
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
Unitary resistance
h = 2.0 m
R
T
M
/
V
M
[
N
/
m
]
3
0.0
5.0
10.0
15.0
20.0
25.0
30.0
Unitary resistance
h = 3.4 m
E
L
I
M
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
H
E
L
E
L
I
B
3
R
T
M
/
V
M
[
N
/
m
]
3
Fig. 6. The bow block coefcient C
BE
(the bows are ranked on the basis
of their hull resistance values at the water depth of 3.4 m, see Fig.4) .
Bow block coefficient
0.0
0.2
0.4
0.6
0.8
1.0
0
.
6
9
9
0
.
6
7
5
0
.
6
9
5
0
.
6
6
3
0
.
7
0
5
0
.
7
0
9
0
.
7
4
2
0
.
5
8
9
0
.
7
0
0
0
.
7
8
2
0
.
8
2
3
C
E
L
I
M
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
H
E
L
E
L
I
B
3
Bow
B
E
Fig. 4. Resistance of the two-barge-train model at the water depth of 3.4 m .
E
L
I
M
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
H
E
L
E
L
I
B
3
E
L
I
M
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
H
E
L
E
L
I
B
3
E
L
I
M
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
H
E
L
E
L
I
B
3
E
L
I
M
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
H
E
L
E
L
I
B
3
E
L
I
M
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
H
E
L
E
L
I
B
3
E
L
I
M
E
I
I
B
H
E
I
I
B
V
2
E
I
I
B
E
I
B
M
W
A
L
C
W
A
L
E B
H
E
L
E
L
I
B
3
10.0
12.0
14.0
16.0
R
P
M
[
N
]
R
T
M
[
N
]
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
0.0
2.0
4.0
6.0
8.0
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
R
V
M
[
N
]
R
B
M
[
N
]
R
M
M
[
N
]
46.0
47.0
48.0
49.0
50.0
51.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
-46.0
-45.0
-44.0
-43.0
-42.0
-41.0
-40.0
-39.0
-38.0
R
S
M
[
N
]
Bow resistance
Stern resistance
Total resistance
Bow
Bow
Pressure resistance
h = 3.4 m
Bow
Midship body resistance
Bow
Viscosity resistance
h = m 3.4
Bow Bow
O
In Fig. 7 and 8 the points are clustered in two groups. The bow forms : HEL, B, WALC and WALE belong to the rst group,
the remaining to the other group. The bows of the rst group have a straight, vertical or only slightly inclined stem, and
greater resistance values as well. This observation suggests that the vertical or only slightly inclined stem is not favourable
from the point of view of pushed barge hull resistance.
NOMENCLATURE
B - hull breadth
C
B
- hull block coefcient
C
BE
- bow block coefcient
Fn
h
- Froude number ( gh V Fn
S h
/ = )
g - gravity acceleration
h - water depth
L
C
- overall length of ship
L
E
- length of bow
T - design draught
V
E
- bow volume
V
EM
- bow volume in model-scale
VM - volume of hull underwater part in model-scale
V
S
- ship speed
RBM - bow resistance in model-scale
RPM - pressure resistance in model-scale
RSM - stern resistance in model-scale
RTM - total resistance in model-scale
RVM - viscosity resistance in model-scale
RMM - resistance of cylindrical midship body in model-scale
Acknowledgement
The research presented in this paper has been nancially supported
by the Minister for Science and Informatics, within the frame of the
research project No. 4 T12C 014 27.
BIBLIOGRAPHY
1. Kulczyk J., Tabaczek T., Werszko R., Zawilak M., Zieliski A.:
Bow forms of inland navigation cargo vessels. 16th International
Conference on Hydrodynamics in Ship Design HYDRONAV05.
Gdask-Ostrda, Poland. 7-10 September 2005
2. Zawilak, M., Tabaczek, T.: Resistance prediction by using
CFD. Report T32-PWR-IREP-Resistance_prediction_CFD of
research project INBAT within 6. Outline Program (G3RD-CT-
-2001-0458). August 2004
Fig. 7. The relationship of hull resistance and the hull block coefcient C
B
.
Fig. 8. The relationship of hull resistance and the bow block coefcient C
BE
.
h = 2.0 m
R
2
= 0.4803
5.0
7.5
10.0
12.5
15.0
0.92 0.94 0.96 0.98
ELI
EIIBV2
WALC
WALE
B
h = 3.4 m
R
2
= 0.333
5.0
7.5
10.0
12.5
15.0
HEL
WALC
WALE B
ELIM
EIIBV2
h = 2.0 m
R
2
= 0.3888
12.5
15.0
17.5
20.0
22.5
25.0
27.5
WALC
WALE
B
ELI
EIIBV2
h = 3.4 m
R
2
= 0.2868
12.5
15.0
17.5
20.0
22.5
25.0
27.5
HEL
ELIM
EIIBV2
WALC
WALE
B
0.92 0.94 0.96 0.98
0.92 0.94 0.96 0.98 0.92 0.94 0.96 0.98
R
T
M
[
N
]
R
T
M
/
V
M
[
N
/
m
]
3
R
T
M
[
N
]
R
T
M
/
V
M
[
N
/
m
]
3
C
C
B
B
C
C
B
B
R
2
= 0.585
5.0
7.5
10.0
12.5
15.0
0.50 0.60 0.70 0.80 0.90
ELI
EIIBV2
WALC
WALE
B
R
2
= 0.2923
5.0
7.5
10.0
12.5
15.0
HEL
WALC
WALE
B
ELIM
EIIBV2
R
2
= 0.4864
12.5
15.0
17.5
20.0
22.5
25.0
27.5
ELI
EIIBV2
WALC
WALE
B
12.5
15.0
17.5
20.0
22.5
25.0
27.5
R
2
= 0.2456
HEL
WALC
WALE
B
ELIM
EIIBV2
0.50 0.60 0.70 0.80 0.90
0.50 0.60 0.70 0.80 0.90 0.50 0.60 0.70 0.80 0.90
R
T
M
[
N
]
R
T
M
/
V
M
[
N
/
m
]
3
R
T
M
[
N
]
R
T
M
/
V
M
[
N
/
m
]
3
h = 2.0 m h = 3.4 m
h = 2.0 m h = 3.4 m
C
C
BE
BE
C
C
BE
BE
15
3. Zawilak, M.: Inuence of waterway depth on pressure
resistance of inland navigation ship (in Polish). Doctoral thesis.
Report of Preprint Series No. PRE 028/04, Institute of Machine
Building and Operation, Wrocaw University of Technology.
Wrocaw, 2004
4. Nussbaum, W.: Entwicklungen der Binnenschiffsformgebung
unter Bercksichtigung der Anforderungen im
Flachwasserseegang, Jahrbuch der STG, Bd 82, 1988
Tomasz Tabaczek, D.Sc., Eng.
Prof. Jan Kulczyk
Maciej Zawilak, D.Sc., Eng.
Institute of Machine Design and Operation,
ukasiewicza 7/9
50-371 Wrocaw, POLAND
HYDROACOUSTICS 2006
On 23 26 May 2006
at Krynica Morska upon Vistula Bay was held :
13
th
SYMPOSIUM ON HYDROACOUSTICS
organized by the Department of Marine Electronic Sys-
tems, Faculty of Electronics, Telecommunication and
Informatics, Gdask University of Technology, under the
auspices of : European Acoustics Association, Hydro-
-acoustics Section of Committee on Acoustics, Polish
Academy of Sciences, and Gdask Division of Polish
Acoustical Society.
The Symposium was commenced
by the key-note lecture on :
Research and development on underwater acoustic sys-
tems of Polish Naval University and Gdask University of
Technology for the Polish Navy by G. Grelowska (Polish
Naval University) and L. Kilian (Gdask University of
Technology).
During 4 plenary session of the Symposium
the following 5 invited papers were presented :
C Science and technology in Polish Ministry of Defense
by W. Drg (Polish Ministry of Defense)
C New scientic multi-beamsystems for shery research
applications by L. Nonboe (SIMRAD, Norway)
C The state of the Baltic Sea hydro-acoustical investiga-
tions (selected problems) by Z. Klusek (Institute of
Oceanology, Polish Academy of Sciences)
C Synthesis and wavelet analysis of side-scan sonar sea
bottom imagery by J. Tgowski (Institute of Oceano-
logy, Polish Academy of Sciences) and A. Zieliski
(University of Victoria, Canada)
C Quadrature phase detection in an acoustic positio-
ning system by A. Zieliski (University of Victoria,
Canada) and Y.Shi (Southwest Jiaotong University,
China)
The remaining 25 papers were presented during 4
panel sessions. Original papers, both theoretical and
experimental, concerning problems of hydro-acoustics
and its applications are published in the annual journal
Hydro-acoustics.
Workshops 2006
Under this name, on 30 March 1 April 2006, Faculty
of Maritime Technology, Technical University of Szcze-
cin, arranged the series of popular scientic lectures and
demonstrations to promote the courses on
Ocean Engineering and Transport
conducted at the Faculty.
Academic lecturers presented the following themes :
The last day the underwater apparatuses built at the
Faculty were demonstrated. The Workshops appeared
very interesting for many visitors hence it was decided to
organize them every year.
* Safety at sea by M. Hann
* Gas an oil mining from sea bed
by W. Chdzyski
* Contemporary maritime industry
and shipping by T. Jastrzbski
* Neural networks by D. Pielka
* Shapes of sound by S. Weyna
* Super-computers and turbulence
by T. Abramowski
* Unconventional energy sources on ships
by W. Zeczak
* Digital evolution by P. Nikoczuk
* Stirlings engine by A. muda
SEM ECO
On 12 May 2006 the scientic seminar on : Ecological
problems in operation of combustion engines, organized by
Prof. L. Piaseczny, was held at Polish Naval University.
The seminar program contained two lectures presented
by the scientic workers from Warsaw University of
Technology, namely :
^ Selected problems of emission of PM10 solid partic-
les from exhaust gas systems of combustion engines
by M. egota
^ Development trends of combustion engines for usage
vehicles by Z. Chopek
Both the topics triggered very interesting discussion
which enriched the knowledge passed on in the presented
lectures.
INTRODUCTION
The aim of the presented work has been to evaluate dis-
placements of the engine crankshaft axis in different working
conditions of propulsion system [3 , 5]. In the shaft-line ali-
gnment methods the crankshaft-shaft line interaction has been
considered in a simplied way so far. The crankshaft has been
modelled as a linear system of cylindrical beam elements,
whereas its thermal displacements and its foundation stiffness
have been evaluated on the basis of simple data supplied by the
producer without considering the type of the ship on which the
engine has to be installed [4]. The goal of the work has been
to improve representation of the boundary conditions of the
marine power transmission system. It is especially important
for the high power propulsion systems since in the literature
many examples of the damage of the rst three main bearings
(counting from the driving end) of the main engine can be
found [1 , 2]. One of the possible causes of such state might
be the imprecise mathematical model of crankshaft, proposed
for the analysis of shaft line alignment.
Within the frame of this work several analyses of the Sul-
zer 7 RTA 84 C engine installed on a big container ship (of
~3000 TEU capacity) were carried out. Also, the computation
of engine body deformation under gravity load as well as
the analysis of its thermal deformation in nominal working
conditions was performed. The static stiffness (horizontal and
vertical) of each of the main bearings were evaluated and then
their dynamic stiffness was determined in the frequency range
of 030 Hz. As the forced vibration analysis was performed
with the use of the modal superposition method, it was ne-
cessary to determine in advance the natural frequencies and
eigenvalues in the frequency range of 070 Hz. The thermal
Stiffness characteristics and thermal
deformations of the frame
of high power marine engine
Lech Murawski
Marek Szmyt
Centrum Techniki Okrtowej S. A. (CTO)
(Ship Design and Research Centre)
ABSTRACT
In the subject-matter literature detail data on stiffness of the crankshaft foundation connected with the
frame of marine main engine are still lacking. Thermal deformation models of the engines casing, propo-
sed by engine producers, are excessively simplied. However the parameters are crucial for the shaft-line
alignment analysis as well as for the analysis of interactions between the shaft-line and engine crankshaft,
especially in the case of high power engines. This paper presents a determination method of the marine
engine body characteristics as well as results of example computations performed for a Sulzer 7 RTA 84 C
engine installed on a ~3000 TEU container ship. It has been demonstrated that the producers assumption
about parallel displacement of the crankshaft axis in thermal working conditions is too rough. The ther-
mal deformation of the engine is of hogging character, which results in signicant change of the moment
load exerted on the crankshaft and shaft line. The stiffness parameters recommended by the producers
for the shaft-line alignment are estimated correctly, however they represent only engines body exibility,
without taking into account ships hull exibility.
Keywords : marine main engine, main bearing, static and dynamic stiffness characteristics of bearing,
thermal deformation, temperature distribution
analysis requires an accurate temperature distribution on the
engine body to be known. Appropriate data were obtained from
comprehensive temperature measurements performed on the
ship and her main engine.
ANALYSIS METHOD
The FEM model of the body of Sulzer 7 RTA 84 C engine
is presented in Fig.1. Fig.2 shows a part of the model represen-
ting the engine main bearing. The engine body model contains
almost 200 thousand plate and solid elements of over 930
thousand degrees of freedom.
The subject of the analysis was the body of Sulzer
7 RTA 84 C engine. The analysis involved, apart from heat ow,
also thermal deformation and stress calculated by means of a 3D
heat transfer model. The method is based on the solution of the
heat ow equation (with variable coefcients). For stationary
heat ow the equation has the following form:
(1)
If the convective boundary conditions on both hot and cold
surface are assumed then the following relations are valid :
(2a)
(2b)
[ ]
( ) 0
z
T
z , y , x k
z
=
+
[ ] [ ]
( ) ( )
y
T
z , y , x k
y x
T
z , y , x k
x
+
( )
y
T
z , y , x k

