SNAME Transactions, Vol. 86, 1978, pp.

197-217

Design of Bulbous Bows

Alfred M. Kracht, 1 Visitor

There is no doubt that bulbous bows improve most of the properties of ships, but the correct design
and power prediction for ships with bulbous bows are still difficult due to the lack of design data. In
the paper, a quantitative design method is presented together with the necessary data providing rela-
tionships between performance and main parameters of ships and bulbs. The data, in the form of
design charts, are derived from a statistical analysis of routine test results of the Hamburg and the
Berlin Model Basins, HSVA and VWS, respectively, supplemented by results of additional tests to
fill the gaps. Three main hull parameters are taken into account: block coefficient, length/beam
ratio, and beam/draft ratio, while six bulb quantities are selected and reduced to bulb parameters,
of which the volume, the section area at the fore perpendicular, and the protruding length of the bulb
are the most important. For power evaluation, the total power is subdivided into a frictional and a
residual part. Depending on bulb parameters and Froude number for each block coefficient of the
main hull, six graphs of residual power reduction have been prepared. Because of the wide range
of block coefficients, there are so many design charts that only one example is presented herein.

Introduction
NEARLY 90 YEARS AGO, R. E. Froude [112interpreted the
lower resistance of a torpedo boat, after fitting of a torpedo tube,
as the wave reduction effect of the thickening of the bow due
to the torpedo tube. D.W. Taylor was the first who recognized
the bulbous bow as an elementary device to reduce the wave-
making resistance. In 1907 he fitted the battleship Delaware
with a bulbous bow to increase the speed at constant power. In
spite of great activities in the experimental field to explore its
potential, 70 years had to pass before the bulb finally asserted
itself as an elementary device in practical shipbuilding. A
suitably rated and shaped bulb affects nearly all of the prop-
erties of a ship. Especially for fast ships, the use of a bulb allows
a departure from hitherto accepted design principles for the
benefit of a better underwater form. The higher building costs
are the only disadvantage.
The protruding bulb form affects hydrodynamically a
variation of the velocity field in the vicinity of the bow, that is,
in the region of the rising ship waves. Primarily the bulb at-
tenuates the bow wave system, which usually is accompanied
by a reduction of wave resistance. By smoothing the flow
around the forebody, there is good reason to believe that the
bulb tends to reduce the viscous resistance too. Therefore, the
beneficial action of a protruding bulb depends on the size, the
position, and the form of the bulb body. See Fig. 1.
The linearized theory of wave resistance has provided the
main contribution to the understanding of the bulb action
(Wigley [3], Weinblum [4], Inui [5, 6]). But it is of no great use Fig. 1 Bow wave pattern of a model without (upper picture) and with
for the project engineer. In the preliminary stage of his project, bulbous bow (lower picture) (Ca = 0.6, FN = 0.29, ABT/AMs =
he needs fundamental information on which to base concrete 0.074).
decisions. Later, in the realization stage, the quantitative as
well as qualitative guidelines are important, because the hy- of action of a bulb and the influence of bulb parameters on
drodynamic phenomena are not describable by few geometric resistance or power reduction, respectively, are described in
form parameters alone. For this reason, in this paper, the mode a qualitative manner; guidelines for bulb design are also in-
troduced. The design charts presented here are the result of
i vws Berlin Model Basin, Berlin, Germany. a research project, that is, of an analysis of routine test results,
Numbers in brackets designate References at end of paper.
Presented at the Annual Meeting, New York, N. Y., November of the Hamburg and the Berlin Model Basins supplemented by
16--18, 1978, of THE SOCIETY OF NAVAL ARCHITECTS AND MARINE results of additional tests to fill the gaps. The charts are new
ENGINEERS. and therefore still in need of improvement and completion.
197

Base
I - T (b) The stem outline is changed by the protruding bulb as
a, A-Type b, O - T y p e c, V-Type
Fig. 2 Bulbtypes
Bulb forms and parameters
For an adequate presentation of the hydrodynamic prop-
erties of bulbs, it is necessary to systematize the different ex-
isting bulb forms [7] by means of geometric parameters. Ob-
viously a definitive description of a bulb shape~ just as for a ship
form, with a finite number of geometric parameters, is i m -
possible. But a rough classification is possible using only few
parameters.
With the shape of the cross section ABr of the bulbous bow
at the forward perpendicular (FP) as the main criterion, one
can differentiate three main bulb types (Fig. 2) [8]:
(a) A-Type: Fig. 2(a) shows the drop-shaped sectional area
A n t of the delta-type with the center of area in the lower-half
part, This shape indicates a concentration of the bulb volume
near the base. The Taylor bulb and the pear-shaped bulbs
belong to this type.
(b) O-Type: This type (Fig. 2b), with an oval sectional area
A n t and a center of area in the middle, has a central volumetric
concentration. All the circular, elliptical, and lens-shaped bulbs
as well as the cylindrical bulbs belong to this type.
(c) V-Type: The nabla-type also has a drop-shaped sec-
tional area ART (Fig. 2c), but its center of area is situated in the
upper-half part, indicating a volume concentration near the
Nomenclature
ABL = area of ram bow in longitudinal plane, m~
ABT = cross-sectional area at forward perpendicular
(FP), m2
AMS = midship section area, m2
Ba = maximum breadth of bulb area Asr, m
BMS = beam, midship, m
Ca, CM, Cp, CWL = block, midship-section,prismatic, and waterline
coefficients, respectively
CpE = prismatic coefficient, entrance
CF, Ca = frictional or residual resistance coefficient, re-
spectively
Cpr~R = residual power displacement coefficient
ACeva ffi residual power reduction coefficient
CAaL = lateral parameter
CABr = cross-section parameter
Css = breadth parameter
CLp. = length parameter
Cves = volumetric parameter
Cza = depth parameter
D, Dwa~e = diameter of propeller, or wake field, respectively,
m V W L = displacement volume, m3
FN = V s / ~ g × Lee = Froude number
Ha = total height of Ant, m
LE, Lee = length of entrance, or between perpendiculars,
respectively, m
LpR = protruding length of bulb, m
Po, PF, PR = delivered, frictional, or residual power, respec-
tively, PS
198 Design of Bulbous Bows
free surface. Because of its favorable seakeeping properties,
this type is the most common bulb.
With respect to the lateral contour of the bulbous bow, two
typical classes are distinguishable:
(a) The stem outline remains unchanged as with the Taylor
bulb. These bulbous bows do not have favorable properties and
are no longer built today.
with all modern bulbous bows.
In addition to these classification criteria, quantitative bulb
parameters are necessary for delineation of the bulb form. The
author is of the opinion that six parameters are sufficient for
all practical purposes. Figure 3 shows the three linear and three
nonlinear geometric bulb quantities that are reduced to the bulb
parameters, that is, normalized by the main dimensions of the
ship, as described in the following.
The three linear bulb parameters are
1. The breadth parameter, that is, the m a x i m u m breadth
BB of bulb area A s r at the FP divided by the beam BMs of the
ship
Can = BB/BMs (1)
2. The length parameter, that is, the protruding length Lea
normalized by the Lee of the ship
CLPR = ErR~Lee (2)
3. The depth parameter, that is, the height ZB of the fore-
most point of the bulb over the base divided by the draft TFe
at the FP
Czs = ZB/rFp (3)
The variation of the linear bulb parameters is easily possible
during the project phase. The breadth BB is not necessarily the
maximum breadth of the bulb body that, for hydrodynamic
reasons, can also be located before the FP. The depth pa-
PEF, PER = effective frictional or effective residual power,
respectively, PS
Ae'a = residual power reduction factor
RF, Rr, Rv = frictional,total, or viscousresistance,respectively,
kp
RVR, R w B , RWF = viscous residual, wave-breaking, or wavemaking
resistance, respectively, kp
ARwn, ARwF = bulb effect on wave-breaking or wavemaking
resistance, respectively, kp
ARv = secondary bulb effect, bulb effect on viscousre-
sistance, kp
RN = Vs × Lpe/u = Reynolds number
S, SBtot = ship surface or total bulb surface, respectively,
m2
TFp, TMS = draft at FP or midship, respectively, m
V s = ship speed, knots
V Btot = total bulb volume, ma
V P R = volume of protruding bulb part, m3
ZB = height of the foremost bulb point over baseline,
m
t , w = thrust deduction or wake fraction, respectively
riD = propulsive efficiency
p = viscosity of water, s/m a
p = density of water, kp-sZ/m4
g = acceleration due to gravity, m/s 2
 / ~PR = ~
Lp~
ABT(X)dX
0.3
W
additive bulb ABTIAMs:0.057

thr_uS~tdedudion fraction t

T l
TFT
C B = 0.80
).2 0.1 1 I -t
0 22 Q 24 0 26 FN 28
'Base
F.IP x
~ _BB = = Lp R implicit bulb ABT//~S =0.101

Fig. 3 Linear and nonlinear bulb quantities o21 .~_ w o k e fraction w

rameter is a valuation factor of the beach slope of tbe.bulb top

(thick line inFig. 3). . ).20 fraction t f----,,,
The three nonlinear bulb parameters are
1. The cross-section parameter, that is, the cross-sectional i
%=0.57 \
area ABT of the bulbous bow at the FP divided by the mid- 0.15
shiwsection area AMs of the ship , i28, I I0 ' \-.FN
CABT= ABT/AMs.. (4) Fig. 4 Influence of a bulbous bow on thrust dedqction and wake
2. • The lateral parameter, that is, the area of ram bow ABL fraction (. . . . . with, - - - - without bulb)
in the longitudinal plane norma!ized by AMs
CABL ABL/AMs
---- (5) 0,7 • cB FN ABT .S
8. The volumetric parameter, that is, the volume VpR of A 057 0.322 0.101
the pr6truding part of the bulb divided by the volume of dis- ~< B 0.59 0.271 0085
placement V wL of the ship C 0.80 0.177 0057

CveR= VeR/V,wL (6) \• , \-.mod.C

\
The v o l u m e V/'t{ is the nominal bulb volume. The total or
effective bulb volume 'v' ntot is the sum of ~/PR and the fairing
volume VF, which results from the fairing of thebulb into the
ship hull. N \ "
Finally, a distinction is possible between an additive and an
implicit bulb. An additive bulb ir/creases the displacement
volume V wt of the ship by the effective bulb volume V Btot.
The sectional area curve of•the original hull remains unchanged. mad A im ,cit }
On the other hand, the effective volume V Btot Of an implicit mod.B additive butb
bulb is part of the displacement volume V wc of the main hull mod.C additive
that is shifted from unfavorable.regions and concentrated in
the vicinity of the forward perpendicular. By this process, the ' oJ4 , -oP6 , ,8, '
sectional area curve of the original ship is changed. x =d IDwake
Fig. 5 Influence of a bulbous b o w on the radial distribution of cir-
• Influence of a bulbous bow o n t h e properties of a cumferencial average nominal axial w a k e component (. . . . . . . with,
ship m - - without bulb) .

