Method 369

31.2 Guldhammer and Harvald's

Table 31.1 Range of parameters suitable for Guldhammer and Harvald's method

Permissible range of main dimensions

Parameter
Symbol Unit Range
ship speed n/s Fr<0.33
length-beam ratio L/B 5.0.. 8.0
block coefficient
CB 0.55.0.85
length-displacementratio M = L/VV I-1 4.0.. 6.0

and Mennen (1982) and Hollenbach (1999) are described in some detail after we have
cOvered ship propulsion and propeller selection.

31.2 Guldhammer and Harvald's Method

The methodwas developed in the late 60s and early 70s when most ships did not yet have Origin and
a bulbous bow. In later publications, Harvald (1983) and Andersen and Guldhammer computerization
(1986) updated the procedure and introduced a computation length L which made the
method applicable to ships with bulbous bows. Andersen and Guldhammer (1986)
provide regression formulas to replace the traditional CR-charts and extend the method
to include powering estimates. The regression formulas are the basis for this section, as
they are more applicable to today's computerized work flow than H.E. Guldhammer and
S.A. Harvald's original design charts. However, the design charts provide an important
tool to perform basic error checks during implementation.
In addition, more recent updates reported by Kristensen and Lützen (2012) are refer- Updates to the
enced. The report provides formulas for wetted surfaces of common vessel types and method

an updated bulbous bow correction for the residuary resistance, which promises to
give more accurate results for modern hull forms. Kristensen and Lützen (2012) also
include additional regression formulas for the residuary resistance of ships with high
prismatic coefficient Cp.

31.2.1 Applicability

The method of Guldhammer and Harvald is applicable to single and twin screw dis- Single and twin
placement type vessels. Based on the regression formulas provided by Andersen and screw vessels
Guldhammer (1986) for the residuary resistance, limits of application for the method
shown in Table 31.1. The original charts allow
are
predictions for velocities up to Froude
number Fr = 0.45.

31.2.2 Required input

Guldhammer and Harvald's method requires only the hull form parameters listed in
aDle 31.2. A special computation length L plays a prominent role for modern hull Input
forms.
 Method
370 31 Resistance Estimates
-Guldhammer and Harvald's
for Guldhammer and Harvald's method
Table 31.2 Required and optional input parameters
Remarks
Parameter Symbol
required parameters

length between perpendiculars Lpp

length of aft overhang
in waterline Laft usually LwL =Lpp + Lat
extension of S beyond fore perpend. LEore total extension of wetted surface (usually
computation length equal to Los)

maximum molded beam in waterline

B
molded draft
orvolumetric displacement
block coeficient CB or midship section area AM
prismatic coefficient Cp
at forward perpendicular FP
transverse cross section area of bulb
AgT
optional parameters

propeller diameter Dp for propulsion analysis

LCB or assume optimum position
longitudinal center of buoyancy
wetted surface (hull+rudder) S
bilge keels, stabilizer fins, bossings,
etc.
wetted surface of appendages SAPP
form factors for fore and aft body Fy. Fa-3sFA.Fr S+3
LwL

AP traditional midship Guldhammer/ Harvald

section midship section
waterline
aftmost point of toremost point
displacement of dlsplacement

LPP
/LCBo
LCB
Lfore
0.485 L

length L for Guldhammer

Definition of the midship section and the computational
Figure 31.1
and Guldhammer, 1986)
and Harvald's resistance estimate (Andersen

hull was usually equalto o r

Computation Before 1970, the longitudinal extension of a submerged
waterline length often served basis ed as
length length in the waterline LwL. Therefore, the
resistance computations. However, with the
introduction of bulbous bows, submerg
in
has to be considered
volume is stretched over a longerlength. The increased length
resistance estimate. Andersen and Guldhammer (1986) called it the computationi
L (see Figure 31.1). It is used as reference length throughout the method descr
here. over wetted

For most modern ships, the computation length will be equal to the length ove adding
Lpp Dy
surface Los. It is derived from the length between perpendiculars
 312 Guldhammer and
Harvald's Method | 371

ongth L(ore of the additional wetted parts protruding forward of the forward
tnendicular FP (mostly for bulbous bows) and the length Lat of wetted surtace
perpend

ending past the aft perpendicular AP.

