CFD INVESTIGATION OF PENTAMARAN SHIP MODEL WITH CHINE HULL

FORM ON THE RESISTANCE CHARACTERISTICS

Yanuar1, Wiwin Sulistyawati2, Kurniawan T. Waskito1

1
Department of Mechanical Engineering, University of Indonesia, Depok 16424, Indonesia
2
Graduate Student Department of Mechanical Engineering, University of Indonesia, Depok
16424, Indonesia

ABSTRACT
Several studies of multi-hulls ship showed that hull resistance could be predicted from
position between the each hull. The design optimization of the hull form could be considered
by minimizing resistance, which is generally the sum of the viscous resistance and the wave
making resistance of the ship model. This paper presents an investigation of pentamaran hull
form with chine hull form to the effects of outriggers position, asymmetry, and deadrise angles
on the resistance characteristics. This research investigated the resistance characteristics by
modeling pentamaran hull form using chine with symmetrical main hull and asymmetric
outboard on the variation deadrise angles: 25o, 30o, 35o and Froude number 0,1 to 0,7. We
examined the calm water resistance characteristics of six pentamaran models with chine-hull
form by variation of deadrise angles using Ansys CFD. Increasing the deadrise angle of chine
hull form contribute to the resistance due to the interaction of the wave systems produced by
each demihull. Strong interaction effect on the resistance coefficient component was found for
change in the hull length as well. Compared to the wigley hull form, the maximum resistance
drag reduction of the chine hull form was reduced by 18.05% in deadrise 250, 16.1% in
deadrise 300, and 18.19% in deadrise 350. While the smallest value of total resistance
coefficient was generated from chine 350 at R/L:1/14 and R/L:1/7. Optimum hull form for
minimum resistance has been obtained, so it is interesting to continue with angle of entrance
and stem angle of hull for further research.

Keywords: pentamaran; deadrise angles; resistance; CFD

1. INTRODUCTION
Many studies established that pentamaran better than monohull, i.e.: large volume and deck,
low resistance at high speed, good stability, and good sea keeping performance. The factors
affected mutual on multihull were viscosity resistance due to the wide variation of the wetted
surface on the changes of pressure and speed, and wave making resistance as effect
engagement of the cross section along the ship and spreads of the wave. Outriggers position to
the main hull contributed to influence on frictional resistance and stability (Oller, et al. 2003).
The Asymmetric outrigger configuration and separation distance resulted change in resistance
and there is no single configuration that consistently outperforms the other configuration
across the entire speed range (Yanuar, et al. 2017). Frictional resistance on pentamaran
increased as a consequence on enhancement of wetted surface area hull form, but wave
making resistance can be reduced with a slender hull form. Ikeda, et al., (2005) defined wave
making resistance reduction and enhancement of frictional resistance were affected by the hull
length to width ratio (L/ B).
Hull form make very significant impact on the ship, then modeling with the precise
configuration on pentamaran was indispensable. Researches on pentamaran generally use
Wigley hull form, this study use chine hull form to investigate and compare with Wigley. The
specialty of this chine hull form in some studies shown advantages such as: reducing
resistance, easier and faster in the building process. Experiment Chengyi (1994) shown
symmetrical chine hull form on catamarans could reduce wave resistance, interference
between hull cause resistance will tend to fall on Fr > 0,5. Blount (1995) found that chine
more profitable on Fr > 0.75 compared with NPL series, while in sea keeping performance;
rounded models are superior in displacement conditions. Moraes, et al. (2004) compare chine
for catamarans and Wigley model by a computer program SHIPFLOW. Wigley hull form in
deep waters have a higher value Cw (wave coefficient) at 0.3 < Fr <0.6, and in shallow waters
the highest Cw value at Fr 0.4. While the chine hull form, Cw tend to fall at 0.6 <Fr <0.7 in
line with the decrease in depth, and the highest value Cw in shallow waters occurred at Fr 0.4.
Ship performance is determined by parameters: L/B, AP / V2/3, LCG, deadrise angle, angle
variation along the hull, and shape of chine (Begovic and Bertorello 2012). Bari (2016) used
multihull with chine get hydrodynamics of the catamaran on deadrise angle 0 o-20o lift
coefficient will improve with decreasing spacing of outriggers and high Froude numbers in
line increase of deadrise angle. But at larger deadrise angles, high Froude numbers and the gap
of outriggers more widen the complex effects occur where the waves flow direction resulting
in pressure on the rear area of hull, and also resulted in lift coefficient. Tarafder, et al. (2013)
configured pentamaran Wigley hull form with numerical method to analyze hydrodynamic
characteristics at speed variation. Specifically of this study is investigating the calm-water
total resistance, frictional and residuary characteristic of a pentamaran with chine hull form at
variation of deadrise angles on 250, 300, 350. The numerical analyzes based on computational
fluid dynamics (CFD) modelling and validation with last experimental data

