Chalmers Formula Sailing

Chalmers Formula Sailing
High Performance Skiff

Marine Design Project 2017
Naval Architecture and Ocean Engineering International Master Programme
Department of Mechanics and Maritime Sciences

C HALMERS U NIVERSITY OF T ECHNOLOGY
Gothenburg, Sweden 2017
Acerbi, Tommaso
Andersson, Rasmus
Eriksson, Eric
Granli, Simon
Jacobs, Eike
Rita, Francisco
Sahlberg, Robert
Werner, Emanuel
Department of Mechanics and Maritime Sciences

Chalmers University of Technology
SE-412 96 Göteborg Sweden
Telephone +46(0)31-772 1000
Printed by Chalmers Reproservice
Gothenburg, Sweden, 2017
ii
Abstract
The 1001VELAcup is a sail race between European universities that takes place annually in
Italy. The current report addresses the design and construction of a high performance sailing
skiff to allow Chalmers University of Technology to participate in the race. The design is
regulated by the Class Rule R3 and the current skiff is mainly inspired by the Olympic class
49er©.
According to the class rules, the hull shall have a content of natural material of at least 70%.
This is achieved by using a sandwich material with a balsa core and a composite from flax (linen)
fibres, and an epoxy resin based on the shell of cashew nuts. Since there are no restrictions in
draft or height of mast, this skiff has a relatively high mast and deep centreboard. All parts
of the design are based on analyses of the race area in Mondello, outside of Palermo, in Sicily.
Thus, the team chose to not use hydrofoils on the centreboard and rudder. The sail area is
maximised according to the class rules.
The work is carried out by eight students on the master program Naval Architecture and Ocean
Engineering with supervision from teachers and experts from the industry.
Keywords: skiff, dinghy, sailing, 1001VELAcup, Chalmers University of Technology, CDIO
iii
Preface and Acknowledgement
The Marine Design Project, MMA 151, is a mandatory 15 credits course within the master
program Naval Architecture and Ocean Engineering at Chalmers University of Technology. The
course is organised by the Department of M2, Mechanics and Maritime Sciences at Chalmers.
The objective of this project is to design and build a high performance racing skiff in order to
compete in the upcoming 1001VELAcup 2018. Furthermore the participants have to manage
the public relation work in addition to the whole process.
The project members would like thank and acknowledge the help, funding, expertise and sup-
port of the following:
To Lars Larsson, Adam Persson, John Mcveagh, Per Wretlind, Henrik Ottoson, Per
Hogström and Rolf Eliasson, thank you for the guidance, knowledge and expertise through-
out the duration of the project.
To Per Hogström and Lars Larsson, thank you for providing us the opportunity to work
with such a unique project and for providing us with Ritsalen. This project would have
been significantly more difficult without it.
To our sponsors and partners in the project: Seldén Mast, North Sails, SSPA, the ITH
project supported by the EU regional fund, Chalmers Foundation "Fond för högskolans
bästa" and Area of Advanced Materials at Chalmers.
To Professor Kai Graf for supplying his VPP.
To SSPA, for general support and SNIC (Swedish National Infrastructure for Computing)
for computing resources
To our classmates, working alongside us and providing the good atmosphere and com-
panionship.
To Chalmers University of Technology.
Project Members
CAD Simon Granli

Francisco Rita
CFD Tommaso Acerbi

Robert Sahlberg
FEM Eric Eriksson

Emanuel Werner
VPP Rasmus Andersson

Eike Jacobs
v
Executive Summary
The following section is a brief outline of the work accomplished in the project. The mission
profile, main particulars for the skiff and a general overview of its rig and sail set up are presented
in this section.
Mission Profile
The mission of this project is to design and build a high performance racing skiff in order to
participate in the 1001VELAcup. The regatta will take place in the bay of Mondello, Italy, at
the end of September 2018.
Main Particulars
The dimensions of the skiff are set by the competition regulations as well as performance and
structural optimisation. Since the goal is to finish in 1st place in the competition the focus has
been on performance and hence the looks of the skiff are purely a result of that.
Type High Performance Racing Skiff

Flag Sweden
Class R3
Hull Dimensions Length over all 4.60 m
Beam of canoe body 1.22 m
Beam 2.10 m
Draft of canoe body 0.17 m
Total draft 1.70 m
Displacement 250 kg
Moment of Inertia of the water-plane area 1.664 · 10−1 m4
Rig Dimensions Mast height 9.50 m
Mainsail 16.0 m2
Jib 6.0 m2
Gennaker/Jib 0 11.00 m2
vi
vii
List of Abbreviations
Abbrevation - Description
CDIO - Conceive, Design, Implement and Operate

FLAT – The flattening of the main sail to empower it
GA - General arrangement
ISPRA - Istituto Superiore Per la Protezione e la Ricerca Ambientale
kn - Knots
MDF - Medium Density Fibreboard
NACA – National Advisory Committee for Aeronautics
ORC - Offshore Racing Congress
PDF - Probability Density Function
PLA - Polylactic acid
Pre-preg - Pre-impregnated
RANS - Reynolds-averaged-Navier-Stokes
RANS - Reynolds-Avergaed-Navier-Stokes
RSM - Reynolds Stress Method
RSM - Reynolds Stress Model
SF - Scale factor
s.f. - Safety factor
SST - Shear Stress Transport
VMG – Velocity Made Good
VOF - Volume of Fluid
VOF - Volume of Fluid
VPP – Velocity Prediction Program
viii
List of Varables
Variable Unit Description
A [m2 ] Planform area

