Design of Hydrofoil Using

Design oI HydroIoil Assisted Catamarans using a .
Jol. 1 / No. 2 / Winter 2005 1

JOURNAL oI MARINE ENGINEERING
Design of HydrofoiI Assisted Catamarans
using a Non-Linear Vortex Lattice Method
Gnther Migeotte
1
, Nikolai Kornev
2
, Karl Gnter Hoppe
3
,
Anna Nesterova
4
1,3-Dept. oI Mechanical Engineering, University oI Stellenbosch
2,4-Marine Technical University oI St. Petersburg
Abstract
Extensive experimental investigations oI hydroIoil-assisted catamarans have identiIied the main
hydrodynamic parameters aIIecting the perIormance oI these vessels. Design principles oI existing vessels
are discussed in the light oI these hydrodynamic Iindings. The experience gained Irom experimental testing
has provided the basis Ior the development oI a non-linear vortex lattice method Ior design purposes. The
method is validated by comparison with experimental results and applied to the problem oI optimizing a 40-
knot, 40m hydroIoil-assisted catamaran with a hydroIoil system developed Ior modern high-speed
catamarans. It is shown that the method can be successIully applied to optimization problems oI these vessels
as it successIully captures the complex hull-Ioil interactions. Results oI the numerical investigation
perIormed, indicate that optimizing the mutual distance between the hydroIoils can signiIicantly improve the
liIt to drag ratio oI such a vessel.
Keywords:
Introduction
The development oI hydroIoil assisted
catamarans started in the late 70`s with
developments in the Soviet Union and in
the early 80`s with developments in the
USA and South AIrica. Early
developments Iocused on the design oI
smaller planing type catamarans with
asymmetric hulls using up to three Ioils
mounted in tandem between the demi-
hulls. Over the last decade, the
application oI hydroIoil assistance has
extended to include the more modern
high-speed semi-displacement type
catamarans used mainly Ior Iast Ierry
applications. While the application oI
small hydroIoils Ior these vessels is well
known and popular Ior sea-keeping
improvements, the use oI larger hydroIoils
Ior improvements in resistance is less well
known and is the Iocus oI this paper.
Since the early 80`s, the University oI
Stellenbosch has been involved with
development oI hydroIoil-assisted
catamarans oI sizes ranging Irom 5m to
40m. A practical and successIul Ioil
system was patented and named
'Hysucat, Hoppe (1982). Hysucats were
developed through systematical towing
tank tests and planing craIt theory. Later a
mathematical model was developed using
Savitsky`s (1964) Iormulations and
hydroIoil theory as applied to hydroIoil
craIt, Hoppe (1991, 1995). Good results
were achieved and later Hysucats were
designed and optimized by use oI the
mathematical model alone without the
need Ior model tests. The experience built
up in the design oI Hysucats has more
recently applied to catamarans with semi-
displacement hull Iorms. A new, patented
Jol. 1 / No. 2 / Winter 2005 2
JOURNAL of MARINE ENGINEERING
Iranian Association oI
Naval Architecture & Marine Engineering
Ioil system named 'Hysuwac, Hoppe
(1998), was speciIically developed Ior
larger semi-displacement type catamarans.
Application oI Hysuwac and other Ioil
systems to semi-displacement type
catamarans is still relatively new, with
only a hand Iull oI vessels that have been
built worldwide. This paper describes
some oI the hydrodynamic design aspects
oI hydroIoils Ior modern high-speed
catamarans based on experimental results,
prototype data and numerical calculations
carried out on Hysuwac and other
hydroIoil conIigurations.
Hydrodynamic Principles
A hydroIoil-assisted catamaran
achieves superior perIormance over a
conventional catamaran by unloading the
hull and liIting it partially out oI the water.
This has two major inIluences on
resistance: reduced Iriction resistance due
to a decreased wetted area and secondly,
reduced hull wave making. In addition to
the positive eIIects introduced by the Ioil
assistance, one has to consider the added
resistance oI the Ioils and the interIerence
resistance coming Irom hull-Ioil and Ioil-
Ioil interactions. Figure 1 shows a
resistance breakdown Ior a hydroIoil-
assisted catamaran. The Iigure attempts to
breakdown the resistance components into
their well-known parts and indicate how
the diIIerent components interact with
each other. A detailed explanation oI
these is given in Migeotte and Hoppe
(1999).
TotaI, C
7
Friction, Cf Residuary, Cr
HuII,
Rn|10
HuII
HydrofoiI,
Rn|10
HydrofoiI
Viscous
Interference
Wave
Interference
Free Surface Effects
Drag due to
Lift
Wetted Area
and L
:/
reduction
Wave-
making
Separation
and Spray
wave-making
FoiI, Cf HuII, Cf HuII, Cr FoiI, Cr
+XOO
8QORDGLQJ
Figure 1 - Resistance Components for Hydrofoil Assisted Vessels
Jol. 1 / No. 2 / Winter 2005 3
Due to the hull being constantly unloaded
with increasing speed, determining the
resistance components is not an easy task.
The foil resistance components are the
easiest to calculate, as Ioil theory is well
developed. There is some measure oI
uncertainty in all calculations due to hull-
Ioil interIerence eIIects. These are mainly
wave-making interIerence, disturbed
inIlow conditions to the Ioils and the end-
plate` eIIect oI the hull on the Ioils.
Hoerner (1965) and Tao (2000) give some
equations on how to estimate these
empirically. Alternatively, one can apply
CFD methods to determine these eIIects.
Determining the hull resistance
