Professional Documents
Culture Documents
and Propulsion
Second Edition
J S Carlton
Global Head of Marine Technology and
Investigation, Lloyds Register
Copyright 2007, John Carlton. Published by Elsevier Ltd. All right reserved
The right of John Carlton to be identied as the authors of this work has been
asserted in accordance with the Copyright, Designs and Patents Act 1988
Permissions may be sought directly from Elseviers Science & Technology Rights
Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333;
email: permissions@elsevier.com. Alternatively you can submit your request online by
visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting
Obtaining permission to use Elsevier material
Notice
No responsibility is assumed by the published for any injury and/or damage to persons
or property as a matter of products liability, negligence or otherwise, or from any use
or operation of any methods, products, instructions or ideas contained in the material
herein. Because of rapid advances in the medical sciences, in particular, independent
verication of diagnoses and drug dosages should be made
ISBN: 978-07506-8150-6
Printed and bound in Great Britain by MPG Books Ltd, Bodmin Cornwall
06 07 08 09 10 10 9 8 7 6 5 4 3 2 1
The resistance extrapolation process follows Froudes
12.3.3 Direct model test hypothesis and the similarity law is followed. As such
the scaling of the residual, or wave making component,
Model testing of a ship in the design stage is an import- follows the similarity law
ant part of the design process and one that, in a great
many instances, is either not explored fully or is not RWship = RWmodel 3 (s /M )
undertaken. In the authors view this is a false economy,
provided that VS = VM , where = LS /LM .
bearing in mind the relatively small cost of model testing In general, the resistance is scaled according to the
as compared to the cost of the ship and the potential costs relationship
that can be incurred in design modication to rectify a
S
problem or the through life costs of a poor performance Rs = [RM RFM (1 + k)]3 + RFs (1 + k) + RA
optimization. M
S
General procedure for model tests = [RM FD ]3 (12.30)
M
Whilst the detailed procedures for model testing differ
in which
from one establishment to another the underlying gen-
2 M
eral procedure is similar. Here the general concepts are FD = 12 M VM SM (1 + k)(CFM CFS ) RA /3
discussed, but for a more detailed account reference can S
be made to Phillips-Birt (Reference 11). With regard to that is,
resistance and propulsion testing there are fewer kinds 2
FD = 21 M VM SM [(1 + k)(CFM CFS ) CA ]
of experiment that are of interest: the resistance test, the
open water propeller test, the propulsion test and the (12.31)
ow visualization test. The measurement of the wake The term FD is known as both the scale effect cor-
eld was discussed in Chapter 5. rection on resistance and the friction correction force.
The term RA in equation (12.30) is the resistance com-
Resistance tests ponent, which is supposed to allow for the following
factors: hull roughness; appendages on the ship but not
In the resistance test the ship model is towed by the
present during the model experiment; still air drag of
carriage and the total longitudinal force acting on the the ship and any other additional resistance component
model is measured for various speeds (Figure 12.22). acting on the ship but not on the model. As such its
The breadth and depth of the towing tank essentially non-dimensional form CA is the incremental resistance
governs the size of the model that can be used. Todds coefcient for shipmodel correlation.
original criterion that the immersed cross-section of the When (1 + k) in equation (12.30) is put to unity, the
vessel should not exceed one per cent of the tanks cross- extrapolation process is referred to as a two-dimensional
sectional area was placed in doubt after the famous Lucy approach since the frictional resistance is then taken as
Ashton experiment. This showed that to avoid bound- that given by the appropriate line, Froude at plate data,
ATTC or ITTC 1957, etc.
ary interference from the tank walls and bottom this
The effective power (PE ) is derived from the resist-
proportion should be reduced to the order of 0.4 per cent. ance test by the relationship
The model, constructed from parafn wax, wood or
glass-reinforced plastic, requires to be manufactured PE = RS VS (12.32)
to a high degree of nish and turbulence simulators
placed at the bow of the model in order to stimulate the Open water tests
transition from a laminar into a turbulent boundary layer The open water test is carried out on either a stock or
over the hull. The model is positioned under the carriage actual model of the propeller to derive its open water
and towed in such a way that it is free to heave and pitch, characteristics in order to derive the propulsion coef-
and ballasted to the required draught and trim. cients. The propeller model is tted on a horizontal
In general there are two kinds of resistance tests: driveway shaft and is moved through the water at an
the naked hull and the appended resistance test. If immersion of the shaft axis frequently equal to the
diameter of the propeller (Figure 12.23).
appendages are present local turbulence tripping is
The loading of the propeller is normally carried out by
applied in order to prevent the occurrence of uncon- adjusting the speed of advance and keeping the model
trolled laminar ow over the appendages. Also the revolutions constant. However, when limitations in the
propeller should be replaced by a streamlined cone to measuring range, such as a J -value close to zero or
prevent ow separation in this area. a high carriage speed needed for a high J -value, are
reached the rate of revolutions is also varied. The meas-
ured thrust values are corrected for the resistance of the
Ship resistance and propulsion 305
Figure 12.23 Propeller open water test using towing tank carriage
Propulsion tests of the propulsion test. In a similar way the local vari-
In the propulsion test the model is prepared in much the ation test can be interpolated to establish the required
same way as for the resistance test and turbulence stimu- torque and propeller rotational speed at self-propulsion
lation on the hull and appendages is again applied. For for the ship.