a [m] Distance between forces
AR [−] Aspect ratio
AW A [◦ ] Apparent wind angle
AW S [m/s] Apparent wind speed
b [m] Distance between forces
B [m] Beam
Bmax, canoe body [m] Maximum beam of canoe body
BS [m/s] Boat speed
C̄ [m] Mean chord length
C1 [m] Upper chord length
C2 [m] Lower chord length
CB [−] Block Coefficient
CD [−] Drag coefficient
CD,2D [−] Profile drag coefficient
CDI [−] Induced drag coefficient
Cf [−] Skin friction coefficient
CL [−] Lift coefficient
CL,2D,1◦ [−] 2D lift coefficient at 1◦ angle of attack
CL,3D [−] 3D lift coefficient
CE [m] Centre of effort
CLR [m] Centre of effort of the appendages
CoG [m] Centre of gravity
ELT [M P a] Elastic modulus in the longitudinal-transverse plane
ET V [M P a] Elastic modulus in the transverse-vertical plane
EV L [M P a] Elastic modulus in the vertical-longitudinal plane
EL [M P a] Longitudinal elastic modulus
ET [M P a] Transverse elastic modulus
EV [M P a] Vertical elastic modulus
FDI [N ] Induced drag force
Fx,aero [N ] Aerodynamic force in x-direction
Fx,hydro [N ] Hydrodynamic force in x-direction
Fy,aero [N ] Aerodynamic force in y-direction
Fy,hydro [N ] Hydrodynamic force in y-direction
Fz,maststep [N ] Force in z-direction at the mast step
FD [N ] Drag force
FL [N ] Lift force
Fx [N ] Force in x-direction
Fy [N ] Force in y-direction
Fz [N ] Force in z-direction
Fn [−] Froude Number
g [m3 /(kg · s)] Gravitational constant
GM [m] Metacentric height
Ix [m4 ] Moment of inertia of the cross section
ix
IxBS [m4 ] Moment of inertia of the cross section of the bowsprit
I49er [m4 ] Moment of inertia of the water-plane of the 49er©
IW P [m4 ] Moment of inertia of the water-plane area
kDY N [−] Dynamic factor
L [m] Length
l [m] Distance on racks
LBS [m] Length of the bowsprit that is unsupported
LCF [m] Longitudinal Center Flotation
LCG [m] Longitudinal Center of Gravity
LOA [m] Length over all
LW L [m] Length of Water Line
mcrew [kg] Mass of the crew
mtot [kg] Total mass
Mx [N m] Moment in x-direction
My [N m] Moment in y-direction
Mz [N m] Moment in z-direction
P [N ] Point load applied at the racks
PBS [N ] Point load applied at the tip of the bowsprit
pDY N [N/m2 ] Dynamic pressure
pf w [N/m2 ] Flat water pressure
p0 [N/m2 ] Uniform pressure
ρBS [m] Density of the material composing the bowsprit
R1 [N ] Reaction force on the centreline
R2 [N ] Reaction force at the gunwale
rBS [m] radius of the bowsprit’s cross section
Re [−] Reynolds number
Rex [−] Reynolds number at distance x from the stagnation point
RM [N m] Righting moment
RM˜ [N m] Righting moment approximation
RMR [N m] Righting moment from the crew on the racks
RM˜ R [N m] Righting moment approximation from the crew on the racks
RMT [N m] Righting moment from the crew in the trapeze
RM˜ T [N m] Righting moment approximation from the crew in trapeze
σBS [P a] Stress at a bowsprit section
σu [P a] Ultimate stress
S12,max [M P a] Maximum shear stress in the XY-/LT-plane
S12,min [M P a] Minimum shear stress in the XY-/LT-plane
SL,max [M P a] Maximum stress in the longitudinal direction
SL,min [M P a] Minimum stress in the longitudinal direction
SL [M P a] Stress in the longitudinal direction
SLT,max [M P a] Maximum stress in the longitudinal-transverse plane
SLT [M P a] Stress in the longitudinal-transverse plane
ST,max [M P a] Maximum stress in the transverse direction
ST,min [M P a] Minimum stress in the transverse direction
ST [M P a] Stress in the transverse direction
ST V [M P a] Stress in the transverse-vertical plane
SV [M P a] Stress in the vertical direction
SV L [M P a] Stress in the vertical-longitudinal plane
SF [−] Safety factor
T [m] Draft
t [m] Thickness of the boundary layer
x
tBS [m] Thickness of the bowsprit
Tmax, canoe body [m] Maximum draft of canoe body
Tmax [m] Maximum draft
TK [−] Draft of the centreboard
T CG [m] Transverse centre of gravity
T CGcrew [m] Transverse centre of gravity of crew
TR [−] Taper ratio
TWA [◦ ] True wind angle
TWS [m/s] True wind speed
U∗ [m/s] Friction velocity
U∞ [m/s] Freestream velocity
U2,max [mm] Maximum deflection in the transverse direction
U2,no17 [mm] Deflection in the transverse direction of layup No. 17
U2,no18 [mm] Deflection in the transverse direction of layup No. 18
U3,i [mm] Deflection of node i in the vertical direction
U3,max [mm] Maximum deflection in the vertical direction
U3,no17 [mm] Deflection in the vertical direction of layup No. 17
U3,no18 [mm] Deflection in the vertical direction of layup No. 18
U1 [mm] Maximum deflection of 1st model
U2 [mm] Maximum deflection of 2nd model
U3 [mm] Maximum deflection of 3rd model
V [m/s] Speed
V CG [m] Vertical Centre of Gravity
xLOA [m] X-position of the maximum beam
y [mm] Distance to the nearest wall
y+ [−] Non-dimensional distance from the wall
zB [m] Z-position of centre of buoyancy
zG [m] Z-position of centre of gravity
zCE [m] Vertical centre of effort of the sails
zCLR [m] Vertical centre of effort of the appendages
α [◦ ] Leeway angle
∆ [kg] Displacement
∆S [m] Cell spacing
µ [kg/(m · s)] Dynamic viscosity of the fluid
ν [m2 /s] Kinematic viscosity of the fluid
νLT [M P a] Poisson’s ratio in the longitudinal-transverse plane
νT V [M P a] Poisson’s ratio in the transverse-vertical plane
νV L [M P a] Poisson’s ratio in the vertical-longitudinal plane
ρ [kg/m3 ] Density
τwall [M P a] Wall shear stresses
xi
Contents
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
Preface and Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
Mission Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
Main Particulars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
List of Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii
List of Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv
List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii
1 Introduction 1
2 1001VELAcup 3
2.1 Class Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2 The Race . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3 Design basis 7
3.1 Wind Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.2 References for Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.2.1 Hull . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.2.2 Rig and Sails . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.2.3 Appendages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2.4 Racks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.3 Overall Design Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4 Theory 17
4.1 Computer-Aided Design (CAD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.1.1 Hull Fairing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.2 Computional Fluid Dynamics (CFD) . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.2.1 Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.2.2 Governing Equations Solved . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.2.3 Turbulence Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.2.4 Numerical Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.2.5 Grid Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.2.6 Aerodynamic Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4.2.7 CFD Analysis Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4.3 Finite Element Method (FEM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.3.1 Prestudy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.3.2 Hull Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.3.3 Centreboard and Rudder . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
4.3.4 The Sandwich Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
4.3.5 FEM Analysis Set Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.4 Velocity Prediction Program (VPP) . . . . . . . . . . . . . . . . . . . . . . . . . 38
4.4.1 Input and output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.4.2 Solved equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.4.3 Centreboard and Rudder . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.4.4 Sail Set Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
xii
Contents
5 Design process 47
5.1 Hull . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.1.1 Hull Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.1.2 Hull Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
5.1.3 Hull Design Optimisation . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
5.1.4 Final Hull Geometry and Hydrodynamics . . . . . . . . . . . . . . . . . . 58
5.1.5 Deck . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
5.1.6 Structural Hull Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
5.2 Sail and Rig . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.2.1 Sail Set Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.2.2 Rig . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.3 Appendages and Racks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
5.3.1 Centreboard and rudder . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
5.3.2 Racks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
6 Balancing of Final Design 97
7 Building process 99
7.1 Templates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
7.2 Planking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
7.2.1 Planking the hull and deck . . . . . . . . . . . . . . . . . . . . . . . . . . 100
7.2.2 Planking the internal structure . . . . . . . . . . . . . . . . . . . . . . . . 101
7.3 Lamination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
7.3.1 Laminating the hull and deck . . . . . . . . . . . . . . . . . . . . . . . . . 102
7.3.2 Laminating the internal structure . . . . . . . . . . . . . . . . . . . . . . . 103
7.4 Further Building . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
7.4.1 Rudder and Centreboard . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
7.4.2 Racks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
7.4.3 Bowsprit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
8 Future Work 105