Before discussing the influence of the bulbous bow on the

ship's resistance and required power, respectively, we should
mention other important hydrodynamic qualities which play
a r01ein the decision whether a bulb should be used or not. The
change of resistance influences the thrust loading of the pro-
poller and, consequently, other propulsive Characteristics of the
ship; fbr example, the quasi-propulsive coefficient, the wake,
and the thrust deduction fraction [9-11]. Figure 4 shows this
indirect influence of a bulbous bow on {hrust deduction and
.wake fraction. Both are increased by an additive as well as by
an implicit bulb, if the bulb ship has a loWer total resiStance than
the bulbless form. But there is also a direct influence of the
bulbous bow on the wake distribution in the propeller •plane..
In Fig. 5 the radial distributions of axial wake components Of
ships with and without a bulb are compared. Within the
propeller disk area the axial wakes £f all bulb ships are higher
Design of Bulbous Bows
than the wakes of the bulbless ones. 'The reason for this is t,he
change of flow around forebod~¢and bilge, which is obser,~able
in the model case up to the propeller disk [12, 13]. But in the
•correlation of model test and full-scale results, scale effects play
a very important role [10] and it is not certain if this bulb effect
is also found at the ship.
Although unfavorable effects are possible, bulbousbows in
general do not influence the course stability or the maneuver-
ability [14]. No significant changes of the overshoot angle or
the period in zigzag tests could be established. The bulb is an
ideal place for the arrangement of bow thrusters and acoustic
sounding gears.
The seakeeping qualities of a ship are a special problem, and
a very broad field, which will be discussed her~ only briefly.
199
 r 2/3-,. Resistance and bulb effect
/I k= 2T/g The most important effect of a bulbous bow is its influence
• T 2~ / 3_~ TI=I.0 *_- on the different resistance components and consequently on
_ D 2~C]T2 co the required power. Although the design charts represent the
power reduction due to a bulb, for a better understanding
O2_- - 1/2 " "
the hydrodynamic, phenomenon shall be discussed by means
of the influence of the bulb on the resistance. For this purpose,
• 1/3 1 .
the following subdivision of the total resistance is used .
Rr = Rv + RwF + RwB = Rp + Rvn + RwF + RwB (7)
OJ
Where -

\ Rv = viscous resistance
RF = frictional resistance
RvR = viscous residual resistance
RWF = wavemaking resistance
0 Rwn = wave-breaking resistance
Fig. 6 Damping coefficient D* of partially immersed bulbous cylinders The latter two components are related to wavemaking. Their
as function of wave number k [18] contributions to the total resistance are very different for Ships
with different block coefficients and speeds. Here, an expla-
nation is to be found, for the fact that the resistance reduction
Except for the relative foreship motion against the water sur- due to a bulb for full, slow ships can exceed the wave resistance
face, the bulbous bow has no unfavorable influence on either alone, which at FN < 0.2 is a negligible part of the total resis-
the remaining motions or on the maximum bending moment tance.
in the midship section [15, 16]. In spite of the higher relative The additional bulb surface always increases the frictional
motion of a bulb ship, the danger of slamming of a well-shaped resistance RF, which is the main part of the viscous Component
bulb is no higher than with the bulbless ship [17]. In detail, the Rv.. Up to now, it is not quite clear whether the bulb affects
bulb mitigates the pitching motion of the ship by its higher the viscous.residual resistance RvR due to the variation of the
damping. It should be mentioned here that the damping velocity field in the near bow range [25, 26]. But in the rean-
coefficient of bulb cylinders in a two-dimensional case vanishes alysis of test data based on Froude's method, presented here,
for a certain wave number [18] as shownin Fig. 6. Since •this open point isof no account.
nonbulbous cylinders do not show this quality, the damping There is no doubt concerning the influence of the bulbous
effect of a bulbous bow, for example, of O-type, can vanish for bow on wavemaking resistance Rw~. The linearized theory
definite wave numbers, and the bulb.ship moves like a bulbless of wave resistance has rendered the most important contribu-
one. tion to the clarification of this problem [3, 4]. According to
this theory, the bulb problem is a pure interference problem
In regular waves, model tests show that the critical Froude
number FN from which the bulb ship begins to be superior of the free wave systems of the ship and the bulb. Depending
increases with increasing ratio X/L [15]. As a function of k/L on phase difference and amplitudes, a total mutual cancellation
of both interfering wave systems may occur. The position of
at constant Froude number, the resistance of a bulb ship in-
creases more rapidly than that of the bulbless form. Therefore, the bulb body causes the phase difference, while its volume is
most of the smooth-water advantages of bulb ships vanish above related to the amplitude. The wave resistance is evaluated by
about X/L = 0.9. In irregular waves, nearly all bulbous ships analysis of the free wave patterns measured in model experi-
have disadvantages above Beaufort 8. Since in the North At- ments [5, 26-28].
The wave-breaking resistance Rwn depends directly on the
lantic the probability of the occurrence of wind intensity more
rising and development of free as well as local waves in the vi-
than Beaufort 8 is only about 10 percent, and up to X/L - 0.8
the bulbous ship is the best ship [20] regardless of seakeeping cinity of forebody and is a question of typical spray phenom-
enon. Understandingof the .breaking phenomenon of ship
aspects, the bulb design consequently may be carried out in
waves is important for the bulb design for full ships. Rw8 in-
view of the smooth-water performances only.
cludes all part.s of the energy loss by the breaking of too-steep
In navigation in ice, the bulbous bow has proved to be ad- bow waves. The main part of this energy can be detected by
vantageeus. Its form enables a tipping of ice floes coming from wake measurements [23, 26]. The local wave system contrib-
the front in such a way that they glide along the hull of the fo~ utes the main part to this resistance component. This wave
reship with their wet-side friction coefficient, which is small. System consists mainly of the two back waves of bow and Stern
Due to this effect, the speed loss of a ship with a bulbous bow which are generated by deflection of the momentum. The
is smaller than that of a bulbless one. deflection rate of the flow is a degree of the steepnessof these
If from the beginning a bulbous bow is included in the back waves, of which only the bow wave is of a practical im-
shaping of the underwater hull form, then for fast ships espe- portance in bulb design.
cially it is possible to leave the traditional recommendations on The wave-breaking resistance can be diminished only insofar
the fullness of the forebody and the unavoidable abaft position as it is possible to prevent the breaking of bow waves. Ac-
of the center of buoyancy. -Without disadvantages, the bow cording to the reason of its creation, this is only possible by
bulb allows a fuller foreship form and therefore better trim and changing the deflection of momentum or the bow near the
stability properties. Using an implicit bulb, a more slender velocity field, respectively. In principle this may be achieved
aft-body is possible at constant total block coefficient with not only.by a bulbous bow, butby suitable hydrofoils as well
improved propulsive performance. Without increase of the [29]. A theoretical treatment of the linearized problem has
resistance, a greater angle of entrance for the waterlines can recently begun [23, 80].
be used as compared with the accepted practice so far [22]. The effect of the bulb on the differeht resistance components

200 Design of Bulbous Bows

 L

can be discerned by taking the differences of the corresponding ARWF

resistance components of the ship without (index o) and with ARwF * ARwB
bulb (index w):' -, .. .
,< o3 ) !
• " A R T = RTo-- RTw = z~Rv + ARWF + ARwB- (8): / / ~, . f t h e o r y • ARwF
Consequently it is possible to define three different bulb effects. ~/'/ 0.2R / / -' ' , .
In any case, a positive bulb effect ~R meansa resistance re-
duction, and .vice versa• The two latter terms in 'equation (8) //i
account for the primary bulb effect, which for bulb design are " //.
the most important. The difference of the v~ave resistances.
-O. 1 / / C 25 '~!0 O ,05 0.1 "
" ARwF = RwFo - RwFw = R w I + Rwb (9)' . /~1". F. ~P. ,----#~,' x
!\ \.~y./ . ~!\,
is the interference effect, Which is the sum of the interference
resistance Rwl and the wave resistance of the bulb body Rw/,
// I -o.1
experiment l
alone. Its contribution to the total bulb effect can be estimated
ARwF÷ ARw8 -0.2
by an analysis of the interfering free wave patterns [26]. Ac-
cording to Froude's law, it can be scaled directly to the full-scale ,CLp R = x * 0.013
ship. Only for slender fast ships does it give the maid propor-
tion to the tot&l bulb effect, where the amount depends es~n- Fig. 7 Dependence' of interference and primary bulb effect on the
tially On the bulb volume and the sign on the longitudinal po- length parameter CLpa = LpR/Lpp. Comparison of theory and eXperi-
sition of the bulb center, ment with an elementary ship of theform (2,4,6,0.72,1.0) [31].
The difference between the•wave-breaking resistances-
ARwR = RWao RWBw (10) Because of the doubts connected with the secondary bulb effect,
the following consideration is confined to both parts of the
is the breaking effect, which is the main contribution to the total primary effect.only--to the interference and breaking effects,".
bulb effect forfull, slow ships. Its contribution is: the bigger respectively. The relationship between the two and the
the bulb, the better the deflection of the flow in the vicinity of • magnitude of their contributions to the total bulb effect are not
the bow region. This means for the bulb form an optima] discussed at this stage " . . .
distributed bulb volume in the longitudinal direction to mini~ " According to linearized theory, the interference effect de-
mize gradients of the hull surface in the region of rising bow pends on the volumetric parameter C v e R = V e a / V w L in a
waves. Using geosim model tests, Taniguchi [12] and Baba [23] .quadratic manner-f82]. C'¢en is a measure of the amplitude.
have. shown that the wave-breaking resistance, and conse- of the wave pattern. •The breaking effect shows a similar de-
quently the breaking effect, is Froude number dependent. pendence. With increasing bulb volume, both effects increase
The difference between the viscous resistance parts up to a maximum with a subsequent decrease. The optimal
ARv Rvo - Rvw = (RFo - RF~) + (Rvao - RvnW) (ll) bulb volumes corresponding to the maximum Values of the
different bulb effects do not, in general, coincide'. For the.
is the secondary bulb effect, Which is of minor importance for interference effect, the optimal Volume can.be estimated'for "
the total bulb effect. Due to the larger surface of bulb Ships, agiven ship-bulb combination by the wave cut method [26]:
the fricti0na[ resistance term of equation (11) is always negative .In a similar way, the interference effect depends on the. breadth~
and diminishes the bulb action.; For reasons mentioned before, ' ' and cross:section parameter.
the contribution of the difference of viscous residual resistance For a constant bulb volume and depth, the length parameter
is not taken into account. -. Ct, en = L e R / L e e has a great influence on the interference ef-
. Finally, the question has to be answered whether equivalent fect. As it is a measure for the phase relation of the fr.ee Wave
variations of the ship may result in the Same improvements as, systems of ship and bulb, typical maxima and minima appear
an addition of a bulb, for example; an increase of block coef- as a direct consequence of the interfering waves, b~sshown i n
ficient C~ corresponding to the bulb volume, or an elongation Fig. 7, this tendency is confirmed by model experiments ]81]: ~
of the ship length corresponding to the bulb length Lea. It has The influence of the length'parameter on the breaking effect
been shown by model experiments [14] and by linear theory [22] can be caught intuitively by its mode of action. With in-
that the effect of a bulb cannot be achieved by form variations . creasing CLea, this effect increases at first and after a maximum
[0].. " ' ' "
decreases monotonically to zero, due to the fact that the de-
flection of momentum in the vicinity', of the bow is hardly al-
Influence of bulbparameters on bulb~effeet. tered by a very 10ng cylindrical bulb. BeSause the lateral pa-
rameter is strongly related to the length parameter, its influence
At constant Froude number EN, the bulb effect.is a function on the bulb.effect is'similar•
of'all six bulb parameters: The dependence of the interference effect on the depth pa-
AR = F ( C v e a , CABr, CABL, CLeR,CBB, CZB) (12) -rameter CZB = Zn/.TFe is described simply by linear theory
because the term.ZB of a spherical bulb coincides with the
This multidimensional relationship complicates the under- center of the sphere. If such a bulb of constant volume and
standing of the dependencies on single parameters, the longitudinal position is moved from infinite depthup to the
knowledge of which is very helpful for bulb design. Unfor- water, surface, the interferential effect increases! at first
tunately, a quantitative description o f these dependencies is monotonically from zero to a maximum, decreases subse-
possible in only a few cases; because systematic model experi-~ quently, and finally becomes negative due to an increase of the
ments are too expensiveand some parameters e`annot be varied resistance of the emerging bulb bo_dyl The breaking:effect
independently. On the basis of linearized wave resistance behaves similarly, but it can become positive again~fflZB'~'.>%TFe,
theory, however, a qualitative picture can be developed, sup- as with the V-type. In this case the behavior of a.:sh![i-'bulb
ported by special model experiments,i for example, the wave • Combination is similar to a longer main hull increased in length
cuts[26[, which not only prove the tendency of the dependence. by LeR.
Design of.Bulbous Bows 201
 longitudinal position, the optimal bulb volume increases with
~;~ ~ Cp 0.72 increasing depth position ZB. Moreover, both figures clearly
o~, i i.~,< L lT.s /'1~\, ///i// indicate the enormous influence of ship speed on optimal bulb
volume, which increases in an undulating manner with in-
D4 ~ v ~ X;. ,.11,/ creasing speed. These theoretical results are upper limits for
bulb posit. Z~=O Lpt=r / / ""--J /" / I the actual effects. While their absolute values are hardly of
elementary ship I.I / .... ...0.64' , practical interest, the tendency of the dependence of bulb
0..33_ ( 2,4,6,C,,1.0) ' ~/f-~ "'~'-...I" /"/i"',, volume on speed, block coefficient, length of entrance, and bulb
position is useful for the actual design.
• kl/i . . . . . <,o.:6o .,." Since the wave system is created only by the nonparallel part
02 , /I// ""--._J
of the main hull, and in the real fluid the forebody makes the
main contribution, the length LeM of the parallel middlebody
....... .o o ..,-." has hardly any influence on the bulb size and, therefore, the
same holds for the length/beam ratio Lee/BMs too. For ships
o , ll , i l l l l l l , I I with Lpe/B~4s > 5.0, the wavemaking resistance is only a
0.20 025 030 " 035 FN function of LE/BMs, as shown by the upper limit curves in Fig.
10 (see also Baba [23], Fig. 10). The beam/draft ratio BMs/TMs
Fill. 8 Optimal bulb volume V k of a ship-bulb combination as a has a great influence on bulb effect, bulb size, and draft pa-
function of Froude number FN. The prismatic coefficient Cp of the rameter CzB (upper limit curves in Fig. 11). Consequently,
elementary ship form (2,4,6,Cp,1.0) is the parameter of the curves for dimensioning of a bulbous bow, the main-hull parameters
[3t]
are Ce~, BMS/TMs,and LE/BMs. Unfortunately, in the pre-
liminary design phase, Cp~, and often LE too, is unknown.
Therefore, the following design guidelines are mainly based
only on block coefficient CB and beam/draft ratio BMS/
~-L J - f /
elementary ship. / f -"\ZB/TMs=-0.25 / TMS.
006 ' ( 2,4,6,0.72,1.0 ) / / \ \ /