2. METHODOLOGY
Experimental and numerical modeling techniques are very important to get hull form with
good hydrodynamic, performance and safety. Experimental for multihull ships were first
performed by Hsiung and Xu (1988) on catamarans to obtain optimal hull with minimum
resistance. Then Tuck and Lazauskas (1998) used thin ship theory to determine the optimum
distance multihull configuration for get minimum wave resistance. Peng (2001) with
numerical computational approach based on Michell theory (1898) determined resistance and
motions on catamaran, trimaran, quadrimaran and pentamaran using Wigley hull form on
several configurations. He mentioned that the waves making resistance on multihull always
the biggest component as the speed or value of Fr increase.
Based on the hypothesis of Froude that the total resistance consists of two components, i.e.:
frictional resistance (Rf) and the residuary resistance (Rr).

Rt =Rf + R r (1)

The analysis data experiment using ITTC57 methodology for calculate the total resistance
coefficient (Ct):
 Ct =C f +C r (2)

Rn2 (3)
log



0.075
Cf=

Pentamaran consists of one main hull, two inners and two outers hull, with different Reynold
number, then the frictional resistance coefficient calculated by:

S main 2 side1 2 Side 2

C f =C fmain +C f 1 +C f 2 (4)
ST ST ST

Then the residual coefficient was given by:

Cr=CtCf (5)

2.1. CFD Method

This study was investigation a K- model transformed the outer boundary layer and the free
stream (Menter, 1994). In a large variety of different flow, K- model has been used
successfully. The K- model (Versteeg, 2007) is numerical integrated equation of viscous sub
layer used in Computational Fluid Dynamics (CFD) to simulate characteristics of
turbulent flow. It was generally used to mixing-length model, to find algebraically turbulent
length scales in high complexity flows. K is turbulent kinetic energy determines in the
turbulence, . A nonlinear boundary condition on allows using of a homogeneous boundary
condition.
The mathematical model for free surface flow was based on the homogenous multiphase
EulerianEulerian fluid approach. This approach share the same velocity field and other
relevant fields such as temperature, turbulence, etc. Numerical method was derived from
Navier Stokes equations where using a Volume of Fluid (VOF) approach. Incompressible
Reynolds averaged Navier-Stokes (RANS) expressed (Wilcox, 2006).

Ui U i P T ij
+ U j = + (6)
t x j xi x j

( t +U x )+ Ux =0
j
j i
(7)

Equation (6) as momentum equation (second Newtonian law for fluids) and equation (7) as
continuity equation. Where, : fluids density, U i : velocity components, P: pressure, and
T ij : components of viscous stress tensor as follows:
 T ij =(2 Sij U iU j) (8)

1 Ui U j
S ij = ( +
2 x j xi ) (9)

Ui
where =0 as effect of turbulence on the flow by Reynolds Stress (RS) uiu j . And
xi
the equation of volume fraction transport (Ferziger and Peric, 2002):

c (c U j ) (10)
+ =0
t xj

c; volume fraction, ; fluid density, ; viscosity is calculated as:

= air c + water (1-c) (11)

The turbulence model wall laws have restrictions on the y+ value at the wall. The wall y+, a
nondimensional distance from the wall to the first grid point was similar to local Reynolds
number represented in Equation 12. Its determining whether the influences in the wall-
adjacent cells are laminar or turbulent. Turbulence models deal with the flow in the boundary
layer can be further subdivided into a laminar (viscous) sublayer and a fully turbulent region.