8.1 Remaining Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
8.1.1 Sails . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
8.1.2 Local Reinforcements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
8.1.3 Structure of Centreboard and Rudder . . . . . . . . . . . . . . . . . . . . 106
8.1.4 FE-Analysis of Racks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
8.1.5 Structure of Bowsprit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
8.1.6 Deck Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
8.2 Future Investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
8.2.1 Hull Fluid Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
8.2.2 Wing Sails . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
8.2.3 Foiling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
8.2.4 Rotating Racks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
8.2.5 Rotating Bowsprit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
References 111
Appendix A - Drawings I
Appendix B - Plots VII
Appendix C - Rules IX
xiii
Contents
xiv
List of Figures
2.1 Map of Mondello, Sicily . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2.2 2017 Regatta track, wind from left side of figure . . . . . . . . . . . . . . . . . . 5
3.1 Weibull probability distributions . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

3.2 Weibull probability density functions . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.3 Weibull probability density functions for different time steps . . . . . . . . . . . . 9
3.4 Model of a 49er©, (Wikipedia, 2010) . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.5 Reference hull concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.6 Bowsprit reference concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.7 Racks reference concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.1 Keel line curvature graph for final hull model . . . . . . . . . . . . . . . . . . . . 18

4.2 Water line curvature graph for final hull model . . . . . . . . . . . . . . . . . . . 18
4.3 Chine line curvature graph for final hull model . . . . . . . . . . . . . . . . . . . 19
4.4 Computational domain setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.5 Numerical ventilation error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.6 Volume mesh growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4.7 Mesh detail for the bow in the CFD-simulations . . . . . . . . . . . . . . . . . . . 27
4.8 Mesh detail for the aft in the CFD-simulations . . . . . . . . . . . . . . . . . . . 28
4.9 Sandwich constituents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.10 The meshing of the hull surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.11 Shear stresses in core . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.12 Simplification of forces acting on the skiff considered in the VPP . . . . . . . . . 40
4.13 Description of planform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.14 Profile drag for different profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.15 Profile drag for different thicknesses . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.16 Lift for different profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.17 Forces acting on a sail . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.18 Lift and Drag coefficients as a function of the angle of attack α . . . . . . . . . . 46
5.1 Transonic hull© . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