components is diIIicult without the use oI
model tests. An important point to note is
that the hull oI a Ioil-assisted catamaran
diIIers Irom a conventional catamaran
hull, in that being constantly unloaded, it
has no Iixed displacement, so one is not
able to rely on catamaran model test
databases to predict perIormance.
Unloading the hull by eIIects the two most
important parameters governing hull
resistance:
- The hull length displacement
ratio, L/V
1/3
, where V is the
volumetric displacement oI
the submerged part oI the
hull.
- The hull Froude
displacement number, Fn
V
,
again based on the load
Iraction oI the hull.
Values oI both parameters are lower Ior
the hulls alone than that Ior the complete
vessel. More oIten than not, this means
that the available experimental results Ior
high-speed catamarans is not applicable as
either oI these two parameters Ialls out oI
the range oI data. The Ioils additionally
have a signiIicant eIIect on the pressure oI
the hull. Depending on the Ioil
conIiguration these eIIects may be positive
or negative. II the hull is in close
proximity to the suction side oI the Ioil,
(Ior example a Ioil located slightly below
the keel) the hull will suIIer Irom the Ioil`s
downwash and pressures will be lower
than normal in that area. The inverse
obviously also applies: the Ioil can be
positioned so that the hull beneIits Irom
the up wash oI the Ioils. A typical
example is a Ioil mounted in the tunnel
between the demi-hulls. The vortex wake
oI the Ioil will induce up wash on the
sections oI the hull aIt oI the Ioil. Such
dynamic eIIects can have important
inIluence on the behavior and resistance oI
the vessel.
Analysis oI the large body oI experimental
data collected on hydroIoil-assisted
catamarans at the University oI
Stellenbosch has allowed certain
tendencies to be established that have
proven useIul in the design process.
Hydrodynamically, one can identiIy three
phases oI operation oI hydroIoil-assisted
catamarans (Figure 2): the
displacement/semi-displacement phase,
the transition Irom semi-displacement to
planing and Iinally the planing phase.
Jol. 1 / No. 2 / Winter 2005 4
Figure 2 - Hydrodynamic Phases of hydrofoil Assisted Catamarans
At present hydroIoil-assistance does
not make sense Ior vessels operating
below the displacement hump (Fn
V
- 1.5).
Since the weight oI a craIt is proportional
to the cube oI a linear dimension and the
liIt to the square oI the linear dimension,
the required size Ior a Ioil system out
grows practical sizes Ior these speeds. To
the knowledge oI the authors, Fn
V
1.5 is
currently the lower limit Ior which
hydroIoil assistance has been designed to
provide a useIul improvement in
resistance (Hoppe and Migeotte (2000)).
The transition phase (approx.
2.0Fn
V
3.0) is the most complex
hydrodynamically. The increasingly
dominant hydrodynamic Iorces (as
opposed to hydrostatic Iorces dominating
at lower speeds) acting on the hull cause
the hull to suck itselI down, opposing the
Ioil liIt. It is common Ior non-Ioil-assisted
high-speed catamarans to submerge
slightly with speed, as the hydrodynamic
suction Iorces on the hull increase. The
sudden disappearance oI these suction
Iorces on the hull signiIying the beginning
oI true planing are the cause oI the large
and sudden reductions in resistance in the
transition phase. Existing hydroIoil
assisted catamarans (a review oI existing
vessels is given in Migeotte and Hoppe
(1999) and Hoppe (1991, 1995, 1999)
operating in this range try and avoid these
sudden changes due to suction by either
making the hulls very slender (e.g. the
Hitachi 'Superjet catamarans, Miyata
(1989)) or designing them so that they
develop mostly positive dynamic liIt (e.g.
Kvaerner 'FoilCat, Jorde (1991)).
Experimental investigation oI various Ioil
positions is usually needed to determine
the optimal Ioil design to minimize these
eIIects.
True planing speeds (approx. Fn
V
~3.0)
are the easiest to design Ior, as both hull
and Ioils operate eIIiciently in this speed
range and positive interIerence eIIects can
more easily be utilized in the design.
Jol. 1 / No. 2 / Winter 2005 5
Simple planing hull lines can be utilized
with smaller Ioils mounted out oI harms
way in the tunnel between the demi-hulls.
Hydrofoil Assisted Catamaran
Designs
Examining existing vessels provides
some example oI the various options that
are available. Figure 3 summarizes the
established Ioil conIigurations currently in
use on existing high-speed semi-
displacement vessels:
Avion or
Hysucat
Mono-Foil
Tandem
Canards with
Interceptors
Standard Canard
Hysuwac
Figure 3 - Foil Configurations for Hydrofoil Assisted Catamarans
The tandem and standard canard
conIigurations are the most popular Ior
semi-displacement type craIt, the bulk oI
existing craIt making use oI these two
systems. This is mainly due to their large
liIting capability allowing the complete
vessel to be liIted out oI the water iI
needed. This makes the system less
sensitive to LCG shiIts and overloading.
The avion system has Iound application on
a large number (200) oI planing
catamarans ranging Irom 5 to 20m but
limited application to semi-displacement
type vessels (one 36m vessel) due to its
limited liIt capability at semi-displacement
speeds. Similarly, the simpler mono-Ioil
has also only seen one successIul
application Ior a 40m semi-displacement
type vessel. The canard only system,
usually coupled with a transom