( )
a s
T T h
( )
z
T
z , y , x k

( )
a s
T T h
17
No heat ow is assumed on other surfaces of boundary
planes of the considered hull segment, which leads to the
equation:
(3)
The FEM thermal analysis was performed with the use of
MSC NASTRAN software. MSC PATRAN software was used
as a pre-and post-processor for the calculation results of the
stress under thermal load.
Before starting the thermal deformation analysis of the
engine it is necessary to determine temperature distribution on
its body. The temperature map was created on the basis of the
measurements carried out on a marine main whose size and
structure was similar to those of the engine in question.
TEMPERATURE DISTRIBUTION
MEASUREMENT ON ENGINE BODY
The temperature distribution measurements on the engine
body were performed on the Sulzer 8 RTA 96 C engine during
sea trials. The engine load was kept stable in nominal working
conditions. Alfa-Tech Rytek MT 4 pyrometer was used for the
measurements. The example layout of measurement points (on
the port side) is shown in Fig.3. The results of measurements
in those points are presented in Tab.1. On the starboard side as
well as on the fore and aft end of the engine the measurement
points were distributed in a similar way.
Fig. 3. Layout of the measurement points on the port side
of the main engine body .
Tab. 1. Results of the temperature measurements performed
on the port side of the main engine body .
Port side of the Engine
Measure-
ment
point No.
Measured
tempera-
ture [C]
Measure-
ment
point No.
Measured
tempera-
ture [C]
Measure-
ment
point No.
Measured
tempera-
ture [C]
1 37 8 52 15 51
2 44 9 49 16 50
3 52 10 52 17 57
4 53 11 49 18 60
5 54 12 50 19 63
6 49 13 54
7 53 14 54
THERMAL ANALYSIS
OF ENGINE BODY DEFORMATION
In the numerical thermal analysis the value of heat conduc-
tivity coefcient for steel was assumed equal to 42.9 [W/mK].
The heat ow analysis was performed for the condition of the
hold being hot and thermally balanced. Because of lack of more
precise data the heat transfer coefcients were assumed as for
heated cargo in accordance with DNV Classication Rules.
The values of heat transfer coefcients are presented in Tab.2.
The thermal expansion coefcient of the engine body was
assumed equal to = 1.610
-5
. The temperature distribution
on the engine body, shown (in C) in Fig.4, was analogous to
that obtained from the measurements.
Tab. 2. Assumed values of heat transfer coefcients [W/m
2
C] .
From air in the hold to the inner bottom structure 58.1
From air in the hold to the side structure 58.1
From air in the hold to the deck structure 58.1
From outboard water to the hull shell 7400
From air to the hull shell 23.2
From air in the double bottom to the hull structure 0
Fig. 2. Main bearing frame of Sulzer 7 RTA 84 C engine .
x
T
0
Fig. 1. FEM model of Sulzer 7 RTA 84 C engine .
Fig. 4. Temperature distribution on the engine body, assumed
for the analysis . Note : On this and all next gures
the SI standard units (e.g. m, Pa) are applied .
In Fig.5 the thermal deformation of the engine body is
presented in the form of fringe plot. From the point of view of
the propulsion system and main engine shaft line interaction,
displacements of the engine main bearings are most important.
The values of the displacements are presented in Tab.3. The
bearings are numbered beginning from the driving end of the
shaft line (the right-hand side of Fig.5).
Fig. 5. Thermal deformation of the body of Sulzer 7 RTA 84 C engine .
Tab.3. Thermal displacements of the main bearings
of Sulzer 7 RTA 84 C engine .
Main bearing
No.
Vertical
displacement
[mm]
Horizontal
displacement
[mm]
Axial
displacement
[mm]
1 0.756 -0.004 1.905
2 1.086 -0.001 1.327
3 1.428 0.002 0.604
4 1.564 0.005 0.290
5 1.627 0.009 0.056
6 1.605 0.011 -0.180
7 1.498 0.010 -0.476
8 1.252 0.006 -0.948
9 0.645 -0.005 -1.986
For the crankshaft axis translation the main engines
producer recommends to use the following formula :
(4)
where :
h
e
displacement of the main bearing axis [mm]
dh thermal expansion coefcient of the engine
body [mm/K]
T
s
service temperature of the engine [K]
T
a
air temperature in the engine room [K].
For the examined Sulzer 7 RTA 84 C engine
the translation of the main bearing axis amounts to :
(5)
The numerically computed value of the translation of the
shaft line axis (Tab. 3) is greater than that recommended by
the producer despite the fact that the measured temperature
of the real engine is slightly lower than its specied service
temperature. The difference for the rst bearing is not par-
ticularly large (less than 20%), but other bearings are much
more displaced. It seems that the producers assumption on
the parallel translation of the crankshaft axis is incorrect. The
hogging deformation of the crankshaft results in a signicant
change of the moment transferred from the shaft line. Its effect
calculated by using a precise shaft line alignment analysis
would be considerable.
ANALYSIS OF STATIC STIFFNESS
OF MAIN BEARINGS OF THE ENGINE
Determination of the static stiffness consists in applying
unitary forces equal to mass forces and radial gas forces (of
750 kN), to each of the main bearings, one by one, rst in verti-
cal and then in horizontal direction. The achieved displacements
serve to calculate the local static stiffness. The quantities are
very important for shaft line alignment analysis since applying
only the ships hull stiffness may be insufcient. The main
engine producers usually provide (on request) information on
the crankshaft foundation stiffness but without dividing the
quantity into that concerning engine body alone and ship hull.
Some time ago this parameter was assumed innitely large,
now it is considered as large as 6.010
9
N/m. The CTO Co.
gained vast experience concerning the stiffness of ship hulls
of many types but no main engine body stiffness has been so
far examined by this company.
The deformation of the main engine body under only gravity
load was computed rst. It was observed that the deformation
due to gravity load was several times smaller (of the order of
only 0.01mm) than that due to thermal load, cylinder mass and
( )
a s e
T T dh h =
( )
mm 607 . 0 20 0 . 53 10 4 . 18 h
3
e
= =