5.2 Scaled 49er© hull . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
5.3 Straight vs flared freeboard sprays differences . . . . . . . . . . . . . . . . . . . . 52
5.4 Max Beam position vs resistance. Red stars mark measured data; Blue line is
the fitted resistance curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5.5 Max Draft position vs resistance. Red stars mark measured data; Blue line is
the fitted resistance curve. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
5.6 Resistance curve for final canoe-body . . . . . . . . . . . . . . . . . . . . . . . . . 58
5.7 Wave pattern generation at 6 kn speed . . . . . . . . . . . . . . . . . . . . . . . . 59
5.8 Location of hull nodes investigated . . . . . . . . . . . . . . . . . . . . . . . . . . 62
5.9 Location of deck nodes investigated . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.10 Overview of internal stiffening structure . . . . . . . . . . . . . . . . . . . . . . . 68
5.11 Comparison of stiffener design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
5.12 Schematic flange layup design of stiffener . . . . . . . . . . . . . . . . . . . . . . 69
5.13 Comparison of model designs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5.14 Comparison of model deflection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
xv
List of Figures
5.15 Deformations magnified 50:1 for hull-with-trapeze load case . . . . . . . . . . . . 71

5.16 Deformation of hull magnified 50:1 for hull-with-trapeze load case. Hull plating
and deck are hidden . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
5.17 Distortion of elements at forestay magnified 50:1 for hull-with-trapeze load case . 72
5.18 Bulkhead element for which stress is plotted in Figure (5.19) . . . . . . . . . . . 73
5.19 State of stress in bulkhead element shown in Figure (5.18) . . . . . . . . . . . . . 73
5.20 Strain plotted for bulkhead element in Figure 5.18 . . . . . . . . . . . . . . . . . 74
5.21 Hull element for which stress is plotted in Figure (5.22) . . . . . . . . . . . . . . 74
5.22 State of stress in hull element shown in Figure (5.21) . . . . . . . . . . . . . . . . 75
5.23 Final sail area and mast height dimensions . . . . . . . . . . . . . . . . . . . . . . 80
5.24 Final sail area and mast height dimensions using the Jib 0 . . . . . . . . . . . . . 81
5.25 Final sail plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.26 Highlighted are the Jib 0 (red) and Jib (green) . . . . . . . . . . . . . . . . . . . 85
5.27 Schematics of the attachment of the bowsprit to the hull . . . . . . . . . . . . . . 85
5.28 Different levels of extension of the bowsprit, with the fixed conical section high-
lighted in red . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
5.29 Rotating bowsprit concept model . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
5.30 Final rig design, with the diamond rig . . . . . . . . . . . . . . . . . . . . . . . . 89
5.31 Planforms comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
5.32 Analysed load case for the aft support of the rack . . . . . . . . . . . . . . . . . . 93
5.33 Planforms comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
5.34 Analysed load case for the aft support of the rack . . . . . . . . . . . . . . . . . . 96
6.1 General arrangement drawings of the skiff . . . . . . . . . . . . . . . . . . . . . . 97
7.1 Dog bones in one of the deck’s template . . . . . . . . . . . . . . . . . . . . . . . 99

7.2 Templates models in Rhino 3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
7.3 Template assemblies used for planking . . . . . . . . . . . . . . . . . . . . . . . . 100
7.4 Planking with divinycell blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
7.5 Faired hull . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
7.6 Bulkhead template completely covered with balsa for the core . . . . . . . . . . . 101
7.7 Finished half-bulkhead core . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
7.8 Layup procedure of stiffener component. . . . . . . . . . . . . . . . . . . . . . . . 103
B.1 Max beam position vs resistance. Red stars mark measured data; Blue line is
the fitted resistance curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII
B.2 Max draft position vs resistance. Red stars mark measured data; Blue line is the
fitted resistance curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII
xvi
List of Tables
3.1 Mean values and Weibull distribution fit for inspected wind statistics data . . . . 7
3.2 Mean values and Weibull distribution fit for different time steps . . . . . . . . . . 8
3.3 Important dimensions of the 49er© . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.4 Important dimensions of Federica (DeSantis, 2016) . . . . . . . . . . . . . . . . . 13
3.5 Comparison of rudder areas and lead for different skiff classes . . . . . . . . . . . 13
4.1 Computational domain boundaries . . . . . . . . . . . . . . . . . . . . . . . . . . 22