interceptor, has seen two applications both
designed by the patent owners, MTD-St.
Petersburg Ior 30m high-speed planing
Ierries. The Hysuwac system to date has
been applied to three vessels. The
Hysuwac system utilizes a high aspect
ratio Iront Ioil together with a rear Ioil
mounted in the tunnel. The system has
been applied to vessels ranging Irom 20m
to 40m and successIul designs completed
Ior vessels up to 72m.
One can presently distinguish between
two kinds oI vessel designs: those oI new
vessels with hull and Ioils optimized to
match and hydroIoil retroIits to existing
catamarans, where Ioils are matched to an
existing hull. The design oI new vessels
allows the designer more Ireedom in
avoiding potential problems oI
Jol. 1 / No. 2 / Winter 2005 6
interIerence, making the design somewhat
simpler iI one has experience in these
eIIects. Positive eIIects such as wave
cancellation and taking advantage oI the
up wash Irom the hydroIoils to increase
the hull liIt can be eIIectively used to
minimize resistance. Utilizing these
eIIects allows one to achieve the resistance
tendencies considered optimum as
indicated in Figure 2.
HydroIoil retroIits tend to be more
diIIicult because a hull not designed to be
dynamically unloaded risks being unstable
in some way (directional stability being
the most common problem). Experience
with Hysuwac type retroIits to existing
hulls has shown that the main diIIiculty
lies in achieving suitable transition phase
resistance and behavior. It is also more
diIIicult to take advantage oI positive
interIerence eIIects between hull and Ioils
as the Ioil design is dictated more by
Iactors such as stability and structural
requirements. HydroIoil retroIits, while
not optimal, can nevertheless signiIicantly
improve the vessel eIIiciency and speed.
This has been proven on Hysuwac Ioil
retroIits to commercial craIt perIormed by
the University oI Stellenbosch.
Worldwide, initial development and
design oI new hydroIoil assisted semi-
displacement catamarans has relied
heavily on experimental model testing in
the design process. Such experiments are
invaluable Ior identiIying the main
hydrodynamic problems and Iinding
suitable solutions Ior these. A number oI
publications, based on experimental
investigations, are available that provide
some experimental data (see Ior example
(Hoppe (1984), Miyata (1990), Shimizu
(1994), Tao (1998, 2000), Migeotte and
Hoppe (1999)) useIul Ior design purposes.
From such experimental investigations,
enough experience has been built up at the
University oI Stellenbosch making it
possible, with the aid oI the CFD
knowledge oI the Marine Technical
University oI St. Petersburg, to develop a
suitable computational method to aid in
the design process and reduce the required
towing tank time. As an example oI the
application and useIulness oI the method,
the design oI Hysuwac hydroIoils Ior a
benchmark size passenger catamaran will
be considered. Such a vessel has typical
dimensions as given in Table 1. This
design was developed by use oI the
mathematical model Ior Hysucats, Hoppe
(1991, 1995) and systematical towing tank
tests at the University oI Stellenbosch.
Table 1: Catamaran Main Particulars
Overall Length 40 m
Displacement 170 t
Installed Power 2 x 2200 kW
Speed w/o Ioils 36 knots
Speed with Ioils Approx. 45 knots
Application oI the method is described
Ior speeds in the upper end oI the
transition phase and higher. This covers
the design speeds oI the vessel with and
without Ioils. This requires one to
consider numerical methods that can
Jol. 1 / No. 2 / Winter 2005 7
adequately model semi-planing and
planing in combination with hydroIoils.
The theoretical method implemented in
the commercial package AUTOWING,
developed by Dr. Kornev (1998) at the
Marine Technical University oI St.
Petersburg, is particularly suited to Ior
these types oI calculations. Combined
with the expertise gained Irom
experimental work, the method was
Iurther developed Ior application to
hydroIoil-assisted catamarans.
Theoretical method
Hydrodynamically, the hydroIoil-
assisted catamaran at high speed can be
considered as consisting oI two parts: the
Iront Ioil and the aIt hull plus rear Ioil.
The parts are considered separately. The
Iront Ioil is calculated as a wing moving
under the undisturbed Iree surIace. The
hydrodynamic eIIects oI the aIt hull and
rear Ioil on the Iront Ioil are neglected.
This has been Iound to be a valid
assumption iI the Iront Ioil is located
Iorward near the bow. The aIt hull and the
rear Ioil are treated as a wing system
moving in the vortex-wave wake shed
Irom the Iront Ioil.
Treatment of the free surface
problems
To calculate the wave surIace and
vortex wake aIter the Iront hydroIoil, a
three-dimensional Nonlinear Vortex
Lattice Method (NLVM) was developed.
The intensity oI the discrete vortices on
the Iree surIace is Iound Irom the Iree
surIace dynamic boundary conditions
(both linear and nonlinear) whereas the
Iorm oI the Iree surIace is sought Irom the
Iree surIace kinematic boundary
conditions.
For a short description oI the NLVM
application to Iree surIace problems, we
consider a hydroIoil advancing at constant
Iorward speed V
in an incompressible,
inviscid and irrotational Iluid. We use a
Cartesian coordinate system Iixed to the
Ioil; x points Iorward, y upward and z in
transverse direction. The Iree surIace is
modeled by a vortex sheet with unknown
intensity, J
. Without loss oI generality,