19
gas forces. Therefore in further analysis the inuence of gravity
load may be neglected.
Next, was carried out the static analysis consisting in
application of the load rst in vertical and then in horizontal
direction (18 load cases). The estimated radial forces were
distributed on the main bearing surface. Tab.4 contains the
static stiffness values for particular main bearings (numbered
beginning from the aft driving end of the crankshaft). The
selected deformation values and stress fringe plots are pre-
sented in Fig. 68.
Tab. 4. Static stiffness of the main bearings of Sulzer 7 RTA 84 C engine .
Main bearing No.
Vertical stiffness
10
9
[N/m]
Horizontal stiffness
10
9
[N/m]
1 6.615 10.942
2 6.828 11.232
3 7.317 12.229
4 7.357 12.300
5 7.364 12.325
6 7.360 12.307
7 7.329 12.248
8 7.188 12.020
9 6.312 9.689
Fig. 6. Deformation of the main engine body
under vertical load applied to 6
th
main bearing .
Fig. 7. Deformation of the main engine body
under horizontal load applied to 1
st
main bearing .
Fig. 8. Stress in the main engine body under vertical
and horizontal load applied to 3
rd
main bearing .
On the basis of the performed analysis it can be stated
that the static stiffness values specied by the producer are
properly evaluated, however they represent only the exibility
of the engine body without taking into account the ship hull
exibility. In the authors opinion in the analysis of shaft line
alignment the sum of both exibility parameters should be taken
into account. It can be observed that the engine body stiffness
is very high and the cylinder mass and gas forces acting on
a single cylinder have a little inuence on displacements of
other main bearings.
It means that it is not required to take into account any
coupling between particular bearings hence there is no neces-
sity to determine the equivalent stiffness reecting the engine
structure integrity. It is intriguing that the horizontal stiffnesses
are higher than the vertical ones, not so as in the case of most
other marine structures. The stress level in the main bearing
structure is not high as it does not exceed 15 MPa for horizontal
load and 22 MPa for vertical one.
ANALYSIS OF DYNAMIC STIFFNESS
CHARACTERISTICS OF BEARINGS
The determination of the dynamic stiffness characteristics
consists in applying unitary forces dened as a function of
excitation frequency, to each of the main bearings, one by one,
rst in vertical and then in horizontal direction. The obtained
displacements serve to calculate the local dynamic stiffness for
each excitation frequency, independently. These quantities are
important for shaft line lateral vibration analysis. The data on
the dynamic stiffness of main engine body are not available
neither from literature nor from the producers.
The analysis of the dynamic stiffness of Sulzer 7 RTA 84 C
engine body was carried out by using the MSC Nastran FEM
solver and modal superposition method. The stiffness characte-
ristics were determined in the frequency range of 030 Hz; this
way the full spectrum of possible propulsion system excitation
frequencies was covered. In such case according to the authors'
experience, it is recommended to calculate normal modes in
the range of natural frequency, taken at least twice as wide as
that. The eigenmodes were determined for frequency values up
to 70 Hz. The eigenvector maps for the most important normal
modes are shown in Fig. 911. The applied denotation for engi-
ne body deformations, commonly used by engine producers, is
as follows: H lateral deformation, X twisting deformation,
C bending deformation in vertical plane.
Fig. 9. H- type normal mode at the natural frequency of 32.90 Hz .
Fig. 10. X- type normal mode at the natural frequency of 45.30 Hz .
It is important that the rst signicant normal modes have
their natural frequencies above the range of excitation frequ-
encies of the propulsion system. And, the signicant normal
modes are only a few and those of interest concern the whole
engine body. No signicant normal modes were found in the
region of the engine main bearings. It speaks well for the correct
design of the engine body, i.e. of sufciently rigid structure.
In such case the resonance discontinuities of main bearing
exibility are not expected and the characteristics appear close
to linear. For this type of structure it is allowed to use only
static stiffness as the expected dynamic amplication should
be insignicant.
Next step was to compute forced vibration by applying rst
vertical load and then horizontal one to each of the engine main
bearings, one by one (18 load cases). The analysis was performed
within the frequency range of 030 Hz. Tab.5 and 6 contain
dynamic stiffness values for characteristic excitation frequencies
and particular bearings (numbered beginning from the aft driv-
ing end of the crankshaft). The nominal rotational speed of the
examined engine was 100 rpm. The engine had seven cylinders,
and the propeller - ve blades. For such conguration the basic
excitation frequencies were 8.33 Hz and 11.67 Hz.
Tab. 5. Vertical stiffness of main bearings of Sulzer 7 RTA 84 C engine .
Main
Bearing
No.
Stiffness
values
at 0 Hz
10
9
[N/m]
Stiffness
values
at 8.33 Hz
10
9
[N/m]
Stiffness
values
at 11.67 Hz
10
9
[N/m]
Stiffness
values
at 25 Hz
10
9
[N/m]
1 6.632 6.611 6.591 6.442
2 6.846 6.824 6.804 6.650
3 7.335 7.321 7.308 7.203
4 7.375 7.362 7.350 7.254
5 7.382 7.370 7.357 7.264
6 7.378 7.365 7.353 7.259
7 7.348 7.334 7.321 7.219
8 7.206 7.190 7.174 7.053
9 6.327 6.301 6.274 6.309
Tab. 6. Horizontal stiffness of main bearings of Sulzer 7 RTA 84 C engine .
Main
bearing
No.
Stiffness
values
at 0 Hz
10
9
[N/m]
Stiffness
values
at 8.33 Hz
10
9
[N/m]
Stiffness
values
at 11.67 Hz
10
9
[N/m]
Stiffness
values
at 25 Hz
10
9
[N/m]
1 10.969 10.924 10.878 10.447
2 11.260 11.208 11.155 10.616
3 12.260 12.195 12.128 11.406
4 12.331 12.266 12.199 11.444
5 12.356 12.290 12.222 11.435
6 12.338 12.271 12.201 11.388
7 12.279 12.210 12.138 11.303
8 12.050 11.974 11.895 11.010
9 9.714 9.643 9.572 8.847
Fig. 11. C- type normal mode at the natural frequency of 69.95 Hz .
21
The sample plots (for the main bearing No.3) of displacement of main bearing foundation in function of vertical (and subsequ-
ently horizontal) frequency are presented in Fig.12. Selected fringe plots of the engine body deformation in the region of the
main bearing No. 3 for excitations in both directions are shown in Fig.13.
0
-2.00-05
-4.00-05
-6.00-05
-8.00-05
-1.00-04
-1.20-04
1.00+01 0 5.00+00 1.50+01 2.00+01 2.50+01 3.00+01
D
i
s
p
l
a
c
e
m
e
n
t
t
r
a
n
s
l
a
t
i
o
n
a
l
Frequency
0
-1.50-05
-3.00-05
-4.50-05
-6.00-05
-7.50-05
-9.00-05
1.00+01 0 5.00+00 1.50+01 2.00+01 2.50+01 3.00+01
D
i
s
p
l
a
c
e
m
e
n
t
t
r
a
n
s
l
a
t
i
o
n
a
l
Frequency
Fig. 12. Displacement of the main bearing No. 3 in function of excitation frequency .
Fig. 13. Deformation of the main engine body under vertical and horizontal excitation applied to the main bearing No. 3. at the frequency of 8.33 Hz .
The dynamic stiffness values for the excitation frequency
of 0 Hz are almost identical with the static stiffness (their
differences are at 3
rd
decimal place see Tab. 4, 5 and 6). It
speaks well for the correctness of the dynamic analysis. As it
was expected after performance the analysis of normal modes,
the dynamic stiffness values did not signicantly differ from
the static ones. The stiffness decrease by 2% in vertical direc-
tion may be observed and that in horizontal direction by less
than 7%. Such change of dynamic stiffnesses can not have any
signicant effect on the analysis of shaft line lateral vibration.
In standard (commercial) analyses the stiffness evaluation
may be limited to only a static quantity which can be assumed
constant in the domain of excitation frequency.
CONCLUSIONS
O The numerically computed shaft line displacement is greater
than that recommended by the producer despite the fact
that the measured temperature of the real engine is slightly
lower than its specied service temperature. For the rst
bearing (counting from the shaft line side) the difference
is not particularly big (lower than 20%), but other bearings
are much more displaced. Hence the producers assumption
on the parallel translation of the crankshaft axis seems to
be incorrect. The hogging deformation of the crankshaft
results in a signicant change of the moment load exerted
by the shaft line. It may be expected that the effect result-
ing from the precise shaft line alignment analysis would be
considerable.
O On the basis of performed analysis it can be stated that the
static stiffness values specied by the producer are properly
evaluated, however they represent only the engine body
exibility without taking into account the ships hull ex-
ibility. In the shaft line alignment analysis the sum of both
exibility parameters should be taken into account. It can
be observed that the engines body stiffness is very high
and the cylinder mass and gas forces acting on one cylinder
have a little inuence on displacements of other main bear-
ings. Therefore it is not necessary to take into account any
coupling between particular bearings.
O It is important that the rst signicant normal modes have
natural frequencies above the range of excitation frequ-
encies of the propulsion system. Moreover the signicant
normal modes are few and those of interest concern the
whole engine body. No signicant normal modes were
found in the region of the engine main bearings. It means
that the engine body in question is a well-designed rigid
structure. In such case the resonance discontinuities in the
exibility of main bearings should not be expected and their
characteristics should be close to linear.
O The dynamic stiffness values differ insignicantly fromthe
static ones. The stiffness decrease by 2% can be observed
in vertical direction and that by more than 7% in horizontal
direction. Such change of dynamic stiffness values can not
have any signicant effect on results of the analysis of shaft
line lateral vibration (whirling). In commercial analyses
evaluation of stiffness may be limited only to determining
the static value which in the domain of excitation frequ-
ency can be assumed constant.
O The direction of research in question looks very promising.
It would make it possible to introduce such improvements
to high power propulsion systems as to avoid failures of
the engine's main bearings. The presented method may
be also used for more advanced and complete numerical
computations carried out for main engines of other types,
installed on ships having specic hulls.
O The next step in developing the proposed method of pro-
pulsion system analysis should be incorporation of a more
complex crankshaft representation based on its full 3D
characteristics. The crankshaft springing effect on the shaft
line alignment should be also examined.
Acknowledgements
The described project has been nanced from the budget
of Ministry of Science and Informatics, allocated to the Ship
Design and Research Centre for its statutory activities in the
year 2005. Execution of measurements on the real object and
access to reliable engines data were kindly made possible
by Gdynia Shipyard Co. The authors are very grateful to all
