1 0 JAN. 1974 RCHIEF DOKTORSAVHANDLINGAR Lab. v. Scheepsboiniikunt Technische Hogeschigit, Deift VD CHALMERS TEKNISKA. HOGSKOLA Nr 11 9 .OHALMERS.-- UNIVERSITY OF TECHNOLOGY: GOTEBORG.- 'SWEDEN'. A STUDY OF CRASH STOP TESTS WITH SINGLE SCREW SHIPS RALPH A. NOR RBY INSTITUTIONEN FOR SKEPPSHYDROMEKANIK 1972 C74613,...
125
Embed
A STUDY OF CRASH STOP TESTS WITH SINGLE SCREW SHIPS
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1 0 JAN. 1974
RCHIEFDOKTORSAVHANDLINGAR Lab. v. Scheepsboiniikunt
Technische Hogeschigit,
Deift
VD
CHALMERS TEKNISKA. HOGSKOLANr 11 9
.OHALMERS.-- UNIVERSITY OF TECHNOLOGY:GOTEBORG.-'SWEDEN'.
A STUDY OFCRASH STOP TESTS WITH
SINGLE SCREW SHIPS
RALPH A. NOR RBY
INSTITUTIONEN FOR SKEPPSHYDROMEKANIK1972
C74613,...
A STUDY OF
CRASH STOP TESTS WITH
SINGLE SCREW SHIPS
RALPH A NORM
1972
A STUDY OF
CRASH STOP TESTS WITH
SINGLF SCREW SHIPS
BY
RALPH A NORRBY
AKADEMISK AVHANDLING
SOM MED TILLSTAND AV CHALMERS 'TEKNISKA HOGSKOLA
FRAMLAGGES TILL OFFENTLIG GRANSKNING FOR
TEKNOLOGIE DOKTORSGRADS VINNANDE FREDAGEN DEN.
26 JANUARI 1973, KL 10.00 A PALMSTEDTSSAL-FN VID
CHALMERS TEKNISKA HOGSKOLA, SVEN HULTINS GATA,
-GOTEBORG
SYNOPSIS
Over the next ten year period the volume of transport by
ship is expected to be doubled. The number of vessels will
increase which will accentuate the already difficult
situation in congested waterways. The medium and maximum
size of tankers will tend to increase as will the speed of
container ships. There is a definite requirement to
improve the manoeuvring qualities and also to simplify the
data available for appraisal of same on board. The paPer
deals with the stopping qualities of ships.
Means and methods to improve the stopping qualities are
discussed and devices such as brake flaps and controllable
pitch propellers as well as the method of rudder cycling
are dealt with. It is expected that modern ships will to
an increasing extent be equipped with devices to improve
the stopping qualities.
The data available on board usually consists of recordings
of stopping results at one or two displacements. This data
is difficult to interpret for intermediate conditions. The
Oosterveld diagrams are discussed as those are considered
rather practical. However, they only give the master
information on the ship's expected mean stopping qualities.
A large number of methods to calculate the stopping data
for ships both with fixed pitch propellers and controllable
pitch propellers have been suggested. Complicated conditions
prevail during a stopping manoeuvre and a number of simpli-
fications have therefore been introduced in the calculating
methods. In spite of these simplifications it is claimed
by several authors that the results from actual crash stop
manoeuvres conform well with calculated results.
1
In this work crash stop manoeuvring has been treated on the
basis of data from full scale tests. A relatively large
2
number of tests from ships with fixed pitch as well as
controllable pitch propellers form the basis of the
investigation, covering the range from coastal vessels
to large tankers. The data has been received from ship-
yards in various countries. Simple statistical methods
have been used to. deal with the material.
Observations from crash stop tests suggest that the _vessel's
course stability during the manoeuvre is of importance.
A. course stability criterion is developed for trimmed and
untrimmed vessels during crash stops. The influence,,of
propeller type, wind and direction of propeller rotation
on the sheer of a ship during this manoeuvre is discussed
as is also_the probability of ships sheering against each
other when making the crash.st0P manoeuvre. The turning
-angle for-ships with .fixed pitch propellers and with
controllable pitch propellers is investigated..
To determine track reach and stopping time simple relations
are given for the retardation factor. The influence of
retardation from ship hull resistance, turning, wind,
propeller and uncertain influences is discussed. Non-
dimensional numbers for the stopping distance and time
have been evaluated.
Propeller characteristics at
results based on atmospheric
are collated. Tests from the
introduced for evaluation of
of ships._
crash stop from published
tests at bollard pull astern
KMW cavitation tunnel are
the stopping characteristics
In evaluating. the stopping distance and time the statistical
material is treated with linear regression analysis. As
a result diagrams for estimating and appraising crash stOp
performance have been obtained for ships with fixed pitchand controllable pitch propellers.
In order to examine the proposed method to calculate the,:
stopping distance and time a comparison has been made
using a number of wellknown methods. For these the same
statistical material has been used throughout. It will be
seen that the proposed method, based on statistical
analysis, is at least as accurate as the best of the
other methods used for the stopping material investigated.
Stopping equations are developed which state the probability
that the stopping performance is inferior to the calculated
values. Head reach as well as lateral reach are discussed.
The use of practical stopping diagrams is demonstrated.
Several authors have made comparisons of stopping perfor-
mance of ships with fixed pitch propellers and with control],-1-
able pitch propellers. This data is compared with the
results based on the statistical material.
SYMBOLS
Symbol Definition Physical dimension
A frontal area above water
AEblade area
A constant, es' (41)
Atconstant, eq (42)'
A0propeller disc area
a constant, eq (44),
at constant, eq (45)
BM'moulded beam
Bs constant, eq (41)
Bt.conStant eq (42)
bs constant, eq (44)
btconstant, eq'(45)
4
Symbol Definition Physical dimension
centrifugal force,-2
longitudinal component
astern thrust constant,eq (33)
block coefficient
CQPtorque coefficient, eq (32) -
C thrust coefficient, eq (31) -TP
C constant, eq (3) -Y.v
C1constant, eq (4)
C2 constant, eq (4)
c1constant, eq (34)
c wind resistance coefficient, -X longitudinal
@ Froude resistance constant -
CPP controllable pitch propeller -
D propeller diameter
DE diesel engine
,d mean draught L
dAaft draught L
dFforward draught L
F total braking force MLT-2
FPP fixed pitch propeller -
acceleration of gravity LT2-,
,
h water depth L ,
I immersion of propeller shaft L
KQtorque coefficient *wn2D5
KTthrust coefficient Thivn2D4
MIM
kxquotient between longitudinally"added mass" and the vessel'smass
quotient between athwartships"added mass" and the vessel'smass
Lpp length between perpendiculars
-5-
Symbol Definition Physical dimension
N total number ofobservations
-1Nr moment-yaw velocity MI2 T
derivative
N' Nr/0,5pwlapp5d
Nvmoment-sway velocity MITderivative
N;N' Nv/0,5pwV1Ipp2 d
constant, eq (20) -
number of observations -
T-1propeller shaft sped-
11Areversed propeller. shaft T1speed
propeller shaft speed,0 design
tOshaft speed at stop order. Tn
propeller pitch L
PAastern propeller pitch at L
0,7 radius
17 statistical m.v.of PA LA
propeller design pitch at L0 0,7 radius
-1 --
PVvapour pressure of water. MI T2
Shaft horse power at ML2T3-PSt0 approach speed
PT .. port '
--
-static pressure at propeller M12 T3shaft
R propeller torque MI2T-22 -2
QAmachinery torque, astern MI T
Qt0machinery torque ahead at MI2T-2stop order
Q0 machinery torque, max 2T2continuous.continuous
radius of stopping track-
R ship hull resistance MLT2
R0R at approach speed MLT2-
r correlation factor
rs
SC
U-
V
V
W-
VRO
Symbol Definition Physical dimension
correlation factorfor eq (46)
correlation factorfor eq (47)
distance
head reach1:1
SLlateral reach
STtrack reach
Sy one standard error ofestimate
STc STcalculated
SB starboard
ST. steam turbine
propeller thrust
TApropeller thrust astern
astern thrust at ,dead inTAis the Water
Student's t
thrust deduction factor -
t time
tRtime.to reduce shaft, speed.or pitch to zero
Student's t for eq (46)
tastopping time
tt. StUdent'-s t for eq (47).
tscalculated_
MIT-2
MLT-2
MIT-2
-term of uncertainty MIT2
ship speed
approach speed LT-1
-1relative longitudinal wind LT
speed
absolute longitudinal wind 1JT-1
speedLT-1VR at stop order
wind resistance-2MLT
- 7 -
Symbol Definition Physical dimension
0
wake fraction -
x longitudinal distance betweencentres of origin of coordi-nate system and gravity
-
Yrforce-yaw velocity derivative MLT1
Y' Yr/0,5i3w171Tp2 d
Yvforce-sway velocity derivative MT-1
Y' Yv/0,5p,NVIppd
number of blades
a retardation factor
a mean value, m..v.amean
a due to turning,aC meanaa due to ship hullR "-mean .resistance,
a due to propeller thrust,mean McV.
a meana due to uncertain influences
U
mean a due to wind resistance, m.v.aW
.0 correction factor, eq (15)-2
A weight displacement MIT
V volume displacement3
1Sshafting efficiency
1R0hull efficiency at stop order
TIP()open water propellerefficiency at stop order
1R0relative rotative efficiencyat stop order
1tot toverall propulsive efficiency0 at stop order
de course stability numberduring crash stop,
V Pearson's coefficient _
of variation 1)
PAmass density of air MI-3
Pw mass density of water MI-3
W at stop order MLT-2
8
Symbol Definition Physical dimension
a'cavitation number at dead -pull astern, 2)
Y turning angle, yaw angle deg
statistical m.v. of T deg
Y angular Velocity, m.v. T-1
1) one standard deviation 100/arithmetic mean value
2)(P - Pv)/ 0,5 pw /12.11 D2
INTRODUCTION
The volume of transport by ship is expected to more than
double during the next ten years. There will be an increase
in the number of ships. HAINES 1971 estimates that 300-400
ships per day pass through the English Channel. For the
Uraga Channel, the entrance to Tokyo Bay, the number of
passages daily was 700 in 1968 and according to an EDITORIAL
1970 in Zosen, 50 ferryboat crossings of the main sea lane
should be added to this figure. HAINES 1971 reports 1300
vessels per day for the Inland Sea. Among other regions
with high density traffic are the straits of Malacca and
Singapore, the Persian Gulf, the Danish Straits, the St
Lawrence Seaway and the Panama Canal entrances.
The continuously existing requirement to improve the economy
of transport has resulted in an increase in the size of
ships and an increase in the speed for some types of ship.
The number of tankers above 200.000 dwt and high speed
container ships increases rapidly. As a result of the
building up of complete container handling systems at
large cost, the existing severe requirements to maintain
time schedules will be further accentuated. This can force
the master to maintain a higher speed in congested waters
than safe manoeuvring permits. According to HAINES 197111
every year approximately 7 % of all vessels above 500 grt
9
are involved in collisions. To this figure must be added
an allowance to cover groundings due to faulty manoeuvring
when the vessels are in the collision risk zone. It can
be estimated that 10 % of all vessels collide or run
aground. Naturally the majority of these casualties occur
in sea lanes with high density traffic, KOSTILAINEN 1971
and PORRICELLI et al 1971.
Recently there has been discussions in many places and
papers have been published on the question of the
manoeuvrability of large tankers especially having regard
for the risk of oil pollution. A case can be made for
giving special attention to large tankers but this way of
dealing with the problem of ship manoeuvrability is
insufficient. HAINES 1971 makes a survey of 40 collisions
in the Dover Strait. Of the 80 vessels involved 65 were
not tankers. The result of a collision between a tanker
and a general cargo liner or a ferry can be of the same
degree of severity as between two tankers. The risk of
loss of life is of most importance, pollution is a
secondary risk. Further, a ship which has run aground or
is party to a collision is usually involved in cost
consuming assistance and repairs. To this should be added
the cost of freight loss and increased insurance. Con-
sequently, it is of tha greatest importance to be able
to estimate the manoeuvring characteristics of different
ship types under varying conditions.