U y (12)
+=

y

2.2. Last Experimental Result

The results of the experimental data by Yanuar (2015) was used to validate the CFD analysis
on Wigley pentamaran hull form with asymmetric inner side-hulls. The experiment had done
to 9 configurations at variations Fr 0.1 - 0.7. On model IIA with configuration S/L (separation/
length of main hull) 3/16 dan R/L (stagger/ length of main hull) 1/ 20 at Fr 0.35 the resistance
was reduced by 25%. At Froude numbers ranging from 0.2 to 0.7, the resistance coefficients
have a downwards trend as shown in Figure 1.
This research implements a numerical simulation by CFD on the configuration of the
pentamaran experimental, Yanuar (2015). The results of IIa configuration as initial modeling
then performed with pentaraman chine hull fom at variation of deadrise angles: 25 o, 30o, 35o.
The first step was modeling an initial pentamaran which used wigley linesplan of Yanuar
experimental. The second, modified the wigley hull form with chine form in the same S/L
position. The specications of both hull form were shown in Table 1. Midship section of
Wigley pentamaran and chine hull forms were designed as shown in Figure 2 and 3.
Transformation Wigley to chine form with some differences main hull L/B: 12; 13.75; 16.5 in
constant B/T: 2.9 and H/T: 2.1, there are 6 new configurations from the initial configuration.
In each deadrise angle of chine hull form there are two configurations as shown in Figure 4.
Each configuration at each chine hull form is a change in the position of inner and outer side
hull from its original position.
 I.a S/L : 75/200, R/L : 5/20
I.b S/L : 65/200, R/L : 5/20

Total R esistance C oefcient (C T)

0.006 I.c S/L : 55/200, R/L : 5/20
II.a S/L : 75/200, R/L : 1/20
II.b S/L : 65/200, R/L : 1/20
0.005 II.c S/L : 55/200, R/L : 1/20
III.a S/L : 75/200, R/L : -3/20
III.b S/L : 65/200, R/L : -3/20
0.004 III.c S/L : 55/200, R/L : -3/20
Monohull

0.003

0.002

0.001

0.1 0.2 0.3 0.4 0.5 0.6 0.7

Froude number (Fr)

Figure 1. Total resistance coefcient for all congurations (Yanuar, 2015)

Table 1. Model hull specification
Main Wigley Chine 250
Dimension Main Inner side Outer side Main Inner side Outer side
LOA 2.00 m 1.00 m 0.75 m 2.40 m 1.02 m 0.93 m
B 0.20 m 0.10 m 0.07 m 0.20 m 0.10 m 0.07 m
T 0.07 m 0.07 m 0.06 m 0.07 m 0.07 m 0.06 m
H 0.15 m 0.15 m 0.14 m 0.15 m 0.15 m 0.14 m
Cb 0.57 0.59 0.58 0.59 0.67 0.55
WSA 0.39 m2 0.18 m2 0.11 m2 0.55 m2 0.30 m2 0.21 m2
tot 24.5 kg 24.5 kg
Main Chine 300 Chine 350
Dimension Main Inner side Outer side Main Inner side Outer side
LOA 2.75 1.08 m 0.93 m 3.30 m 1.12 m 0.93 m
B 0.20 m 0.10 m 0.07 m 0.20 m 0.10 m 0.07 m
T 0.07 m 0.07 m 0.06 m 0.07 m 0.07 m 0.06 m
H 0.15 m 0.15 m 0.4 m 0.15 m 0,15 m 0.14 m
Cb 0.52 0.65 0.55 0.44 0.62 0.55
WSA 0.61 m2 0.31 m2 0.21 m2 0.69 m2 0.32 m2 0.21 m2
tot 24.5 kg 24.5 kg

Figure 2. Pentamaran Wigley

 Figure 3. Pentamaran chine

25 deg, S/L: 3/16, R/L:1/16 25 deg, S/L:3/16, R/L:1/33 30 deg, S/L: 3/16, R/L:1/8) 30 deg, S/L:3/16, R/L:1/14