we can decompose J
at each point on the

Iree surIace within the computational
domain as the sum oI two components
tangential to the Iree surIace: J
,
is the
component perpendicular to the x-axis, J
is perpendicular to J
,
. The governing
equation Ior the perpendicular component,
J
,
can be obtained Irom the dynamic
boundary condition (see Kornev and
Taranov (1998)):
ox
v
v
Fn
y
v n v 2 2
4
1
) (
2
2
0
2
0
+ + + = J J J
]

(1)
Where,
= J / J J ,
= J v v /
0 0
, the
wave ordinate L y y
v v
/ = ,
gL J Fn /
= are non-dimensionalized
with respect to speed oI motion, V
and
wing chord length, L. The velocity on the
Iree surIace,
0
v
induced by the Iront Ioil

and the vortex sheet J
can be calculated
using Biot-Savart's law. Together with the
Jol. 1 / No. 2 / Winter 2005 8
zero-divergence condition and conditions at inIinity,
, , 0 , , 0 = V : for J J
0 ) , ( , 0 ) , ( = + = + : : y
v
J
(2)
(1) represents a complete system oI
governing equations Ior the vector
intensity, J
.
In the numerical implementation, the
Iree surIace is modeled within a rectangle
deIined by x
0
, x
1
, z
0
and z
1
. As usual in the
vortex lattice method, the surIace Vorticity
J
is represented by a number oI closed

vortices (see Fig. 4). Thus, the Iirst
equation in Eq. (2) is satisIied
automatically in the integral sense. For the
panelization oI the Iree surIace along the
x-axis two procedures were investigated: a
procedure proposed by Thiart (1997) and
simple exponential stretching
})| ) ( 1 . 0 exp 1 ( 1 |
2
1 1
X x N x x x
f f f
+ A =

(3)
where X is the abscissa oI the leading
edge oI the hydroIoil and Ax and N are
Iree parameters. Surprisingly we have
obtained almost the same results Ior both
panelizations. In the lateral direction the
wave surIace is divided into panels
uniIormly between struts and the simple
stretching (3) is used outside oI ship`s
width.
The vector equation (1) is satisIied at
the center oI each vortex lattice. The
vector intensity, J
, necessary Ior
calculations oI the RHS oI Eq (1) is sought
by smoothing the discrete vortices
according to simple Iormula:
+ + I I + I I =
+
) /( | ) ( ) |(
10 01
11
11
1
00
00
1
m m
m
m
m
if
m
if
m
m
m
if
m
if
m
R R
R
R
R
R

J
) /( | ) ( ) |(
11 00
01
01
1
10
10
1
m m
m
m
m
f i
m
if
m
m
m
if
m
f i
R R
R
R
R
R

+ I I + I I
+
(4)
Where:
m
if
I is the strength oI ij-th discrete
vortex and m is the number oI iterations.
As soon as the vortex intensity,
1 + m
J
is
Iound Irom Eq. (1), the continuous vortex
sheet is redistributed among the discrete
closed vortices
2 / ) (
10 01
1 1
1
1 m m m m
if
m
if
R R

+ + I = I
+ +
+
J (5)
The kinematic condition on the Iree
surIace is used to calculate the shape oI
the Iree surIace and to adjust the height oI
the vortex lattice. More precisely the Iorm
Jol. 1 / No. 2 / Winter 2005 9
oI the Iree surIace is determined Irom the solution oI the streamline equations
o: oy x
v
d:
v
dy
v
dx
= =
0
(6)
The velocity at each knot oI the vortex
lattice are Iound using the rule
(designations see in the Fig.4)
4 / ) (
+ + + +
+ + + = J J J J W