persons and institutions which supported this research project
with really reliable data.
NOMENCLATURE
dh thermal coefcient of expansion of the engine body [mm/K]
h
e
vertical displacement of the main bearing axis [mm]
h thermal heat transfer coefcient [W/(m
2
K)]
k thermal conduction coefcient [W/(mK)]
T temperature [K]
T
a
air temperature in the engine room [K]
T
s
service temperature of the engine [K].
BIBLIOGRAPHY
1. MAN B&W Diesel A/S : Bearings. Copenhagen. 2000
2. MAN B&W Diesel A/S : Elasto-hydro-dynamic evaluation of
main bearing performance. Copenhagen. 2002
3. American Bureau of Shipping: Guidance notes on propulsion
shafting alignment. Houston. 2004
4. MAN B&W Diesel A/S : Shafting alignment for direct coupled
low-speed diesel propulsion plants. Copenhagen. 1995
5. Wrtsil: Sulzer RTA-C. Technology Review, Helsinki. 2003.
Lech Murawski, D.Sc., M.E.
Marek Szmyt, M.Sc., M.E.
Centrum Techniki Okrtowej S. A.
Rzeczypospolitej 8
Mourning
9 August 2006 was the time of a deep sorrow
for the circle of Polish shipbuilders as
Professor Jerzy Doerffer
passed away this day.
Graduate of Shipbuilding Faculty of University
of Glasgow and Gdask University of Technology.
Professor of Gdask University of Technology
where he worked since 1948.
The organizer and the rst Head of the Department
of Ship Technology and its auxiliary unit.
The Dean of Shipbuilding Faculty
in the years 1953-54 and 1958-64.
The Rector of Gdask University of Technology
in the years 1964-67.
The Chairman of the Forum of Shipbuilding
and Ship Repair Industry.
The creator of the scientic school in the domain
of shipbuilding technology.
He was a worldwide recognized authority
on this domain.
The co-author of novel design solutions
and manufacturing techniques in ship technology.
Doctor Honoris Causa of Gdask University of Tech-
nology, Leningrad Shipbuilding Institute, University of
Glasgow, University of Rostock, Polish Naval University
and Technical University of Szczecin.
He was also honoured with William Froude Medal
by the Royal Institution of Naval Architects, London.
Prof. Doerffer was not only an outstanding engineer-in-
ventor but also a moral authority of wonderful personali-
ty. Due to his attitude and work he won respect from the
side of the whole circle of Polish shipbuilders.
He was a gure-symbol of whom the circle
of Polish shipbuilders boasts.
Humanist and teacher. He promoted a whole genera-
tion of Polish engineers in shipbuilding technology.
A member of prestigious scientic
institutions in Poland and in the world.
23
INTRODUCTORY COMMENTS
Contemporary sea-going ships are tted as a rule with a rich
set of control and measuring instruments. Commonly are in use
analogous or digital measuring systems as well as those based
on computer technique. They serve both for carrying out direct
measurements and data processing in the systems for remote
and programmed control of ship operation, main propulsion
system and auxiliary shipboard devices.
On ships a special high-accuracy measuring instrumentation
is also applied. Such instrumentation is installed only for the
time of measurements and it does not belong to ships standard
equipment.
Within the frame of the KBN research project no. 9 T12D
033 17 some energy measurements were performed on the trai-
ning-research ship Horyzont II in the period from 23.06.2001
to 14.07.2001. The measurements covered the quantities
associated with operation of ship main propulsion system in
various service conditions, namely main engine (ME) power
output, ship speed, fuel consumption by the main propulsion
system. Additionally were measured selected parameters of ME
operation, interesting from the point of view of loading state of
main propulsion system, e.g. exhaust gas temperature, super-
charging air pressure and temperature. Also, some parameters
associated with ME operation such as exhaust gas content in
various loading states, were investigated. For carrying out the
measurements standard measuring instruments installed on
the ship as well as special measuring devices installed by the
research team, were used.
Usefulness assessment of standard measuring
instruments installed on sea-going ships
to perform energy measurements
Zygmunt Grski,
Romuald Cwilewicz,
ABSTRACT
The presented work is a contribution to discussion on usefulness of application of measurement instrumen-
tation used on sea-going ships for energy measurement and scientic research purposes. Contemporary
sea-going ships are equipped as a rule with up-to-date measurement instrumentation usually based on
electronic data processing and computer technique. These authors have made many times use of such
instruments in their research work. This way it was not necessary to install any special instruments, that
signicantly reduced measurement cost. In such cases to obtain a sufcient accuracy of measurements
constitutes a crucial problem. In this paper was presented an analysis of measurement errors of some
operational parameters of ship and its main propulsion system, elaborated within the frame of the KBN
research project no. 9 T12D 033 17. Results of the analysis conrmusefulness of the standard measurement
instrumentation installed on ships, and its sufcient accuracy.
Keywords : ship, standard measurement instrumentation, accuracy of measurements.
The investigations were aimed among other at deter-
mination of ship propulsion characteristics. To this end the
measurements of ship speed, ME power output and its fuel
consumption at various rotational speeds and propeller pitch
settings, were performed.
Specication of the standard measuring instruments installed
on the ship and used for the research in question is as follows :
for the measuring of : propeller pitch setting, rotational
speed of ME and its load index the integrated computer-
-based system for measuring operational parameters and
control of settings of ships main propulsion system
-Wichmatic 2 Propulsion Control System, Graphic System
WM 11G ECR, WM2 12G ECR, WRTSIL NSD Norway
AS, having digital read-out system
for the measuring of ship speed : the SAL R1 U/N 701488C
electro-magnetic log of Consilium Marine Co, having di-
gital read-out system
for the measuring of : exhaust gas temperature at outlets
from engine cylinders, rotational speed of turbo-blower,
exhaust gas temperature before and behind the turbo-blower,
supercharging air pressure and temperature, ship draught
the integrated computer-based system for measuring and re-
cording operational parameters of ship power plant devices
- NORLIGHT Integrated Ship Monitoring and Control, I/O
Station A 01 ER, NORIS Tachometer GMBH & Co.
for the measuring of ME fuel consumption a calibrating
tank of 11dm
3
capacity being an element of the standard
equipment of ship power plant, and a stop-watch.
By making use of the standard measuring instrumentation
its usefulness had to be assessed regarding its accessibility
and easiness of read-out taking, and measurement accuracy of
selected parameters.
The accessibility and easiness of read-out taking is obvious.
The standard measuring systems are tted with easily readable
indicators, usually in an analogue or digital form or in the form
of messages displayed on a computer monitor.
In the case of investigations of such complex technical ob-
jects as ships, indication and measurement accuracy should be
considered in a way different from that used in direct metrolo-
gical measurements. In experimental investigations of complex
objects the variability of measured quantity is assumed to be
itself a basic source of errors [2]. The difference results from
an inuence of disturbing factors on the investigated object.
Additionally, errors of the measuring method itself (including
read-out errors) should by taken into consideration. In Fig.1
is shown the example distribution of measurement results
within a sample, location of the mean value of the sample, as
well as the range of changes resulting from variability of the
object and inaccuracy of the measuring method. In the case of
long-lasting measurements (e.g. of fuel consumption or a large
number of repetitions) one should also analyze errors resulting
e.g. from a change of service conditions during measurement
taking. The whole range of values of resultant factors is called
the variability measure of results [2].
The errors resulting from the above mentioned sources can
be numbered among random ones. As usually assumed, they
are normally distributed that makes it possible to state that in
accordance with the Gauss law the average from the sample is
the most probable value of the resultant factor obtained from
multiple measurements :
(1)
where :
z
j
- a single measurement result
r - number of repetitions of a measurement.
Fig. 1. Diagram of position of variability measures
and measurement error for an example sample of results :
z
j
- a single measurement result,
m
- measurement method error,
o
- variability measure of an object,
zw
- variability measure of results,
z
j
- average value from a sample.
Possible systematic errors resulting from individual features
of measuring instruments should be eliminated by introducing
some corrections to read-out values on the basis of calibration of
the measuring instruments. Also, should be rejected the results
deemed to be loaded by rough errors, i.e. those signicantly
differing from the average from read-out values, resulting from
mistakes or other hardly identiable causes.
The basic variability measure of measurement results is the-
ir variance [2]. In the case in question, is of interest the variance
of the whole set of results coming from realization of the scope
of experiment. For the same number of repetitions, r, for every
point of the scope the following can be written [2, 3, 4] :
(2)
where :
m number of measurement points
r number of repetitions in a measurement point
) u (
j
z result of a single measurement
) u (
z average value in a measurement point.
The root-mean-square error of results or standard
deviation is equal to the square root of variance [2] :
(3)
The standard deviation of the average, i.e.
the mean standard error of the average
is described by means of the following expression :
(4)
and the relative mean square error, i.e. the variability
coefcient of results by the expression :
(5)
where :
) u (
r
z
average from a sample of results.
In the case of the direct measurements of the independent
quantities x
1
, x
2
, ... x
n
, the root- mean-square error
is equal to [5] :
(6)
whereas the mean square error of the functions
of direct measurements of the kind : z = f(x
1
, x
2
,... x
n
,),
is equal to [5] :
(7)
Discussion on usefulness of the standard measuring in-
strumentation intended for energy measurements on ships was
carried out on the example of measuring the ship speed and ME
fuel consumption. Accuracy of measurements and measurement
method was assumed an assessing criterion.
The measurements were performed in accordance with
a static, determined, multi-factor, rotational and uniform
program of the technical experiment, described in [2], which
contains 13 measurement points. The measurement accuracy
was assessed in compliance with the above presented principle
of interpretation and by using the formulae (2 16).
=
=
r
1 j
j j
z
r
1
z
= =