oo, i/
/
/

0.04
0.03
0.02 / \\
• /
0.01 ~ , ,
"'--"
0_ ~ I I I I I
020 0 25
Fill. 9 Optimal bulb volume V k of a ship-bulb combination with CP
= 0.72 as a function o! Froude number FN. Depth position ZB is the
parameter of the curves [31] : Since the bulbous bow affects primarily the wavemaking
' I n f l u e n c e of s h i p m a i n p a r a m e t e r s on b u l b Size
and bulb effect
Linear theory permits comment on the influence of some
ship main parameters on bulb size. The theoretical results
relate to the interference effect only, but have general validity
throughout and, in particular, are applicable to the breaking
effect. For discussion, the most suitable case is the optimal
bulb, which minimizes the wave resistance of a ship-bulb
combination. With elementary ships of the form (2,4,6,Ce,1.0)
[31], it may be shown that an increased prismatic coefficient
Cp or block coefficient.Cn, respectively, is associated with in-
creasing volume of the optimal spherical bulb. Figure 8 gives
an impression of this fact. From this it follows, for ships with
long parallel middlebody, that with increasing CeE,LpM/Lep
and decreasing LE/BMs, the optimal bulb volume increases
(Yim [22], Fig.'4). The depth of thebulb has a similar influ-
ence. As shown in Fig. 9, at constant Froude number FN and
202
f
/ \ / D e s i g n g u i d e l i n e s for b u l b o u s b o w s •
It is well known that the existing design me[hods, for exam-
ple, the classical method by Taylor, are not sufficient for power
estimation of a bulbship and for modern bulb design. To fill
this gap in the design guidelines, a large number'of routine test
"65g- results of ships without and with bulb, carried out by the two
German model basins, have been reanalyzed in a research
project. The design guidelines derived, the design charts, and
a computer program have been successfully applied on various •
Occasions. From the multitude of diagrams developed in this
paper, only one example is depicted. For complete informa-
tion, reference has to be made to FDS Bericht No. 36/197318]
and VWS Bericht No. 811/78138]. It is emphasized that the
bulb position LpR = r information content of the design diagrams cannot be better
I i i i I i ] I I than that of the original data base, especially in the cases with
0 30 0.35 FN very small data collection. • 7
Reanalysis of routine test results
resistance, the design guidelines should correctly be related to
the wave or residual resistance. During preparation of the
research work, it became evident, however, that most of the
usable data were propulgion rather tha n resistance test results.
Since in principle it makes no difference whether the bulb ef-
fect is derived from resistance or propulsion tests, a power
specific bulb effect, or power reduction factor, respectively,
was defined:
Ap* = 1 . 0 - Pw/Po (13)
In this form the bulb effect is the power difference of the ship
without Po and with bulb Pw related to the power of the
bulbless ship. According to this definition, a positive bulb effect
corresponds to a power reduction, and vice versa.
In order to separate the different friction resistance com-
ponents of ships without and with bulb in accordance with
Froude's method, the total delivered power PD
PD =-Pa + PF (14)
is regarded as composed of a frictional part (index F) and a
residual part (index R). If the propulsive efficiency 71o is
known, the residual power can be calculated as the difference
Design of Bulbous Bows
 f p p e r limit .~upper" h'mif "-"--" 0. 7per limit
o ~1 ~ ~ ~- ~c o. ~ _ [ ~ - - ~ - - ~ ~ - ~ " ~=--
-IFN=~0.2~ - .~'~-~ ..~'-C .- ~ .-~0:2s/ 0:2¢ -I- \ :<
I o.-' " ' . %->~~.-~t-~" 0~1 ~ "~' 0.~s.,

- J" ° \ /

°'1 / I .I I o ° I I I I:1"'~
e.l( 0 L'/ B 0 OSSCLpR
. . C~¢p R
Fig. 10 . Dependence of the residual power reduction coefficient on the length/beam ratio, the volumetric parameter,
and the length parameter, respectively. Curve parameter is the Froude number. The parameters CABr, CABL,CBB,
' CzB, and BMS/TMSare constant

I
0.~ ACpTR 0.51 =TR 0,5 A'CpvR

upf =.rlimit -' upf. er limif upper lira, ~t

0.~,-
• t. .. FN-0.21 . "" ''"""

0.3 ,,'~b o ,,y/o22 .,," 3 . ~,/o. ~2,.

.<..0.20 ///,. 0.23 ~, ~"
0.2_ , ~-,~ \ 0 ,- ,\ .2 ,"~23 0. 2 0 \ \
\
0.1- - . 0 '.1 xO.
19,,,
~ • \

O_ I I I I
,"x "1 I I
\- 0 I X .I
.2.25 2.5 2.75 3.0 0.016 0.020 0.02$ 0.7 0.8 0.9-
. BIT c~oR CzB
Fig. 11 Dependence of the residual power reduction coefficient on the beam/draft ratio, the volumetric parameter,
and the depth parameter, respectively. Curveparameter is the Froude number. The parameters LppIBMs, LJBMs,
CABT,CAROL,CLpRand CBBare constant

between total and frictional power, and a residual power re -~ With equations (16) to (18), the residual power reduction factor
duction factor becomes
zxe~ = 1.0 - Paw~PRo AP~ = 1.0 P o w n O w . - PEFw X ~O....~o (15a)
= 1.o - , ( C o w - e , ~ w ) / ( e o o - e~o) (15) PDonDo -- PEFo nDw ."

can be defined. A separation of the various bulb effects is not possible in this
The relationship between effective PE and delivered power way.
is The propulsive efficiency no is a function of hull form and
speed and should be known for the analysis Of the model test
PD - PE./~O = (PER + PEF)/nD . (16) results.. Moreover, for ships without and with bulb, the pro-
With the frictional power PEF calculated by the International pulsive efficiencies are generally not equal, but it is nDw > nDo
Towing Tank Conference (ITTC) 1957 line in the beneficial speed range of the bulb ship. Unfortunately,
from most of the propulsion test results, nD could not be esti-
mated. Therefore, as a first step in simplification of the
PIu~=CI~(.2V~s'..S ) (17) reanalysis, a constant VD has to be chosenfor ships without and
the residual power is with bulb. The mistake is small if in the calculation of the re-
sidual power reduction factor by equation (15a), the condition
PR = PEa~no = e o - PE'F/nD . (18) nDw nDo isassumed. In general, the relations
"~

Design:of Bulbous Bows• 203

 0,5
° m

,4Cm7R
o,z._
0.66
0,3 . 0.68
,~" 0.70
0.72 •
o,2_ i +10°/o
~-.~._ ~._ _.~.. ~ ÷5%

o,L
I : 5,.%
- .

0.18 0.19. 0.20 0.21 0.2 0.23 0.21,, 0.25 "0.26 gN

Fig. 12 Influence of different propulsive coefficients on the residual power reduction coefficient,
shown with the ship-bulb combination No; 7 of Table 2

Therefore, the residual power reduction factor used here is

0.~ i

Ap~ = 1.0 -- (Pow - PEFw/~D)/(Poo -- Pp.fo/~O) (15b)

Z~C
case~._1_~ \
Because a few propulsive efficiencies are known only in. the
collected routine test results, in a second step of simplification

Zz,o ,v an ~?Dhas to be defined that should be constant for all ship-bulb

combinations within the whole range of block coefficient Cn.
From the scarce expe~'imental results and practical experi-
ence
0.2 ~?O = 0 . 7