To separate the vectors
]
J
and
[
J
, as it
requires the numerical implementation oI
the Eq. (1), the vortex lattice on the Iree
surIace should satisIy the Iollowing
conditions:
the lateral sections oI every closed
vortex lie in the plane xconst, (7a)
the longitudinal sections lie in the plane
zconst. (7b)
Certainly aIter solution oI the streamline
equations we obtain the lattice violating
the condition (7b). ThereIore, a special
procedure is used to build a necessary
orthogonal grid on the base oI calculated
one at each iteration step. This algorithm
works only iI the waves do not break.
Both J
and y
w
are assumed zero at both
side boundaries oI the modeled Iree
surIace region.
Figure 4 - Vortex Lattice of the free surface and vortex wake behind the hydrofoil
Calculation of the hydrofoil
The hydroIoil and its vortex wake are
treated using the common NVLM method
described comprehensively in many
textbooks and special papers. The
numerical instability, which is typical Ior
problems including the dynamics oI thin
vortex sheets, was avoided using the
concept oI the cut-oII radius. The
thickness oI hydroIoil is accounted Ior by
Jol. 1 / No. 2 / Winter 2005 10
a source distribution with strength equal to
the thickness gradient in downstream
direction.
The calculation oI the hydroIoil under
the Iree surIace is perIormed using a
special iterative technique. Even iI linear
boundary conditions are used, the iterative
technique is necessary because the
position oI the vortex wake is a very
important Iactor aIIecting the wave
deIormation and it can only be Iound
iteratively.Divergence oI the iterative
technique was observed in three Iollowing
cases:
- Wave breaking
- Very long computational
domain (about 50 chords)
- Froude number is less than
0.8
The method Ior modeling the Iree
surIace was successIully tested by
comparisons with measurements Ior
various hydroIoils (see Kornev and
Taranov (1998)).
Calculation of the planing hull
The calculation oI the hull is done by
assuming planing conditions. This makes
the method valid Ior the upper end oI the
transition phase and higher as given in
Figure 2. The planing hull moving in the
vortex wave wake oI the Iront Ioil is based
on the wing analogy discovered by
Wagner (1932). Each liIting surIace is
represented via a set oI thin cambered
longitudinal strips with rectilinear leading
and trailing edges. Such an approximation
allows one to model planing surIaces and
hydroIoils with curved leading edges,
including warped hulls with variable
deadrise. II the wetted area oI the planing
surIace is known, the pressure distribution
and Iorces are obtained by using the
conventional Vortex Lattice Method
(VLM).
A special technique, which in its idea is
very similar to the simpliIied method
described by Zhao and Faltinsen (1996)
and Mei et al. (1999) Ior two dimensions,
is applied to estimate the wetted area oI a
three dimensional planing surIace with an
arbitrary length to beam ratio. This
method is rather simple and can be written
in a general Iorm that covers complex
planing geometry moving in the vortex-
wave wake. The essential points oI this
approach are described shortly below.
Consider a longitudinal strip oI a
planing surIace (see Fig.5). The wetted
chords oI strip are calculated iteratively. In
the Iirst iteration, the wetted chord is
determined as a length measured Irom the
trailing edge to the point oI intersection
between bottom contour and the Iree
surIace either not disturbed or disturbed by
the Iront Ioil. The Iree surIace deIormation
caused by the planing surIace itselI can be
written in the n-th iteration in the Iorm oI
the streamline equation
}