=
m
1 u
r
1 j
2 ) u ( ) u (
j
2
) z z (
) 1 r ( m
1
) z (
= =

= =
m
1 u
r
1 j
2 ) u ( ) u (
j
2
) z z (
) 1 r ( m
1
) z ( ) z (
= =

=
m
1 u
r
1 j
2 ) u ( ) u (
j 2 z
) z z (
) 1 r ( r m
1
r m
) z (
% 100
z
) z (
) u (
=
r
) x ( ... ) x ( ) x ( ) z (
n
2
2
2
1
2
+ + + =
) z ( =
) x (
) x (
f
) x (
) x (
f
2
2
2
2
1
2
2
1
+
( )
( )
) x (
) x (
f
...
...
n
2
2
n

+
( )
25
ACCURACY ASSESSMENT
OF SHIP SPEED MEASUREMENTS
The accuracy assessment of ship speed measurements was
performed for the results obtained from two measurement series
(no. 2 and 4 according to the same notation as that used in me-
asurement reports from shipboard tests), carried out in similar
sea conditions, which are presented in the table below:
No. of
measurement
Ship speed v[kn]
Series no. 2 Series no. 4 On average
u
) u
1
( ) u
1
(
z v =
) u
2
( ) u
2
(
z v =
) u ( ) u (
z v =
1
2
3
4
7.0
8.4
8.9
11.3
7.0
8.7
8.6
11.0
7.00
8.55
8.75
11.15
5
6
7
8
10.5
7.4
10.4
7.8
10.5
7.0
10.2
7.6
10.50
7.20
10.30
7.70
9
10
11
12
13
9.2
9.1
9.1
9.1
9.0
8.9
8.95
9.0
8.95
8.9
9.05
9.03
9.05
9.03
8.95
) u (
r
z 8.94

After calculations by means of the formulae (2), (3), (4)
and (5) the following results were obtained :
variance of measured values of ship speed
2
(v) = 0.0258
standard deviation of (v) = 0.16 knot
standard deviation of the average value
z
= 0.032 knot
variability coefcient of measured values = 1.8%.
ACCURACY ASSESSMENT OF ME FUEL
CONSUMPTION MEASUREMENTS
The accuracy assessment of fuel consumption measure-
ments was performed for the results obtained in average sea
conditions, presented in the table below:
No. of
measurement
Fuel consumption per hour B
h
[kg/h]
u
) u ( ) u (
h1
z B =
1
) u ( ) u (
h2
z B =
2
) u ( ) u (
h
z B =
1
2
3
4
50.22
75.27
114.13
183.57
51.84
83.66
110.13
182.09
51.03
79.47
112.13
182.83
5
6
7
8
169.91
52.04
137.25
74.02
170.36
55.44
134.85
74.31
170.14
53.74
136.05
74.17
9
10
11
12
13
102.33
101.20
100.27
101.20
100.91
102.43
101.25
98.91
101.25
102.32
102.43
101.23
99.59
101.23
101.65
) u (
r
z 105.05
After calculations by means of the formulae (2), (3), (4)
and (5) the following results were obtained :
variance of measured values of fuel consumption

2
(B
h
) = 4.33
standard deviation of (B
h
) = 2.1 kg/h
standard deviation of the average value
z
= 0.41 kg/h
variability coefcient of measured values = 2.0%.
ACCURACY ASSESSMENT
OF THE MEASUREMENT METHOD
FOR SHIP SPEED
As the ship speed measurement is direct, its accuracy
depends on accuracy of measuring instrument and external
disturbances inuencing the investigated object. The mean
square error of measurement method is determined by using
the relation (6). For a single measured quantity (ship speed v)
it takes the form :
(8)
hence it is equal to the square root
from variance of the measured values :
(9)
where :

To calculate the mean square error, were taken the measu-
rement results at the ME rotational speed n = 830 rpm and
propeller pitch H=80% for the sea conditions in centrum of
measurement series 2 and 4, presented in the table below :
No. No. of measurement series v [kn]
1 9.2
2 2 9.1
3 9.0
4 8.9
5 4 9.0
6 8.9
On average v 9.02
Number of repetitions r 6
After calculations by using the formulae (9) and (5)
the following was obtained :
* the mean square error of the ship speed measurement
method (v) = 0.12 kn
* variability coefcient of measurement results = 1.3 %.
ACCURACY ASSESSMENT
OF THE MEASUREMENT METHOD
FOR ME FUEL CONSUMPTION
The ME hourly fuel consumption is the direct measurement
function described by the formula :
[kg/h] (10)
where :
V = 11 dm
3
volume of calibrating tank
t
=
15
(t 15)0.00064 [kg/dm
3
] fuel density
in temperature of measurements
15
= 0.875 kg/dm
3
fuel density at 15C (standard density)
) v ( ) v (
2
=
=
r
1 i
2
i
) v v (
1 r
1
) v (

=
t
h
V 3600
B
v
i
result of a single read-out of ship speed
v average speed in a measurement point
r number of repetition in a measurement point.

After inserting constant values the formula takes the form :
[kg/h] (11)
The mean square error of the fuel consumption measure-
ment method is calculated on the basis of the relation (7). In
the case in question it is described as follows :
(12)
Values of the partial derivatives of the function (11), with
respect to fuel temperature and measurement time, after inser-
ting the average measured values (see the table below), are as
follows, respectively :
(13)
(14)
The mean square error
of fuel temperature measurements is equal to :
(15)
where :
The mean square measurement error of consumption time
of fuel contained in calibration tank volume is equal to :
(16)
where :
i
result of a single read-out of consumption time of fuel
contained in the calibrating tank volume
average consumption time of fuel contained in calibrating
tank volume
r number of repetitions in a measurement point.
To determine the mean square error of fuel consumption the
measurement results at the ME rotational speed n = 830 rpm
and propeller pitch H = 80% for the sea conditions in centrum
of 2 and 4 measurement series, presented in the table below,
were applied :
No.
No. of
measurement
series
t B
h
min sec C kg/h
1
2
5
5
5
32.4
39.2
36.9
40
40
41
102.33
2 100.27
3 100.91
4
4
5
5
5
31.8
43.6
32.2
40
40
41
102.52
98.91
102.32
5
6
On average 5 36.0 40.3 101.21
Average time [s] 336.0
Number of repetitions r 6
After the calculations by using the formulae (12) and (5)
^ the mean square error of the fuel consumption measurement
method (B
h
) = 1.44 kg/h
^ variability coefcient of results = 1.4 %.
ACCURACY ASSESSMENT
OF MEASUREMENTS
OF SHIP PROPELLER SHAFT TORQUE
Torque measurements were carried out with the use of
special instruments installed by the measuring team. For the
measurements was used the PHILIPS PR 9914/01-NC 9408499
strain-gauge torque meter with analogue read-out system.
Accuracy assessment of torque measurements was per-
formed for the results obtained from two measurement series
(no. 2 and 4 - according to the same notation as that used in
the measurement reports from shipboard tests) carried out in
similar conditions, presented in the table below :
No. of
measurement
Torque M[Nm]
u
) u ( ) u
1 1
(
z M =
) u ( ) u
2 2
(
z M =
) u (
) u (
z M =
1
2
3
4
3460
5895
6023
10509
3332
5831
5575
10317
3396
5863
5799
10413
5
6
7
8
8971
3845
8843
4357
8715
3909
8843
4229
8843
3877
8843
4293
9
10
11
12
13
6216
6248
6280
6280
6280
6216
6184
6151
6087
6023
6216
6216
6216
6184
6152