PAFFETT 1971 (I) gives a valuable general survey on
various aspects of safe manoeuvring, in which both the
human and technological influences are discussed. It is
necessary to improve the manoeuvring qualities of ships
and also to supply the master with reliable and simple
means to increase the knowledge of manoeuvring in
critical situations.
The decision to steer away from a risk zone or to carry
- 10 -
out a crash stop can only be made by the master dependent
on the circumstances prevailing at the moment when action
is required to be taken. To some extent TANI and FUJI
1970 deal with these questions. KENAN 1972 in an interest
ing approach investigates the manoeuvring qualities when
combining rudder and propeller for two Mariner, ships.
Thereby graphs of critical ranges for a number of vessel
manoeuvres are developed. However simplified assumptions
have been introduced such as for instance that the
propeller and rudder do not interact in a manner more
complicated than they do at 15 knots even at a full
astern manoeuvre. It is thus assumed that the rudder is
effective also when the propeller is reversed which is
not-supported in practice.
The stopping characteristics of ships will be considered
in this study.
MEANS AND METHODS OF IMPROVING THE STOPPING QUALITIES OF.. SHIPS
It is expected that in the future especially single screw
ships to an increasing degree will be designed to carry
equipment for improving their stopping qualities. Stopping
-methods by the use of rudder and propeller have been
investigated by among others JOURDAIN 1965, GROSSMANN 1971
and CLARKE et al 1972.. Multiple screw ships usually have
a better stopping performance than single screw ships.
Only the latter type will be discussed here.
To stop the ship more effectively, so called braking
flaps in the bow have been suggested by JAEGER 1963 and
JAEGER and JOURDAIN 1968. CLARKE and WELLMAN 1971 descripe
brake flaps placed in the afterbody of the ship. MITSUBISHI
1970 discuss the use of underwater parachutes to stop
large vessels. Mitsubishi have made tests in model, half
and full scale. DUPORT 1968 and ENGLISH 1968 have applied
for patents on Y-shaped canals in the bow. Duport suggeSts
an active duct i.e. including a pump, while English's
solution is a passive duct. The latter system is also
described in ENGLISH 1971. The above mentioned types of
manoeuvring devices are restricted insomuch as being
primarily efficient when proceeding at high speed. Further
they are relatively complicated to build into a vessel
or to operate. Although a relatively large number of
tests as well as theoretical studies have been made, the
basis for a practical solution is still considered to
be lacking. DICKSON 1971 is of the opinion that it is
not necessary to use this type of equipment. At NFL such
systems have been studied. On the basis of these findings
PAFFETT 1971 (II) states that: "The verdict in all cases
was that the devices undoubtedly worked, but the improve-
ments achievable would be relatively modest".
In order to reduce the time for shaft reversal for super-
have been. .carried out. As a result a method for shortening
tt.e time for shaft reversal has been developed based on
the decompression method Klaunig's discussion of ILLIES
1969; °NISH' et al 1970.
CLARKE et al 1972 report from rudder cycling tests with
a 193.000 dwt turbine tanker. The rudder cycling programme
which BSRA conducted from an approach speed of 12,7 knots,
resulted in a track reach of 1.850 m compared to approxi-
mately 3.000 m at an ordinary crash stop manoeuvre with
the same approach speed. The method required five rudder
manoeuvres syncronized with five changes in shaft speed
and three alterations in the course of the vessel. Under
discussion with some shipowners, masters and shipyards
the method is considered to be complioated and as a
result there is a risk of making a faulty manoeuvre, see
also JOURDAIN 1965. The advantage of the method is that
the vessel is under steering control during the main
part of the retardation. pROSSMANN 1971 deals with rudder
cycling with an appreciably'smaller steam turbine vessel
and claims a reduction in stopping distance and time with
rudder cycling. However, the time for shaft reversal
during the ordinary crash stop tests was comparatively
- 12 -
long; 95 and 120 seconds. Until more tests are carried
out and have been evaluated, care must be taken when
comparing the method of rudder cycling with that of an
ordinary crash stop even assuming that the above mentioned
syncronization is managed successfully. A rudder cycling
test according to the BSRA method and an ordinary crash
stop test were carried out with a fully loaded motor
tanker of 150.000 dwt from an approach speed of 15 knots
The track reach with rudder cycling was 3.700 m and with
crash stop 2.600 m.
It is now generally accepted that a ship With controllable
pitch propeller (app) can have significantly better
stopping qualities than an equivalent ship with fixed
pitch propeller (FPP). Among others this has been stated
by GETZ and REFSNES 1958, HOOFT and VAN MANEN 1968,
ILLIES 1968, VAN GUNSTEREN 1970, NORRBY 1970, RITTERHOFF'
1970, INUBUSHI and SAKAI 1971, MASUYAMA 1970 and 1971, 1
OKAMOTO et al 1971 and AUCHER 1972. The reason for the
better stopping ability with the CPP vessel is that a
higher astern power can be used and utilized rapidly
after the stop order is given.
AVAILABLE DATA ON BOARD FOR ESTIMATING SHIPS' STOPPINGQUALITIES
Stopping tests are usually carried out in connection withthe trial trip. At some shipyards an appreciable time
for the stopping tests is reserved. Then crash stop from
full, half and slow speed ahead and a coasting test can
be included.
At the trial trip the stopping test is usually carried
out in non-restricted waters. Those engaged in the test
are well prepared. Consequently, the stopping order is
executed without loss of time and with due consideration
for the characteristics and condition of the propulsion'
machinery. The results thus added to the performance data
for the vessel are usually more favoUrable than what is
- 13 -
achieved in practice. Then the ship must be stopped in
the shortest possible time and distance in a critical
situation for which sometimes no preparations have been
made. This explains a statement made by CHURCH 1957:
"Even so, ships usually stop as well as can be expected,
except on occasions when collisions occur".
As a rule the ship is equipped with a plotted crash stop
track including a graph of ship speed and shaft speed
over time. For, a ship equipped with CPP the propeller
pitch is also included in the graph. Usually for a
tanker these results are for full load condition and for
a bUlk carrier or general cargo liner they are for some
specified ballast condition. This type of information
is difficult to interpret and use when differing condi-
tions of displacement, trim and approach speed appertain.
A more practical graph has been suggested by Oosterveld
in the discussion of TANI 1968 and is also shown in
SHELL 1968. TANI 1969 also uses 'this type of graph. IMCO
have recommended that such a diagram, shown in figure 1,
be added to the manoeuvring data for "large ve6sels and
those carrying bulk chemicals", PRICE 1970. The majority
of crash stop tests have been carried out at full
astern shaft speed. Consequently, the diagram, based
mainly on crash stopping tests, is not so accurate at
lower shaft speeds and care must be taken when estimating
the performance under those conditions. Further the
diagrams which have been presented only deal with full
draught and these should be complemented by diagrams
under part load conditions. In SHELL 1968 the graphs
are based upon the mean stopping values for each class
of tankers. From the safety point of view it would be
better to supply the ship with a diagram showing a so
called risk distance and risk time at a crash stop.
It is also valuable for the master to be able to estimate
the probability of the vessel turning to starboard (S13)
or to port (PT) during a stopping manoeuvre. The general
14 -7
opinion is that the vessel turns to SB but there is no
stated degree of probability for turning SB versus PT.
There are different ideas of how a ship with CPP reacts
during a stopping manoeuvre as regards turning.
METHODS Fait. CALGULATING STOPPING PERFORMANCE
It is necessary to clarify how the results of a stopping
test should be appraised. The scatter in performance is
large and sometimes difficult to explain. The appraisal
is comPlicated by the influence arising from turning
forces, suction effects, wake variation, cavitation,
air drawing, oblique flow to the propeller and hull
resistance during retardation, DAILY and HANKEY 1953,BINDEL and GARGUET 1962, JAEGER and JOURDAIN 1962,SAUNDERS 1965, HARVALD 1967 and LOVER 1969. The picture
is rendered still more complex due to the transient
nature of the stopping manoeuvre.
A large number of stationary and quasistationary methods'
for calculating track reach and stopping time have been
developed for ships with FPP based on relatively simple
1931, ROBINSON 1938, HORNE 1945, CHASE and RUIZ 1951,HEWINS and RUIZ 1954, CHASE et al 1957, MINIOVICH 1960,LINDGREN and NORRBIN 1962, TREFETHEN 1962, SAINSBURY
1963, BOATWRIGHT and TURNER 1965, CRANE 1966, HARA et al
1966, TANI 1966, CRANE 1967, HARVALD 1967, TANI and
ISHIKURA 1967, GOODWIN et al 1968, GROBE 1968, ILLIES
et al 1970, RUMS 1970, TANI and ENOKIDA 1970,
CLLRandWELLMAN 1971 and EHRICHE and GROSSMANN 1971. In all
these methods it is assumed that the course of the vessel
is constant during the stopping manoeuvre, the propeller
characteristics from tests at atmospheric pressure are
assumed valid with no allowance for the influence from
cavitation or air drawing and with the exception for
HARVALD 1967 it is assumed that the suction and wake
factors are constant. In spite of these simplifications,
-215-
and of the fact that the methods differ somewhat one from
another, several authors claim good conformity with results
from actual crash stop manoeuvres.
For ships with CPP also methods have been suggested to
calculate track reach and stopping time, CHASE et al 1957,
KITO 1959, MARTIROSOV 1962, GOODWIN et al 1968 and OKAMOTO
et al 1971. The method used for calculation is the same
as that for ships with FPP with the exception that at
stalled conditions a CPP has other characteristics than
an FPP.
COLLECTION AND PROCESSING OF CRASH STOP TEST RESULTS
In order to shed more light over the stopping charac-
teristics of ships, a collection of crash stop results
and ship data has been made with the aid of shipyards,
shipowners, universities, private persons and literature.
Test results from the following references in the
literature have been used: HEWINS et al 1957, GETZ and
REFSNES 1958, HEBECKER 1961, WILSE 1962, MOCKEL and
HATTENDORFF 1966, PARKER et al 1966, HARA et al 1966,
CHIHAYA 1967, HOOFT and VAN MANEN 1968, JAEGER and
JOURDAIN 1968, JOURDAIN 1969, JOURDAIN and PAGE 1969,
SCHIELE et al 1969 and .editorials from HANSA 1962,
SHIPPING WORLD AND SHIPBUILDER 1967, SHIPPING WORLD
AND SHIPBUILDER 1969, SWEDISH SHIPPING GAZETTE 1969 and
ZOSEN 1969. For every test a table has been filled in.
An example is shown in table 1. The test results are
from 25 shipyards in Denmark, England, France, Germany,
Japan, Norway, Sweden and USA. In total 330 crash stop
tests in ballast and full load have been collected,
of which 255 are for ships with FPP and 75 for ships
with CPP of KaMeWa design. The distribution of the
tests is shown in figure 2 on the basis of the deadweight
tonnage. The material has been collected without reference
to any special size or' type of ship. The purpose has been
- 16
to cover the whole range of single screw ships. As the
tests originate from a large number of shipyards in
various countries, it can be assumed that they constitute
a representative average and therefore the material
should not be influenced to any great extent by the
special practices of any single shipyard. 60 % of the
tests have been carried out with vessels with a dead-
weight above 20.000 tons.
Some Of. the tests are incomplete. In these, instances
only part results have been used in a wider context.
FUrther it has been possible to exclude some obvious
faults made during Certain particular tests When this
data has been plotted.together With the rest of'the
material. The number of tests included in the various
parts of the study is .stated in each case
Track reach ST is the distance. along the stopping track.
It has been calculated as the area below the speed-time
curve. Thus ST is based on speed log recordings and not
on fixed plotting from for instance, Decca. In this
instance the former method is to be preferred as ST
is then not influenced by current. The ST value is the
distance travelled through the water. The accuracy of
the log method is + 0,5 - 1 % down to 4 - 5 knots. At
lower speeds the accuracy is less; + 1 - 4 % due to
disturbances from propeller wash and oblique flow in
the last period of the manoeuvre. However, the recording
of the track reach is only influenced by this to a
small degree. The stopping time has been taken from
fixed plotting which is more accurate than speed log
recording at low speeds.