+ + =
n
x
n
x
n
y
n
d v J v y y
0
)) ( /( ) ( ) ( [ [ [
(8)
where v
x
and v
y
are components oI
velocity induced by the entire planing
surIace and hydroIoil in the n-th iteration.
They can be obtained by using the Biot-
Jol. 1 / No. 2 / Winter 2005 11
Savart law. In the iterations the leading
edge
n
o
x is shiIted towards the positive
direction oI the x-axis by adding A
x
along
the bottom contour until the intersection
point between the streamline Eq. (8) and
the bottom contour oI the hull (i.e. the
critical point) will be lay within the chord
x
0
x
int
x
1
. PerIorming calculations Ior all
sides and strips and connecting
corresponding points one obtains the
wetted area oI the planing surIace.
Obviously, the jet Ilow near the leading
edge and sides oI the planing hull is
neglected in this approach. It should also
be noted that the integral Eq. (8)
converges only in the three-dimensional
case.
Figure 5 -Vortex Lattice of the free surface and vortex wake behind the hydrofoil
Influence of the free surface
deformation on the hydrodynamic
forces of the catamaran
One oI the peculiarities oI the present
theory is consideration oI the Iree surIace
deIormation. The Iront Ioil disturbs the
Iree surIace, producing a wave whose
length is proportional to the Froude
number. A reasonable question is whether
the Iree surIace deIormation plays a
signiIicant role and whether it is possible
to neglect the Iree surIace eIIects saving
the computational time and simpliIying
the design process. Fig. 6 below allows
one to answer this question.
The results indicated as 'w/o
deIormation were obtained without
consideration oI the Iree surIace
deIormation. Both the planing hull and the
Iront Ioil move on and under the
undisturbed Iree surIace. As seen Irom the
Iigures, the theoretical results obtained
with account Ior the Iree surIace
deIormation converge to measurements
when the Froude number is increased. On
the contrary, a neglect oI the Iree surIace
eIIects, leads to a large discrepancy
between measurements and the numerical
prediction. The reason is clear. The larger
the Froude number the longer the wave
aIter the Iront Ioil. The wave trough
reaches the planing hull and, thereIore, the
inIluence oI the Iree surIace deIormation
becomes larger.
Jol. 1 / No. 2 / Winter 2005 12
Optimization of Hydrofoil
Assisted Catamarans
The developed method was used Ior the
investigation oI hydrodynamics oI a
Hysuwac type HydroIoil assisted
catamaran with particulars as given in
Table 1. The investigation was directed
at Iinding the optimal way to retroIit
existing catamarans with hydroIoils. Focus
was thereIore on investigating diIIerent
Ioils and the mutual position oI the Ioils
and the hull, while not changing the hull
geometry.
The numerical calculations showed that
among the geometric parameters
investigated, the most eIIective tool to
optimize the catamaran is the lateral
distance between the Ioils. The eIIect oI
other parameters was comparable with the
accuracy oI the numerical approach. The
calculation included both the
determination oI the hydrodynamic Iorces
and attitude oI the ship.
Figure 8 and 9 display results. Three
diIIerent cases, schematically sketched in
the Fig.7, are considered. Firstly, the
basic variant oI the Hysuwac arrangement
(Case 1), secondly, the same arrangement
but with the rear Ioil shiIted towards aIt by
3 m (Case 2). Finally, the arrangement oI
the Hysuwac with hull and the rear Ioil
shiIted aIt by 7 meters with respect to the
Iront Ioil or, in other words, the Iront Ioil
shiIted Iorward towards the bow by 7 m
(Case 3). A clear tendency emerged: that
increasing the distance between Ioils
results in an improvement oI the L/D ratio.
The results presented in the Fig.8 were
obtained without considering wave
resistance. As seen in the Iigure, the best
case is the Case 2, Iollowed by Case 3 and
then the Case 1.
2.10 2.20 2.30 2.40 2.50 2.G0 2.70
)Q
145.0
150.0
155.0
1G0.0
1G5.0
170.0
175.0
180.0
185.0
190.0
/
L
I
W
W
R
Q
([SHULPHQW
ZLWK GHIRUPDWLRQ
ZR GHIRUPDWLRQ
2.10 2.20 2.30 2.10 2.50 2.0 2.0
)Q
9.0
9.5
10.0
10.5
11.0
11.5
12.0
12.5
13.0
5
H
V
L
V
W
D
Q
F
H
W
R
Q
([SHULPHQW
ZLWK GHIRUPDWLRQ
ZR GHIRUPDWLRQ
Figure 6- Lift and resistance versus the Froude number Fn.
Jol. 1 / No. 2 / Winter 2005 13
Figure 7- Schematic sketches of the arrangement in three cases