) u (
M
r

6332

After calculations by using the formulae (2), (3), (4) and (5)
* variance of measured torque values
2
(M) = 18003.8
* standard deviation of (M) = 134.2 Nm
* standard deviation of the average value
z
= 26.3 Nm
* variability coefcient of results = 2.1%.
USEFULNESS ASSESSMENT
OF STANDARD INSTRUMENTS
FOR ENERGY MEASUREMENTS
O The obtained results of the assessment of measurement ac-
curacy, performed with the use of ship standard instrumen-
tation, are contained within the interval of 2% variability
coefcient, and the assessment of accuracy of measurement
methods of 1.5%. It makes it possible to conclude that the
standard measuring instruments are fully useful for carrying
out energy measurements on ships.
O The variability coefcient of measurement results obtained
by means of the special instruments for torque measuring,
amounts to 2.1%, hence it is not greater than that in the case
of the standard measuring instruments installed on ships.
=
t 34 . 25 6 . 35030
B
h
) (
B
) t (
t
B
) B (
2 h 2
2
h
h

=
( )
2
( )
34 . 25
t
B
h
= 0.07542
2
h
6 . 35030 t 34 . 25 B
= 0.3012
=
r
i
i
) (
r
) (
1
2
1
1
=
r
1 i
2
i
) t t (
1 r
1
) t (
t [C] temperature of fuel
[s] consumption time of fuel contained
in calibrating tank volume.
t
i
result of a single read-out of fuel temperature
t average result of fuel temperature measurement
r number of repetitions in a measurement point.
27
O the accuracy of the measurement methods, which is higher
than that of measurement results, conrms that the reaso-
ning presented in this paper is correct.
Similar energy measurements were carried out in the labo-
ratory of combustion engines of Ship Power Plant Department,
Gdynia Maritime University. As expected, the obtained results
were found even more accurate in view of the stable operatio-
nal conditions and loading state of the laboratory engine. For
instance the variability coefcient value of measurement results
of hourly fuel consumption, equal to 0.65% was obtained.
To conclude one should state that the standard measurement
instrumentation which is installed as a rule onboard contempo-
rary ships, can be successfully used for energy measurements.
In particular, the measurements performed with the use of the
standard instrumentation are of the following features :
C a sufcient accuracy
C no special instruments are required to be installed
(a decrease of cost)
C they can be carried out by ships crew itself hence no ad-
ditional measuring team is required (a decrease of cost)
C measurement results may be used in operational and dia-
gnostic analyses of technical state of ships propulsion
system.
NOMENCLATURE
B
h
[kg/h] - hourly fuel consumption
H [%] - pitch of CP propeller (Controllable Pitch Propeller)
KBN - The State Committee for Scientic Research
M [Nm] - torque
n [rpm] - rotational speed
r - number of measurement repetitions
t [
o
C] - temperature
u - number of a measurement point
v [kn] - ship speed
V [dm
3
] - volume
z
j
- result of a single measurement
j z - average value from a sample of measurement results
) u (
j
z - result of a single measurement point
z
(u)
- value in a measurement point
) u (
z
- average value in a measurement point
) u (
r
z - average value from a sample of measurement results
m
- measurement method error
o
- variability measure of an object
zw
- variability measure of results
[kg/dm
3
] - density
- relative mean square error
2
- variance
(z) - mean square error
z
- standard deviation of average value
[s] - time
BIBLIOGRAPHY
1. Adamkiewicz H.G.: Statistics, applications to economy (in
Polish). The Centre for Personnel Advising and Improving Co
(Orodek Doradztwa i Doskonalenia Kadr Sp. z o.o.). Gdask.
1996
2. Polaski Z.: Planning experiments in engineering (in Polish).
State Scientic Publishing House (Pastwowe Wydawnictwo
Naukowe). Warszawa. 1984
3. Strzakowski A., liyski A.: Mathematical methods for
elaboration of measurement results (in Polish), State Scientic
Publishing House (Pastwowe Wydawnictwo Naukowe).
Warszawa. 1969
4. Szydowski H., Kaczmarek W., Kotowska M., Kozak A.,
Kudyska J.: Theory of measurements (in Polish). State
Scientic Publishing House (Pastwowe Wydawnictwo
Naukowe). Warszawa. 1981
5. Taylor J.R.: Introduction to measurement error analysis (in
Polish). Scientic Publishing House (Pastwowe Wydawnictwo
Naukowe PWN). Warszawa. 1999
6. Volk W.: Applied statistics for engineers (in Polish). Scientic -
-Technical Publishers (Wydawnictwo Naukowo - Techniczne).
Warszawa. 1985
7. Grski Z.: A method to choose settings of the elements of ship
main propulsion system tted with controllable pitch propeller
in the aspect of operational effectiveness of ship power plant
(in Polish). Doctoral thesis under supervision of Prof. Romuald
Cwilewicz. Mechanical Faculty, Gdynia Maritime University.
Gdynia. 2003
Zygmunt Grski, D.Sc., Eng.
Assoc. Prof. Romuald Cwilewicz
Faculty of Marine Engineering,
Morska 81/87
81-225 Gdynia, POLAND
CYLINDER 2006
On 26-28 September 2006 in Zakopane, Polish town at
the foot of Tatra Mountains, was held 16th country-wide
technical scientic conference on :
Testing, construction, production and
operation of hydraulic systems
during which were presented 28 papers whose authors and
co-authors were research workers from 13 scientic research
centres and 5 industrial institutions. The largest share in re-
alization of the Conference program (6 papers) was brought
in by the authors representing KOMAG, the Centre for
Mechanization of Mining in Gliwice. Foreign guests from
the LUBRICANT, a Czech rm, presented one paper.
Photo : Cezary Spigarski
SHIP PROPULSION SYSTEM
Screw propeller
Ship propulsion system is represented rst of all by a propel-
ler cooperating with propulsion engine. To calculate instanta-
neous ship service speed it is necessary to know hydrodynamic
characteristics of the propeller (derived from model tests or
calculated by using approximate methods) and the characteri-
stics which describe engine load area.
In the calculations were used the characteristics of B-Wa-
geningen screw propellers [4,5], from which propellers thrust
and torque were calculated.
Thrust of the propeller placed behind the ship hull
is expressed as follows :
(27)
where :
R
C
total resistance of ship sailing in rough waters
t thrust deduction factor.
Thrust of free propeller can be calculated from the formula :
(28)
where :
Calculation of the mean long-term
service speed of transport ship
Continuation ot the rst part of the paper published
in the Polish Maritime Research No.4(50), October 2006
Part II
Service speed of ship sailing
on regular shipping route in real weather conditions
Tadeusz Szelangiewicz
Katarzyna elazny
Szczecin University of Technology
ABSTRACT
Service speed obtainable by a ship in real weather conditions when sailing on a given shipping route, is
one of the major parameters which have great impact on ship operation costs. The so far used, very ap-
proximate method of service speed prediction based on service margin, is very little exact. In this paper
a new method based on additional ship resistance dependent on mean statistical parameters of wave and
wind occurring on a given shipping route, is presented. The mean long-term service speed is calculated on
the basis of the calculated additional resistance and the screw propeller and propulsion engine parameters.
Also, a new denition of service margin and a way of its calculation is presented apart from the results of
the mean service speed calculation depending on ships type and size and shipping route.
Keywords : ship service speed, wind, waving, shipping route, service margin, long-term prediction.
K
T
thrust coefcient which for typical B-Wageningen
propellers of given values of the pitch ratio (P/D),
blade area ratio (A
E
/A
0
), number of blades (Z) is
approximated by using the expression :
(29)
where :
A
0
, A
1
, A
2
, A
3
coefcients of the polynomial describing
thrust characteristics, depending on (P/D), (A
E
/A
0
), Z
J advance coefcient :
(30)
w(V) wake fraction depending on the ship speed V.
The propeller when working generates the torque Q :
(31)
where :
K
Q
torque coefcient which like the thrust coefcient
can be expressed for a given propeller
in the following form :
K
Q
= B
0
+ B
1
J + B
2
J
2
+ B
3
J
3

(32)
where :
B
0
, B
1
, B
2
, B
3
coefcients of the polynomial describing
torque characteristics, depending on (P/D), (A
E
/A
0
), Z.
t 1
R
T
C
=
2
p
4
p w T
n D K T =
3
3
2
2 1 0 T
J A J A J A A K + + + =
( ) [ ]
p p
n D
V w 1 V
J