Head reach, SH, is the distance measured along the
ship's initial course from the bow to the point in the
stopping track which is most distant in this direction.
Lateral reach, SL, is the distance from the ship's centre
line measured perpendicular to the initial course to the
With very few
in open water
by the effect
the few tests
- 17 -
most distant point on the Ship during the manoeuvre,
see figure .3 for definitions.
Practically in all tests the rudder has been kept amid-
ships in accordance with the usual practice. In a few
tests the rudder has been used but this has not influenced
the result. A general observation is that during an
ordinary crash stop the ship cannot be steered due to the
very low or no rudder effectiveness, AUCHER 1972.
A crash stop manoeuvre is characterized in that the
reversing of the propeller shaft or with CPP, reversing
of the propeller pitch, is carried out in the shortest
possible time. This usually means within 15 - 45 seconds
for FPP, diesel engine (DE) and 30 - 75 seconds for FPP,
steam turbine (ST). The corresponding figure for CPP is
approximately 15 seconds. After the reversal the highest
possible astern shaft speed is used. For the collected
material on FPP DE this means an astern shaft speed of
approximately 65 % of the full ahead value and for
FPP ST 50 - 55 %. With CPP between 90 and 100 % is
utilized when stopping.
exceptions the tests have been carried out
whePe,h/d>4. Thus they are not influenced
of shallow water, FUJINO 1969. When plotting
where h/d7. 2 these fall well within the
mean values of the remainder of the observations and
have consequently been included in the study.
The design power used is the maximum continuous rating
of the machinery. If the CPP ship is equipped with
auxiliary shaft driven devices the stopping test has
been carried out with these working. This means that
the auxiliary power is deducted from the power of the
main machinery. It has been estimated that the auxiliary
power used during the test is 70 % of the design power
of the auxiliary machinery.
18
All FPP are righthand rotating while for CPP both right
and lefthand rotation exists. The reason why a CPP is
frequently supplied lefthanded is that thereby it will
act on the vessel with the same force direction during
a stopping manoeuvre as an FPP. The reason why lefthand
rotation is not always specified is that the majority
of main propulsion machineries are built for righthand
rotation.
In the purpose of the study of the various combinations,
it has been found convenient to divide the statistical
material from the stopping tests into three.groUpa:
Fixed pitch propeller, diesel engine (FPP DE)
Fixed pitch propeller, steam turbine (FPP ST)
Controllable pitch propeller, (CPP DE ST)diesel engine and steam turbine
Each group consists of tests with ballasted as well
as fully loaded ships and also with approach speeds
of design value and down to 3 knots.
So far as is possible the material has been converted
into nondimensional form in order to eliminate the
influence of size to some extent. Finally the material
has been dealt with and appraised by simple statistical
methods described in MORONEY 1962 and EZEKIEL and FOX
1967.
COURSE STABILITY DURING_THE STOPPING MANOEUVRE
Before considering the stopping performance of ships,
the shape of the stopping track and its depehdence
on the ship's course stability at a stopping manoeuvre
will be studied.
It is a matter of general knowledge that a fully.
loaded vessel sheers more during a stopping manoeuvre
than a ballasted vessel. During such a manoeuvre when
- 19 -
the propeller thrust is reversed the effect of the
rudder is eliminated. This occurs in the early stages
of the manoeuvre, i.e. before the reversing of the
shaft or the pitch. HOOFT 1970 indicates that the rudder
effect is decreased when the speed of the propeller jet
is reduced and is zero after the reversal. Based on PMM
tests with an 8 m model of a 190.000 dwt tanker in full
load it is confirmed by WAGNER SMITT and CHISLETT 1972
that the rudder efficiency is negligible during a
stopping manoeuvre.
When the propeller drives the ship ahead a force in the
stern of the ship perpendicular to its longitudinal
axis is developed due to the local loading of the
propeller. This is here called the propeller rotation
effect and is described in detail by SAUNDERS 1957.
For a righthanded propeller this results in the side
force being directed towards SB. According to Norrbin's
discussion of CLARKE and WELLMAN 1971 it is some
3 - 6 % of the ahead propeller thrust. The side force
directed to SB gives the ship a tendency to sheer to
PT. When steering on a straight course the side force
is compensated for by some degrees of SB rudder angle.
For the sake of simplicity it has been assumed that
the influence of wind, waves and current is negligible
here.
During ,a stopping manoeuvre with an FPP the side force
on the stern will change direction due to the reversal
of the propeller thrust. The stern is then forced to
PT and as the rudder has no effect in this condition
the vessel usually sheers to SB. On the other hand
for a vessel with OPP no side force reversal takes
place when the propeller thrust is reversed. From the
model tests by WAGNER SMITT and CHISIETT 1972 it can
be estimated that the side force from the propeller
on the hull att full astern shaft speed is in the order
of 10 - 20 % of the ahead design speed propeller thrust
for this vessel.
- 20-
With a lefthanded CPP when the pitch is reversed, the
direction of the side force is the same as for a right-
handed FPP. For the remainder of this study a ship with
a lefthanded CPP which sheers to SB will be designated
a SB sheer. The same is valid for a ship with a right-
handed CPP sheering to PT. If these ships sheer to the
opposite side they will be designated PT sheering.
Usually with a CPP the thrust is reversed in a consider-
ably shorter time than with an FPP and consequently the
influence on the sheering of the vessel due to the
propeller rotation effect will appear at an earlier stage.
The sheering is not always directed to SB. The influence
of outer forces on the ship results in some instances
in sheers to PT. This is the case especially at, or
close to, the full load condition when the ship usually
is directionally unstable during a crash stop. It is then
more sensitive to disturbances from wind and current.
Also extensive air drawing and production of eddies at
the manoeuvre can influence the direction and amount
of sheer, BINDEL and GARGUET 1962. Full scale observa-
tions have shown that air drawing exists whether the
propeller is partly or completely submerged.
However, air drawing has only secondary influence on
the direction of the sheer. The essential is the
influence of external forces mainly due to wind and
eddy formation. and internal forces on the vessel during
the first phase of the stopping manoeuvre when the
rudder effect is lost. These forces directly in4luence
the direction of the sheer. In the second phase
propeller rotation effect dominates so that an alrqady
initiated SB sheer is increased while the PT sheer is
decreased.
In ballast., when the vessel as a rule i8 stable on
course, it is less sensitive to outer disturbances.
In the majority of cases the vessel then sheers SB,
21
i.e. in the second phase the propeller rotation effect
again dominates.
A"COURSE STABILITY CRITERION FOR STOPPING'MANOEUVRES'
The difference in the sheering tendency between a fully
loaded and a ballasted vessel whilst stopping is to be
further investigated having regard for course stability.
The criterion for dynamic stability in straight line
motion for a vessel with its rudder in the neutral
position can be written
Y (N CB BMd2
xGV) - Nvr - pw CB Bm d2 V)> 0 (1)
The deduction is shown by a.o. ABKOWITZ 1964 and further
work to obtain a practical criterion has been carried out,
WAGNER SMITT 1970 and 1971. Based on 55 PMM tests with
35 different models varying from trawlers to large tankers
force and moment coefficients Y;., Y,,N; and N. have been
evolved as a function of d/Iipp. Using these values in
the above.expression (1) Wagner Smitt writes the criterion
as
41 3,88C Bu L /( L ) 1.5'23
+ 0,0050 ( ) >0pp
Wagner Smitt states that the criterion is not valid
for trimmed vessels and shows that approximately there
is a linear relation between the force and moment
coefficients and the square of the aspect ratio i.e.
(d/LF2)2. When test results with trimmed models are
included the scatter is relatively large.
A more correct expression for aspect ratio is 2d2max/
lateral area. Assuming that d =dmax A
the expression
for aspect ratio can approximately be written
44A/[Iippi(dF/dA) + . To adapt the criterion to
(2)
br....,-"....-4Vreammrom.,.."....
- 22-
trimmed vessels the force and moment coefficients are
plotted as a function of the latter expression for aspect
ratio. Then the expression's for force and' moment
coefficients will be of the type
[v - YvL +
dF
pp dA
dA 2
WhereC is constant. -Thus the inequality (2) can be
written v
LV PPL72.2)3
< Ci d 2A. Ad + 1)
A
Lacking coefficients for trimmed vessels Wagner Smitt's
values for vessels on even keel have been used here.
This gives
5,4L L 3
<, V + 0,0013 (U)dF 2 dA
dA3.
(cr + 1)A
The conditions for dynamic stability during a stopping
manoeuvre &re considered to be more complicated than for
steady propulsion ahead. To some extent this question
has been dealt with by NORRBIN 1964 and HOOFT 1970.
WAGNER SMITT and CHISLETT 1972 show that from the above
mentioned PMM tests the dynamic stability during a crash
stop Is closely the same as for steady ahead propulsion
and uninfluenced by the speed of propeller or ship.
Howev-er, the question requires further scrutiny because
with the existing material it is difficult to make
general forecasts regarding sheering tendencies during
stopping manoeuvres.
The inequality (5) has been used to further' studythe
tendency to sheer when stopping. If Wd3A is' less than
the value' in (5) thenthe vessel is stable as usually
defined for steady propulsion ahead. From 'the. stopping
(3)
(5)V
"fir
23
, 1/3tests SL/ V and Y for sheer SB and PT have been
plotted as a function of (V/d3A)/(v/d3A)c where index
C indicates a critical value based on inequality (5).
This has also been done for sheer against and with the
wind, see figures 4, 5, 6 and 7. It is evident that a
definite boundary value exists for (V/d3A)/(V/d3A)c.
For (N7/d3A)/(V/d3A)c < 1,1 the vessels sheer moderate-
ly while for values > 1,1 the sheer and its scatter is
appreciably greater. Obviously the dynamic stability
during a crash stop has a critical value around 1,1.
A course stability number is now introduced. The
critical stability value during a crash stop is for
simplicity put at 6f = 1, i.e. where (V/d3A)/(V/d3A)c
= 1,1. Based on expression (5) and WAGNER SMITT and
CHISLETT 1972 a criterion for dynamic stability during
a stopping manoeuvre can be written thus
5 9 L L 3
(34)dF
PP + 0,0014 (--P-2)2 dA
(6)d' CA
A d, (7 1),A
This is shown in figure 8. In order that the vessel shall
be dynamically stable during a crash stop its V/d3A
must be less than the value in equation (6). If for
example a ship has dp/dA = 0,95 and Lpp/dA = 18 then
(V/d3.)= 36. The actual value for V/d3A is 39A CA
which means that the ship must be considered to be
dynamically unstable during a crash stop; the value
is > 1. By dividing the material from the crash stop
tests into two parts, ee>1 and dlo<1 a Clearer picture
of the sheering tendencies is obtained.
SHEERING TENDENCIES
In order to study with what probability a vessel sheers
to SB or PT, with or without the influence of wind,
figures 4, 5, 6 and 7 have been investigated. To this
material has been added results from stopping manoeuvres
24
where the direction of sheer and relative wind are known
but not SL and LP . The result is divided into the two
groups 4f > 1 and g < 1 and is shown in figures 9, 10
and 11. To simplify the comparison in figures 9 to 11 it
has been assumed that for ahead propulsion all FPP are
right hand rotating and all CPP are left hand rotating.
As mentioned above this implies that the side thrust
on the vessel from the reversed CPP then has the sae
direction as from the reversed FPP. Thus if the CPP is
right hand rotating the numbers for SB and PT are to
change places in figures 9 to 11. The windward side
must also be-changed.
SHEER WIN, TYPE OF 'PROPELLER, MACHINERY AND PROPELLERROTATION EFFECT ARE TAKEN INTO CONSIDERATION BUT.EXCLUDING INFLUENCE: OF wIya
For df>1 figure 9 shows SB sheer dominant with approxi-
mately 60 % for FPP DE and ST and with 68 % for OPP.
The tendency mentioned above for SB sheer has further
Increased for H < 1. There the SB sheer accounts for
70 % with PPP ST, 78 % with FPP DE and 82 % with CPP.