0.1 J
/
model
5hip
appeared to be a very good mean value. For normal ships, ~?D
is between 0.6 and 0.8. Therefore, it is important to test the
consequences of divergence of 7/0 from the mean value 0.7 on
the residual power reduction coefficient ACeva. AS shown
in Fig. 12, a change in 7/0 of 4-3 percent affects at low Froude
numbers, FN, a change in ACpvR of only 4-4 percent; at higher
rN FN of only 4-1.5 percent.
Further analysis is facilitated by the introduction of di:
mensionless coefficient~ as follows. To eliminate influences
O.17 O.18 O.19 O.20
• of the ship hull form, the power displacment coefficient
Fig. 13 Comparison of the residual power rc:~ductioncoefficients from
full-scale and model-scale measurements (for main particulars, see cp~, = e / ( p / 2 v ~ 3,FTC[~) (19)
Table 1)
is chosen. This leads to a residual power coefficient
Ct, vn = P o / (p/2V~ 3v"-~WL) -- C~S] (~IO~/-~LWL) (20)
P~Fw > PEFo ibecause Sw > So)
and with friction law
PDw ,< Poo
CF = O.075/(logRN --2.0) 2 , (21)
TIDw > ~Do
to the bulb effect related to the residual power coefficient, that
and is, to the residual power reduction coefficient
PDwrlDw > PDw'qDo ACpv R = 1.0 -- Cp~TR wtth/fP~7R urithout (22)
hold. Consequently the numerator difference of equation Finally, it should be mentioned that in the reanalysis a form
(15a) is factor k is not considered.
The residual power reduction coefficient can be scaled di-
PDw~Dw -- PEFw > PDwT1Do --PEFw~ rectly to full-scale ship: Figure 13 shows the results of mea-
but because of surements at ship [35] and model scale. The slight'sea-wave
influence on full-scale measurements is eliminated. That the
~Do/~Ow < 1.0 amounts do not coincide totally might depend on the different
it is trims of the forms without and with bulb. Table 1 gives the
main particulars of the ships. In case 8, the delivered power
(PowitDw - PEFw) ~lO.._..~o~ ( P o w n D o - PEFw) of the bulbless ship is converted to a draft T = 6.57 m (21.55 ft)
YDw by aid of the Admiralty formula.
204 Design of Bulbous Bows
Design charts Table 1 Main particulars of ships without (Ship 1) and with bulbous
bow. (See Fig. 13; ATTILpp is therelated trim
The analysis of the many experimental results is sorted o u t
by CB-collectives in which many bulb ship forms with nearly case i case 2
the same block coefficients are collected. There are so many
ship 1 ship 3 ship i. ship 2
design Charts that only the case of CB = 0.7 is presented here.
The variation of the residual power reduction coefficient by. T., l,~ l 5,98 5,57 5,9b 5, C ~
equation (22) of each ship form nTT / Lpp o, 129% 1, 17°/o 0, 1 2 9 % i, 05 %
~rpp q "184.2 O, "1~59 0,7~;2 0,7813
ACpvR = F(FN, bulb form) " '(23) Yv O,8370 o, 7857 0,~370 0,7832
t ~.L I B~s 6,g33 7,o92 6,933 Z 087
has been calculated and presented as a function Of Froude Bns/Tns 4,550 #, 1#0 4, 550 ¢,6#5
number F~. The curve parameter is the bulb form (Fig. 14). i
From these curves it is possible to derive cross curves for/~ll six LpR/L#~ - o, o2o5 - qO20s
bulb parameters, which are collected in Table 2 together with B~ I Br~5 • " -- O, 1397 -- 0,1397
some other main parameter s of the ship-bulb combinations. ZB / TFp - 0,5570 - 0,5260
•The derivation and fairing of these cross curves (Fig. 15-20) A s r / An5 - 0,09¢8 = ,0,0952,
naturally are not totally without problems, because at only one A BL IA ~s - 0, 1055 - 0, 7070
variable bulb parameter the other five parameters are regarded - q 1725 - 0,1773
as constant. This assumption is correct in only a few cases for
some parameters; in most of the cases, all parametersalter si-
multaneously. Therefore, each diagram contains an upper results which are evaluated in the same manner are compared
limit curve which indicates the maximum possible improve- [:31]. Evenif the optimistic theoretical results are not achieved
ment due to a bulb. Figures 21 and 22 show that theway of in the experiments, the tendencies are at least represented
analysis and construction of design diagrams 'is justified. In correctly.
these figures, theoretical and the corresponding experimental The diagrams of Figs. 16-20, which are derived from the

0.5

aCpv R
o.#
>.f"
':.-: .-..-el_-_..
.,>
0.3

0.2

0.1

0 I I I i
0,18 0.20 ' 0.22 0.2~ 0,26 FN
Fig. 14 Residual.power reduction coefficient of 14 ship-bulb combinations as'a function of
Froude number. Basic diagram for Figs. 15-20 (for main parameters, see Table 2). " Curve
•parameter is the bulb form

Table 2 Main parameters of ship-bulb combinations of the data collected with Ca = 0.7 (see Fig, 14)

Nodel 1 2 3 4. 5 6 ,7 8 9 10 11 12 13 14
C9 0,68(,6 0,689 0,6891 0,6918 0,692~ 0,6963 O,6967 0,696? 0,69"70 0,7004- 0,7033 0`7165 0.9227 0,7266
C wL 0,8173 0,8389 0,7931 0.7762 0,8058 0.8(,49 O,7994
~cM 0.9829 0,9868 0,8785 0,9938 038#5 0.9916 0,9733 0.9733 0.9868 0.9920 0,9829 0,994`9 0,9845 0,9816
"~ C pE 0,6865 " 0,7211 0,6979 o.699LI -o.714`8
- O, 7234 0,7234 0.6703 0,64`52 0,6662 0,6985 0,6994` O,6~
.~ LISL/B ?,54.89 6.088 5,293 6.4997 6.oo61 5.702 6.094. 6,094, 7 2094 7.315 8.171 6,74`14. 6.7170 6.517
B/T 2.1539 3.006. 3.153 2.8285 3.0402' 2.972 3.114 3,11(, 2.25 "15 2.491 2. 168 2.3259 3.0#02 2.72¢
L ~ I B 3.774. .3.0#4. 2.0#72 2.880 12A561 12,855 2.978 2,978, 3.163# Z907 3.501 2.9250 2.4561 2.51¢
C8 0.68~,6 O.6935 0.6977 0,7069 ~0,7026 0.7055 0.7028 O:7062 0,70(,7 0.7070 0,7194. 0,7373 0,7317
L ~ / B 3.774. 22#4. 3.4`639 2.6587 3.063 3.2#0 3:197 3M62 3.1100 3.673 3.4,639 2.6587 2.722
C LP~ 0.0370 0.0299 0.04`33 0.0329 00330 0.0363 0,0440 0.0366 0,0381 0.0251 0.0258 0.0329 0.0295 0,0330
C89 0.155~- 0.084`0 0.1762 0.1821 0.1798 0,171"1 0,209I 0,1766 O,1538 0.14`63 0.1351 0,1821 03798 0.1734.,
CzB 0.6377 0.6500 0.5612 0.6730 0.9333 O.5938 0.5587 0,5852 0.5810 0.6309 0.4`267 0.5530 0,9333 0.5652'
..c. _.C AST 0`1008 0.0465, 0`It03 0.105610.1090 o,1035 0,1367 0,1061 o. 1032 0.0956 0,0802 0`0879 O,1090 0.1008
C A#L 0.1.829 0,1258 0,1286 0,14`98;0,15/6 0.1237 0.1696 0,1284` o. 1832 0.1096 0.1268 0.1230 0.1516 0,1157
°1oCv p~ 0,3356 0,1350 0,#015 0,3929 0.3606 0.2871 0,5878 0,3618 0,3120 0.22(,6 0,1819 0,3118 0,3096 0,2975
Z C~, tot L2026 0.3782 0,6.{66 0,8712 0.3713 0.9071 1.2547, 0.81-62 I, 3183 0.6207 0.51"I? 0.6936 2.0363 0,6957
%C5 tot 3.05?2 1.4`159 2.1905 3.0228,2.8112 &0283 3.0692 24268 2.5054. 2.3123 1.875"1 Z6458 2.¢5(,5 Z3753

Design of Bulbous Bows 205

 0.5-- 0.5 -I 1•
upper hmit
~CpGR i L: Cp~7R
0,~- 0.4-
/ '~//

~ / ' ( 8 = 0.7
/.:/"
o.3 . . . .
0,3- / o ~D : 0.7
ti,',l ',
CvpR ii IJ
\\ X
,'/"/ I . I "
0,2- / ~ , .mZ o.~ 0.2- ,;/"" I
~" °° o " - .... "- n'~ --
Y~ ',Ll
•' tJ = 0.7
C,,
' '

0.1- ~r°/°, , o 0.2 o.1- /

/ 7a " 0.7
f" /
/
! /.
0 # 1 ,0- I c' ~'PR
0 01 . . 0II 0 0.I 0.2 03 O.Z~ 0.5 0.6 P A l
. .1 Ct~B tot
Fig. 15 Maximum residual power reduction coefficient as a function Fig, 16 Residual power reduction coefficient as a function of the
of the volumetric parameter of total bulb volume (deri#ed from Fig. . volumetric parameter (derived from Fig. 14). Curve parameter is the
14) Froude number

upper hmit
5-
! ;, c p
CB = 0.7
04--
I. ~o = o.7
Is,'sl
0.3- I

O.2-
j /
J"
0.1 -- i

0 --
I
0.04, 0.05 0.06 " 0.07" 0.08 009 0.10 Oil 0.12 CAB T
Fig. 17 Residual power reduction coefficient as a function of the cross-section parameter
(derived fromFig. 14). Curve parameter is the Froude number

0.5-
6 Cp~TR upper limit~,
0/+- I /i

/" .j
0.3 --
t
JJf ~ .
0.2- .... \ /
CB = 0 7
0.1-. ,°.
~0 = o,7

0 -- J I
0.I0 O.I1 O.12 0.13 014 0.15 0./6 ( ABL 0./8
Fig. 18 Residual power reduction'coefficient as a function of the lateral parameter
(derived from Fig. 14). Curve parameter is the Froude number

206 Design of Bulbous Bows

 L

0:5-- f

., . ,,KS

))'/ -
0.3 --

....

0.2 --

. \
CB = 0.7
01--
~o ° 07

0 -- I I I I I I I I I I I I I I [ 1 J.
I
0 )25 0.03 0.035 0.0¢ CLpR
Fig. 19 Residual power reduction coefficient as a function of the length parameter (derived from
Fig. 14). Curve parameter is the Froude number

05i
0.4-
A CPVR
" " >.1
upper limit
/" .
~~.~0.
/ - ~ ' ....... " : .... i -"0.
F

. ../, _.::~/-" ' ~, . \0,

0.3

• :11 .5/" ct3 = 0.7

~'" 7?0- 0.7
0.2 I
Off

0
0.12 0,13 0.1~
I

0.15 0.16
[ I

0.t7.
I

0.18 0.19
,! 20 0.2f
I.
EBB
Fig, 20 Residual Power reduction coefficient as a function of the breadth parameter (derived
from Fig. 14). Curve parameter is the Froude number

diagram of Fig. 14 for CB = 017,rePresent the correlations
between bulb parameters and power gain. Regions may clearly
be recognized in which certain bulb parameters are unfavorable
and which are to be avoided.
The use of thediagrams follows from their derivation. In-
terpolations are permitted inside the parameter ranges shown,
while extrapolations should be avoided. If the main dimensions
of a ship are fixed and a bulb is to be fitted, then the estimation
of bulb parameters and thepower redu£tion due to this bulb
is possible by means of the design charts of the corresponding
block coefficient Ca. For this purpose, only the main-hull
particulars, Lep, BMs, TMS, and" CB as well as" the Froude
number FN, are required. For a given bulb parameter--which
can be any of the six parameters--it is possible to read in the
respec[ive design chart at the curve of the known F~'oude
number FN the residual power reduction factor ACeva. With-
this ACevR, the remaining bulb parameters are estimated in
the other diagrams atthe corresponding FN-curve. Except for
the configuration of the bulbous forebody, the problem is
solved, because, for FN. = constant, all six bulb.parameters are
assigned to a constant ACeva..
If Ca, Let,/BMs, and BMs/TMs of the bulble~s ship are inside
the ranges of the analyzed ship-bulb combinations, for example,
Design of Bulbous Bows
of Table 2, then the residual power reduction factor ExCevR
yields the wanted residual power gain. If departures from the
main-hull parameters appear, the ACevR is a good approximate
value. Except for the beam/draft ratio, the'influence of Ca
and Lee/BMs deviations is small, so that a special correction is
not necessary for these parameters. A general correction for-
mula so far does not exist.
If the delivered power for the bulbless ship is known, then
the required power of the bulb shipcan be calculated by the
following formula:
eo = {(1.0 - ACevR)CevRo
+ [CFS/(rlD~)]}2 V~ ~ (24)
where the residual power coefficient Ce~t~oof the bulbless ship
has to be estimated with 7o - 0.7 according to equation (20).
For unknown delivered power of the bulbless ship, in the project
phase, the reguired power can be calculated bythe following
formula:
Po = [(1.0 - ACevn)Ca + CF + ACE] P,c s (2s)
207
 .\ c,,,Bfo,=a2z°/o .

. // ~ ' < " . z "z~* / / , . HitcheltTh¢ory

/ .

~,. . . . . j exp~"lrr~

,. / . 25 0 30 0 35 F N

"/'/ 4CR, w = lO-CR,ww,thout /CR, la/with bulb

Fig. 21 Bulb effect on residual or wave resistance coefficient, respectively, as a function of
Froude number. Comparison of theoretical and experimental results of elementary ship form
(2,4,6,0.72,1.0). Curve parameter is the volumetric parameter of the total bulb volume. The
experimental results are evaluated with a form factor k = 0.205

- /,0__ From Figs. 28 to 26, the bulb parameters can be estimated

if at first instancethe power consumption is of no interest.
Z~Cw These figures show the relationship of the different bulb pa-
AC R rameters to each other for the ship-bulb forms which have been
~ upper hmit analyzed within the research project. The usual bulb param-
eter ranges are marked by upper and lower dotted lines.