35 deg, S/L: 3/16, R/L:1/5 35 deg, S/L:3/16, R/L:1/7

Figure 4. Six new configurations of chine hull form

2.3. Meshing Technique
Blazek (2001) and Ferziger (2002) revealed that mesh quality is very important to perform a
reliable CFD. Total element number in the mesh should be large enough to represent the
geometry and flow phenomena in the domain. Physical domains at free surface model for
domain boundaries in detailed are provided in Table 2. The mesh settings for element size was
repeated 5 times (4.5M, 7M, 9M, 10M and 11M) to evaluate convergence. For verication, i.e.
convergence and the assessment of numerical uncertainty, processing mesh has been carried
out to the total resistance (force) which element size of boundary 0.1 and the hull of 0.005 to
0.003. The meshing results on the model shown in Figure 5, which produced about 10 million
elements. ICEM meshing method was selected to produce the mesh automatically. The mesh
generated in ICEM using a low transition options to refine the mesh gradually. Unstructured
tetrahedral elements were selected in the entire domain and hull.

Table 2. CFD simulation properties

Property Fine Mesh (half hull)
Type of mesh Structured (tetrahedral, mixer)
Homogeneous water/air multiphase, K-, automatic wall function,
Domain Physics
buoyancy model density difference, standard free surface model
Boundary physics:
Inlet Inlet, volume fraction, turbulence intensity 0.05
Outlet Outlet, relative pressure
side wall Wall, free slip condition
Top Opening, relative pressure 0 Pa
Hull/ Bottom Wall, no slip condition
Symmetry plane Along centerline of the hull
Solver settings:
Convergence criteria Residuary type: RMS, Target: 10-e4
Multiphase control Volume fraction coupling

Total number of elements, mesh size as well as value of Y+ consideration for several variations
of mesh. The initial mesh created 7M elements then increased to 10M elements with Y+ on the
 hull > 55. In multiphase flow, volume fraction is considered to prevent a large residual.
Repeated convergence was assumed through normal residuals. The residual RMS of domain
required on 1E -04. The control time step for steady state problems, CFX uses a robust
implicit formulation allowing the large time step can be determined, and accelerate
convergence. Time step permits fixed time step size used for the entire flow domain. The flow
was dominated by advection (horizontal mass movement which resulted in the change) the
size of the time step should be scale length divided by the speed scale (Seo et al., 2010). A
steady state calculation usually requires between fifty to a hundred time steps to achieve
convergence. But overall a great time step make an unstable convergence, otherwise if time
step too small would be very slow convergence.

3. RESULTS AND DISCUSSION

Analysis of Wigley pentamaran and six chine hull form configurations with difference Lpp
demanded to change the domain, particularly, the stern and bow hull forms adjust to the ship
length. In addition to domain size, mesh settings and CFD setup follow the previously
described arrangement, i.e.: mesh size, boundary conditions, time step and convergence
controls. From the results of the CFD analysis and comparison with experimental data was
shown in Figure 6, which both graphs of total resistance coefficient (Ct) show similar trend.
The difference between the experimental result and CFD was obtained by the equation (13).
The difference average result of the cases were 3.6%, which shows CFD setup well-defined.

C T .exp CT .CFD (13)

C T= x 100
CT .exp

a)Meshing domain b) Hull meshing detail

Figure 5. CFD mesh analysis results

The resistance coefficient results of 6 chine hull forms Ct, Cf and Cr were shown in Figure 7
9, respectively. The results of Ct showed the smallest value was generated from chine 35 0
R/L:1/14 and R/L:1/7 with similar trend. The highest average of all chine hull form for Ct
occur in 0.2-0.4 Fn. Then, by increase of Fr the Ct values gradually began to fall. At deadrise
angle 350, it significantly has the smallest Ct compared to deadrise 250 and 300. It indicates
that increased deadrise angle caused decrease the Ct value. The average reduction of Ct value
as shown in Figure 7 with chine 30 0 to 350 was about 9.75%, 250 to 300 was about 2.9%, and
250 to 350 was about 12.3%. The frictional resistance acting on a hull form for experimental
refers to equation 4, while in CFX-Post could be calculated by performing an area integral of
the wall shear in the x-direction. In Figure 9 shows Cf graph has similar trend to Ct, where the
lowest Cf value generated by chine 350 as well. While the greatest value of residual resistance
(Cr) was shown by chine hull form 350 at R/L: 1/14 and R/L: 1/7. It means this model has
bigger wave resistance component than the other forms.
The Ct graph of Wigley hull form appeared hump and hollow phenomenon, while the chine
hull form there is no significantly increase or decrease of Ct in Fr variation. The drag
reduction of Wigley pentamaran to chine hull form was represented by the equation (14).
Percentage of drag reduction results of 6 chine hull forms was shown in Figure 10. While the
maximum drag reduction successively to chine form set out in Table 3 below.