The explanation is that the directionally stable ship
maintains roughly the initial course during the first
phase of the stopping manoeuvre and in the second phase
when the speed is low is influenced principally by
the propeller rotation effect which tends to turn the
ship to SB. Further the influence from propeller and
machinery is larger when the ship is in ballast than
in full load.
SHEER WHEN TYPE OF PROPELLER, MACHINERY, PROPELLERROTATION EFFECT AND WIND DIRECTION ARE TAKEN INTO
CONSIDERATION
With of>1 and the wind on the SB side approximately
75 % of the ships with FPP turn SB. The corresponding
value for CPP ships is 80 %, however care must be
- 25 -
taken with this value because the number of observations,
is only 5. With the wind from PT, ships with FPP ST..and
DIE turn SB, 40 % and 50 %respectively. The CPP ship falls
Off from the wind and sheers SB in 65 % of all cases.
Evidently the sheer of the CPP ship is mainly influenced
by the direction of the rotation of the propeller and
not by the direction of the wind wheh it is on the PT
side. For FPP ST and Eg the influence of the PT wind is
strongest. With the wind on the SB side the effect of
wind and propeller rotation add up resulting in an
increase in the sheerability to SB.
With al?<1 and SB wind, 75 % and 86 % turn SB for FPP ST
and DE respectively. For CPP the value was expected to
be between 80 % and 90 % but in this study is not more
than 57 %, probably due to the study being based on 7
observations only and in consequence not being reliable.
The tendency to luff into or fall off from the wind can
be noted in, for instance, SHEARER and LYNN 1960 or
GOULD 1967. There it is shown that vessels on even keel
with the superstructure aft have a positive wind moment,
i.e. they have a tendency to luff for a total sector
of a relative wind direction of 220° - 280° from astern.
For ballasted vessels trimmed on the stern and for vessels
with the superstructure midships or forward the "luffing"
sector is smaller, approximately 180°.
This survey of the sheering tendencies can be used for
estimating approximately the probability that a vessel
will sheer in a certain direction when making a crash
stop manoeuvre. The values for sheer given in figures
9 to 11 indicate the sheering probability. By use of
the multiplication law the probability may be estimated
as to what extent two ships meeting on counter course
will sheer against each other for instance.
- 26-
Assume that two ships on counter course meet in a
passage and keep on the SB side. One of the ships is
a fully loaded tanker (0 >1) with FPP ST and the other
is a general cargo liner in ballast (0 < 1) with FPP DE.
The tanker has the wind from PT and the liner from SB.
The probability that both ships sheer PT in a stopping
manoeuvre is 0,60. 0,14 = 0,08. This means that in 8
cases out of 100 the ships can sheer into the centre of
the passage thereby increasing the risk of collision.
If the two vessels steam the passage on the wrong side
(PT) which can occur, the risk that both. sheer SB
during a stopping manoeuvre is Markedly increased;
0,40 0,86 = 0,34. In this particular case the risk that
the vessel will Sheer against each other is approximately
four times as great.
Generally it Can be concluded for the. material that
if two vessels on counter, course steam a passage
keeping'SB the probability that both will sheer. PT
does not exceed 0,20. If the vessels steam the PT side
of the passage when making a.crash stop the corresponding
value for both vessels sheering SB is up to 0,67.
THE TURNING ANGLE
From the view point of handling the ship it is of
interest to know the final turning angle y the vessel
will reach during a stopping manoeuvre. In figure 12
the mean turning angle for FPP and CPP at 0> and < 1
is shown with approach speeds above 13 knots. Sectors
for + one standard deviation have been drawn approxi-
mately. Above 13 knots y is not to any measurable
degree dependent on speed or ship size while for lower
approach speeds V decreases markedly. As there is no
significant difference between the turning angle for
FPP DE and ST these results are presented together in
the figure.
- 27 -
At FPP >1 the difference in LP-SB and PT is not
significant. On the other hand the difference in
turning angle SB and PT is highly significant for
FPP,dT < 1. From the plotting of the course during
the stopping manoeuvre it is seen that in the second
phase the vessel turns SB due to the propeller rotation
effect whether its final sheer is SB or PT. This
explains the difference SB and PT in the mean turning
angle i-7 which is 13° for df >1 and 29° for g < 1.
For CPP, d-p >1, SB sheer Ti) is 133° and its standard
deviation is 24°. The .corresponding values for FPP are
125° and 49°. The smaller scatter for the vessel with
CPP is explained by the fact that the crash stop as a
rule is performed more uniformly than for a vessel
with FPP. The difference 8° in mean turning angle is
not significant. At oe< 1, SB sheer the turning angle
and scatter are approximately equal for FPP and CPP.
For FPP as well as for CPP there is a highly significant
difference for 4) between d-P > and < 1.
DETERMINING TRACK REACH AND -STOPPING TIME
During a crash stop manoeuvre the vessel along its
longitudinal axis is mainly retarded by the ordinary
diameter ratio, blade thickness and thickness distribu-
tion. From VAN LAMMEREN et al 1969 can be seen that the
CA value is practically independent of the number of
propeller blades.
The relation, between CA and PA/D is Well establiehed
for both FPP and CPP. The difference between FPP and
CPP is explained mainly by the fact that the blades of
a reversed OPP are working, under even more stalled'
conditions than the blades of a reversed FPP, NORDSTROM
1945..
Mean curves from the KMW cavitation tests for FPP and
QPP are shown in figure 16 and these CA values will be
Used Iereafter.
. .
ESTIMATiNG STOPPING QUALITIES WITH THE AID OF REGRESSIONANALYSIS
By putting equations (37) and (38) into logarithmic form
linear relationships are obtained approximately. This
is shown by plotting the stopping results in loglog
diagrams. Analytical expressions have thus been obtained
by using the method of least squares. These relations
can be expressed as
P1 ST
10l10 AOlog
logAs
logAA
Vo
g t1 s 10°log - log At
Vo
where A and B are constants.
Based'on equations (23) and (35) an attempt has been Made
to exchange QA against Q0 also for PPP ST. This simplifies
the procedure When estimating the stopping performance.
PAn10log
QA CA
40
Otherwise the relation nA/no must be known or estimated.
Comparisons show that the accuracy in ST and ts is not
influenced whether the estimation is based on QA or Qo
considering the uncertainty in estimating nA/no. There-
fore hereafter when estimating or checking stopping
performance QA will be exchanged for Q0.
For FPP it is assumed that PA Po. Initially for CPP
PAG/Q0CA was calculated with the actual mean value for
PA from recordings made during the stopping tests.
However, this method is inconvenient as the mean PA
value is not known in advance. Thus it is desirable to
be able to estimate stopping performance for CPP also
when PA is unknown- From the CPP statistical material
the mean value 7 is obtained asA
17- 0,75 PA 0
This implies that the statistical mean value of the astern
pitch during a crash stop is 75 % of the desigh pitch
ahead. Utilizing this relationship about the same accuracy
in the result of the regression analysis has been obtained
as when using actual mean pitch values. Therefore
equation (43) will be used throughout for CPP.
It is now possible to estimate mean vaUes for stopping
performance for ships with FPP as well as CPP Simply
by using values for Po, 6 and Qo. Equations (41) and
(42) will then be slightly modified.-
10 g 10loglog = log as
Vo
g ts10log
10- log at +
0
PAb 10log
Q C0 A
10log
where a and b are constants.
In linear form the expressions are
CA
( 43 )
(44)
(45)
g tsV0
PZ b( ) S
OA
P b+
=at A- )t Q0 CA
The result of the regression analysis for the three
groups FPP DE, FPP ST and CPP DE and ST is shown in
table 2. As mentioned above the statistical material
consists of data from full load and ballast, approach
speed at design value as well as appreciably below.
The correlation factors r are high for the three groups
and Student's t-test shows that the values of the
correlation factors are extremely significant. In
addition the values for the logarithm of one standard
error of estimate S have been given. Based on the r-
and t-values the result in table 2 is considered to be
significant. The exponents bs and bt are highest for
FPP ST and lowest for CPP DE and ST. The steepness
indicates the sensitivity in stopping performance for
changes in PA, L or Q0. The results show that ships
with CPP are the least sensitive to those changes.
From the table it is also seen that the scatter is largest
for FPP ST and smallest for CPP. This result is to be
expected because the material from the stopping tests
indicates an appreciably larger scatter in tR for FPP
than for CPP. The value of tR directly influenoes the
stopping result. The time for changing the direction of
thrust' is closely related to tR. From an examination
of trial results of several ships HEWINS and RUIZ 1954
state that the time to develop maximum astern thrust
TAt is equal to tR. This relation can also be seen
in sJAEGER and JOURDAIN 1968. From the extensive tests
reported by HARA et al 1966 the astern thrust at tR
usually ranges between 0,3 and 0,7 TA. Further in the
discussion of SMITH 1937 Dodson shows from cavitation
tests with a model propellex. at low ahead revolutions
acting as a brake in stopping a vessel that although
- 41 -
42
the propeller cavitates heavily in some conditions the
astern thrust is appreciable. During a crash stop the
astern thrust can reach considerable values when shaft
speed or-propeller pitch still are positive. For a CPP
the astern thrust is high already early during the pitch
reversal independently of the momentary torque reduction
due to the braking effect of the blades. The frequency
distributions of tR are shown in figure 17.
In this connection it is of interest to study the
frequency distribution of the mean time value for nA/no
for the three combinations. Figure 18 gives the distri-
butions. In table 3 the arithmetic mean value and the
Pearson coefficient of variation V for the various
combinations in the figure are shown. The Pearson
coefficient of variation which is expressed in per cent
is one standard deviation times 100 divided by the mean
value. From table 3 it is obvious that nA/no for FPP DE
does not have the high values sometimes stated (5.0 85)
whilenA /n0
for FPP ST is close to 0,5 which is a
wellknown figure. The scatter is slightly smaller for
FPP ST than for FPP DE. With CPP the relative scatter
is less than half the aforementioned values and the
shaft speed astern is close to no. The reason for nA
not being equal to no is that not all CFP's are
reversed at full shaft speed. In 37 % of the tests with
CPP shaftdriven auxiliaries have been in use with full
shaft speed during the crash stop manoeuvre.
Equations (46) and (47) have been drawn in loglog
diagrams, figures 19 and 20, where also the statistical
material has been plotted. The mean lines of the three
groups diverge for increasing PAZ/Q00A for gST/V02 as
well as gts/Vo. This is shown in figure 21 for gST/V02
where equation (46) has been drawn in a linear diagram.
The same tendency is valid for gts/Vo. The curves level
out with increasing PALVQ0CA i.e. for larger vessels.
Based on figures 16, 19 and 20 it can be seen that
- 43 -
superior stopping performance is achieved with ships
equipped with CPP. This is treated below. With a CPP an
appreciably higher stopping energy can be transformed
than with an FPP. With FPP DE the stopping performance
is usually somewhat better than with FPP ST. However,
a strict comparison between FPP DE and FPP ST can only
be made using actual values of PAA/Q0CA. The relation
between these is such that P LS/Q C for DE is equal to,A 0 A
or larger than, the value for ST when comparing ships
of the same size and power.
The scatter of the statistical material in relation to
the respective mean curves has been studied for the
three groups. The y2 test shows that the data approxima-
tely follows the normal curve. The frequency distributions
do not differ significantly from their respective normal
distributions. With the aid of the S value which is the
one standard error of estimate, lines for + S and + 2 Sy
have also been drawn. Between the lines + 5y and + 2 SY
respectively 68 and 95 % of all stopping tests should
fall. This is valid for material strictly following the
normal distribution, see figures 22 to 27. For the
statistic material this is approximately valid, table 4.
The per cent deviation for one standard error of estimate
is shown in table 5.
Diagrams of the type shown in figures 22 to 27 can be
used for appraising stopping test results which usually
are only vaguely commented upon. It is then possible to
estimate on a statistical basis, if the stopping test is
successful or otherwise.