.O 2.O B.O 4.O 5.O 6.O 7.O
CC measured from the transom (m)
4.5
5.O
5.5
6.O
6.5
7.O
L
f
D
basic variant
The rear foil shifted towards aft by 3 m The rear foil and hull shifted aft by 7 m
Figure 8 - The L/D ratio versus the position of the center of gravity for three cases . The wave
resistance is not taken into account. Diamonds- 170 t, bars- 180 t, circles- 190 t, stars- 200 t
Jol. 1 / No. 2 / Winter 2005 14
11.0 12.0 13.0 14.0 15.0 16.0 17.0
&* PHDVXUHG IURP WKH WUDQVRP P
12.0
12.5
13.0
13.5
14.0
14.5
/
'
The rear foil shifted towards aft by 3 m
basic variant
The rear foil and hull shifted aft by 7 m
Figure 9 - The L/D ratio versus the position of the center of gravity for three cases . The wave resistance
is taken into account. Diamonds- 170 t, bars- 180 t, circles- 190 t, stars- 200 t
To explain this result, consider the
Table 2 and Figures 10-12 illustrating the
three results obtained Ior the center oI
gravity position at the point securing the
best L/D ratio Ior given weight: 170 ton at
a speed oI 40 knots. Table 2 gives the
results Ior Cases 2 and 3 as percentage
values oI Case 1, which is considered to
be 100.
Table 2: Calculated results for Cases 2 and 3 in relation to Case 1
Basic
variant
The rear Ioil is shiIted
by 3 m towards aIt
Both the rear Ioil and the hull are
shiIted by 7 m towards aIt
Case 1 Case 2 Case 3
Pitch angle 2.147 2.079 1.652
Submergence -0.476 m -0.54 m -0.704 m
LiIt coeII. oI the Iront
Ioil
100 83.1 92.5
L/D oI the Iront Ioil 100 98.6 63.8
LiIt coeII oI the rear
Ioil
100 106 107.4
L/D oI the rear Ioil 100 107 136.9
LiIt coeII. oI the hull
the rear Ioil
100 102 103.5
L/D oI the system
Planning hullrear Ioil
100 109 101.8
Wave Resistance ton 100 87 79.8
L/D oI the ship 100 105 104.3
Jol. 1 / No. 2 / Winter 2005 15
As seen Irom the Fig.12, the larger the
distance, between the Iront Ioil and the
rear Ioil, the larger the up wash induced by
waves. The reason is clear. In the Iirst case
the rear Ioil is located close to the wave
trough. The Ioil located behind the wave
trough utilizes the positive up wash
induced by waves (see Fig.11). That is
why the liIt coeIIicients oI the rear Ioil and
the rear liIting system (the rear Ioilhull)
are the maximal Ior Case 3 (see Table 2).
The L/D ratio oI the rear Ioil is also
maximal Ior Case 3. Because the liIt and
the L/D ratio are increased when we shiIt
the rear Ioil aIt, the L/D ratio is larger Ior
Cases 2 and 3 than Ior Case 1.
The increase in liIt oI the rear liIting
system leads to the ship rising Iurther out
oI the water (see Table 2). The wetted
area is decreased being minimal Ior Case 2
(see Fig.10). The increase oI liIt leads to a
decrease oI the pitch angle and
consequently the liIt oI the Iront Ioil (see
Table 2). ThereIore, the intensity oI the
tip vortex shed Irom the Iront Ioil is also
decreased, weakening the downwash
induced by the tip vortex on the rear Ioil
(see Fig.12). Again, it leads to an increase
oI the liIt on the rear liIting system.
Despite oI the Iact that the liIt decreases
with decreasing pitch angle, the liIt oI the
rear liIting system becomes larger. The
two eIIects increasing the liIt (i.e. less
downwash induced by the weaker tip
vortex and increased up wash induced by
wave crest) prove to be stronger eIIects.
Because the wetted area is maximal in
Case 3, this arrangement has the largest
Iraction oI the hull Iriction resistance.
This is why the L/D ratio oI Case 3 is less
than that oI Case 2. However, the positive
inIluence oI shiIting the Iront Ioil Iorward
resulting in higher liIt Ior the rear liIting
system proves to be suIIicient to improve
the L/D ratio oI Case 3 compared to that
oI Case 1. Accounting Ior the wave
resistance (see Fig. 9) makes the results
Ior the Cases 2 and 3 closer. The wave
resistance was calculated using Michel`s
theory as implemented in the program
Michlet 6.06 developed by Dr.Lazauskas
(2000). The wave induced by the Iront
Ioil is not considered in this approach.
Comparison oI computed results with
experiments Ior the basic variant showed
that the overall resistance prediction is
within 5 oI measured values.
Figure 10 - Wetted area of the catamaran hull for three cases.
Jol. 1 / No. 2 / Winter 2005 16
Fi
gure 11- The mutual positions of the rear foil and the wave surface at 40 knots.