DR ( )=| Ct wigleyCt chine

Ct wigley |
Table 3. Maximum drag reduction
25deg 25deg 30deg 30deg 35deg 35deg
chine
R/L 1/16 R/L 1/33 R/L 1/8 R/L 1/5 R/L 1/14 R/L 1/7
DR (%) 18.05 17.80 16.10 14.76 18.02 18.19

-3
4.5x10
Total R esistance C oefcient (C t)

Ct-Exp
-3
4.0x10 Ct-CFD

-3
3.5x10

-3
3.0x10

-3
2.5x10

-3
2.0x10
0.1 0.2 0.3 0.4 0.5 0.6 0.7
Froude Number (Fr)

Figure 6. Comparison experimental and CFD on Ct of Wigley hull form

(14)
 25deg R/L 1/16
-3
25deg R/L 1/33
Total R esistance C oefcient (Ct)
3.4x10 3.00x10
-3
30deg R/L 1/8
30deg R/L 1/5
-3 35deg R/L 1/14

F rictional R esistance (C f)
-3
3.2x10 2.75x10 35deg R/L 1/7

-3
-3 2.50x10
3.0x10
-3
-3
2.25x10
2.8x10
25deg R/L 1/16 2.00x10
-3

-3 25deg R/L 1/33

2.6x10
30deg R/L 1/8 -3
30deg R/L 1/5 1.75x10
2.4x10
-3 35deg R/L 1/14
35deg R/L 1/7 1.50x10
-3

0,1 0,2 0,3 0,4 0,5 0,6 0,7 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Froude Number (Fr) Froude Number (Fr)

Figure 7. Result on total resistance coeff. (Ct) Figure 8. Result on friction resist.coeff. (Cf)
25deg R/L 1/16
25deg R/L 1/16 20 25deg R/L 1/33
25deg R/L 1/33 30deg R/L 1/8
30deg R/L 1/5
30deg R/L 1/8
-3
1.05x10 35deg R/L 1/14
30deg R/L 1/5 16 35deg R/L 1/7
35deg R/L 1/14
Drag Reduction (%)
Residual Resistance (Cr)

-4 35deg R/L 1/7

9.00x10
12
-4
7.50x10

8
-4
6.00x10

-4 4
4.50x10

-4
3.00x10 0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Froude Number (Fr) Froude Number (Fr)

Figure 9. Result on residual resist.coeff. (Cr) Figure 10. Percentage of drag reduction
The contour of wave volume fraction from CFD simulation at Fr 0.7 for Wigley and chine
forms were shown in Figure 11. There are differences contour of wave volume fraction for
Wigley and each chine, which blue color shows bigger wave value than the dark blue. It seems
that the hull form of Wigley generates bigger wave effect than chine.

Wigley Contours on z/Lpp = 0.035, Fn = 0.7 Chine 250 Contours on z/Lpp = 0.03, Fn = 0.7
 Chine 300 Contours on z/Lpp = 0.025, Fn = 0.7 Chine 350 Contours on z/Lpp = 0.02, Fn = 0.7

Fig 11. Free surface elevation (reverse side), z/Lpp, of global wave pattern

4. CONCLUSION
The comparison of CFD analysis between Wigley to chine form with inner and outer
asymmetric hull showed the effect of increasing deadrise angle was improving in drag
reduction. The highest drag reduction of chine 35 0 due to the big deadrise angle would be
causing unconstructed waves angle between each hull. Some advantages chine hull form
compared to Wigley are:
Ct value tends to be smaller on higher L/B with constant B/T: and H/T;
Increasing Ct value was not fluctuated with increasing Fr;
The greater deadrise made the smaller Ct;
Drug reduction could reach 18.19 % at the largest angle 350 on Fr 0.2.
The implication for further research acknowledged needs to investigate the effect of the angle
of entrance and stem angle of hull form to the resistance characteristic.