4.2 Mesh properties vs ship speed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.3 Balsa wood properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.4 Flax/epoxy laminate properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.5 Load case crew on racks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.6 Load case crew in trapeze . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.1 The simulated speeds for each hull in the CFD-simulations converted from knots
to metres per second. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
5.2 Location of the centre of gravity (CoG) for each speed scaled from the 49er©. . . 48
5.3 Beam fom moment of inertia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
5.4 Resistance Comparison Between Transonic Hull© and Scaled 49er© Hull . . . . . 51
5.5 Positions for the maximum beam variation . . . . . . . . . . . . . . . . . . . . . . 52
5.6 Resistance vs beam position comparison . . . . . . . . . . . . . . . . . . . . . . . 54
5.7 Scale factors to compare which hull to choose for maximum beam position variation 54
5.8 Ranked and scaled comparison between the beam position to evaluate where to
place the maximum beam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
5.9 Positions for the maximum draft position variation . . . . . . . . . . . . . . . . . 55
5.10 Resistance vs draft position comparison . . . . . . . . . . . . . . . . . . . . . . . 56
5.11 Scale factors to compare which hull to choose for draft variation . . . . . . . . . 56
5.12 Ranked and scaled comparison between the beam position to evaluate where to
place the maximum draft . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
5.13 Final hull dimensions for the skiff . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
5.14 Final drag along x-axis for the canoe body . . . . . . . . . . . . . . . . . . . . . . 58
5.15 Drag along x-axis for hull LED3 . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
5.16 Maximum stress for varying core thickness . . . . . . . . . . . . . . . . . . . . . . 61
5.17 Variation of the sandwich skin laminates . . . . . . . . . . . . . . . . . . . . . . . 63
5.18 Deck layup analysis with 8 mm core thickness . . . . . . . . . . . . . . . . . . . . 65
5.19 Deck layup analysis with 10 mm core thickness . . . . . . . . . . . . . . . . . . . 66
5.20 Deck layup analysis with 12 mm core thickness . . . . . . . . . . . . . . . . . . . 67
5.21 Comparison between models with and without flange. . . . . . . . . . . . . . . . 70
5.22 Different concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.23 Systematic variations within different Sail Concepts . . . . . . . . . . . . . . . . 78
5.24 Systematic variations with an centreboard AR = 7.69 . . . . . . . . . . . . . . . 79
5.25 Constants assumed for subsequent computations . . . . . . . . . . . . . . . . . . 86
5.26 Variables used for subsequent computations . . . . . . . . . . . . . . . . . . . . . 86
5.27 Results from the Matlab computations . . . . . . . . . . . . . . . . . . . . . . . . 87
5.28 Leeway angle for different TWS at VMG . . . . . . . . . . . . . . . . . . . . . . . 91
5.29 Forces to dimension against . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
xvii
List of Tables
5.30 Material data for Aluminium EN-AW 6082 T6 taken from Alutrade (2017b) ans
Alutrade (2017a) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
5.31 Dimensions for the aft support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
5.32 Dimensions for the bow support . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
5.33 Dimensions for the beam rack between the support. . . . . . . . . . . . . . . . . 95
5.34 Dimensions of all the parts of the rack . . . . . . . . . . . . . . . . . . . . . . . . 96
xviii
1 Introduction
This project aims at designing and building a high performance racing skiff, the report starts
by explaining the competition context of the project and presenting the class rules that serve
as a starting point for the design. The respective design basis follows, being further developed
in the design method section. Fundamental theory is also presented in order to explain the
main theoretical topics and assumptions that guide all the different sub-groups and their work.
Finally the design process can be described without any additional restrictions. In that section,
as well as the rest of the report, the process is described in terms of the different components
of the boat, namely the hull, sail and rig, and appendages. The subsequent sections deal with
the resulting design and its layout, and the last one with the building process.
The project is inserted in three main contexts. Firstly, it is a mandatory 15 credits course for
the Masters Program of Naval Architecture and Ocean Engineering at Chalmers. The course is
organized by the department M2, Mechanics and Maritime Sciences. Secondly, it is inserted in
the context of Chalmers Sports and Technology, which is where advanced sport-related research
takes place involving engineers, researchers and students with the athletes and coaches, as well
as the enterprise and governing bodies. In this context, one important objective is to improve
the performance alongside safety for the athletes, which are, in this case, sailors. Finally, the
last context this project is included in is the 1001VELAcup 2018, in which the built boat will
be raced against competitors from other universities. This competition is the element linking
all these contexts, being what justifies the building of the boat rather than just its design, as
well as the need for connection with athletes to sail it in the future. These athletes are to whom
the authors turn to for practical advice about several practical usages that are fundamental in
the design process and yet are very difficult to realize if based only on a theoretical analysis.
Naturally with several different contexts come several different goals. First it is intended to gain
valuable knowledge and experience in designing yachts, and all it comprises in all the different
areas of the utmost relevance to Naval Architecture. Furthermore, in the context of Chalmers
Sports and Technology and the 1001VELAcup, the authors gain additional knowledge in the
operation and practical matters of sailing, its equipment, and in boat building.
1
1 Introduction
2
2 1001VELAcup
Born in 2005 from the minds of the Italian yacht designers Massimo Paperini and Paolo Processi,
the 1001VELAcup was brought to the university departments of Naval Architecture, Architec-
ture and Engineering of all Europe, the concept of a Conceive, Design, Implement and Operate
(CDIO),(CDIO, 2017), project similar to the world famous Formula Student, (Imeche, 2017).
It is therefore a project aiming towards the unification and stimulation of academic, as well as
research and innovation purposes. Further aim of the project is the formation of new profes-
sional profiles in the naval architecture world, through the concept design and further building
process of a high performance racing skiff.
The foundation of the project is the Class Rule R3 (A. 1001VELAcup, 2017), kept simple in its
interpretation and minimalistic in order to leave wide margins to the design process. Such rules
are mainly aimed to limit the dimensions of the hulls and the total sail area. The strength of
the class rule lays however in the limitations that concern the materials used for the building
process. It is indeed required that a large portion of the materials used for the hull are of
sustainable, either animal or vegetable, derivation, while the mast, or at least its matrix, is
expected to be realized in extruded aluminum.
The mobilisation of students and universities toward the applied research of new composite
materials applicable to nautical purposes has been without a doubt one of the main results
achieved by the competition. Other relevant factor achieved has been the direct implication
of realities outside the university walls. Collaboration between students, professors and other
experts from the industry made it possible over the years to carry out focused researches in
different branches of the yacht design and develop bilateral partnerships that benefit both the
academic and the professional worlds. The most important achievement of the competition is
however to be identified in the empowerment of students in relation to a real problem, aiming
far beyond obtaining a grade or passing an exam and closing that gap between concept design
and final product that defines unfortunately too often academic projects.
The competition has seen over the years the rise of innovative concepts and the constant growing
interest among Italian and other European universities, a signal perhaps that it is ready to
expand.
2.1 Class Rules
As a proper sport competition, the 1001VELAcup is subjected to a regulation class specifically

created for it, class R3, that allows room for innovation yet limits the differences among the
teams taking part to the race. Such rules affect both the crew, which during the race will be
expected to follow the Racing Rules of Sailing, and the skiff. A full rule book can be examined
in Appendix 8.2.5. In the following section only the main limitations are listed.
As far as the crew is regarded, the regulation states it to be composed of two students currently
registered at the university they race for. A second rule aims toward a fair level of skills among
the sailors, so that crew members that hold a Group 3 Classification (WorldSailing, 2017) are
not allowed to participate.
The class R3 is thought of as a box-rule in which limitations are minimal and principally affect
the main dimensions and properties of the skiff in order to allow different design solutions.
Beam and Length overall are limited respectively to 2.1 and 4.6 m with a tolerance of 15 mm.
3
2 1001VELAcup
No limits are put on draft, displacement nor the freeboard.