Aspects of bulb design

~'--"-~":""
.......
....
o paper.
0.025 0.030 CVBt
Fig. 22 Bulb effect on residual or wave resistance, respectively, as
a function of the volumetric parameter of the total bulb volume (derived
from Fig. 21). Curve parameter is the Froude number
Where Cn can be estimated by one of the common procedures '
(for example, Taylor-Gertler [35], or Guldhammer [36]).
If several bulb parameters are known (this includes also the
judgment of a given bulb), then the estimation of the power
gain by means of the diagrams is problematic, because the
relation of the single values is hardly guaranteed in such a way
that all known bulb parameters result in the same ACevR at
constant Froude number. By using the volumetric parameter
CvPR as the main parameter connected with a possible con-
sideration of CABr and CABL, the diagrams can be used in such
cases for bulb design or for a judgment.
It appears tempting to use the diagrams to design an optimal
bulb by taking the parameters in accordance with their maxi-
mum reduction effect only. This procedure cannot be rec-
6mmended, because all six optimal bulb parameters chosen in
this way do not generally coincide with those of a concrete bulb
of the analyzed test data.
To judge the effect of a bulb change on the required power
of a bulb ship, the diagrams can be used to assist in decision-
making. By their aid, the tendency of parameter changing c a n
be detected which an alteration of shape does or does not
suggest. Even in their imperfect form, the design charts fa-
cilitate the decision for an application of a bulbous bow. But
they do not substitute for the model test, because the actual
power requirement for a project can be measured by experi-
ment only. ' face influences the momentum deflection.
In the preceding sections, it has been shown that a well-
dimensioned bulbous bow improves the performance of a ship
in many ways by smoothing the fl6w~around the forebody and
reducing the wavemaking resistance. In a particular case,
b~e decision for or against a bulbous bow is the matter of a
cost-effectiveness analysis [11 ], which is-not the subject of this
In general it may be stated ihat a hydrodynamically good
main hull with low wavemaking does not need a bulbous bow
in any case. But ships with pronounced bad wavemaking
should always be fitted with a bulb. Practical points of view
will decide whether an additive or an implicit bulb is to be
provided for. For ships already built, an additive bulb will in
general be the best solution, while for a new design an implicit
bulb might be advantageous.
T h e shaping of a bulbous bow, that is, the longitudinal and
depth distribution of the bulb volume in proportion to the bulb
parameters, can be described only in a qualitative way. Al-
though for a concrete longitudinal distribution of bulb volume,
the knowledge of the wave pattern of the bulbless ship is an
important decision-making aid, in the preliminary project phase
this information is lisually not available. But even without this
information it is possible to indicate general guiding rules for
shaping a bulbous bow. Essentially they are related to
• type of ship (full or slender),
.• service speed (slow or fast),
• frequency of draft alteration at FP (large draft variation
• or defined CwL and ballast draft), and
• main operation area of the ship (for example, with much
heavy seaway or drift ice).
A specific ship-bulb form shows optimum performance only
at design conditions. At off-design conditions its performance
may be poor.
For the distribution for the bulb' volume, the following three
rules are important:
1. Deeply emerged volume is of little effect.
2. Longitudinally concentrated volume near to the free
surface increases the interference effect.
8. Longitudinally distributed Volume near to the free sur-

208 Design o f Bulbous B o w s

 06
o •
o
o o
2 . o
0
05 10. CypR o
o
o 0
0
0
CABT 0
0,4 v n o • % 0 O0
--6-0-- ~ --.-.._.... ,.; o ~d
0
o o o
oo ( Oo°°" Oo 0°
d o o Oo : • "o o, o 0 0
0.3 O O C~n O Oo o o
o O 0
v 01 o .o ~,_.~t~o ~.o~,_o0 6
go ° O0 0 0o 0 0 ~.
0,2
B ~ 0
o ~ °~=
O0
o,
.0-
o
•
o
~q~
o o',
o1~Oo
~2
~ogo
eOoOFo o o
-- ~r......_ o, o i=r -
0
o ~ - •
e
0 o o, "
o '~'%
•,0, I 0 g o e
u
O"

0
I [ I I I,] I 11 I'] I I I I I I I I I'I 1 I £2 ~3 04 - (]5 06 tOZ..CVp
IIIII 0
5 I 07 I O~ CB I o R " "

0,2
\
0 x C

0 0 .

o t_. i f , 0.1 °\
o o 0 O0

°00% t~ ~ 0
o 0
O ¢~
0 0 O"
n_o_e
0
O^ r &
o\o
Oo .~ o "~u I: CAB L
o o .oo o ~ o, t#
Cz8 %..o~,,°r-- o o • oo
08 0 0 0 o

Fig: 23 Volumetric and depth parameter versus block coefficient

Fig. 25 Cross-section and lateral parameter versus volumetric

Paramete r

\
o 2O
o

[%
004 • o I
rO/:: g<oo./
7 / ~o
CLPR ,, o
• I , . O
I
O
eO O
e n O
~o~' i o
o . o
002 1.0 o o, . o
~oo po°O o,~
o o

/o o 0

,?
o ~ OOoqO~.~
001 . o oO''~ ¢ oo
0
w vL ul . I
o=o o ,o o
o oo o
~ • 0 o .o O
016 10.CVp R ~b
8 o a
o

' " 0 C2 (3 04- c5 06 U/o] Ct* PR

¢ o o 0'
/.0 o
dP o o
0 oo "6 )"
0.1 o
2.0 ~ o o °°°o @ o
o o -~o vo o "
o mo o :, 8o.`
o ~0 qp o t,~ -o u
o oo
o °° ~86° ° c°'oo-~ "
o " 3.0 o
•
COB o o ~ o ° % 'P~o°T(
v T.o
o
o o
[Olo1
oO 0 0 OPc
r oOl •
0

e°
o o o o
. o
0.2 o og. ~o o,.
0
no
0
4.0 o
0 o

O , O O O
O
• C58 t
o
Fig. 26 Total increase of the volume (C v Bt) and the wetted surface
(CsBt) of the main hull due to a bulbous bow versus volumetric coeffi- r
Length and breadth parameter versus volumetric parameter ". cient [ C y B t = ( Y W L w , " V WLo) / V WLo;CsBt = ( S w ~ "S o ) ! S o ]
Fig. 24

D e s i g n of B u l b o u s B o w s 209
i
 The waterlines of the bulb nose should be streamlined but wave system, thebulb reduces the wavemaking as well as the.
not circular, in order to avoid separation [33]. wave-breaking resistance.
For ships with a strong wavemaking tendency, the bulb Two main bulb effects which are very important for bulb
volume should be concentrated in the longitudinal direction, design are defined as the interference and the breaking effects.
where the upper part of the bulb body at the FP should not be The interference effecl~expresses the resistance change due to
located above the CwL. The integration of bulb and ship can the interfering free wave systems of main hull and bulb.. For
be straight-lined ([3], Rule 4). The fairing volume plays a slender, fast ships, it gives the main proportion to the total bulb
subordinate role. The maximum width of the bulb can be effect. Its amount depends on bulb volume and the longitu-
situated in front of the forward perpendicular. At such'a dinal position of the bulb center. The wave-breaking effect
bulbous bow it may happen, of course, that cavitation occurs. includes the energy loss by breaking of too steep bow waves and
This problem is not treated here. gives the main contribution to the total bulb effect of full, slow
For ships with much wave-breaking, the bulb volume should ships. Its amount depends on well-distributed bulb volume in
be distributed• well in the longitudinal direction. The upper the longitudinal direction. Both bulb effects a r e Froude
part of the bulb body can reach to the peak of the bow back- number dependent.
wave of the bulbless ship [34]. A well-formed laterally and For bulb design, sixbulb parameters are introduced, of which
forward-inclined bulb ridge avoids the breaking of the bow the•volumetric, the cross,section, and the length parameter are
back-wave. The fairing volume plays an important role. The the most important. The influence of bulb parameters on the
upper part of the bulb should be faired well into main hull in different bulb effects is discussed in a qualitative manner,
order that the tail water of the bulb ridge interferes well with supported by the linearized theory of wave resistance and by
the remaining bow wave. In the lower part of the bulb, the some experimental results. This knowledge is important for
waterlines should have small angles of entrance if the bulb is the shapingof the bulb body according to the bulb parame-
to emerge high under ballast conditions. Due to the danger ters. '
of separation in this area, the fairing of the waterlines should A q,uantitative design method is presented together with the
not be too hollow. necessary design data. The data are derived from an analysis
of routine test results of the two German model basins. Most
of the usable data were propulsion rather than resistance test
Bulb-type recommendations results. Therefore, according to Froude's method, a residual
The O-type is suitable for full as well as for slender ships. It power reduction coefficient is defined which can be scaled
fits well with U- and V-types of foreship sections and offers directly to the full-scale ship. The variation of this coefficient
space for sonar equipment. The lens-type should be chosen for each ship-bulb combination has been calculated and pre-
for ships which often operate in heavy seas, because it is less sented as a function of Froude number. From these curves the
susceptable to slamming. design charts are derived--for each bulb parameter, one di-
The A-type is good for ships with large draft variations agram. From the multitude of diagrams, only one example
(tramp ships) and U-type foreship sections. The effect of the is depicted. The calculation of the required power of the bulb
bulbous bow decreases with increasing draft, and vice versa; ship is described. General guiding rules for shaping a bulbous
but in heavy seas the danger of slamming increases with de- bow are given.
•creasing draft. The design guidelines have been successfully applied o n
The V-type can be provided for all ships with well-defined various occasions.,
CwL and ballast draft. It is easily faired into V-shaped fore=
• bodies and has in general good seakeeping performance. In References
the fully submerged condition, its damping effect is very 1 Gawn, R. W. L., "Historical Notes and Investigationsat the
high. Admirality Experiment Works," Torquay; Trans. TINA, 1941.
In all cases, the bulb should not emerge in the ballast condi- 2 Eggert, E. F., "Form Resistance Experiments," TRANS.
tion so that its most forward point, B (Fig. 3), lies on the water SNAME, Vo1.'43, 1935; "Further Form Resistance Experiments,"
, surface. The individual resistance of the bulb body in this • TRANS.SNAME, Vol. 47, 1939.
condition would be higher than its net efficiency. 3 Wigley,W. C. S., "The Theory of the Bulbous Bow and its
Practical Application," Trans. North-East Coast Institution of Engi-
neers and Shipbuilders, Vol. 52, 1985/36.
•S u m m a r y 4 Weinblum,G., Theorie der Wulstschiffe, 1935.
5 Inui, T., Takahei, T., and Kumano, M., "'Wave Profile Mea-
Today the bulbous bow has asserted itself as an elementary surements on the Wave-making Characteristicsof the BulbousBow,"
device in practical shipbuilding. But the existing design Zosen Kiokai, Vol. 108, Dec. 1960; also.Journal of the British Ship
methods are not sufficient for po~,er estimation of a bulb ship Research Association, Vol. 16, 1961.
and for modern bulb design. A well-dimensioned bulb im- Waveless 6 Inui, T. and Takahei, T., "Wave Cancelling Effects of the
Bulb On the High-SpeedPassenger Coaster (Kurenai Maru,
proves most of the properties of a ship. Therefore,qualitative Part I-III)," Zosen Kiokai, Vol. 110, Dec'. 1961.
and quantitative guiding rules are necessary for its beneficial 7 Kerlen,H., "Entwurf von Bugwiilstenfiir v611igeSchiffeaus
application. der Sicht der Praxis," HANSA, No. 10, 1971.
Compared with the indirect influence of a bulbous bow on 8 Kracht, A., "Theoretische'undexperimentelleUntersuchungen
thrust deduction and wake fraction, the bulb also influences f/Jr die Anwendung von Bugw/ilsten,"FDS-Bericht, No. 36 1973.
9 Kracht, A., "Der Bugwulst als Entwur{selementim Schiffbau,"
directly the wake distribution in the propeller plane. Except STG-Jahrbuch, Band 70, 1976.
for the strongly damped pitching motion, the bulb ship has the 10 Takahei,T., "BulbousBow Design for Full Hull Forms," Trans.
same seakeeping qualities as a bulbless ship up to Beaufort 6. Society of Naval Architects of Japan, Vol. 119, May 1966.
Therefore, regardless of seakeeping aspects, the bulb design 11 Schneekluth,-H., "Kriterien zur Bugwulstverwendung,"
may be carried out in view of the smooth-water performances STG-Jahrbuch, 1975.
only.- In navigation in ice, the bulbous bow has proved to be 12 Taniguchi,K. and Tamura, K., "Study on the Flow Patterns
Around the Stern of a Large Full Ship," Mitsublshi Technical Review,
advantageous. Vol. 7, No. 4, 1971.
The most important effect of a bulbous bow is its influence 18 Tatinclaux,J.-C.~"Effect of BilgeKeelsand a BulbousBOwon
ori the different resistance components and, consequently, on Bilge Vortices,'~ IIHR Report No. 107, Feb. 1968.
the required power consumption. By attenuation ofthe bow 14 van Lammeren, W. P. A. and Muntjewerf, J. J., "'Researchon