L =Lpp +
Lat +Lfore (31.3)
S positive, if the submerged hull extends aft of the
aft perpendicular. Lfore S
altse. if the submerged hull extends forward of the fore perpendicular. In
p o s i t i v e ,

Lat may be negative.

or
unusua
cases, Lfore
sC the length in waterline or the aft overhang are unknown, Kristensen and Lützen
(2012) suggest to use the following relationships:
LWL = 1.02 Lpp for tanker and bulk carriers

LWL 1.01 (31.4)

=
Lpp for container ships
Guldhammer and Harvald (1974) assumed that the waterline extends about 3% of Lpp Defhnition of
midship location
aft of the aft perpendicular AP. Therefore, midship was located 48.5% of the length in
waterline LwL aft of the foremost point of displacement (originally equal to FP). Within
the method, this definition of midship location is kept for vessels with bulbous bow.
Nowadays, the foremost point of displacement is the tip of the bulb. As a result, the
midship location moves forward of the usual position at Lpp/2 (see Figure 31.1).

Consequently, the longitudinal location of the center of buoyancy has to be stated with Longitudinal
center of
respect to the midship position as defined by Andersen and Guldhammer (1986).
buoyancy
If the position of the center of buoyancy is given as length LCB, with respect to
the traditional midship position (Lpp/2), its location with respect to Guldhammers
midship definition is

LCB LCB, - 0.015Lpp +0.485Lat 0.515Lfore (31.5)

Obviously, all quantities have the dimension of the selected length unit. The equation
in Andersen and Guldhammer (1986) corresponding to Equation (31.5) has amisprint:
0.15 Lpp instead of 0.015 LPp. Note that in contrast to Andersen and Guldhammer
(1986), a positive LCB indicates that the center of buoyancy is forward of midships
throughout this book.
In early design stages, the wetted surface S is often not yet known. Andersen and Wetted surface
Guldhammer (1986) suggest Mumford's formula (31.6) but it is likely to underpredict
the wetted surface of modern ship hulls with bulbous bow.

S 1.025 Lpp (Cp B + 1.7T)

1.025 V + 1.7 Lpp T Mumford's formula (31.6)

Kristensen and Lützen (2012) modified the constants in Mumford's formula to improve
the wetted surface prediction for modern hull forms. The formulas employ the length
in waterline instead of the length between perpendiculars.

S= 0.99( + 19LwLT)bulk carriers, tankers, vessels with high Cp

(31.7)
container ships (vessels with low to medium C)
S=0995(+19 LwLT
372 31 Resistance Estimates -

Guldhammer and Harvald's Method

twin screw ships (Ro-Ro vessels) with

S=1.53(+0.55 Lw T open shaft lines
twin skeg ships (Ro-Ro vessels with twin
S= 1.2(+ 15 Lw T rudders)
(318)
S=
1.11( 1.7Lw T)
+ double ended ferries

Kristensen and Lützen (2012) also provide two useful equations to estimate the .
in wetted surface for changes in draft. Assuming that the wetted sur
surface at S, ange
draft7
is known, the wetted surface S, at draft T2 is approximately equal to

= S + 2.0(T, - T)(LwL + B) tanker, bulk carrier

S2 = S + 2.4(T-T)(LwL + B) container ships (31.9

31.2.3 Resistance estimate

Total resistance Within the resistance estimation method by Guldhammer and Harvald the
total resis.
tance coefficient is defined as

C = CR+Cf +Ca + CaM + [CAs]

(31.10)
CR is the residuary resistance coefficient. The frictional resistance
is based on the ITTC 1957
coefficient C
model-ship correlation line augmented by the effect of
appendage surfaces not included in the wetted surface S. C stands for a combined
correlation and roughness allowance. As usual, CaA is the air resistance
The steering resistance coefficient Cas coefficient.
originally included in Equation (31.10) has been
dropped by Kristensen and Lützen (2012).
All resistance coefficients have been rendered dimensionless
by division with the dy-
namic pressure pu and the wetted surface S
resistance is computed as usual:
of hull at rest. Therefore, the total
the

Ry p s c (31.11)

Residuary resistance

Residuary Key to all resistance estimates is the derivation of wave or

resistance method does not employ a form factor, hence, the residuary resistance. The
is used. In Guldhammer and Harvald's
residuary resistance coefficient CR
method, it is a function of Froude number Fr,
prismatic coefficient Cp, and the length-displacement ratio (31.12).