=
2
p
5
p w Q
n D K Q =
D
p
propeller diameter
n
p
propeller rotational speed
29
Free-propeller efciency (free from ships hull) is equal to :
(33)
The hydrodynamic characteristics of propeller achieved
from model tests [4,5] are determined for a given Reynolds
number and propeller surface state. In order to use the cha-
racteristics for the behind-the-hull propeller were introduced
relevant corrections associated with inuence of Reynolds
number and real state of propeller surface, which makes it
possible together with knowledge of ship hull surface state
to investigate inuence of ship and propeller ageing on ship
service speed.
The expressions (27) and (28) for propeller thrust are valid
for ship sailing in still and rough waters provided its oscillating
and relative motions are so small that propeller emerging does
not occur. During sailing in waves at large oscillating and rela-
tive motions the propeller operates in highly aerated water or
emerges. It generates thrust variations and drop of mean effec-
tive thrust value relative to that in still water (even if the ship
goes at constant speed and number of propeller rotations).
The thrust decrease is caused a.o. by an inuence of water
particles being in oscillating motion, on wake, as well as by
propeller emerging resulting from large relative motions of the
ship in waves. The thrust decrease due to ship motions in waves
has been discussed in various publications where approximate
formulae for estimating the inuence of relative motions on
operational parameters of propeller are included [3].
The propeller thrust decrease due to ship relative motions in
waves is dened by means of the following coefcient [3] :
(34)
where :
( )
p
Tw
R
h
K thrust coefcient of emerging propeller (the
quantities h
p
and R are shown in Fig.11)
K
T
thrust coefcient of fully immersed propeller.
Changes of values of the coefcient
T

in function of
( )
p
R
h
are shown in Fig.12, [6].
Fig. 11. The propeller draught h
p
as dened in the equation (34) .
Fig. 12. Thrust decrease during propeller emerging [6] .
In simulative propeller thrust calculations, the inuence
of wave parameters on propeller operation effectiveness was
taken into account (detailed analysis of the phenomenon is
given in [9]) whereas the propeller emergence was mitigated
by intentional reduction of ship speed.
Propulsion engine
The behind-the-hull propeller loads the ship engine
with its torque (31). The relation of the propeller torque
and propulsion engine output power is as follows :
(35)
where :
P
D
power delivered to the propeller
n rotational speed of engine (for slow-speed engines
if no reduction gear is applied : n = n
p
) :
P
D
= N
LW

R

P
(36)
Engine output power for a designed ship is so selected as to
obtain the propulsion system working point of the value equal
to 0.85 N
n
at the design (contractual) speed in still water. In
real weather conditions, when wind and waves affect the ship
and the additional resistance R appears, then the propulsion
system working point changes its location within the propulsion
engines load area. By controlling fuel charge (and this way also
rotational speed of engine and propeller) the working point can
be positioned in the maximum continuous rating area or the
limited rating area for engine overloading (Fig.13).
Therefore, in order to nd an instantaneous speed at which
a given ship would sail in considered weather conditions it is
necessary to know the engine load area which is constrained by
relevant characteristics. This area indicates where the propul-
sion system working point is to be located. For instance the
area of a Sulzer engine is shown in Fig.13.
The particular areas are limited by the engine performance
characteristic curves in the following form :
N = k
m
n
m
(37)
where :
N engine output power
k
m
coefcient for a given characteristic curve
n engine rotational speed
=
2
J
K
K
Q
T
( )
T
p
Tw
T
K
R
h
K
=
R
h (t)
p
T
R
h
p
1.0
0.5
0.5 1.0
Thrust decrease due to propeller emerging

Thrust decrease due to :
- propeller emerging
- stern wave system generating
- variable lifting force on propeller blades
n 2
P
Q
D
=
N propulsion engine output power
R
relative rotative efciency
LW
shaft-line efciency
P
reduction gear efciency, if applicable.
m exponent depending on a type of engine of a given
producer :
- e.g. for SULZER RTA 52, RTA 62, RT 72, RTA 84
slow speed engines it is equal to :
m = 0 for constant rated or maximum output power
m = 1 for constant torque characteristic curve
m = 2.45 for overloading characteristic curve.
Fig. 13. The load area of an example Sulzer propulsion engine [2] .
Service speed of ships tted with engines of other producers
can be calculated if engine characteristics limiting
its load area are known.
The searching for of a working point of propulsion system
and its parameters can be effected by taking into account vario-
us criteria e.g. that for maintaining a given or maximum speed,
or also by using the criterion for minimum fuel consumption
or maximum efciency of whole propulsion system.
Intentional reduction of ship speed in view
of hazardous wave-generated phenomena
During ship sailing in rough water ships oscillating mo-
tions and their derivatives i.e. velocities and accelerations are
the direct effect of waves. The secondary phenomena which
accompany the oscillating motions are : shipping of water on
deck, propeller emerging, pounding of wave to ship bottom
and sides (slamming), worsening of ship stability and ma-
noeuvrability, additional dynamic loads on hull structure. The
oscillating motions and accompanying phenomena depend on
parameters of ship hull and waves as well as the ship speed V
and course relative to waves (the angle
w
). The phenomena,
especially if very intensive, may lead directly to averages and
disasters at sea. It is possible to mitigate the phenomena, e.g.
ship rolling - by changing ships course relative to waves (the
angle
w
), reducing ships speed V, or simultaneous changing
ships course and speed.
In predicting the mean service speed was made the assump-
tion that ships speed reduction will be performed if :
(38)
where :
Z
U mean statistical value of the wave-generated phe-
nomenon Z considered hazardous to ship
Zdop
U permissible value of the wavegenerated pheno-
menon Z, at which the sailing ship is still safe.
In the performed calculations the ship course relative to
waves was kept unchanged (though its changing is always
possible), since in predicting the mean long-term ship service
speed it was assumed that the ship heads the course resulting
from the selected shipping route.
In assessing ship performance in waves and making deci-
sion on possible reduction of its speed, the following pheno-
mena were taken into consideration: rolling, pitching, vertical
accelerations, horizontal transverse accelerations, shipping of
water on deck, slamming, propeller emerging. The calculation
methods of the phenomena and the assumed acceptance criteria
are presented in [1, 5, 8].
The points at which values of the parameters of ship sea-
-keeping qualities were calculated, are presented in Fig. 14, and
in Fig.15, in the polar diagrams is shown an inuence of ship
speed and wave direction relative to ship on some phenomena
induced by waves. The zone of unsafe intensity, i.e. where the
sea-keeping criteria are exceeded , is marked red; yellow colour
warns about a real hazard; green colour means the safe zone.
It can be observed that reducing the speed one can mitigate
intensity of the phenomena in question.
Fig. 14. The points at which ship sea-keeping parameters were calculated :
1 for slamming, 2 for water shipping onto deck, and bow accelerations,
3 for propeller emerging, 4 for accelerations at wheelhouse.
Wave :
H
s

= 8.5 m ; T
1

= 10.1 s ; = 0
Water shipping on deck
V = 17 w
course angle = 40
Rolling
V = 15 w
course angle = 135
Horizontal transverse
accelerations at wheelhouse
V = 10 w
course angle = 100
Vertical accelerations
at wheelhouse
V = 17.5 w
course angle = 60
Fig. 15. Inuence course angle and speed of K1 ship on its selected sea-
keeping qualities. Green - safe operation of ship. Yellow - warning on a real
hazard (lower value of a relevant criterion is exceeded). Red - ships safety
is endangered (upper value of a relevant criterion is exceeded) .
Zdop Z U U >
3
1
2
4
E - engine instantaneous operation area ( ) yellow
- engine sea trial operation area ( ) deep blue C
D - engine optimum still-water operation area ( ) blue
B - engine overloading area ( ) red
A - continuous rating area (marked ) green
- nominal torque M
n
Contract maximum continuous rating - CMCR
120
110
100
90
80
70
60
50
40
35
[% CMCR] Relative engine rotational speed
60 70 80 90 100 110 75 65 85 95
[
%
C
M
C
R
]
R
e
l
a
t
i
v
e
e
n
g
i
n
e
o
u
t
p
u
t
p
o
w
e
r
M = const
n
B
D A C
E
A
Propeller curve
31
A METHOD FOR PREDICTING MEAN
STATISTICAL SERVICE SPEED OF A SHIP
SAILING ON A GIVEN SHIPPING ROUTE
Instantaneous service speed of ship
During ship motion in rough water, apart from still-water
resistance also additional forces due to wind, waves and possi-
ble surface sea current act on the ship. These actions generate,
apart from an additional resistance, a transverse force and
a moment tending to rotate the ship around its vertical axis.
The transverse force results in ship drifting, and the moment
in ship course changing. Ships passive rudder is to be laid
appropriately to keep ship course constant under action of the
external rotating moment. Under the assumption that the ship
course has to be maintained , from the solution of the following
non-linear equations :
Y
A
(V) + Y
W
(V) + R
y
(V,) + Y
R
(V,,
R
) = 0
(39)
M
A
(V) + M
W
(V) + M
z
(V,) + M
R
(V,,
R
) = 0
together with the relevant equations describing additional
resistance forces, (presented below), for given values of the
ship speed V, course angle , wind parameters (V
A
,
A
), wave
parameters (H
S
, T, ) and possible sea current parameters
(V
C
,
C
), the following quantities can be obtained :
Schematic diagram of the complete calculation algorithm
of ,
R
and R
C
for given values of the ship motion parameters
(V , ) , wind parameters (V
A
,
A
), wave parameters (H
S
, T, )
and current parameters (V
C
,
C
), was presented in [9].
The instantaneous ship service speed in variable weather
conditions is calculated from the solution of the set of non-
linear equations in such a way as to obtain the propulsion sys-
tem working point laying within the engine continuous rating
zone when total ship resistance and propeller thrust , as well
as propeller torque and driving engine torque become equal to
each other, respectively . Making use of the equations (4), (5),
(27 36) one obtains the set of two non-linear equations :