210 Design of Bulbous Bows

t
Bulbous Bow Ships. Part II. A Still Water Performanceof a 24000
DWT Bulkcarrier with a Large Bulbous Bow," International Ship-
building Progress, Vol. 12, 1965, p. 459.. '
15 van Lammeren, W. P. A.:and Pangalila, F. V. A., "Research
on BulbouS. Bow Ships. Part II B," International Shipbuilding
Progress, Vol. 13, No. 137, Jan. 1966.
16 Wahab, R., "Das Verhalten eines schnellen Frachtschiffes mit
konventionellem Bug und mit Wulstbug im Seegang," STG-Jahrbuch,
Band 60, 1966, p., 283. ' •
• 17 Oehi, K., 'Model Experiments on the Effect of a Bulbous Bow
on Ship Slamming," DTMB-Report 1360, Oct. 1960.
18 Frank, W., "The Heave Damping Coefficients on Bulbous
Cylinders, Partially Immersed in Deep Water," Journal of Ship Re-
search, Vol. 11, No. 3, Sept. 1967, p. 151.'
19 Dillon, E. S., Lewis, E. V., and Scott, E., "Ships With Bulbous
Bows in Smooth Water and in Waves," TaANS. SNAME, V01. 63, 1955,
p. 726.'
20 "Wahab, R., "Research on Bulbous Bow Ships. Part I B,'" In-
ternational Shipbuilding Progress, No. 142, 1966, p. 181.
21 Briihl, W., "Der Bugwulst in der Eisfahrt," Schiffund Hafen,I
Vol. 28, No. 1, 1971, p. 17.
22 Yim, B. and Shih, H. H., "Minimum Wave Bulbous Ship Hulls
with Parallel Middle Body," Hydronautics Inc. Technical Report
117-9.
23 Baba, Eiiehi, "Analysis of Bow Near Field of Flat Ships,"
Mitsubishi Technical Bulletin No. 97, 1975.
24 Baba, Eiichi, ~'Blunt Forms and Wave Breaking," SNAME, First
Ship Technology and Research (STAR) Symposium, Washington, D.C.,
Aug. 1975.
25 Wieghardt, K., "'Viscous Resistance," 18th ITTC, Vol. 1, 1972,
p.S&
26 Eckert, E. and Sharma, S. D., "Bugwiilste fiir langsame, v611ige
Sehiffe," STG-Jahr.buch, 1970, p. 129.
27 Eggers, K., "Uber die Ermittlung des WeUenwiderstandes eines
Discussion
Bohyun Yim, Member
[The views expressed herein are the opinions of the discusser and not
necessarily those of the Department 0fDefense or the Dep~irtment of
theNavy.]
This paper has demonstrated clearly by experimental results
from many-ships that addition of a bulbous bow can reduce a
large portion of the residual resistance of a ship. The charts
with the approximate values of important design parameters
for bulbous-bow ships will be very useful for naval architects.
Although we know the mechanism of the bulb effect in the
theory of wave resistance, the optimum size, shape, and location
of the bulb cannot be obtained accurately because of the
weakness in the theory of wave resistance, w e all know well
that the present linear theory cannot accurately predict the
resistance of oceangoing ships. This is why experimental results
are valuable in ship design. However, it is Obvious that we have
to make full use of theoretical knowledge, in order to have a
better design.
In bulb design,we recognize that there are two t~,pes of bulb: .
•doublet type and source type. The former reduces-the sine-
component of elementary ship waves while the latter reduces
the cosine-component of elementary ship waves. Although
normal ships with relatively large entrance angles have domi-
nating sine waves which require' the doublet-type bulb, there
are ships with hollow waterlines which require the source-type
I~ulb.' That is,the bulb shape depends largely upon the wa-
terline shape near the ship bow. I wonderwhether this effect
• is reflected in the charts.shown in the present paper? Block.
coefficient ~nd Froude number are not enough to represent ship
parameters related to bulb design. In this respect, Yim's3
3 Yim, Bohyun, "A Simple Design Theory and Method for Bulbous
Bows of Ships," Journal of Ship Research, Vol. 18, No. 3, Sept. 1974,
pp. 141-152.
- Design of Bulbous B o w s
Schiffsmodells durch Analyse seines Wellensystems," Schiffstechntk,
Vol. 9, 1962, p. 46. ' "
28 Ward, L. W., "A Method for the Direct Experimental Deter-
mination of Ship Wave Resistance," Doctoral Dissertation submitted
to the Faculty of Stevens Institute of Technology, Hoboken, N. J., May
1962.
29 Schmidt-Stiebitz, H.,"Systematische Erfassung von 6rtlich am
Schiff anzubringende Stau- bzw. Unterdruck erzeugende Elemente
zwecks Verringerung der Wellenh6he und damit des Wellenwiders-
tandes,'" Schiffund Hafen, Vol. 12~ 1960, p. 746. .
80 Dagan, G. and Tulin, M. P., "Bow Waves Before Blunt Ships,"
Hydronautics, Inc:, Research in Hydrodynamics Technical Report
117-14.
81 Kracht, A., "'Theoretische und experimentelle Untersuc.hungen
iiber die Verringerung des Wellenwiderstandes gegeben.er Schiffs-
formen durchden Bugwulst," Bericht No. 166 des Institutes fiir
Schiffbau, Hamburg, 1966. '
82 Kraeht, A., "Lineartheoretische Abhandlung fiber die optimale
Verringerung des Wellenwiderstandes gegebener Schiffsformen durch
einen Wulst in symmetrischer oder asymmetrischer Anordnung,"
Bericht No. 183 des Institutes fiir Schiffbau, Hamburg, 1967.
38 Saunders, M. E., Hydrodynamics in Ship. Design, Vol. 2,
SNAME, 1957.
34 Weicker, D., "Bemerkungen zum Bugwulstentwurf,"
Seewtrtschaft, Vol. 3, No. 11, 1971, p. 827.
:35 Kerlen, H. and Petershagen, H., ."Uber den Einflu/~ eines
Bugwulstes auf Leistung und Geschwindigkeit eines v611igen Fra-
chtschiffes," HANSA, Vol. 105, No. 12, 1968, p. 106,%1067.
36 Gertler, M., "A Reanalysis of the Original Test Data.for the "
Taylor Standard Series," DTMB Report 806, 1954.
87 Guldhammer, H. E., and Harvald, Aa., Ship Resistance, Aka-
demisk Forlag, Copenhagen, 1974.
38 Kracht, A., "Weitere Untersuchungen fiber die Anwendung
von Bugwiilsten,'" VWS Bericht No. 811/78, Berlin, 1978.
simple design method for a bulbous bow for a particul~ar ship
and design speed may be useful.
At the David W. Taylor Naval Ship R&D Center, several
bulbs designed by Yim's method have been tested and satis-
factory results obtained. Because the m e t h o d u s e s linear
theory, the computed bulb size is sometimes slightly larger than
optimum, yet it gives very useful guidance.
Eiichi Baba, Member
The design charts of bulbous bows provided by the author
are suitable especially for ships with relatively small block
coefficients. For those ships, the wavemaking phenomena are
influenced not only by the entrance part but also by the middle
part and the run part, and the form effect on viscous resistance
is relatively small. F o r full forms of CB > 0.8, however, rather
large viscous effects are included in the residual resistance
component. Therefore, the scale effect is not taken into ac-
count in the author's charts. Further, as the author pointed out,
for full forms the wavemaking phenomena are mainly de-
pending on the entrance part. Therefore, the characteristic
length for the expression of Froude number should be entrance
length LE or ship beam BMS instead of ship total length Lee.
The author says that at the preliminary design phase the en-
trance parameters CpE and L~ are unknown. However, if
design charts for full forms based on the entranceparameters
are provided, they may be effectiVely used in the preliminary
design phase. Actually at Mitsubishi Heavy Industries a design
method, of full forms b~sed on entrance parameters has been
developed and since 1968 has served as a routine method as
outlined in reference [24]. Our design method for full forms
is based on the following experimental and practical evidence
derived from the analyses of towing test data of more than 200
full forms.
1. For full forms (v/~TgLee < 0.20, CB > 0.80) with a
211
 well-designed run part, the wave resistance characteristics James H. Robinson, Member "
depend mainly on the geometry of the entrance part, and the IThe views expressed herein arethe opinions of the discusser and not
contribution from the parallel part and run part. is negligibly necessarily those of the Department olDefense or the Department of
small in generating wave resistance. the Navy.] :•
2. The propulsion factors depend mainly on the geometry
of the run part. Despite the wealth of available information on bulbous bows,
3. A reliable formula to estimate the form factor by use of ship designers are still somewhat at a loss when choosing opti-
given geometric parameters has been developed. In this mum bow shapes, and in predicting performance. Dr. Kracht's
method the entrance part and the run part are selected inde- paper is welcome for its practical approach and useful.content,
pendently so as to achieve a desired propulsive performance, and I appreciate this opportunity for brief comments.
and the parallel middle part is designed to satisfy the required With respect to bulbs in general, I note the author's distinc-
displacement. The parameters for wave resistance data are tion between additive and implicit bulbs, and would suggest
the Froude number based on ship beam Bus, that is, v / g ~ M s that many of the examples we see today must be additive--at
and the beam/draft ratio BMs/TMs, and the fineness factor of 'least they look added-on. Would the author indical~e whether
entrance CMHE/BMs,where HE is an effective length repre- he believes one or the other enjoys any inherent advantage in
senting fullness of entrance defined by HE = (1 - CpE).LE. performance? The additive type has an advantage of being
easily adapted to a model test program in which several dif-
Lawrence W. Ward, Member ferent bulb inserts (including a no-bulb bow) can be fitted to
the same basic forebody, tested, and compared.
I congratulate Dr. Kracht and the VWS Model Basin for
presenting us with a design-oriented paper in the area of ship After selecting a large protruding bulb as the best of several
hull form hydrodynamics. This and similar studies have the tested, I sometimes wonder whether a still better solution exists.
potential of obtaining important fuel cost savings for the op- And I would suggest that it maywell involve use of a nonpro-
erator over the lifetime of a ship or class of ships by means of truding smaller bulb on a model whose waterline length is in-
a reasonable extra one-time initial investment of some addi- creased by the amount of bulb protrusion previously foUnd
tional plate and frame bonding in the construction phase. Such desirable. One might ask whether our apparent fixation on
matters have become very important, as evidenced by your and waterline length rather than wetted length is leading us to less
my increasingly difficult task of heating our homes again this than optimum solutions. Let me put it another way: Given
winter for a reasonable percentage of our daily bread--don't a maximum wetted length, does it really make sense to move
we all wish now that our home constructors had worried a little the traditional forward perpendicular.location aft and accept
more about efficiency of operation at that time! an increase in block coefficient, shortening of entrance, increase
of waterline entrance angle, increase of displacement-length
I knew Dr. Krachtwhen I was in Germany in 1965-1966 and ratio, and increase Of speed-length ratio? .
he was immersed in the theoretical study of ship bulbs under
The author discusses briefly the influence of bulbs on wake
the guidance of the late incomparable Professor Georg Wein-