M
(31.12)
For consistency, the prismatic coefficient Cp should be
computation length L. computed on the basis ot the
Residuary
resistance charts
Guldhammer and Harvald (1974)
provided charts for Cp
Figures 31.5 and 31.6. Charts are selected similar to the ones
shown
based on the
length-displacement rat M,
 31.2 Guldhammer and Harvalds Method 373

1es for read from the

CR curve with the correct
and

of Froude number Fr. prismatic coefficient Cp as a

fanction
der to accommodate today's
computerized work
flow, Andersen and Guldhammer Residuary
In nrOvided regression formulas for CR which resistance
(1986)
been redrawn based on the
been redrawr
replace the charts. The
example formula
arts have regression formula (31.13). They represent the
Cal charts fairly well. Although for Froude numbers above 0.3 and
higher prismatic
ficients(Cp> 0.7), resistance coefficients tend to be larger as found in the original
coefh
charts,Andersenaand Guldhammer (1986) state that the regression formulas for CR are
sableforr Froude numbers smaller than Fr <0.33.
Consequently, curves for Fr > the
re printed in light gray in Figures 31.5 and 31.6. For higher Froude numbers, 0.33
are

nal charts should be consulted which provide residuary resistance coefficients up

origina
1o Fr =045 (Guldhammer and Harvald, 1974).
Andersen and Guldhammer (1986) stated the following regression formula for a stan- CRnd, Va
And standard form
dardresiduary resistance coefficient CR.

10'CRa = E+G + H + K (31.13)

The four contributions depend on Froude number Fr, length-displacement ratio M,
and prismatic coefficient Cp:

E = A+ 1.5Fr"+A,FM |0.98+2.5 + (M -5)"(Fr-0.1 (31.14)

(M-2)
with variables

Ag 1.35 -0.23 M + 0.012 M2 A = 0.0011 M°%|

N =
2 M -3.7 M-
G 0.09 M)(5Cp -2.5) (31.15)
600(Fr-0.31s) +1
H =e with H = 8o|Fr - (0.04 +0.59 Cp)-0.015(M-5) (31.16)
K = 180 Fr37 20 Cp-16) (31.17)
The residuary resistance coefficient CR. is valid only for what was considered a standard Standard hull
form
hull form in the 1960s:

1. beam-draft ratio of B/T = 2.5

2. location of LCB at its optimal position (see later)

3. no bulbous bow

4. no appendages like bossings, struts, etc.

For ship hulls that deviate from the standard shape, five corrections have been intro- Cg corrections
duced by Guldhammer and Harvald (1974):
 Method
Resistance Estimates
Guldhammer and Harvald's
374 31

Guidance for the optimum

Figure 31.2 function of Froude standard (optimum) LCB
location of LCB as a
LCB values
number Fr. Here, negative
indicate a location
aft of midship optimum LCB-Corridor

o-G&Hmidships

0.10 0.12 0.14 0.16 0.18 0.20 0.22 0.24 0.26 0.28 030
Froude number Fr -

ratios other than 2.5

1. ACRB/T) for beam-draft
from optimum
2. ACRLCp) for LCB positions deviating
bow and stern that deviate from normal hull forms
3. ACR form) for section shapes at
4. ACRbulb) for bulbous bows

than the typical single rudder and bilge keels

5. ACRapp for appendages other
will be
The adjusted residuary resistance coefficient
CR CRd+ACRB/T) + ACR(orm) + ACR(bulb)
+ACR(LCB) +ACR(app) (31.18)

Beam-draft ratio For vessels with B/T # 2.5 a simple speed independent correction is applied.
correction
10ACRB/T) = 0.16-2.5) (31.19)

Longitudinal In their original work, Guldhammer and Harvald (1974) provided two charts to estimate
center of a correction for LCB-locations which deviate from the presumed optimum. This
DuOyancy correction bears considerable uncertainties, and Kristensen and Lützen (2012) suggest
correction
to drop it altogether. For the sake of completeness, we detail the procedure outlined in
Andersen and Guldhammer (1986).
Standard LCB Figure 31.2 provides an LCB-location corridor which is considered optimal for a given
design speed. A regression formula for the mean line is reported in Andersen and
Guldhammer (1986). The formula below has an additional minus sign, reflecting the
convention that a positive LCBisforward of midships.
LCBdFr)= -(0.44Fr - 0.094) L (31.20)
 Method 375
31.2 Guldhammer and Harvald's