A further geometrical limitation affects the shape of the hull, which has to be of mono-hull type,
symmetric along the centreline and is not allowed to present transverse concavities below the
waterline.
As mentioned earlier, the class rules mainly focus their attention on the materials used for the
hull, setting a 70% minimum limit in terms of weight for the amount of natural material used
for the hull construction. Therefore all titanium, aramid, carbon fibres or other high-modulus
fibres are banned for this component. The most recent update of the regulations introduced in
the 2018 edition of the 1001VELAcup, put different restrictions to the design of the appendages.
In fact, the latest update of the regulations allow the design of the appendages in carbon fibre,
spectra, kevlar or other high-modulus fibres. Finally, the bowsprit, whether needed, can be
realized in carbon fibre.
Similar limitations in terms of materials are applied to the sail plan, so that aramid, carbon
or other high-modulus fibres are prohibited. The total sail area carried onboard is furthermore
limited to not exceed 33 m2 .
Finally, unlike most of the box-rules, no geometric limitations are set on the rig plan, it is
although required for it to be realised out of extruded aluminium.
2.2 The Race
The trophy 1001VELAcup is held every year in Italy during the last week of September. The
2018 edition will be held for the second time in the bay of Mondello in Palermo. A map of the
bay can be appreciated in Figure 2.1.
Historically, the trophy consisted of an open sea regatta distributed in nine races, each consisting
of four legs, over three days of total competition. A sketch of the regatta field, as published in
the Istruzioni di Regata (1001VELAcup, 2017) is shown in Figure 2.2.
Since the 2017 edition, a Midwinter Indoor Race has been introduced with the aim of evaluating
the hulls only on their technical features. The skiff named LED, realised by the students of the
Universita’ degli Studi di Palermo, graduated as champion of the first edition of the Midwinter
Indoor Race. Hosted by the Universita’ di Napoli Federico II, the indoor competition sees the
participant hulls towed at the speeds of 2, 4 and 6 kn and judged based on the total resistance,
with and without appendages, wave generation pattern in calm water, and the seakeeping
properties in head waves.
4
2 1001VELAcup
Figure 2.1: Map of Mondello, Sicily
Figure 2.2: 2017 Regatta track, wind from left side of figure
5
2 1001VELAcup
6
3 Design basis
In this chapter, a prestudy is presented. In this prestudy, all of the skiff’s parts that are designed
are covered, alongside with an investigation of the wind statistics at the race site. Investigation
of similar skiffs, both commercial and those participating in the 1001VELAcup in previous
years are also presented. The limitations and the overall design features of the project are also
identified and stated in the last part of this chapter.
3.1 Wind Statistics
When designing a hull for a competition, it is good practise to optimise it for the racing con-
ditions, therefore, considering where the race will take place, in this case Mondello, Italy. The
design can then be optimised for the expected wind speed. The data needed is thus primar-
ily the mean wind speed together with its statistical distribution. The data is collected from
two different sources, Windguru (Windguru, 2017) and Istituto Superiore per la Protezione e
la Ricerca Ambientale (ISPRA) (R.M.N., 2017), focusing on those data corresponding to the
second half of September. Time period and time frequency of the data extrapolated are differ-
ent for the different sources. Windguru provides data since 2004 to present day with 3 hours
time spans, while ISPRA provides acceptable data from 2009 to 2014 with a time frequency
of 10 minutes. Both data sheets are consequently resumed in order to focus only at the hours
during which the race takes place, therefore only data between 11 am and 5 pm are taken into
consideration.
Analysis of the data provides a mean wind speed at the location of 3.84 m/s from Windguru
and 3.37 m/s from ISPRA. A histogram is thus created to see how the data is distributed.
Figure 3.1a and 3.1b show that trends for the data available well fit a Weibull probability
distribution. Hence, the expected value and variance are finally evaluated, and both probability
density function (PDF) plots are presented in Figure 3.2a and 3.2b. In Table 3.1 below the
expected value and variance are presented.
Table 3.1: Mean values and Weibull distribution fit for inspected wind statistics data
Windguru [m/s] ISPRA [m/s]
Mean Value 3.8438 3.3716

Expected value 3.8616 3.3639
Standard deviation 1.1991 0.4740
Variance 1.4379 1.4379
Further and more detailed investigations focusing on three different consecutive time steps
within which each single race will take place. Such division is needed in order to get a more de-
tailed analyse of the wind conditions during the different races per day as well as to furthermore
decide on the final sail set up. Results obtained from these latter inspections are presented in
Table 3.2 highlighting mean value, standard variation and variance of the data. For each time
step, probability density functions are plotted in Figure 3.3a to 3.3c.
7
3 Design basis
(a) ISPRA data (b) Windguru data
Figure 3.1: Weibull probability distributions
(a) ISPRA data (b) Windguru data
Figure 3.2: Weibull probability density functions
Table 3.2: Mean values and Weibull distribution fit for different time steps
Time step 8-11 [m/s] 11-14 [m/s] 14-17 [m/s]
Mean Value 3.5011 3.7382 4.0253