and thrust deduction, and quotes references showing
blum--as you can see in the References he has continued this trends--which I believe deal mainly with the high-block ships.
study in the intervening years and has become "Mrl Bulb" in In any event, many bulbs have been selected on the basis of
Germany and in the world. We would all do well to study the resistance tests alone, and I doubt that propulsion test com-
results in this paper in detail and to consider Dr. Kracht's rec- parisons would have changed the selection.
ommendations with respect. I plan to, this year especially, in
connection'with a Maritime Administration project at Webb Mr. Kracht's discussion of the various components of bulb-
Institute funded under the newly launched University Research related res!stance is very informative. My impression of his
Program (for which we in the marine industry should thank treatment of wave-breaking phenomena is that it relates mainly
them for initiating). - to high-block ships in ballast, where bulbs can ameliorate the
condition. I would be interested in his comment on the
i have a few comments and questions on some details in the wave-breaking situation with fine-lin6d ships at light draft,
paper itself: where the bulb selected for optimum results under design
I note that the factor ACevR is used for the facl~or of merit conditions is the cause of an increased wave-breaking resis-
in assessing the benefit of a bulb change in most of the figures, tance.
and this factor turns out to be the ratio of the Change (reduction) Turning now to the "Design charts" subsection, I have dif-
of residual power only, not including frictional. Changes in ficulty evaluating the data, and would be yeary of jumping to
residual power due to changes in the wake fraction (due to wave wrong conclusions. For example, Figs. 17 and 20 indicate that
wake changes) are included (although not included in the data one should avoid bulbs having 9 percent transverse area and
reduction in the present case for reasons stated by the author) 17 percent width ratios. Yet Figs. 24 and 25 indicate no
in this factor, and that is a neff idea in the business of bulb as- avoidance of those values. I concur in the author's statement
sessment and one I plan to include in my research. Why not, to the effect that ships with low wa.vemaking resistance do not
m a y I suggest, define an overall power reduction factor need bulbs as much as do ships with high wavemaking resis-
tance, and that may explain why some bulbs appear very
+ Rwl[1- 1 - 1 good--that is, the no-bulb configuration is poor.
=.1 T, -Ew ]t I think the value of the paper would beenhanced ff the au-
thor would include, with his closure, data for ships of other block
that would only ignore changes in the propeller openwater coefficients--especially around 0.6 or less and 0.8 or more.
efficiency, relative-rotative efficiency, and hull thrust deduc- References [8] and [38], which apparently contain this infor-
tion.with or without the bulb? mation, are in German and not readily available.
I have only one other comment, a question: What is the Finally, let me add a strong endorsement of the author's
actual difference in the "Aspects of hull design" section, be- statement regarding use of the design charts, in which he ac-
tween an "additive bulb" and an "implicit bulb"? (The knowledges that actual power requirements can be found only
comments in the "Bulb forms and parameters" section don't from model tests. Theory is a great help, but we still get sur-
seem to help much in this regard.) prises when we test.
2i2 Design of Bulbous Bows J
T. Inui, 4 Visitor, and T. Takahei, Member
The statistical-data on the btilbeus bow are provided for ships
of relatively small block coefficient where the "wavemaking"
plays an important role in designing the bulbs. However, the
'high-efficiency bulb cannot be achieved by the statistical ap-
proach based upon force measurements such as resistance and
propulsion tests, as one of the discussers pointed out in 1962. 5
It is important to find the relationship between (a t ship geom-
etry and (b) ship waves before connecting.them to (c) wave
resistance. The "half-body concept" especially--that is the
treatment of the forebody apart from the afterbody and the
extraction and analysis of bow Waves--is indispensable. For
instance, Figs. 8 and 9, which show the interference of bow and
stern waves are misleading in the evaluation of the optimal bow
bulb volume.
• Wavemaking is sensitive to hull geometry. Thedifferential
sectional area and waterline curves aredecisive influences on
the phase and amplitude of bow and stern waves; the problem
lies, rather, in the configuration of the main hull prior to the
desig n of the bulb. It is regrettable that the author's funda-
mental concept on these points is not clearly shown in the
paper.
In the design charts of'Figs. 15-20, the residual power re-
duction coefficient is given as a function of six bulb parameters
that supposedly may be chosen, independently, but they are
tightly correlated. For this reason, by taking the parameters
for their maximum reduction effect only, the designer may not
generally obtain a concrete bulb configuration.
Another defect of the charts is that displacement condition
is not designated. The.optimum bulb configuration, particu-
larly for moderate- and low-speed ships, varies with the dis-
placement condition. This fact was depicted early in 1966. 6
The author admits' that there is the necessity of a special cor-
rection of the charts for the beam/draft ratio and yet a general
correction formula has not been obtained.
The geometrical relations between the bulb parameters are
also constrained for getting a concrete bulb configuration.
Figures 23-26 are presented supposedly, to show their con-
strained conditions, but the data exhibit quite a large scatter.
Rearrangement of these data with respect to a specific C8 and
displacement condition would yield plainer and more infor-
mative design figures. This further work would make the
design process more straightforward.
J. J. Muntjewerf, Member, and J. Holtrop, 7 Visitor
The author has provided a very ~¢aluable survey of the many
aspects to be considered in the design 0f bulbous bows and he
should be congratulated on his success in delineating a purely
graphical method for the design of a bulbous bow and estima-
tion of its effects on power requirements.
Although the graphical approach may have the advantage
of yielding to the researcher a clearer picture of the physical
phenomena involved, the enormous amount of design charts
to be handled is cumbersome for the project engineer andthe
present paper can not be used by him without access to refer-
ences [8] and [38].
Taking into account the fact that the basic data are scarce and
the number of variables large, a more appropriate way to es-
tablish the functional relationships, in our opinion, might
therefore have been the careful use of nonlinear multiple re-
4 The University of Tokyo, Toky0, Japan.
[nui, T., ",Wave-Making Resistance of Ships," TRANS•SNAME,
Vol. 70, 1962.
6 Couch, R. B. and Moss, J. L., "Application of Large Protruding
Bulbs to Ships of High Block Coefficient," TRANS.SNAME, Vol. 74,
1966.
7 Netherlands Ship Model Basin, Wageningen, The Netherlands.
Design of Bulbous Bows
gression analysis with a restricted number of variables. The
results of Such calculations could have more easily been incor-
porated in a design computer program for practical application
in preliminary.ship design work.
This latter approach has b e.en followed by the Netherlands
Ship Model Basin s (NSMB), and it is interesting to compare the
NSMB data with those of Fig. 14 of Dr. Kracht's paper. From
Fig. 14 it can be deduced that for Models Nos. 5 a n d l 3 in the
low-speed region negative ACevR-values can be expected.
This would involve an increase of the residuary resistance
component at low Froude numbers by fitting the bulb. From
inspection of the bulb parameters it appears that the coefficients
Czn are very high, and it can be assumed that the location of "
the volumetric center of the bulb in the vicinity of the surface
will yield flow separation or local Wave breaking or both.
Calculations by the statistical method developed at NSMB, in
which this addition~il pressure resistance component of a bul-
bous bow that protrudes the water surface is. incorporated,
showed curves of the same character arid equal order of mag-
nitude as those in Fig. 14~ From the design graphs presented,
Figs. 15-20, however, it is not clear how the vertical distribution
of the volume of thebulb influences the power reduction fac-
tors. Moreover, the oscillating character of most of the curves
in these figures suggests that a phenomenon as indicated in the
foregoing has been associated with the other general bulb pa-
rameters as well.
Hence, we emphasize that in establishing the functional re-
lationship between the power reduction factors and the bulb
parameters, indirect incidental correlations should b e
avoided. ' ~-
On the theoretical foundation of the paper, we wish further
to make the following remarks:
1. Dr. Kracht's statement that it is not quite clear Whether
the bulb affects the ~iscous residual resistance RvR is probably
not correct, since Steele 9 clearly demonstrated, from local skin
friction measurements on a liner model with and without bul-
bous bow, that a bulb improves the flow around the hull and
thereby reduces the frictional component of resistance, or in
Dr. Kracht's split-up of components, the residuary viscous
component.
2. The use of FN based on Lpe instead of on length of en-
trance Le seems to be a step backwards from the work by Baba
[24] and Sharma [26] in the case of full ships with a long parallel
midbody. So, for the block 0.70 ships with probably a 10
percent parallel midbody or higher as taken for the examples
in the paper, we should like to know whether the use of entrance
length was tried out and, if so, what was the result of the anal-
ysis.
Finally,. We should like to mention that we appreciate the
author's warning that his diagrams can be used only to assist in
preliminary design work and we fully endorse his statement
that they (and other calculation methods) do not substitute for
the model test, because the actual power requirement for a
project can indeed be determined accurately by exper!ment
only.
David W. Byers, Member , •
[The views expressed herein are the opinions of the discusser and not
necessarily those of the Department of Defense or the Department of
the Navy.]
Speaking as a naval architect concernedwith hull form de-
sign of U.S. naval ships, I welcome Dr. Kracht's paper as a major
s Holtrop, J. and Mermen, G. G. J., "A Statistical Power prediction
Method," International Shipbuilding Progress, Vol. 25, No. 290, Oct.
1978.
9 Steele, B. N. and Pearce, G. B., "Experimental Determination of
the Distribution of Skin Friction on a Model of a High-Speed Liner,'"
Trans. RINA, 1968.
213
 1.2 i i i I I I i I J