The method of stopping a ship by rudder Cycling has been
mentioned previously-. In figure 23 two crash stops and
the best rudder cycling result with the sate Value of
P Ls/Q C have been plotted for the 193.000 dwt steamA 0 Aturbine. tanker "Esso Bernicia", CLARKE et al 1972. The
44
crash stop results are of medium class and the rudder
cycling result is at the same level as the best crash
stop results for FPP ST. From figures 16, 23 and 24 can
be deduced that the best rudder cycling manoeuvre for
"Esso Bernicia" with 93 % probability will result in a
longer track reach than if the ship is equipped with a
CPP and makes a crash stop. The "Esso Bernicia" best
stopping distance when rudder cycling is 25 % longer
than the mean crash stop value for a corresponding CPP
ship. In figure 22 another previously mentioned compari-
son has been made between crash stop and rudder cycling
with a 150.000 dwt motor tanker. Its crash stop result
is better than the expected dean value and the rudder
cycling is not so good as the mean crash stop value.
COMPARISON OF VARIOUS CALCULATION METHODS FOR ESTIMATINGTI STOPPING PERFORMANCE .
In order to examine the proposed method for estimating
the stopping performance the following'weliknown methods
have been studied; CHASE et al 1957, SAINSBURY 1963,TANI 1968, ILLIES et al 1970 and CLARKE and WELLMAN 1971.
These. methOds assume throughout that the vessel keeps its
course during the stopping manoeuvre. Manifestly the result
should be a longer stopping distance and Stopping time than
the real values. Hereby a safety, factor is said to be .
introduced.
Further it is assumed that the astern thrust is constant
and that it has the same value as the thrust developed
by the screw when backed in open water at the appropriate
torque and at zero speed of advance. The propeller
characteristics are based on model tests in atmospheric
conditions. From trials with the "Esso Suez" and a
number of model tests CHASE et al 1957 use a CA value
of 5,5 for FPP. For CPP in the same reference a CA value
of 4,33 is used. This latter figure is taken from two
- 45 -
full scale observations and three Model tests. When
calculating the stopping performAnce according'to
SAINSBURY 1963 it has been assumed that the TAis Values
from CEASE et al 1957 are valid. CLARKE and.WELLDWN 1971
use KT/KQ2/3 2,25 as an average for tankers. The.
overall propulsive efficiency neCessary to know when
calculating according to ILLIES et al 1970, has been
estimated frOm curves received by letter from
Mentzendorff in1971.. TANI 1968 reduces TA.t.ty 7,5%,while in reality the astern thrust cannot ti realized
itMediately at the stop order. 0LARKE.and WEIJI4AN'
-1971 introduce a thrust deduction of 7 % for the bollard
Pull astern value, based on measurements of thrust in
fUll scale. ILLIES et al 1970 euggests a thrust deduc-
tion coefficient of 15 %.
When calculating according to these methods, except that
of TANI 1968, it has been assumed that the QA/Qo-values
according to CHASE et al 1957 are valid. These are
based on the Troost B 4-40 series with variation in P/D.
The torque limit line of 0,8 Qo at 0,5 no and 1,0 Q0
at 1,0 no is used for FPP ST and 1,0 Q0 for FPP DE
between 0,5 and 1,0 no. For all five methods it is
required to know or estimate the value of nA/no at dead
in the water. The measured nA values for that condition
have been used in the calculation.
The time from the stop order till full astern thrust
has been put at 20 seconds when calculating according
to CHASE et al 1957. This. value has been suggested' in
that paper when actualfigures are not available.
TANI 1968'does not mention the reversing time but
presumably the method takes account of ityorthe.other
three methods the real t values have been Used ih the
calculation.
SAINSBURY 1963 assumes-more correctly that the actual
46
ship speed when the reversal has been effectuated shall
be put into the calculation instead of the approach
speed. This is a complication as the value of the speed
is not known and must be estimated. ILLIES et al 1970suggest a simple formula for estimating the ship speed
at the reversal. The SAINSBURY 1963 method has been
checked with that formula and the result is that the
scatter is appreciably greater than if the approach
speed value is used. Further from the crash stop
recordings it is seen that the actual ship speed when
the propeller is reversed is very close to the approach
speed. Therefore when calculating according to
SAINSBURY 1963 the approach speed has been used. This
is also the case for the remaining four methods.
The ship resistanee is assumed to be proportional to V2
which is approximately valid for the greater part of the
speed range. Ro; i.e. the ship resistance at the stOporder, has been calculated according to SAUNDERS 1957for CHASE et al 1957 and SAINSBURY 1963. TANI 1968 bases
R0 on model tests with tankers above 40.000 dwt and.
CLARKE and WELLMAN 1971 apply a Froude resistance
coefficient . 0,65 as an average for tankers.
According to CHASE et al 1957, SAINSBURY 1963 and CLARK
and WELLMAN 1971 the "added mass" is 8 %. ILLIES et al.
1970 use 10 % of L and TANI 1968 probably uses results
from MOTORA 1960 for "added mass".
ILIJIES et al 1970 utilize the energy relation to solve
the expressions for ST and ts. The Other four methods
use the equation of motion (7). Based on these
assumptions -
9
kx) V:U e RoS = V log. T 0 g
1 +TAt
)2 R0
+ kx
V Rtan-1 V RO
= tRT0 Ats TAts
- 47 -
With the aid of the basic equations (48) and (49) the
special assumptions valid for the various methods have
been applied resulting in different expressions for
STand ts derived by the above authors. For the actual
expressions reference is made to the respective method.
The methods according to CHASE et al 1957, SAINSBURY
1963 and ILLIES et al 1970 make no restrictions
regarding.type or size of ship. On the other hand the
method of OLARKE and WELLMAN 1971 has been detiVed.
especially for tankers and that of TANI 1968 is ,.valid
only for tankers above 40.000 dwt. The, greater part of
the material on which the latter method is based lies
'within the boundaries 40.000 to 70.000 dwt.
The stopping performance for the statistical material
has been calculated by these methods on the above
assumptions. The calculated result has then been
compared With the corresponding real. values. In table
6 a comparison is shown where the results of the above
suggested calculating method are also included. A
graphical 1.epresentation is given in figures 28 to 32.
For FPP tankers and bulk carriers 91 tests have been
carried out in full load. For various groups of tests
the average arithmetic values as well as the relative
variability or scatter expressed by Pearson's,
coefficient of variation are shown for STc/ST and
t /ts The same stopping test material has been usedsc
throughout. The table shows that the degree of accuracy
forSTc /ST
is higher than that for tsc /ts in the five
methods. Further the estimating method suggested here
has approximately the same degree of accuracy regarding
STc/ST as the most accurate of the other methods. When
compared to other methods tsc/ts has significantly less
scatter. In addition the calculation for vessels with
CPP is more accurate than for those with FPP. Three of
the five methods indicate shorter-stopping distances
than those recorded although no resistance increase
due,to turning is taken into account in any of the methods.
- 48 -
Another comparison 'between the various methods is shown
131 figures 33 to 37. There the material, has been divided
into two groups namely
g >1, Vo 13 knots
a() < 1 or Vo < 13 knots or af < 1 and Vo < 13 knots
From the above it is seen that the sheer of the vessels
is significantly less for the latter than for the
former group. According to figures 33 to 36, for
vessels which do not sheer appreciably, the five methods
give estimates of track reach and stopping time which
in approximately 60 to 80 % and 60 to 70 % respectively
are lower than the average values for the combination
"all ships" in table 6. The corresponding figure based
on the method suggested above is approximately 50 %.
The reason for the calculated values being lower than
those measured can mainly be attributed to that the
propeller thrust in reality is influenced by cavitation
and air drawing not taken into account in the calcula-
tion. For the cathegory > 1, Vo? 13 knots on the
other hand the majority of the estimated values fall on
the high side of the mean value for "all ships". This
is due to the turning resistance not being considered,
the influence of which can be large. Again for the
method introduced here approximately the same number
of estimated values are on both sides of the mean value.
The influence of the resistance increase from the turning
motion as well as the time for reversal, tR, is inherent
in the method suggested here. Further its application
is relatively simple as only the following factors
require to be known or estimated:
approach speed Vo
propeller design pitch ahead Po
propeller diameter D
ship displacement
propulsive machinery design torque Q0
- 49 -
For CPP, Q0 is to be corrected for shaft driven
auxiliaries if any.
TEE STOPPING EQUATION FOR ONE PER CENT RISK
The principle used based on statistics makes it possible
to calculate a curve for which the crash stop distance
is not exceeded in more than for instance 1 % of all
cases. This is considered to be a reasonable figure.
In other words there is a 99 % probability that the
vessel will stop in a shorter distance. Based on the
normal distribution this corresponds to a curve whichis 2,33 S above the mean curve. These so called risk
curves which in the log-log diagrams are parallell to
the mean lines, will have the as and at values shown in
table 7. The bs and bt values are identical to those in
table 2. In figures 22 to 27, the risk line and a line
which indicates that 1 % of all crash stops will be
shorter, have been drawn.
In practice, it can also be of interest to know the
1 % risk value for SH. Such risk lines for head reach
have been calculated, using results from tests where
g < 1. As mentioned above the sheer for e < 1 is
appreciably less than for g >1. This implies that the
SH/ST values are significantly closer to unity fore < 1.
Aboard it is not a simple task to determine if o > 1
or < 1 and consequently when estimating a risk distance
along the initial course, SH for g< 1 has been chosen.
The result of the calculation is that the 1 % risk
SH-line is close to the 1 % risk S -line but it is not
so accurate. Studies of the frequency distribution from
plotted values of SH/ST above 14 knots, show-that there
is close agreement between the groups PPP DE, FPP ST and
CPP DE and ST. Below 14 knots the Si/ST values are
dependent on approach speed and are appreciably higher
because at reduced approach speed the sheer is much lower.
The material from the three groups has been put into two
sL -25V
- 50-
frequency distributions, namely with g > 1 and e < 1the characters of which are distinctly different, see
figure 38. The median value of SH/ST in the figure is
approximately 0,68 and 0,94 for 0 > 1 and <1 respec-
tively. Thus for g < 1 50 % of all SH-values are within
6 % of ST. Therefore and due to the larger scatter in
the SH-values, it is considered preferable to use the
STrisk line in practice when estimating the risk
distance ahead.
In the same way as for ST and SH, attempts have been mad
to calculate the equations for gSL/V02 as a function of
PAA/Q00A. This, however, resulted in a too large1/3
scatter. Instead the quotient STIP has been studied.
From figure 4 it is seen that the highest Slip1/3
values are reached for Qtf > 1. Since the purpose is to
determine the lateral risk distance, the material for
g > 1 has been selected. At full load and with an1/3
approach speed close to the design value, SL/ is
not dependent upon size. Frequency distributions for11/v do not show any characteristic difference
between FPP DE and ST or between SB or PT sheer. There-
fore it has been considered acceptable to treat FPP DE
and ST, SB and PT sheer as one material. There is a
dependence on approach speed up to approximately 13
knots which is seen in figure 39 where SL/s71/3 is
plotted as a function of Vo for FPP DE and ST, SB and
PT, g > 1. Above 13 knots there seems to be no speed
influence. The frequency distribution above 13 knots
has been studied. Its mean value is expressed as
(V-0 13 knots) (50)
and the coefficient of variation is 45 %. The value
SL- 51 (Ir 13 knots)
will not be exceeded with more than 1 % probability.
( 51 )
5 1
For Vo < 13 knots it has been assumed that
"%0VO
Then the following is obtained:
SL
773 0,30 V20-
This relation is suggested to give the lateral dis-
placement for 1 % risk.
FQT CPP On the basis of the same line of argument
the mean value is
sL
7 (v0 13 knots)7177 -4
The coefficient of variation is 32 % and the 1 %
risk value is
SL773 - 30V
( < 13 knots)
(1/.0 2_13 knots)
( 5 2 )
( 5 3 )
( 5 4 )
( 5 5 )
In the statistical material there are no stopping
tests with CPP and 4P> 1 for V0 K'13, knots. Using the
SaMe reasoning as above for FPP; the following relation
is valid for CPP at approach speeds below 13 knots.:
V1 / 3 '18 V2 (56)
The difference In SL/ mean values between FPP and
CPP is significant.