Standard deviation 1.5952 1.4041 0.8439
Variance 2.5446 1.9715 0.7122
8
3 Design basis
(a) 8-11 am time step data (b) 11-14 am time step data
(c) 14-17 am time step data
Figure 3.3: Weibull probability density functions for different time steps
3.2 References for Design
To avoid starting from a clean sheet, it is decided to start the design process with some ref-
erences. The overall reference skiff used in this project is the 49er©. The 49er© is a high
performance sailing dinghy which was designed in the 1990’s (Bethwaite, 2010). However, some
other references are used as well for the hull shape and the appendages.
3.2.1 Hull
When starting the design process for the hull shape, it was decided to start from a couple
of reference hulls. This section presents the two hull shapes that are used as references and
explains why they were chosen.
9
3 Design basis
3.2.1.1 49er© Hull

A new hull, based on the 49er© (see Figure 3.4), is scaled down to 4.60 m LOA according to
the regulations of the competition. The main design properties of interest from the 49er© are:
• Maximum beam far aft
• Flat bottom in the aft
• Sharp bilge above the waterline
• Slender and v-shaped bow
The design objective of the 49er© was to make a dinghy of less than 5 m that performs well
with no "hump" in the drag/speed curve. This means that there should not be a peak in the
resistance during the transition between the displacement mode and the planing mode. Since
a planing hull was desired, the focus for inspiration was on ’light dinghies’ only. The main
inspiration for the 49er© was the 18-footer which at that point of time was a modern recently
developed dinghy (Bethwaite, 2010). The 49er© is still today one of the highest performing
skiffs in its class and is consequently chosen as the main reference hull.
Figure 3.4: Model of a 49er©, (Wikipedia, 2010)
3.2.1.2 Transonic Hull ©

The second reference hull that is used is the so-called ’transonic hull’©, see Figure 3.5b (Calderon
& Maskew, 2015). The hull has a double wedge shape with a vertical stem, entirely flat bottom
from the deepest point at the stem to the widest point at the stern, which has zero submergence
and the sides are vertical. The submerged part of the hull, thus, forms a tetrahedron). The
basic idea of a transonic hull© is that it has a low wave generation when compared to the more
regularly shaped hulls. Since the bow is very sharp, the bow wave will in principle be very
small, and with no curvature along the hull there will be no pressure peaks and therefore no
waves generated along the hull. However, at the stern there is a very small pressure rise due to
10
3 Design basis
the water that is leaving the flat and rather horizontal bottom. Thus there will be very small
waves generated, but they will practically not cause any negative interference between the bow
and the stern waves. Most hulls have a wave resistance hump at a Froude number of 0.4-0.5,
this is caused by a wave crest in the bow and a wave trough at the stern. Since the stern wave of
a transonic hull© is very small, this effect will be very small. Because of this a transonic hull©
will not have a specific "hull speed" for which the created wave system that is most favourable.
Furthermore, since in principle this shape allows the hull to pierce the waves, it translates
in sailing with very small pitch, and is therefore almost free from slamming problems in the
structure.
(a) Scaled 49er©-based hull (b) Transonic hull©
Figure 3.5: Reference hull concepts
3.2.2 Rig and Sails
The rig and sails are one of the most important parts of a racing skiff. Both can be adjusted
iterative over and over again. However, to start designing, a basis needs to be defined. This
is done by using well known references, among other things. The following sections contain
descriptions of references used for the rig and sail design process.
3.2.2.1 Rig
The rig reference is the traditional 49er© rig, see Figure 3.4. It contains a 8.3 m high mast made
out of carbon fibre as its boom. The rig has two spreaders for the lower and upper shrouds
each. These shrouds are attached at the hull shortly behind the mast used for pretensioning the
rig. The forestay goes on top of the bowsprit system attachment. Two trapezes are mounted on
top of the upper shrouds pulled back to the racks. They are used for provision of the righting
moment via hiking.
3.2.2.2 Bowsprit
Using a third additional sail requires a bowsprit separating the sails that are set. The 49er©
serves as inspiration not only for the sail and rig design, but also for the bowsprit. Furthermore,
some opponent’s dinghies are also taken as a reference for the design of this component. Both
systems can be seen in the following Figures 3.6a and 3.6b.
As shown in the figures, both bowsprits are retractable, meaning that they can be pulled inside
the skiff when not in use. Therefore both references have installed a system connected to the
sail setting. A reinforcement at the bow prevents the bowsprit of falling out as well as taking
the bending force of the sail and serving as an attachment for the forestay. The material of the
bowsprits is usually carbon fibre. Apart from a small piece sticking out in the bow, the length
of the bowsprit is smaller than the distance between forestay and mast attachment.
11
3 Design basis
(b) Previous competitor in 1001VELAcup, pic-

ture taken by Adam Persson (2017)
(a) 49er©, picture taken by Eike Jacobs (2017)
Figure 3.6: Bowsprit reference concepts
3.2.2.3 Sails
For the sails, the 49er© is used as the main reference as well. The current 49er©model, Section
3.4, has a total upwind sail area of 21.2 m2 . This area is divided into a main sail area of
16.1 m2 and a jib sail area of 5.1 m2 . For the downwind course, the 49er©has an additional
38 m2 gennaker to hoist giving it a total of 59.2 m2 .
Table 3.3: Important dimensions of the 49er©
Dimension
Main Sail 16.1 m2