span (n)
A
v
e
r
a
Q
e
O
O
c
w
n
w
a
s
h
*
2
C
&DVH
WRWDO
GXH WR
WKH WLS
YRUWH[
t
Figure 12 - The downwash (up wash if positive) averaged along the chord B.
The Iirst tendency observed by shiIting
only the rear Ioil aIt in Case 2 was clearly
conIirmed in the tests perIormed at the
University oI Stellenbosch. The second
tendency Iound in Case 3 (shiIting the
Iront Ioil towards bow) is rather an
unexpected one because, as a rule, an
increase in wetted area oI that magnitude
leads to an overall increase in resistance.
ThereIore, the result presented here, is
material Ior discussion and needs Iurther
experimental investigations.
Conclusions and future work
The main hydrodynamic design
principles oI hydroIoil-assisted catamarans
have been presented, and existing designs
reviewed. The importance oI the
Jol. 1 / No. 2 / Winter 2005 17
interactions between the Ioils and hull are
illustrated through calculations with the
proposed mathematical model. The
mathematical model provides an eIIective
basis Ior design oI hydroIoil-assisted
catamarans. The method was thoroughly
tested Ior the Hysuwac design based on
model test results. The model has no
restrictive Iactors, typical Ior instance Ior
the slender hull theory and other
approaches, and can be generalized Ior
small Froude numbers accounting Ior the
gravity eIIects.
A technical measure Ior optimization,
Iollowing Irom this purely numerical
investigation would be to increase the
distance between the rear Ioil and the Iront
Ioil so that it the rear liIting system
beneIits Irom the up wash oI the Iront Ioil.
Experimental tendencies validate the
result. Further experimental investigations
will be conducted to investigate variations
in Ioil placements Ior the Hysuwac
system. This data will be used to validate
and extend the theoretical method into a
versatile design tool Ior hydroIoil-assisted
catamarans.
References
HOERNER, S. (1965), Fluid Dynamic Drag, Published by Author, Bricktown, New Jersey.
HOPPE, K.G. (1992), Boat, HydroIoil supported Catamarans, S.A. Patent No. 82/3505 and Foreign Patents.
HOPPE, K.G. (1991), PerIormance Evaluation oI High Speed SurIace CraIt with ReIerence to the Hysucat
Development, Fast Ferry International, Vol. 30(No.s 1 & 3), Jan. & Apr.
HOPPE, K.G. (1995), Optimization oI HydroIoil Supported Planing Catamarans, FAST`95, Third
International ConIerence on Fast Sea Transportation, pp. 307-317.
HOPPE, K.G. (1998), S.A. Patent App. 98/3763 and U.S. Patent No. 6,.164,235 oI 26/12/2000.
HOPPE, K.G. (1999), HydroIoil Catamaran Developments in South AIrica, HIPER`99, First International
ConIerence on High PerIormance Marine Vehicles, Zewenwacht, South AIrica.
HOPPE, K.G.; MIGEOTTE, G. (2000), Model Tests on the Voyager II Foil Assisted Catamaran, ITM Report,
Dept. oI Mech. Engineering, University oI Stellenbosch. Not available to the public.
JORDE, J.H. (1991), The Development oI a 50 knots, 40m FoilCat, IMAS`91 High Speed Marine
Transportation ConIerence, pp. 79-84.
KORNEV, N.V. (1998), The Computational Method oI Vortex Elements and its Application to Hydro-
Aerodynamics, 2
nd
Doctor Thesis, Marine Tech. Univ. oI St. Petersburg, Dept. oI Hydromechanics. In
Russian.
KORNEV, N.V.; TARANOV, A., (1998), Investigation oI the Vortex-Wave Wake behind a
HydroIoil, Ship Technology Research 46, pp. 8-13.
LAZAUSKAS, L., (2000), Michlet 6.06, Ireeware on the Internet:
www.maths.adelaide.edu.au/Applied/llazausk/hydro/hydro.htm
MEI, X.; LUI, Y.; YUE, D.K.P. (1999), On the Water Impact oI General Two-Dimensional Sections, Applied
Ocean Research 21, pp 1-15.
MIGEOTTE, G.; HOPPE, K.G. (1999), Development oI hydroIoil Assisted Catamarans with Semi -
Displacement Hulls, FAST`99, FiIth International ConIerence on Fast Sea Transportation, pp. 631-642.
MIYATA, H., (1989), Development oI a New Type HydroIoil Catamaran, J. oI Ship Res. 33 (2).
SAVITSKY, D. (1964), Hydrodynamic Design oI Planing Hulls, Marine Technology, October, pp. 71-95.
SHIMIZU, K.; MASUYAMA, K.; YAGI, H. (1994), Hydrodynamic Design Philosophies oI the Hybrid
HydroIoil Catamaran, Techno-Ocean`94 Proceedings, pp. 265-270.
TAO, M. (1998), Development oI the Foil Augmented Wave Piercing Catamaran, RINA International
ConIerence oI Fast Freight Transportation by Sea.
Jol. 1 / No. 2 / Winter 2005 18
TAO, M. (2000), Resistance Prediction Method oI Foil Augmented Cat, HPMV2000, Third International
ConIerence on High PerIormance Marine Vehicles, pp. 172176.
THIART, G. (1997), Vortex Lattice Method Ior a Straight HydroIoil near a Free SurIace, Int ernational
Shipbuilding Progress, 44 (5), pp. 5-26.
WAGNER, H. (1932), ber Stoss-und Gleitvorgnge an der WasseroberIlche von Flsigkeiten, Z. Angew.
Math. Mech. 12(4), pp. 193-215.
ZHAO, R.; FALTINSEN, O.; AARSNES, J. (1996), Water Entry oI Two-Dimensional Sections with and
Without Flow Separation, Proceedings 21
st
Symposium on Naval Hydrodynamics.

Design of Hydrofoil Using

Transféré par

Informations du document

Description originale:

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Design of Hydrofoil Using

Transféré par

Droits d'auteur :

Formats disponibles

Design oI HydroIoil Assisted Catamarans using a .

Jol. 1 / No. 2 / Winter 2005 1

. Without loss oI generality,

at each point on the

induced by the Iront Ioil

is represented by a number oI closed

Vous aimerez peut-être aussi