PRACTICAL CRASH STOP DIAGRAMS
The expressions for 1 % risk can be used on board so
that the vessel can be handled with small risk of
collision._Such diagrams, approximately valid in full
load as well as in ballast, are shown in figures 40 to
52 --
43 for two sisterships, bulk carriers of 75.000 dwt,
one with FPP and one with CPP. For the ship with CPP
the power at the propeller is 3 % lower than for the
ship with FPP as in the former ship a shaft driven
generator is used. For instance at full load, 95.000
tons and 16 knots approach speed the FPP ship in 99 %
of all crash stops will have a stop track of less than
2,7 nautical miles. The corresponding distance for the
CPP ship is 1,6 miles. Thus there is more than 1 mile
of extra margin for the CPP ship. For lateral deviation
with the CPP ship the 1 % risk distance is 0,7 miles and
for the FPP ship 1,2 miles.
It is suggested that the SVCsdiagrams shown in figures
40 to 43 be used for instance as follows for the two
ship case, see figure 44: The master of ship A deter-
mines the distance AB to ship B. Master A shall assume
that ship B will not stop in a shorter distance than
ship A. Master B is to reason corresPondingly that ship
A will not stop in a shorter distance than ship B.
Suppose for example that A and B are the sisterships
mentioned above and that. ship A has OPP and ship B FPP.
Further. assume that both are fully loaded. Master A
whose ship, at the radar contact with ship 33 makes 12
knots, can keep this speed as long as AB:>2 STmin = 1,8:
When,AB closes up to 1,8'master A can order a crash stop.
Ship B which makes 12 knots at the radar contact is
manoeuvred correspondingly when AS 3,0'.'
On board it is considered important-to be able to
estimate the stopping time. The longer ts is, the longer
it will be impossible to manoeuvre the ship. The slip
cannot be steered during the stopping manoeuvre and it
is natural to try to minimize the period the vessel is in
this condition. Figures 45 and 46 show practical tWs-
_diagrams for the stopping time of the mentioned 75.000
dwt ships. In 99 cases out of 100 the ship with OPP is
- 53 -
expected to craSh. stop from 12 knots fully loaded in
less than 8 min 45 sec. For the FIT ship the corresponding
time at risk is 5 minutes longer.
Earlier it was mentioned that Oosterveld has developed a
practical stopping diagram, figure 1. Diagrams of this
type have been distributed to various classes of Shell
tankers and are said to give representative mean values.
From SHELL 1968 the stopping results from design speed
and full astern have been plotted in figures 22 and 23.These show that the crash stop values are on the mean
lines for the respective groups. This implies that in
50 % of all cases the crash stop result can be longer
than that indicated in the data for the Shell tankers.
It will be appreciated therefore that this type of
diagram does not give the master sufficient information
on the distance and time at risk.
COMPARISON OF STOPPING PERFORMANCE WITH FPP AND CPI'
SHIPS
Amongst those who have compared the stopping performance
-between.ships:with FPP and CPP are HOOFT and VAN MANEN
1968, RITTERHOFF 1970, MASUYAMA 1970 and 1971 and OKAMOTO
et a1 1971. In table .8 results from those studies have
been collated. The results vary appreoiably.
The study by HOOFT and VAN MANEN 1968 is based on
stationary model tests with reversed propeller. In
addition in that investigation, the full scale propeller
design shaft speed for ST as well as DE has been put at
85 revs/min. Further the diameter for the FPP is 9,193 m
and 9,480 m for the CPP. The comparison is not considered
to be fully realistic but not withstanding this it has
been referred to and treated by several authors, among
others by RITTERHOFF 1970 who interprets the results
somewhat differently, see table 8. The latter reference
for crash stop (quick reversal) gives relatively high
ts CPP1,245
ts FPP
- 54 -
figures of ST app/ST Fpp and ts cpp/ts Fpp for DE, as
do HOOFT and VAN MANEN 1968. MASUYAMA 1970 bases the
figures on full scale observations as well as theore-
tical calculations but in the 1971 reference uses only
full scale results. The statement by OKAMOTO et al 1971
is founded on full scale crash stop tests with two high
speed CPP DE cargo liners and a CPP ST OBO-carrier of
130.000 dwt. In this latter reference it is claimed that
a CPP only improves the stopping qualities for ships
with DE and not for those with ST: "The difference of
stopping performance between FPP and CPP is expected to
be small". This statement is not in accordance with
observations from ship trials.
With the aid of the mean curves based on the statistical
material, equations (44) and (45) the relations
ST CPP/sT FPP and ts C/tPPs FPP have been studied. It
has been assumed that the following factors are
reciprocally equal: P0, A , Q0, no and D. Further the
relation -FA = 0,75 Po has been used. The material covers
the size range from 570 to 134.000 dwt. Actual data for
ship, machinery and propeller have been utilized for
full load and ballast and the result has been divided
into DE and ST groups. A suitable way to present the
above relations is to use (PAL/Q OCA)FPP as a basis.'
For the vessels studied the mean value of (PAL/Q0CA)cpp/
(PAZN/Q0CA)Fpp is 1,33 for DE and 1,31 for ST. With these
figures and equations (44) and (45) ST app/ST Fpp as
well as ts cpp/ts Fpp can approximately be calculated
as a function of (PALI/Q0CA)Fpp. Thus the following
indicative curves are arrived at
go C 0,068 'sST CPP_ 1,045 ( (57)
sT FPP A FPPDE
0,088
FPP(58)
s FPP
These equations are presented in log-log diagrams,
figures 47 and 48 together with plottings for ships in
full load to indicate the influence of size. The
numbers at those points represent dwt/1.000. Due to the
divergence of the mean curves the difference in stopping
performance increases with increasing PAL/Q0CA, whioh
generally means increasing ship size. For the statistical
material, up to approximately 130.000 dwt, the mean data
in table 8 is valid.
For the lateral displacement also a comparison can, be
'made between TPP and CPP ships. From above the mean of
SL/ \-71/3 is 25 and 17 for FPP and CPP respectively.
Thus SL cpp/SL Fpp"--= 0,68 as a mean value for e > 1.
SUMMARY
- 55-
An attempt has been made to solve the stopping distance
and time with the aid of a large number of observations
from crash stop tests. The material studied consists of
330 tests from single screw vessels built at 25 shipyards
in 8 countries. 75 of the tests are from ships with CPP.
The size and type range from dry cargo vessels of 600
dwt to tankers of 230.000 dwt in ballast as well as full
load, figure 2. The statements made refer to the
material used in this study.
Based on the criterion for dynamic stability at straight
line propulsion ahead. by WAGNER SMITT 1970 and 1971, an
approximate stability criterion for vessels at crash
stop has been developed, figures 4 to 8 and equation (6).
ST CPP (
(
c. 01 ,54)
> ST
(59)
(60)
1,592T FPP
ts CPP
pu
A FPP
Q,0 cA 0,154- 1,493 )FPP
56
This criterion indicates whether the sheer Of the vessel
will be moderate or whether there is risk of large
deviation from the Original course. The criterion is
valid for trimmed as well as untrimmed vessels. It some
parts of the study the Criterion is used to divide the
material into stable and Unstable vessels.
The probability of sheer SB and PT is given for three
categories, namely vessels with FPP DE, FPP ST and CPP
DE ST, figures 9 to 11. The influence of dynamic
stability, direction of wind, propeller type and
machinery has been considered.
The turning angle at crash stop is shoWt for stable and.
unstable vessels with FPP or:CPP, figure 12. The diffe-
rence in mean value of turning angle between Ve'Ssels
with.FPP and OPP is not significant.
The influence on retardation during crash stop from
factors is discussed. It is shown that for practical
estimation of crash stop performance it is sufficient
to put ST and ts as functions of Vo, PA,A , Q0 and
CA.The astern thrust constant CA is given for FPP and
CPP as a function of PA/D from open water tests as well
as from KMW cavitation tests, figure 16. It is suggested
that the curV'es representing cavitating conditions be
used.
The stopping qualities have been studied with linear
regression analysis for the groups FPP DE, FPP ST and
CPP DE ST. The relations have been put in logarithmic
fort and it is seen that there is approximate linearity
for the logarithmic expressions, equations (44) to (47),
table 2, figures 19 and 20. The scatter of the statistical
material in relation to the respective mean curves has
been studied for the three groups, tables 4 and 5. The
57
2'X test shows that the data approximately follows the
normal distribution. Curves for +1, 2 and 2,33 standard
errors of estimate (Sy) have been evaluated. Material
for estimating and appraising crash stop performance for
FPP and CPP has thus been obtained, figures 22 to 27.In these diagrams results from rudder cycling tests
can be compared with results from crash stop tests.
The curve with + 2,33 S shows that the crash stop
result can be longer than the diagram value with 1 %
probability. Practical diagrams for each ship can thus
be designed based on the 1 % risk curve, equations (46)
and (47), table 7, figures 40, 41, 45 and 46.
The expressions on ST and ts from equations (46) and
(47) have been compared with measured values from the
crash stop tests. This has also been done for five well
known methods, namely those by CHASE et al 1957,
SAINSBURY 1963, TANI 1968, ILLIES et al 1970 and CLARKE
and WELLMAN 1971. Frequency diagrams of STc/ST and
tsc/ts are shown, figures 28 to 37. Mean values of
STc/ST and tsc /tshave been calculated as well as the
relative variation, table 6. It is seen that for the
material investigated (156 tests) the estimating method
developed in this study is the most accurate. This is so
although for the five methods with which the comparison
was made, the measured nA values at dead in the water
and for three of the methods the measured tR values
were used in the calculation.
Head reach, SH and lateral reach, SL have been discussed.
For safety reasons it is suggested that the track reach
values ST be used as head reach. Expressions for
S /pi are given for FPP and CPP vessels', equations
(50), (51), (53), (54), (55) and (56). Individual
diagrams for lateral reach for each ship, based on the
1 % risk value, are shown, figures 42 and 43.
58 -
Comparison of crash Stop performance is made between
vessels equipped with CPP on the one hand and FPP DE
as well as FPP ST on the other, based on mean value8,
equations (46) and (47) and table 2. Relations for the
comparison are given, equations (57) to (60). The crash
stop qualities for vessels with CPP are clearly superior
to those for vessels with FPP. The reason for this is
that with a CPP the thrust reversal is quick and the
backing energy is higher than with an FPP. The
difference becomes larger with increasing ship size,
figures 47 and 48.