21.3 Bulbous bow COrrections to the standard

Table
and Guldhammer, 1986) residuary resistance coefficient (Andersen

Fr 0.15 0.18 0.21 0.24 0.27 0.30 0.33

0.5 +0.2 +0.2 +0.2 0.0 -0.2 -0.4 -0.4
0.6 +0.2 +0.2 +0.2 0.0 -0.2 -0.3
0.7 ACR(lable) +0.2 +0.2 0.0 -0.2 -0.3
-0.3
-0.3 -0.3
0.8 +0.1 0.0 -0.2 -0.2 -0.2 -0.2 -0.2

Rased on the actual longitudinal center of

buoyancy location LCBactual of the
ship
iect,
proj
the relative deviation from the standard value for each
speed determined:
is

ALCB(F) Bactual LCB(Fr)

=
(31.21)

The actual LCBactual is of course considered constant. Effects due to sinkage and trim
are ignored.

A correction for the residuary resistance coefficient is computed, if, for a given Froude LCB-correction

number Fr, the actual LCB is forward ofthe standard LCB location, i.e. ALCB> 0.

if Fr <(-C+1.1Cp - 0.0875) or ALCB 0

10' ACR(LCB) Fr ALCB else

C+1.1Cp -0.0875
(31.22)
version of the Guldhammer
Again, this correction has been dropped from the newest
and Harvald method (Kristensen and Lützen, 2012).
stations Shape factors
A standard hull form has neither pronounced V- nor pronounced U-shaped
bulbous bow and a moderate
in entrance and run of the hull. A raked stem without a
Modern bulk carriers and tankers with high block
cruiser stern are considered normal.
Fast
coefficient often feature extreme U-shaped stations in the fore body (entrance).
stations. Guldhammer
container ships often have entrances with pronounced V-shaped
and Harvald (1974) assigned fore and aftbodyshape factors Fr and FA between-3 and
The value -3 is for extreme V-shaped stations, 0 for normal stations,
+3, respectively.
and +3 for extreme U-shaped stations.
and correction to residuary resistance Form correction
Once values have been assigned to Ff Fa» a

coefficient is computed from

(31.23)
10ACRdorm)
deemed beneficial whereas V-shaped
Clearly, U-shaped stations in the fore bodythe aft
are

stations increase resistance. Vice-versa

for body
resistance is derived by interpolation Bulbous bow
bow residuary
Ihe influence of a bulbousbetween
on
and correction
Fr- and Cp-directions the values provided in Table 31.3 (Andersen
in
Guldhammer, 1986).
376| 31 Resistance Estimates Guldhammer and
-

Harvold's Method

Original bulbous The full correction is

bow correction applied,
if the transverse cross section area of
forward perpendicular AT is larger than 10% of the midship sectionthe hul
ulb a
ART 0.1AM correction is linearly scaled down to zero.
the are a A.

from Table 31.3 if

10'ACR(bulb)
10'ACRable AgT20.1AM
10 ACRable) 10AAMRT ifAgT0.1AM (31.24
New bulbous Kristensen and Lützen (2012) analyzed model tests of 27 newer tanker
bow correction and bulk.
hull forms and 21 newer hull
forms of vessels with lower block coefficients. car
by Andersen and Guldhammer (1986), they concluded that the As inde
correction is too original bulbned
pessimistic. Experience in the design of bulbs and optimizab
their shape has made them more
effective. The effectiveness of a bulbous bow
on different
factors and its exact shape should be determined by d0
shape optimi
process. Design guidance can be found in Kracht (1978). mization
later in the design
To reflectthese improvements, Kristensen and Lützen (2012)
proposed new
making a distinction between vessels with high (Cp * 0.8) and low block correr ons hi..

(C0.6). coeficiients

for
tankers and bulk
max(-0.4,-(0.1+ 1.6Fr) carriers (high C)
10'ACRbulb (250Fr-90) for container ships (vessels
100 10°CR(no bulb) with low Cg)
For vessels with high block coefficient (and ME [4.4,5.2] and (31.25)
Cp E
[0.78,0.87), the
correction is mostly speed dependent. More slender vessels have
larger variations of the
length-displacement ratio M and prismatic coefficient Cp. Therefore, the correction is
expressed as a fraction of the residuary resistance coefficient without a bulb.