(40)
where :
J advance coefcient (30)
R
C
a function of total ship resistance, depending on the ship
speed V, ship course angle , wave parameters H
S
, T,
and wind parameters V
A
,
A
N driving engine output power determined by its charac-
teristics valid within respective intervals of its rotation
number n.
To solve the set of equations (40) and determine instan-
taneous ship service speed it is necessary to know the total
ship resistance R
C
which depends not only on the statistical
parameters of waves and wind occurring on a given shipping
route but also on current ship speed, course and drift angle.
Since all the quantities depend on random parameters of waves
and wind the total ship resistance should be calculated for all
statistical parameters of waves and wind occurring on a given
shipping route. The calculation algorithm of instantaneous ship
speed is presented in [9].
Mean statistical service speed of ship sailing
on a given shipping route
During the long-term sailing of ship on an assumed ship-
ping route the additional resistance due to wave action will
depend not only on a wave height (and period) but also on
geographical directions: of wave, , and ship course , . Also
the additional resistance due to wind will depend, apart from
the wind speed V
A
, on the directions
A
and . It means that
the additional resistance and also ship speed will depend on the
values of the parameters of waves (H
S
, T
1
, ), wind (V
A
,
A
)
and ship motion (V, ), which may occur on a given shipping
route within a long period of time.
In the case in question the surface sea current is consi-
dered to be a determinate phenomenon of the mean speed V
C

and direction angle
C
. If a ship is assumed to enter a region
of large-scale surface currents then the current action will be
included into ship still-water resistance.
Therefore the occurrence probability of the additional
resistance R of a given value, as well as the speed V which
can be reached at occurrence of that additional resistance,
depends on :
C shipping route and probability of staying the ship in parti-
cular sea areas
C the statistical parameters of waves , (H
S
, T
1
, ), and wind,
(V
A
,
A
), and probability of occurrence of the parameters
in given sea areas
C probability of occurrence of the ship motion parameters
(V, ).
The probability of being the ship in a given situation
when sailing in waves on a given route, is as follows :
p
w
= f
A
f
S
f
f
HT
f
V
f
(41)
where :
f
A
probability of staying the ship in the sea area A
f
S
probability of staying the ship in the sea area A du-
ring the season S
f
probability of occurrence of the wave direction in

the sea area A during the season S
f
HT
probability of occurrence of the wave of the para-
meters (H
S
, T
1
), propagating from the direction
f
V
, f
probability of the event that the ship moves with the

speed V and heads the course , respectively.
In a similar way can be expressed the probability p
A
of being
the ship in a given situation associated with wind state. In the
calculations of additional resistance due to wind and waves it
was assumed that wind speed and wave height are mutually cor-
related, hence p
w
= p
A
. As the event of being the ship in a given
situation described by (41) will result in generating an addi-
tional resistance and achieving a determined speed, hence :
p
w
= p
R
= p
V
(42)
where :
p
R
partial occurrence probability (in given conditions) of
additional resistance
p
V
partial occurrence probability (in given conditions) of
instantaneous ship service speed.
Values of the additional resistance due to wind, R
XA
, and
that due to waves, R
XW
, depend on random parameters of wind
ship drift angle
R
ship passive rudder angle
R additional ship resistance due to wind,
waves, current and passive rudder
R
C
total ship resistance to motion.
( )
0
n D 2
N
J B J B J B B
3 5
p w
P R LW 3
3
2
2 1 0
=

+ + +
( )
0
t 1
R
n D J A J A J A A
C 2 4
p w
3
3
2
2 1 0
=
+ + +
and waves. Therefore the same values of R
XA
and R
XW
can oc-
cur for different values of the parameters V
A
,
A
, H
S
, T
1
, ,
V, . For each of the values of this way calculated additional
resistance, a value of ship speed is calculated (the criteria con-
cerning sea-keeping qualities are simultaneously examined to
execute possible speed reduction in order not to violate them
at given wave and wind conditions).
The total probability P
TV
of achieving the ship speed V at
a given value of the additional resistance R, is as follows :
(43)
where :
V
i
(R
i
) instantaneous ship service speed in function
of instantaneous additional ship resistance
n
A
, n
S
, n
, n
HT
, n
V
, n

numbers of sea areas (crossed by
a given ship), seasons, wave directions, wave parameters,
ship speeds and courses, respectively.
By calculating the distribution function of occurrence
probability, f(V
i
), of the instantaneous ship speed V
i
the mean
long-term ship service speed for a given shipping route, can
be determined as follows :
(44)
where :
n
V
number of intervals containing similar values
of the instantaneous ship service speeds.
On the basis of the presented calculation formula for the
mean ship service speed, (44), were performed relevant cal-
culations and analyses for example ships and shipping routes,
whose results will be presented in the 3
rd
part of the paper.
NOMENCLATURE
( )
0
A
E
A
propeller blade area ratio
A
0
, A
1
, A
2
, A
3
coefcients of the polynomial describing thrust
characteristics
B
0
, B
1
, B
2
, B
3
coefcients of the polynomial describing torque
characteristics
D
p
propeller diameter
f
A
probability of staying the ship in the sea area A
f
HT
probability of occurrence of the wave of the parameters
(H
S
, T
1
), propagating from the direction
f
S
probability of staying the ship in the sea area A during the
season S
f
V
, f
probability of the event that the ship moves with the speed
V and heads the course , respectively
f
A
probability of occurrence of a given wind direction
H
S
signicant wave height
J advance coefcient
K
T
thrust coefcient
K
Tw
thrust coefcient of emerging propeller
K
Q
torque coefcient
k
m
coefcient of a given characteristic curve of engine
performance
N propulsion engine output power
N
n
nominal output power of propulsion engine
n rotational speed of engine
n
n
nominal rotational speed of engine
n
p
rotational speed of propeller
( )
D
P
propeller pitch ratio
P
D
power delivered to propeller
P
TV
combined probability of reaching a given value of
instantaneous ship service speed at occurrence of a given
value of instantaneous additional ship resistance
p
w
partial probability of staying the ship in a given situation
p
R
partial occurrence probability of instantaneous additional
resistance (in given conditions)
p
v
partial occurrence probability of instantaneous ship
service speed (in given conditions)
Q propeller torque
R
C
total ship resistance to motion
R
xA
, R
yA
, M
zA
mean wind-induced forces exerted to going ship
(R
xA
- additional ship resistance due to wind)
R
xR
, R
yR
, M
zR
passive rudder forces (R
xR
- additional ship
resistance due to rudder)
R
xW
, R
yW
, M
zW
mean wave-induced drift forces (R
xW
- additional
ship resistance due to waves)
T free-propeller thrust
T
1
mean characteristic wave period
t thrust deduction factor
U
z
mean statistical value of the wave-induced phenomenon Z
V ship speed
V
A
wind speed
V
C
sea current speed
V
E
mean statistical ship service speed
w wake fraction
Z number of propeller blades
ship drift angle
T
coefcient of propeller thrust decrease
A
geographical direction of wind
C
geographical direction of sea current
R additional ship resistance due to weather conditions
R
passive rudder angle
0
free-propeller efciency
P
reduction gear efciency
R
propeller rotative efciency
LW
shaft-line efciency
rudder aspect ratio
geographical direction of wave
w
water density
geographical direction of ship course
BIBLIOGRAPHY
1. Dudziak J.: Theory of ship (in Polish). Maritime Publishing
House (Wydawnictwo Morskie), Gdask, 1988
2. SULZER : General Technical Data for Marine Diesel Engines,
1986
3. Holtrop J.: A Statistical Re-analysis of Resistance and
Propulsion Data. International Shipbuilding Progress, No. 363,
1984
4. Holtrop J., Mennen G.G.J.: An Approximate Power Prediction
Method. International Shipbuilding Progress, Vol. 29, No. 335,
1982
5. Karppinen T.: Criteria for Sea-keeping Performance
Predictions. Technical Research Centre of Finland, Ship
Laboratory, ESPOO, Helsingfors, 1987
6. Minsaas K.J., Thon H.J., Kauczyski W.: Inuence of Ocean
Environment on Thruster Performance. Proc. of. Int. Symp.
Propeller and Cavitation, Supplementary volume. Shanghai
1986. The Editorial Ofce of Shipbuilding of China
7. Oosterveld M.W.C., van Oossanen P.: Further Computer-
Analyzed Data for the Wageningen B-Screw Series. International
Shipbuilding Progress, Vol. 22, No. 251/1975
8. Szelangiewicz T.: Ships operational effectiveness factor as
criterion of cargo ship design estimation. Marine Technology
Transactions. Polish Academy of Sciences, Branch in Gdask.
Vol. 11/2000
9. elazny K.: Numerical prediction of mean long-term service
speed of transport ship (in Polish). Doctoral thesis. Faculty
of Maritime Technology , Szczecin University of Technology.
2005.
( ) [ ]

= = = = =
=
A S HT V
n
1 A
n
1 S
n
1
n
1 T , H
n
1 V
n
1
i i Vi TV
R V P P
( )
=
=
=
=
V
i
V
i
n
1 i
TV
n
1 i
E
i TV
P
const R V P
V

Polish Maritime Research

Uploaded by

Copyright:

Available Formats

Polish Maritime Research

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Polish Maritime Research

Uploaded by

Copyright:

Available Formats

Address of Publisher

& Editor's Ofce :

4 POLISH MARITIME RESEARCH, No 1/2007

Thrust decrease due to propeller emerging

probability of occurrence of the wave direction in

probability of the event that the ship moves with the

You might also like