REFERENCES
ABKOWITZ, M A: "Lectures on ship hydrodynamics - steer-ing.and manoeuvrability", Hydro- and AerodynamicsLaboratory, Denmark, Report No Hy-5, 1964
AKEN, J A VAN and TASSERON, K: "Comparison betweenthe open-water efficiency and thrust of the 'Lips-Schelde' controllable pitch propeller and those of'Troost' - series propellers", InternationalShipbuilding Progress, 1955
AKEN, J A VAN and TASSERON, K: "Results of propellertests in the 'astern' condition for comparing theopen-water efficiency and the thrust of the 'Lips-Schelde' controllable pitch propeller and the 'Troost' -series propellers", International Shipbuilding Progress,1956
AUCHER, M: "Stopping of ships. A general survey",13th International Towing Tank Conference, Berlin-Hamburg, Report of Manoeuvrability Committee, 1972
BINDEL, S and GARGUET, M: "Quelques aspects du fonc-tionnement des h4lices pendant les manoeuvres d'arretdes navires", ATMA Bull, 1962
BOATWRIGHT, G M and TURNER, J J.; "Effect of shipmaneuvers on machinery component design", Bureau ofShips Journal, 1965
CHASE., H J and. -RUIZ, A L: "A theoretical study 'of theStopping of ships", SNAME Trans, 1951
CHASE, H J et al: "Guide to the selection of backingpower", SNAME Techn and Res Bull No 3 - 5 New York,1957
- 59 -
CHIHAYA, M: "'Idemitsu Maru' completed" Japan Shipping
& Shipbuilding, 1967
CHURCH, J E: "The stopping of ships in an emergency",
IME Trans, 1957
CLARKE, D and WELLMAN, F: "The stopping of largetankers and the feasibility of using auxiliarybraking devices", RINA Trans, 1971
CLARK, D, PATTERSON, D R and WOODERSON, R K:"Manoeuvring trials with the 193.000 tonne d w
tanker 'Esso Bernicia'", RINA Trans, 1972
CONN, J F C: "Backing of propellers", IESS Trans, 1934
CRANE, C L: "Response of slowly moving ship to propellerand rudder actions", Stevens Institute of Technology,Davidson Laboratory, Report R-1169, 1966
CRANE, C L: "Methods to improve ship stopping performance"Stevens Institute of Technology, Davidson Laboratory,Report R-1208, 1967
DAILY, J W and HANKEY, W L: "Resistance coefficientsfor accelerated flow through orifices", MassachusettsInstitute of Technology, Hydrodyanmics Laboratory,
Report No 10,1953
DICKSON, A F: "What the engineer should know aboutshiphandling problems", Institution of Civil Engineers,Proceedings, 1971
DRESSLER: "Ermittelung des Stoppweges", Schiffbau, 1912
DUPORT, J: "Device for steering and/or braking a shipwith water jets", Patent application, France 5.236, 1968
EDITORIAL: "Cedros'. Combination oil/bulk cargocarrier", Shipping World & Shipbuilder, 1967
EDITORIAL: "Trials data for the 'Athenic'",Shipping World & Shipbuilder, 1967
EDITORIAL: "'Energy Transport' delivered by Sasebo",
Zosen, 1969
EDITORIAL: "'Energy Transport'. Results of sea trials
for 212.000 dwt tanker", Shipping World and Shipbuilder,1969
EDITORIAL: "'John A Mc Cone's' manoveregenskaper",Swedish Shipping Gazette, 1969
- 60-
EDITORIAL ."Safety traffic measures to be taken for.heavy congestion in Tokyo Bar, ZoSen 1970 .
EHRICKE, W and GROSSMANN, G: "Beitrag zUr'VoraUs-Berechnung von StoppmanOvern grosser Schiffe",Hansa, 1971.
ENGLISH, J. W: "Device for reducing speed or stop aship", Patent application, England No 3762/68, 1968
ENGLISH, J W: "The bow duct stopping and manoeuvringdevice", IMCO, DE/48, 1971
EZEKIEL, M and FOX, K A: "Methods of correlation andregression analysis", New York, 1967
FROUDE, 111! "On experiments with HMS 'Greyhound'",INA Trans, 1874
FUJIN°, M: "Studies on manoeuvrability of ships inrestricted waters", Japan Shipbuilding and MarineEngineering, 1969
GETZ, J R and REFSNES, B: "Experience with controllablepitch propellers in the Norwegian Coastal ExpressService", IME Trans, ,1958
GOODWIN, A J H, IRVINE, J H and FORREST, J:"The practical application of computers in marineengineering", II E Trans,. 1968
GOULD., R W "Measurements of the wind forces on aseries of models of merchant ships", NPL, Aero Report-1233, 1967,
GROSSMANN, G: "ManOverversuche mit der 'Otto Hahn'",Hansa, 1971
GROBE, H: "Vorausbestimmung der Stoppzeiten undAuslaufwege von Schiffen mit verschiedenenAntriebsanlagen", Schiffbauforschung, 1968
GUNSTEREN, L A VAN: "Hydrodynamics of controllablepitch propellers", Design and economical considerationon shipbuilding and shipping, Report of the PostGraduate Course 1969, Wageningen, 1970
GUTSCHE, F and SCHROEDER, G: "Freifahrversuche anPropellern mit festen und verstellbaren FlUgeln'voraus' und 'riickwarts", Schiffbauforschung, 1963
HAINES, R G: "Collisions in the Dover Strait.",Tahket & Bulk Carrier, 1971
TIARA, S et al: Y"Study on backing power of gearedturbine driven vessels", Shipbuilding ResearchAssociation of Japan, Report No 57, 1966
HOOFT, J P:manoeuvre",
HORNE, L R:
- 61 !-
HARVALD, S A: "Wake and thrust deduction at extremepropeller loadings", SSPA Publication Nr 61, 1967
HEBECKER, 0: "Stoppen aus voller Fahrt mit vollerRUckwartsleistung", Schiff und Hafen, 1961
HEWINS, E F, CHASE H J and RUIZ, A L: "The backingpower of geared turbine-driven vessels", SNAME Trans,
1950.
HEWINS, E F and RUIZ, A L: "Calculation of stoppingability of ships", SNAME Techn and Res Bull No 3 - 4,
New York 1954
HOOFT, J P and MANEN, J D VAN: "The effect of propellertype on the stopping abilities of large ships",RINA Trans 1968
"The steering of a ship during the stopping'International Shipbuilding Progress, 1970
ILLIES, K: l'Wechselwirkungen zwischen Maschine-uhdSchiff", STG Jahrb, 1969
ILLIES, K,, =ERSE, G, MENZENDORFF, H and RITTERHOFF, J:"Analyse des Verhaltens Von Schiff und Maschine bei -
UMsteuern"_, Schiff und Hafen, 1970.
INUBUSHI, M and SAKAI, Y; "The braking performance of
gas turbine powered ships", Japan Society of MechanicalEngineers, 1971
jAEGER, H E and JOURDAIN, DT: "Le freinage de grand8navires", ATMA Bull, 1962
JAEGER, H E: "The braking of large vessels", InternationalShipbuilding Progress, 1963
JAEGER, H E and JOURDAIN, M: "The braking ofvessels", SNAME Diamond Jubilee Proceedings,
JOURDAIN, M: "Le freinage de grands navires,Utilisation de la barre", ATMA Bull, 1965
JOURDAIN, M: "Le freinage de grands navires (VI).
Essais sur un navire a moteur diesel", ATMA Bull,1969
JOURDAIN, M and PAGE, J-P: "Essais d'arret sur le'Magdala'. Effet de la profondeur", ATMA Bull, 1969
large1968
-
KEMPF, G and HELM, .:. "Auslaufmessungen am Schiff undam Modell DaMPfer:'HaMbUrge", WRH.,'1928_
KENAN, G: "Collision avoidance between surface. shipsat short ranges", SNAME, Northern California Section,1972
KITO, F: "On the manoeuvre of a controllable pitch -
propeller", Keio University, Fujihara Memorial Facultyof Engineering, Proceedings, 1959
KOSTILAINEN, V: "Analysis of causalties to tankers inthe Baltic, Gulf of Finland and Gulf of Bothnia in 1960 -1969", Helsinki University of Technology, Ship Hydro-dynamics Laboratory, Report No 5,1971
LAMB, H, "Hydrodynamics", New York, 1945
LATINEREN, W P A VAN, MAN, J D VAN and OOSTERVELD M W C:!!The Wageningen B-screw series", SNAME Trans, 1969
LEWIS,. F. M.: "The inertia of water surrounding avibrating ship", SNAME Trans, 1929
LINDGREN, H and NORRBIN, N H: "Model tests and shipcorrelation for a cargo liner", RINA Trans, 1962
LOVER, E P: "Stopping of ships using propellers",12th International Towing Tank Conference, Rome,Proceedings, 1969
MARTIROSOV, G G: "Calculation of the reversing of aship with controllable pitch propeller", Sudostroenie,1962
MASUYAMA.,. T: "Summary of running results of steamturbine plant provided with OP propeller", Societyof Naval Architects of Japan, JOurnal, 1970
MASUYAMA, T: "Marine turbine plant with direct-drivenauxiliaries machinery and controllable pitch Propeller",Mitsubishi Heavy Industries Technical Review 1971
MEYNE, K J: "Umsteuereigenschaften von Schiffs-propellern", Schiff und Hafen, 1964
MINIOVICH, I Y: "Investigation of hydrodynamiccharacteristics of screw propellers under conditionsof reversing and calculation methods for backing ofships", Bureau of Ships, Translation No 697, 1960
MITSUBISHI HEAVY INDUSTRIES: "Research project on theuse of underwater parachutes to bring mammoth shipsto an emergency halt", Tokyo 1970
MOCKEL, W and HATTENDORFF, H G: 'Untersuchung vonStoppmanOvern in Gefahrenfalle", Schiff und Hafen, 1966
- 63-
MORGAN, W B: "Open water test series of a controllable
pitch propeller with varying number of blades",DTMB Rep No 932, 1954
MORONEY, M J: "Facts from figures", London, 1962
MOTORA, S: "On the measurement of added mass and added
moment of inertia of ships in steering motion", FirstSymposium on Ship Maneuverability, Proceedings, DTMT,
Rep No 1461, 1960
NORDSTROM, H F: "A study on the interaction between
the engine, the screw propeller and the ship", IVA
Proceeding Nr 115, Stockholm, 1931
NORDSTROM, H F: "Propellers with adjustable blades",SSPA Publication Nr 4, 1945
NORDSTROM, H F: "Screw propeller characteristics",
SSPA Publication Nr 9,1948
NORRBIN, N H: "Steuern bei geringer Fahrt", Hansa, 1964
NORRBIN, N H: "Theory and observations on the use of
a mathematical model for ships manoeuvring in deep
and confined waters", SSPA Publication Nr 68, 1971
NORRBY, R: "Notes on ships with controllable pitchpropellers with special emphasis on speed and
manoeuvring qualities", SNAME, Northern California
Section, 1970
OKAMOTO, H, TANAKA, A, NOZAWA, K and SAITO, Y:"Stopping abilities of ships equipped with controllable
pitch propeller", Kawasaki Technical Review, 1971
ONISHI, T, TOKIZANE, T and TAKASAKI, K: "Investigation
on crash astern performance of 4-cycle medium speed
geared diesel engine in emergency", Mitsubishi HeavyIndustries Technical Review, 1970
PAFFET, J A H: "Technology and safe navigation",Trans IESS, 1971 (I)
PAFFET, J A H: "The National Physical Laboratory'scontribution to efficient ship design", BritishShipbuilding Today, 1971 (II)
PARKER, M N, PATTERSON, D R and SLATER, C:"Manoeuvring trials with the 50.000 ton deadweight
tanker 'British Bombardier', Part 1. Stopping trials",
BSRA Rep NS - 141, 1966
PORRICELLI, J D, KEITH, V F and STORCH, R L:
"Tankers and the ecology", SNAME Trans, 1971
PRICE, R I: "Manoeuvring data", Marine Technology; 1970
- 64-
RITTERHOFF, J: "Beitrag zur Erhohung der Sicherheitvon SChiffsantriebsanlagen durch Untersuchung ihtes.Manoververhaltens", Schiff und Hafen, 1970
ROBINSON, S M: "Stopping, backing and turning Ships";ASNE Journal, 1916
ROBINSON, S M: "The stopping Of ships", ASNE Journal,1938
RUMS, C J:"Braking and reversing ship dynamics",aval Engineers JOurnal 1970
ODENBERG, R: "The transient performance of,propellers-and Ships during backing and reversal", FranklinInstitute Journal, 1945
SAINSBURY, J C: "Stopping the ship", Ship and BoatBuilder, 1963
SAUNDERS, H E: "Hydrodynamics in ship design", Vol Iand II, SNAME, New York, 1957
SAUNDERS, H E: "Hydrodynamics in ship design",Vol III, SNAME, New York, 1965
SHEARER, K D A and LYNN, W M: "Wind tunnel tests onmodels of merchant ships", NECIES Trans, 1960
SHELL INTERNATIONAL MARINE LIMITED: "Tanker manoeuvringcharacteristics", MRT/31, London, 1968
SHIELE, M, KQNIG, H, POTER, E0 STEFFENS, W and SCHWELLE,"209.000 - tdw - Turbinentanker 'Texaco Hamburg.",Hansa, 1969
SMITH, L P: "Cavitation on marine propellers",ASME Trans, 1937
SMITH, S ".BSRA resistance experiments on the'Lucy Ashton'", INA Trans, 1955
STEINEN, C VON DEN: "Uber Versuche zur Messung derscheinbaren Masse von Schiffsmodellen", WRH, 1933
TANI, H: "On the reverse stopping of ship's", Societyof Naval Architects of Japan, Journal, 1966
TANI, H and ISHIKURA, H: "Approximate calculationdiagram for the reverse stop ability of a ship",Nautical Society of Japan, Journal, 1967
TANI, H: "The reverse stopping ability of super-tankers",Institute of Navigation, Journal, 1968
TANI, H: "Approximate values of the reverse stoppingdiStance8 of filly-loaded large tankers", NauticalSociety of Japan, Journal, 1969.