10CR(no bulb) = 10°(CRg + ACRB/T) + ACR(LCB) + ACR(form))

(31.26)
Although not specifically mentioned, it seems safe to assume that the new correction
values (31.25) have to be scaled for bulbous bows with
Agr less than 10% of Ay.
from Equation (31.25) if
10 R(bulb)
10 AC R(bulb) Agr20.14M (31.27)
10A BT
10 ACR(bulb) AM ifAgr 0.1AM
Once all adjustments have been evaluated, the final residuary resistance coeficient can
be computed from Equation (31.18).

Appendage Guldhammer and Harvald (1974) also include a correction for appendages. Etects
correction for CR ofsingle rudder and bilge keels are included in the standard CR value. However,
the vessel features bossings, exposed shafts, and shaft brackets they may affect wave
and viscous pressure resistance which warrants additional corrections to the residuay
resistance coeficient.
 377
31.2 Guldhammer and Harvald's Method
bossings: add.3%-5% of CR..,d for full ship forms
shafts and brackets: add 5%-8% of CR., for fine hull forms
Andersen and G
Guldhammer (1986) caution that the correction
tions should be small com-
pared to the overall CR value.

Frictional resistance

The frictional resistance coefficient is computed with the ITTC 1957

model-ship corre
lation line. Frictional
resistance
0.075 coeffhcient
C log Re-2
(2.18)
Acain, a form factor is not used in Guldhammer and Harvald's method.

In case the wetted surtace S does not include the surface of appendages, the frictional
coefficient should be augmented by the ratio of total wetted surface S+ SAPP and
wetted surface of the hull S.

CF S+SAPP
S
C (31.28)

Additional resistance components

The incremental resistance coeficient represents a combined correlation and roughness Incremental
resistance
allowance CA Guldhammer and Harvald (1974, (page 5) stated only a few discrete
coeffhcient
values depending on vessel length. In their update to the procedure, Andersen and
Guldhammer (1986) provided the formula (31.29).

10 C = 0.5 logo(V) - 0.1(logio(V)) (31.29)

t considers that the roughness effect tends to decrease with increasing ship length
and size. The input is unfortunately dimensional, and the displacement V must be
entered in cubic meter [m']. This equation should potentially be replaced with the
roughness allowance ACp from the ITTC 1978 performance prediction procedure (see
Equation (30.2).
Air resistance coefficient CaA and steering resistance coefficient CaS are assumed to be Air and steering
Andersen and Guldhammer (1986). resistance
constant in Guldhammer and Harvald (1974) and
Suggested values are
10CAA 0.07 (31.30)
10 CAs = 0.04 (31.31)

by moving
caused the superstructure of
he air resistance represents the resistance winds must be considered separately
the vessel through air at rest. Resistance high
in
1or the selection of a service margin and maneuvering assessment. Steering resistance
may occur in vessels which need constant rudder action to counter the wheel effect of
a single propeller
 31 Resistance Estimates Guldhammer and Harvald's Method
378|
Table 31.4 Air resistance
Vessel type 10CAA
coefficients for different types of small tanker O07
vessels (Kristensen and Lützen, 2012) Handysize tanker
O.07
Handymax tanker
0.07
Panamamax tanker
0.05
Aframax tanker 005
Suezmax tanker 0.05
crude carrier (VLCC)
very large 0.04
container ship seeEquation (31.32
resistance estimate example
Table 31.5 Principal dimensions for
1000 TEU container ship
145.0 m
length between perpendiculars LpP
3.3 m
extension of bulb forward of FP Lfore
2.7 m
extension of waterline aft of AP Laft
B 24.0 m
beam
T 8.2 m
draft
18872.0 m
displacement
estimated wetted surface S 4400.0 m2
ABT
ABT 14.0 m2
bulbous bow cross section area at FP
wetted area of appendages SAPP 52.0 m2
Lpp) Cp 0.6783
prismatic coefficient (based on

location of (with respect to Lpp/2) LCB% +0.4 % of Lpp

US 17.5 kn
design speed
standard shaped fore and aft body

Kristensen and Lützen (2012) analyzed data for

different types of vessels and recommend
based
the constants in Table 31.4. For container ships they developed a simple formula
on the container carrying capacity of the vessel

((0.28 TEU-0.126), 0.09) (31.32)

10 CAA = max

container. The steering

TEU stands for Twenty-foot Equivalent Unit, i.e. a 20 ft long
resistance is no longer used in their update to the Guldhammer and Harvald metno
Andersen and Guldhammer (1986) complemented the resistance estimate with eq ua
34
these in Chaptet
tions for the hull-propeller interaction parameters. We will discuss