this test, however, the model is tted with the propeller In the extrapolation of the propulsion test to full scale
used in the open water test together with an appropri- the scale effects on resistance (FD ), on the wake eld
ate drive motor and dynamometer. During the test the and on the propeller characteristics need to be taken
model is free to heave and pitch as in the case of the into account. At some very high speeds the effects of
resistance test. cavitation also need to be taken into account. This can
In the propulsion test the propeller thrust TM , the pro- be done by analysis or through the use of specialized
peller torque QM and the longitudinal towing force F facilities.
acting on the model are recorded for each tested com-
bination of model speed VM and propeller revolutions
nM . Flow visualization tests
Propulsion tests are carried out in two parts. The rst Various methods exist to study the ow around the hull
comprises a load variation test at one or sometimes more of a ship. One such method is to apply stripes of an
than one constant speed whilst the other comprises a especially formulated paint to the model surface, the
speed variation test at constant apparent advance coef- stripes being applied vertical to the base line. The model
cient or at the self-propulsion point of the ship. The is then towed at Froude identity and the paint will smear
ship self-propulsion point being dened when the tow- into streaks along the hull surface in the direction of the
ing force (F) on the carriage is equal to the scale effect ow lines.
correction on viscous resistance (FD ), equation (12.31). In cases where the wall shear stresses are insuf-
The required thrust TS and self-propulsion point of cient tufts are used to visualize the ow over the hull.
the ship is determined from the model test using the In general, woollen threads of about 5 cm in length
equation: will be tted onto small needles driven into the hull
surface. The tufts will be at a distance of between 1
and 2 cm from the hull surface and the observation
TM 3 S
TS = TM + (FD F) (12.33) made using an underwater television camera. The inter-
F M action phenomenon between the propeller and ships
hull can also be studied in this way by observing the
In equation (12.33) the derivative TM /F is determined behaviour of the tufts with and without the running
from the load variation tests which form the rst part propeller.
Ship resistance and propulsion 307
Model test facilities facilities but is included here to give an idea of the range
of facilities available.
Many model test facilities exist around the world almost
all of which possess a ship model towing tank. Some of Two-dimensional extrapolation method
the model facilities available are listed in Table 12.7;
this, however, is by no means an exhaustive list of This as discussed previously is based on Froudes ori-
ginal method without the use of a form factor. Hence
Table 12.7 Examples of towing tank facilities around the the full-scale resistanceis determined
from
world (Reproduced with permission from Reference 55)
RS = (RM FD )3
S
Three-dimensional extrapolation method According to the ITTC 1987 version of the manual for
the use of the 1978 performance reduction method, the
The three-dimensional extrapolation method is based
relationship between the ship and model Taylor wake
on the form factor concept. Accordingly the resistance
fractions can be dened as
is scaled under the assumption that the viscous resist-
ance of the ship and its model is proportional to the w TS = (t + 0.04) + (wTM t 0.04)
frictional resistance of a at plate of the same length (1 + k)CFS + CF
and wetted surface area when towed at the same speed,
the proportionality factor being (1 + k) as discussed in (1 + k)CFM
Section 12.2. In addition it is assumed that the pres- where 0.04 is included to take account of the rudder
sure resistance due to wave generation, stable separation effect and CF is the roughness allowance given by
and induced drag from non-streamlined or misaligned
appendages follow the Froude similarity law. ks 1/3
The form factor (1 + k) is determined for each hull CF = 105 0.64 103
LWL
from low-speed resistance or propulsion measurements
when the wave resistance components are negligible. In The measured relationship between the thrust and
the case of the resistance measurement of form factor torque coefcient is corrected for the effects of friction
then this is based on the relationship: over the blades such that
R KTS = KTM + KT and KQS = KQM + KQ
(1 + k) = lim
Fn 0 RF where the factors KT and KQ are determined from
In the case of the propulsion test acting as a basis for the ITTC procedure as discussed in Chapter 6.
the (1 + k) determination then this relationship takes The load of the full-scale propeller is obtained from
the form the relationship
KT S CTS
F T /(T /F) =
(1 + k) = lim J 2 2D (1 t)(1 wTS )2
2
Fn 0 (F|T =0/R)RF
The low-speed measurement of the (1 + k) factor can and with KT /J 2 as the input value the full-scale advance
only be validly accomplished if scale-independent pres- coefcient JTS and torque coefcient KQTS are read
sure resistance is absent, which means, for example, off from the full-scale propeller characteristics and the
that there is no immersed transom. In this way the form following parameters calculated:
factor is maintained independent of speed and scale in (1 wTS )VS
the extrapolation method. nS =
In the three-dimensional method the scale effect on JTS D
the resistance is taken as KQTS
PDS = 2D5 n3S 103
2
FD = 21 M VM SM [(1 + k)(CFM CFS ) CA ] R
(12.35)
KT 2
in which the form factor is normally taken relative TS = 2 JTS D4 n2S
to the ITTC-1957 line and CA is the shipmodel cor- J
relation coefcient. The value of CA is generally based KQTS 5 2
QS = D nS
on an empirically based relationship and additional R
allowances are applied to this factor to account for
extreme hull forms at partial draughts, appendages The required shaft power PS is found from the deliv-
not present on the model, contract conditions, hull ered power PDS using the shafting mechanical efciency
roughness different from the standard of 150 m, S as
extreme superstructures or specic experience with
previous ships. PS = PDS /S
In the three-dimensional procedure the measured
relationship between the thrust coefcient KT and the
apparent advance coefcient is corrected for wake scale
effects and for the scale effects on propeller blade fric-
tion. At model scale the model thrust coefcient is
dened as
KTM = f (Fn , J )M
whereas at ship scale this is
1 w TS
KTS = f Fn , J + KT
1 w TM