-657
TANI, H and ENOKIDA, M: "A practical method of analysingthe acceleration and deceleration of ships", NauticalSociety of Japan, Journal, 1970
TANI, H and FUJI, A: "On crash stopping of ships",Society of Naval Architects of Japan, 2nd Symposium onShip Manoeuvrability, Proceedings, 1970
TREFETHEN, L: "The use of reversed propellers to stopbodies moving in a fluid", Naval Engineers Journal, 1962
WAGNER SMITT, L: "Steering and manoeuvring of shipsfull scale and model tests", European Shipbuilding, 1970and 1971
WAGNER SMITT, L and CHISLETT, M S: "Course stability whilestopping", International Symposium on Directional Stabilityand Control of Bodies Moving in Water, IME, 1972
WAGNER SMITT, L and LANDSBURG, A C: "Note on added massin surge measurements with model of large tanker",13th International Towing Tank Conference, BerlinHamburg,Materials for Reports, 1972
WILSE, T: "PrSvning air nye skip", NSTM, Stockholm, 1962
TABLE 1
Data for evaluating the stopping_performance ofns
Yard number /23Ship type .7.--0M/KgR
LOA(length over all) 26441M
L between perpendiculars)
4,1B (moulded beam)PA
M d (mean draught) ./.6/ 6 f IIdwt (tons dead weight) ./.Aq .Q9.Q..V (volume displacement at mean draught)
.24/6OOOOOOOOOOO
cf3
P Type of machinery (steam turbine or diesel engine) .4.99 SHP (shaft horse power, design. For CPP ship also design
0 power of shaft driven auxiliaries) t211 009
Type of propeller (FPP or CPP) F PPRight- or lefthanded propeller R/6.HTZ (number of blades)
di
D (propeller diameter) 1.4:679./. O
k P0 pitch at 0.7 radius)0 uH
AE /A0(expanded area ratio) .P4444;4
T-1 /7a RPM° (propeller revolutions per minute, design) .... OOOO
0P V (speed of ship, design) Adir. c?!".crnsa D
E (distance from propeller shaft centre to ship's base
line) A.A-47.4?.!1. OOOOO
dF (draught at forward perpendicular) 614416.1 OO .
dA (draught at aft perpendiCUla±)
.(weight displacement) 9.4??. .........
Wind speed and direction . .....
Sea .....
Waterdepth 170
p., Vo (approach ship speed at stop order
0 t (stopping time) acsi o'rpti
-P sV = fi (t) and RPM f2 (t)
Ship's course during stopping
TABLE 2
Result of regression analysis of equations (46) and (47) for mean track reach and stopping time
NaS
bS
rS
tS
10g
lo
s Y S
at
bt
rt
1°log
Sy t
FPP, DE
82
4,44
0,684
0,93
22
0,094
7,80
0,693
0,93
23
0,090
FPP, ST
79
2,88
0,770
0,86
15
0,100
6,44
0,759
0,89
17
0,086
CPP, DE and ST
46
3,89
0,616
0,97
25
0,066
8,17
0,605
0,97
25
0,067
TABLE 3
Mean value and variation of reversed shaft speed
Pearson'scoefficient ofvariation %
TABLE 4
Distribution in % of the total number of observations .
within the limits + 1 and + 2 standard errors of estimate
respectively
gST/v20 gt vo
+ S +.2 S + S + 2 S_ _ _
TABLE 5
Variability in % for one standard error of estimate
g 5T1vo2g t5/v0
FPP, DE 0,65, 16
FPP ST 0,53 13
CPP, DE and ST 0,96 5
PPP, DE 67 95 .66 95
FPP, ST 63 96 68 96
CPP, DE and ST 70 96 67 93
FPP, DE 22 21
FPP, ST 23 20
CPP, DE and ST 15 15
TABTR 6
Relation between calculated
and real values of stopping
distance and time
Number
Combination
of tests
8Tc/
8TPearson's
Average
coefficient
value
of variation %
tB
Cit
Average
value
BPearson's
coefficient
of variation
PPP,
all Ships
156.
0,76
825
,80,
882
30,9
FPP, tank + bulk
113
0 81
624
,50,
959
28,6
CPP
, all
ship
s29
0,91
822
,31,
034
26,0
PP, tank + bulk
250,
941
21,8
1,05
526
,2
FPP, all ships
-
156
0,94
825
,30,
925
28 4
-
FPP, tank + bulk
11.3
0,94
724
,10,
963
28,5
FPp, all ships
156
0,93
138
,51,
188
49,2
FPP, lank + bulk
113
1,02
834,0
1,31
345
,8
PPP, tank + bulk
95
1,04
035
,51,
341-
47,9
>40.000 dwt
FPP, all ships
156
0,74
437
,80,
964
39,8
FPP, tank + bulk
113
0,81
932
,71,
075
34,3
FPP, all Ships
156
1,09
025
,01,
167
31,4
YPP
1tank + bulk
113
1,12
0.23
,61,
249
30,1
PPP,
all
ship
s15
,6,
0,99
523
11,
026
20,0
FPP, tank + bulk
113
0,99
9:2
3,1
1,01
520
,6
CPp, All
ship
s46
1,01
115
,31,
007
15,5
OPP, tank+ bulk
321,
003
17,4
0,99
416
,6
TABLE 7
The
1 % risk curve coefficients for track reach
and stopping time in equations
(46)
and
(47)
bt
bS
0,69
3
0,75
9
- 0,
605
FPP,
DE
7,35
0,68
412
,65
FPP,
ST
4,93
0,77
010
,21
CPP, DE and ST
5,55
0,61
611
,70
1>crash stop
2?
up to r-4130.000
dwt
TABLE 8
Comparison of estimates of stopping performance between PPP and OFF ships
Figure 9. Probability of sheer to SB or PT during a crashstop, taking into consideration type of propeller,machinery and propeller rotation effect .(circled.figures indicate number of cases investigated).Assumption: Propeller is left hand rotating afterreversal.
32% 68% 18% 82%
FPP ST
FPP DE
CPP DE ST
26% 74%
wind
24% 76%
20%
50%
wind wind
60% 40%
50%
35% 65%
wind
Figure 10. Probability of sheer to SB or PT during a crash stopfor be> 1 taking into consideration type of propeller,machinery, propeller rotation effect and directionof wind (circled figures indicate number of cases inves-tigated). Assumption: Propeller is left hand rotatingafter reversal.
FPP ST
FPP DE
CPP DE ST
25% 75%
14%
wind
86%
43% 57%
wind
<
60%
70%
92%
Figure 11. Probability of sheer to SB or PT during a crash stopfor 4t4.1 taking into consideration type of propeller,machinery, propeller rotation effect and direction ofwind (circled figures indicate number of cases investi-gated). Assumption: Propeller is left hand rotatingafter reversal.
PT 041N = 13
T's = 24°
SB
N = 44
T= 53°
PT
N=3
af<1
Figure 12. Turning angle
Voat 13 knots
-
40-values are
FPP
CPP
.PT eb1N = 33
T.= 112°j
PT
N= 5
4' for FPP- and CPP-ships,
Approximately 67 % of all
in the shaded sectors.
0
0 ts
Figure 13. Relation between ship speed and time duringa crash stop. Track reach ST is representedby shaded area below 1,7"-t curve.
11-1. -
centrifugal force
Figure 14. Derivation of the longitudinal component ofthe centrifugal force on a turning ship.
awmean/amean
0,20
0,15
0,10
0,05 0
95%
05
ia15
Vwind speed mjsec
Figure 15. Influence of wind resistance on total retardation during crash
stop
.Wind on bow, no turning, Vn = 16 knots. 67 and 95 % of all wind speeds
observed are below 7 and 1
m/sec respectively.
CA
1111
1111
111R
Nio
nli_
au11
1111
111M
1111
RIM
MII
IIIN
S111
1111
1111
1111
1111
1111
1M
IA 1
1111
111M
ira
El
III
1111
1011
1111
111
EM
I=11
1511
1111
1 M
EM
ME
='A
MI
II N
M=
111
1111
1111
.11
1111
1111
111.
1111
1111
1111
1101
1111
1111
1111
111
1111
1111
1111
1111
1111
11.1
1111
111.
11M
MIN
I
'.
0,4
0,5
0,6
0,7
0,8
0,9
1 , 0
1,2
PA/D
Figure 16. CA as a function of PA/D at bollard pull astern free running testswith models of
FPP and CPP under atmospheric
((A
Pand cavitating
(AA
)conditions respectively.
Mean lines for FPP and CPP based
nnYmw
% N
(to
tal o
bs)
80 70 60 50
FPP
DE
.N
= 8
440 30 20 10 0
030
6090
120
150
tR s
ec
FPP
STN
= 7
4.
030
6090
120
150
180
210
240
tR s
ec
Figu
re 1
7. F
requ
ency
dis
trib
utio
n of
tR f
or F
PP D
E, F
PP S
T a
nd C
PP D
E S
T.
CPP
DE
ST
.N._
= 4
2
r-t
I.0
3060
90
tR s
ec
0,4
0,5
0,6
0,7
0,8
0,9
0,3 0,4 0,5
0,6
0,7
0,8
0,9
1,0
1,1
nA/n0
nA/n0
nA/n0
Figure 18. Frequency distribution of nA/no for FPP DE,
It has been possible to conduct this investigation withthe goodwill and cooperation of the following personsand organizations to whom I wish to express my sincere
Institut de Recherches de laConstruction Navale, Paris
MitsUbishi Heavy Industries, Tokyo
British Ship Research Association,Walls end
Decca Navigator, Copenhagen
Granges Rederi, Stockholm
Lindholmens Vary, Gothenburg
Technische Universitat, Hamburg
HowaldtswerkeDeutsche Werft, Kiel
Gotaverken, Gothenburg
Marcona Corporation, San Francisco
University of Tokyo, Tokyo
Odense Staalskibsvaerft, Odense
Kockums Mekaniska Verkstad, Malmo
Statens Skeppsprovningsanstalt,Gothenburg
Shell International, London
Rheinstahl Nordseewerke, Emden
Uddevallavarvet, Uddevalla
Naval Ship Research and DevelopmentCentre, Washington
Uddevallavarvet, Uddevalla
SOGAARD
T TANAKA
TANI
TANIGUCHI
I E TELYER
T WILSE
Nakskov Skibsvaerft, Nakskov
Mitsubishi Heavy. Industries,Yokohama
Tokyo University of MercantileMarine, Tokyo
Nagasaki Technical Institute,Mitsubishi Heavy Industries,Nagasaki
BP Tanker Co, London
Det Norske Veritas, Oslo
I would also acknowledge the help afforded me by friendsat BirdJohnson in Walpole, Baying in London, KjellbergKabushiki Kaisha in Tokyo and by colleagues on the staffof Karlstads Mekaniska Werkstad in Kristinehamn.
For the permission granted me to publish this reportthanks are due to Karlstads MekaniSka Werkstad.
CONTENTS
Page
Synopsis 1
Symbols 3
Introduction 8
Means and methods of improving the stoppingqualities of ships 10
Available data on board for estimatingships' stopping qualities 12
Methods for calculating stopping performance 14
Collection and processing of crash stoptest results 15
Course stability during the stoppingmanoeuvre OOOOOOOO 4 18
Sheering tendencies 23
The turning angle 26
Determining track reach and stopping time 27
Comparison of various calculation methodsfor estimating the stopping performance 44
The stopping equation for one per cent risk 49
Practical crash stop diagrams 51
Comparison of stopping performance with FPPand CPP ships 53