5. ACKNOWLEDGEMENT
This work has been nancially supported by Lembaga Pengelola Dana Pendidikan (LPDP)
BUDI DN RISTEK DIKTI. The authors would like to acknowledge the support of pentamaran
team of Department of Mechanical Engineering, University of Indonesia.
6. REFERENCES
Bari, Ghazi S, and Konstantin I Matveev, (2016), Hydrodynamic modeling of planing
catamarans with symmetric hulls. Ocean Engineering 115:60-66.
Begovic, E, and C Bertorello, (2012), Resistance assessment of warped hull form. Ocean
Engineering 56:28-42.
Blazek J. 2001. Computational Fluid Dynamics: Principles and Applications. Elsevier Science
Ltd, Oxford England.
Blount, Donald L, (1995), Factors Influencing the Selection of a Hard Chine or Round-Bilge
Hull for High Froude Numbers
Chengyi, Wang, (1994), Resistance charicteristic of high-speed catamaran and its application.
Shipbuilding of China 3:003
Ferziger JH, and Peric M., (2002), Computational Methods for Fluid Dynamics, 3rd ed.,
SpringerVerlag, pp. 157206.
Hsiung. C.C. and Xu. H, (1988), Determining Optimal Forms of a Catamaran for Minimum
Resistance by the Mat hematical Programming Met hodo'. Schiffstechnik Bd.35.
Ikeda, Yoshiho, Emiko Nakabayashi, and Ai Ito, (2005), Concept design of a pentamaran type
fast RoRo ship. Journal of the Japan Society of Naval Architects and Ocean Engineers
1:35-42.
Menter, F.R., 1994. Two-equation eddy-viscosity turbulence models for engineering
applications. AIAA Journal 32(8):1598 605.
Michell, J.H., (1898), The Wave-Resistance of a Ship, Philosophical Magazine, Series 5.Vol.
45, Xo. 272, London, January, pp. 105-1 23.
Moraes, HB, JM Vasconcellos, and RG Latorre, (2004), Wave resistance for high-speed
catamarans. Ocean Engineering 31(17):2253-2282.
Oller, Erik, Vasilios Nikou, and Konstantinos Psallidas, (2003), Focused Mission High Speed
Combatant. DTIC Document.
Peng, Hongxuan, (2001), Numerical computation of multi-hull ship resistance and motion,
Dalhousie University Halifax.
Recommended Procedure and Guidelines, (2002), Testing and Extrapolation Methods in
Resistance Towing Tank Tests, ITTC 7.5-02-02-02.
Seo, J.W, Seol, D.M, Lee, J.H, and Rhee, S.H, (2010), Flexible CFD meshing strategy for
prediction of ship resistance and propulsion performance. Inter J Nav Archit Oc Engng, 2,
pp.139-14.
Tarafder, Md Shahjada, Mir Tareque Ali, and Md Shahriar Nizam, (2013), Numerical
prediction of wave-making resistance of pentamaran in unbounded water using a surface
panel method. Procedia Engineering 56:287-296
Versteeg, Henk Kaarle; Malalasekera, Weeratunge., (2007), An introduction to Computational
Fluid Dynamics: The Finite Volume Method. Pearson.
Wilcox, D.C., (2006) Turbulence Modelling for CFD, Third Edition, DCW Industries.
Yanuar, Gunawan, Kurniawan T. Waskito, and A. Jamaluddin, 2015. Experimental Study
Resistances of Asymmetrical Pentamaran Model with Separation and Staggered Hull
Variation of Inner Side-Hulls, International Journal of Fluid Mechanics Research, Vol. 42,
No. 1.
Yanuar, Ibadurrahman, Kurniawan T. Waskito, S. Karim, and M. Ichsan, 2017. Interference
Resistance of Pentamaran Ship Model with Asymmetric Outrigger Configuration, Journal
of Marine Science and Application, 16:42-47