Bulbs
A Z~

m 1.1
0
z
a.

==
W
~ 1.o

0.9 I I I I I I I
0.10 0.14 0.18 0.22 0.26 .30
Froude N u m b e r , F n
Fig. 2 7 Hull A

o'-'-~,~l I I I I I I . I
Bo,bs
~,f''%~,~ \ A ---- A
1.2
B . . . . . . .
\X~, \ ' c .... []
_ \\~, \, O .... "0 .
x;,\~,,N E . . . . . o
O G . - - o

Z1.1
5

~- 1.0

0.9 t I I 1 I 1 I I I
0.10 0.14 0.18 0.22 0.26 0.30
Froude N u m b e r , F n
Fig. 2 8 Hull B

conti'ibution to the Collection of design aids available for use in
preliminary design. The paper provides not only a method for
designing bulbous bows but also a means for predicting their
effect on powering performance, based on actual model test
results. Modern protruding bulbous bows have proved so ef-
fective in improving the resistance characteristics of naval
auxiliaries in recent years that they are now actively considered
for other classes of naval ships. Dr. Kracht's paper will help
to make this job easier.
Two points which the author makes are worth keeping in
mind when initially considering a bulbous bow, as one recent
design at the Naval Ship Engineering Center demonstrated.
These are that (i) the predicted powering advantages of a bulb
must still be confirmed by model tests and (ii)mhere I quote
from Dr. Kracht's paper--"in general, a hydrodynamically •
good main hull With low wavemaking does not need a bulbous
bow in any case." To illustrate, the Navy recently completed
a series of model tests at the David W. Taylor Naval Ship Re-
search and Development Center involving two alternate hulls
and seven alternative bulb forms for a new ship design. This
ship was of fine form, with a block coefficient of 0.57 and a
Froude number of 0.26 at the design sustained speed. Figures
27 and 28 depict the results, in the form of effective horsepower
ratios of hull with bulb to hull without bulb.
On one hull, designated Hull A in Fig. 27, only one bulb,
identified a's Bulb F, proved advantageous in the speed range
of interest. This bulb was designed for optimum performance
at a Froude number of 0.20 using the wave-cut technique al-
luded to by Dr. Kracht in his paper as reference [26]. It is
significant that the design Froude number for the bulb is lower
than the sustained-speed Froude number. This reflects a
tradeoff in favor• of fuel economy throughout the speed range
rather than a minimum horsepower requirement at sustained
speed.
On the second hull, however, Hull B in Fig. 28, none of the
bulbs, including a different wave-cut bulb specifically designed
for this hull, identified as Bulb G, were fouiad to be beneficial.
Further, the Hull B effective horsepower without bulb was
superior to that of Hull A with Bulb F.
So, in this case at least, the test results confirm the author's
concerns for model test validation and recognition of the fact
that not all hulls benefit from bulbs. Again, I thank Dr. Kracht
for an interesting and valuable paper.•

214 Design of Bulbous Bows

Harrison T. Loeser, Member ' impticit bulb"
i wish to c o m m e n t on e x t e n d i n g the w o r k on bulbs. Well
over 10 years ago, Inui (see footnote 5 of Inui/Takahei discus- 1,0i 17. .=.A(x)/AMs
. __ - ~ . ~ ~
' " .'
sion) showed •that the flow into the bow of a ship and the flow
from the stern were analogous and that a stern bulb would also ~out b,lb
offer reductions in wavemaking resistance.
When I was a naval architect for the Quincy Shipyard of
Genera! Dynamics in 1967, I had some model tests run at t h e
M.I.T. Towing Tank in which we tested a model which had a
bulb at the stern as well as the bow. .0 F.IP.
The results showed that, at certain speeds, the wavemaking
resistance became very low, similar to the results in the Inui Cpw= Cpo - ACp +ACp =Cpo
paper just mentioned. This work was reported to the Maritime
Administration in 1968. I t was not possible to pursue the work sectional area curve is C h a n g e d
further at the time, but I would commend this Subject to in-
terested researchers as a field which can bear valuable fruit.
' additive bulb '

Author's Closure ,r~ = A(x)/AMs

The author thanks the discussers for their contributions, their
compliments, and their constructive criticism. "'
First I would like to respond to some general questions raised
by some discussers.
In explaining the difference between an additive and an
implicit bulb, Fig. 29 will be helpful. The design of an implicit
bulb is connected with a change of the sectional area curve of
the bulbless form over a wide range whereas the prismatic
coefficient Ce is kept constant. In this case the bulb effect
consists of two parts, the firstdue to changing the sectional area
curve by smoothing the shoulder and the second due to the bulb
body itself. The additive bulb is a real added bulb body. The
sectional area curve of the bulbless hull is kept constant wlaereas
the prismatic coefficient changes.. Its effect is a pure bulb ef-
fect.
The design.charts mentioned in reference [38] cover a range
ofblock coefficients of 0.56 < Ca < 0.82 in steps of ACe =
0.02. They willbe available in January 1979. The author is
of the opinion that the handling of the charts is no more cum-
bersome than that of other resistance or power diagrams. The
design charts (Figs. 17-20) indicate indeed unfavorable bulb
parameter ranges which are FN dependent and should be
avoided. The humps and hollows are caused by the interfer-
ence effects between hull and bulb. Theoretical calculations
in which an independent variation of the different bulb pa-
rameters is possible verify the oscillating tendency of the bulb
effects (see Fig. 7) as functions of the position .(represented by
the three linear bulb parameters) and the moment of a doublet
(represented by the volumetric parameter). Figures 17-20
reflect the dynamic behavior of a bulb, and are the real design
guidelines, whereas Figs. 24-26 show only the geometric de-
pendence on the different bulb parameters of the ship-bu!b
forms used in the research work. These figures should be used
only for i'ough estimates of a bulb if the power consumption is
of no interest. They arein need of improvement, for example,
in a manner proposed by Dr. Yim in his own paper.
Dr. Yim's paper has been studied by the author, but unfor-
tunately has not been included in the list of references. The
points of view of Dr. Yim's'paper and the present author's paper
are different. Dr. Yim's simple design method is based on the
theory of wavemaking. According to this method the designer
obtains an optimal bulb radius for a given sectional area curve,
but not the power reduction, which must be calculated or es-
timated from model experiments. From the present design
charts the designer obtains the power reduction for a selected
bulb, which is not necessarily an optimal bulb.
In the author's opinion, one point of Dr. Yim's contribution
is very important--the distinction between doublet and source
Design of Bulbous Bows
1.0
~ +~Cp
Cpw = Cpo .ACp > Cpo
• sectional area curve remains unchanged
Fig. 29
types of bulbs. It is well known that for a bulb form as well as
for all profiles, the configuration of the entrance part--say the
first 15 percent of the body length--is decisive in determining
its hydrodynamic effect. Therefore it is not recommended that
a bulb body be produced from prefabricated parts such as tubes
and spherical sections. Regarding wave resistance, the author
agrees with Dr. Yim that the waterlines of a bulb body should
be ovoidal for ships with low waterline entrance angles .and
nearly spherical with large angles.. The introduction of an
additional parameter related to the waterline.shape and its
quantitative influence on power reduction was too complicated
in this research work. Such finer points were not recognizable
in the results of the unsystematic model experiments used here;
they remain to be clarified by theoretical calculations.
The author thanks Dr, Yim for the detection of the error in
Fig. 21. The correct formula reads
mE*R,w = 1.0 -- CR,W with/CR,w withou'i '"
Dr. Baba's contribution is very instructive and his paper men-
tioned in reference [24] is a practical aid in full-form design.
I agree fully with his three points related to these forms. But
it was not the subject of the present paper to find out rules for
main hull design, especially for afterbody forms.
The resistance problem is very difficult, especially the in-
fluence of the different ship hull parts on the different resistance
components, and in this case the correct choice of a charac-
teristic length. But the designer in practice wants design
guidelines which are easy to use. Therefore only one unique
characteristic length has been introduced and over the whole
Ca range.
In the author's opinion, neither a form factor nor the scale
effect is important in the given definition of the specific bulb
effect.
215

. I restrictedlength i =i
~tst CWL
em contour
WL'A'
Fig. 30
The overall power reduction factor proposed by Prof. Ward
was used in the paper quoted as reference [8]. For reasons
mentioned in the present paper this factor yields unfavorable
bulb effects for ships with long parallel middle body due to the
high viscous resistance component. Therefore the influence
of a bulb may be underrated. The new design charts are re-
lated to the residual power only and are more realistic.
Mr. Robinson mentions the old question of whether a pro-
truding bulbous bow or a longer ship lengthened by the pro-
truding length of the bulb has lower resistance. Here the
statement .of the theory of wave resistance is very clear. The
optimization procedure for a ship with minimum Wave resis-
tance and a restricted length yields waterlines as shown in Fig.
30. The length of the CWL is the shortest one; that is, the ship
with lowest wave resistance is a bulbous ship. Mr. Robinson
is right in that there is no difference whether the bulb selection
is based 'on propulsio n or resistance tests. But in practice it is
better to .base the selection on resistance tests because the pro-
peller is to be adapted.to the wake field which is influenced by
a bulb.
The influence of a bulb on wake or thrust deduction is only
mentioned in the present paper and the author hopes thatProf.
Ward's research program will bring more insight to this
problem.'
The bulb effect depends on speed, size, form, and location
of the bulb body. As shown in Figs. 8 and 9, an optimal bulb
with an optimal effect exists for each speed, or each ship-bulb
form has optimal properties at only one speed. In other words,
depending on FN, the resistance of a ship-bulb form is at first
higher than that of the hulbless ship, and then at a critical speed
this relation changes and the bulb effect becomes posi.tive, in-
creases to an optimal value and then decreases monotonously
to zero as FN increases to infinity.
Indeed the resistance of a bulb body itself is caused by both
wavemaking and wave-breaking, the percentages depending
on the wetted part of the bulb body. Therefore, in the case of
fine ships as well, wave-breaking occurs at the bulb at low speed,
if it emerges so that its most forward point lies in the waterline.
In this case the blunt bulb body behaves like a full ship. At high
speed this unfavorable behavior vanishes if the bul b is fully

~.~.~
. . . . .
without
with bulb
0., Li oo wetted.
There is no doubt that wavemaking and wave-breaking are
very sensitive to hull geometry and the author agrees with the
statements of Prof. Inui and Dr. Takahei that the configuration
0.2
of the main hull is much more important than the bulb design.
But what is one to do with a bad hull form matching the ship-
owner's requirements? Due to different and Often contra-
dictory requirements, the shipbuilder might not design the
underwater hull form with lowest wavemaking. In all model
basins it is well known that to find out the best hull one needs
•three additional model variations, and for the optimal hull form,
Of~ 30, while time and money allow one or two. Therefore a bul-
\
bous bow is often the best alternative means to reducing the
! power consumption of a ship which looks like a box.
' ' 0.10 / As mentioned in the paper, the author is aware of the danger
of independent variation of the different bulb parameters in
order to derive the charts. He does not recommend use of the
w =0.1J charts in the design of optimal bulbs by taking the parameters
in accorclance with their maximum reduction effect only.
The author thanks Messrs. Muntjewerf and Holtrop for their
* I , critical remarks. The use of a regression analysis is certainly
possible and the results could be incorporated in a computer
program. But for a regression analysis the number of the dif-
ferent cases must be large enough. If the number is too small,
the danger exists that no parameter is of importance."
The influence of a bulb On viscous resistance, on the boundary
layer, and on the wake field is a separate research program and
was not the subject of the paper. Relating to the wake field,
it has been observed that the bulb widens the nominal wake
distribution in the propeller plane as shown in Fig. 31. But the
C B = 0.59 author is not able to explain this effect because, due to the lower
FN = 0.27' pressure loss of bulbous ships, the experimental result must be
opposite. It may be that an extension of Hoekstra's 1° calcula-
CAB:O.OS' tion procedure up to the propeller plane would bring light to
this problem.
Wake fro( The author is aware of the step backwards due to using Lee
without ( 1o Hoekstra; M., Boundary-LayerFlow Past a Bulbous Ship-Bow at
Vanishing Froude Number," Proceedings, International Symposium
Fig; 31 on Ship Viscous Resistance, SSPA, G6teborg, Sweden, 1978.
216 /w = Design of Bulbous Bows
instead of Le for full ships. For reasons mentioned by Dr.
Baba, Lee is better for slender ships and Le for full ships. But
two different lengths in power estimation appeared to the au-
thor to be too much of a complication in the procedure.
The author gives special thanks to Mr. Byers for his comment
and for the two figures, which express more than is possible to
explain by words. For a better.explanation of the test results
shown it would be informative to know something more about
the bulb sizes. It may be possible that Bulbs A to'E of Hull A
are too voluminous for the desired design speed. The beneficial
effects of these large bulbs are remarkable at the higher Ft¢
required or, vice versa, for the required FN the bulbs must be
smaller.
Concerning Prof. Couch's question on the reduction of vis-
Design of Bulbous Bows
cous resistance, the author refers to similar questions of other
discussers which have been answered earlier.
Relating to Mr. Guarino's question, it may be stated that the
application of a bulb to tugs and fishing boats is a special
problem because these ships operate at high Froude numbers.
In these cases the wavemaking resistance is dominant and a
suitable or optimal bulb would be too large (see Fig. 8 or Yim's
paper) for practical use. But for reducing pitching motion and
as space for acoustic sounding gear, a small bulbous bow may
be recommended.. Such a small bulb would not influence the
resistance of the ship in an unfavorable manner.
Finally, the author would like to thank the Society for its
interest in his workand for the opportunity of presenting his
.paper.
217