Jib 5.1 m2
Upwind sail area 21.2 m2
Jib to upwind sail area ratio 24%
Mainsail to upwind sail area ratio 76%
Mast height 8.3 m
Mast height to length Ratio 1.69
The main sail is shaped rectangular providing as much sail area up in the top part as possible
using the uninterrupted wind. Furthermore, managing the twist of the main sail is easier. The
sails itself are made out of Mylar and fully battened.
Opponent skiffs from the previous years of 1001VELAcup are also used as references, where it
is common to use 55 % for the main sail and 45 % for the gennaker, and trying to have the
CE (center of effort) as high as possible (DeSantis, 2016). One of the opponents sail plan is
presented in Table 3.4.
3.2.3 Appendages
The design of the appendages, the centreboard and rudder, is based on comparisons between the
appendages of other fast sailing dinghies and skiffs. The recommendations in Larsson, Eliasson,
and Orych (2014) are also evaluated. A comparison between different common wing sections is
also implemented.
12
3 Design basis
Table 3.4: Important dimensions of Federica (DeSantis, 2016)
Dimension
Main Sail 13.794 m2

Jib 4.356 m2
Gennaker 14.85 m2
Upwind sail area 18.15 m2
Jib to upwind sail area ratio 24%
Mainsail to upwind sail area ratio 76%
Mast height 7.74 m
Mast height to length Ratio 1.68
3.2.3.1 Rudder and Centreboard
For the design of the appendages, the most relevant comparison between different skiffs is the
planform area of the centreboard in relation to the upwind sail area. According to Larsson et
al. (2014), the centreboard planform area should be between 3 % and 3.5 % of the upwind sail
area. For the 49er©, this area ratio was found to be
centreboard area 0.3 m2

= = 1.42%, (3.1)
upwind sail area 21.2 m2
where the centreboard area is measured in a CAD file of a 49er©, (Creative, 2012). Furthermore,
this file is used to measure the 49ers© centreboard draft of 1.1 m as well as the 49er’s© rudder
area of 0.2 m2 . Since the recommendations in Larsson et al. (2014) are mainly for keel boats
and not for high performance sailing dinghies, it is more reasonable to use the same centreboard
to upwind sail area ratio as the 49er© rather than the one in Larsson et al. (2014).
The lead, the distance between the sail centre of effort and the centreboard centre of effort as
a percentage of the length of water line (LWL), is investigated. For the design of the of the
rudder the recommendations in Larsson et al. (2014) are disregarded for the same reason as
above. Instead, the properties of several high performance skiffs are investigated and a decision
based on this is made. The results from the study are presented in Table 3.5.
Table 3.5: Comparison of rudder areas and lead for different skiff classes
Boat class Rudder Area [% of centreboard area] Lead [%]
49er© 67 13
29er© 43 8
470© 43 4
505© 45 2
International 14© 60 7
Musto skiff© 68 9
13
3 Design basis
3.2.4 Racks
To maximise the righting moment created by the sailors, in order to carry as much sail as
possible during the races, it is preferred to have the centre of gravity (CoG) as far out from the
centre line as possible. Since it is preferable to have a slender hull, it will most likely not be the
maximum allowed width, so the racks will be used to move the sailors centre of gravity as far
out as possible.
To have some background of how the racks should look like inspiration is taken from the 49er©,
since it is the main reference skiff. The racks on the 49er© are built like wings, which means
that the space between the hull and the outer part of the racks is covered with material, see
Figure 3.7a. Another dinghy that is used as inspiration is the International 14 footer. It has
racks made out of tubes of carbon fibre which makes them light and moves the centre of gravity
far out from the centre line. Finally inspiration is taken from the already existing boats in the
1001VELAcup. These racks are more similar to the International 14 footer, they are made out
of aluminium profiles that give the racks, see Figure 3.7b.
(a) 49er©, picture taken by Kaupp (2007) (b) Previous competitor in 1001VELAcup, pic-
ture taken by Persson (2017)
Figure 3.7: Racks reference concepts
3.3 Overall Design Features
The authors of the project have decided to not go for hydrofoil nor wing sail. It would have
been a great advantage to use hydrofoil since the resistance drops to a minimum due to uplifting
of the hull out of the water. Had this been the group’s choice, the hull would have been shaped
based on different criteria, such as lifting up as fast as possible. Nevertheless, the hydrofoiling
option was disregarded based on the ruling concerning the sailors, as it is very unusual to
have two sailors on a hydrofoiling skiff. It would have been difficult to find sailors who would
do it properly and even then it would have required a disproportionate amount of practice.
Consequently, this would add too much uncertainty to the results of the race, due to factors
that are external to the design team.
Similarly, it would have been a great advantage to use a wing sail since the efficiency of such is
much better compared to a normal sail. Yet this option was also disregarded since the students
would have had to construct the wing sail by themselves without having the possibility to
purchase it from an external company. The handling of a wing sail requires a lot of training.
Therefore, more uncertainty would have been introduced due to alien factors to the design team.
14
3 Design basis
Besides the uncertainties added and the eventual absence of backup hulls during the competition,
there are also severe time constraints that limit the group’s focus, not allowing for proper design
of more complicated concepts. Most of all, since this is the first design for Chalmers University
of Technology, the aim is to achieve a good position in the race as a way to assure the continuity
of this project in the future years.
15
3 Design basis
16
4 Theory
In this chapter, the basic theory for the methods that are used is explained in deeper detail.
The sections focus on the different softwares used and the assumptions made while using them.
The softwares are, Rhinoceros 3D for CAD, STAR-CCM+ for CFD, Abaqus for FEM, and a
VPP-program modified specifically for this project, used for velocity prediction.
4.1 Computer-Aided Design (CAD)
The CAD-software used throughout the whole project is Rhinoceros 3D (McNeel, 2017). It is
a software based on NURBS mathematical model, which is suitable for designing a boat hull
since it can create complex curves and surfaces. The following sections will explain the theory
behind the designing procedure.
4.1.1 Hull Fairing
Fairing of the hull surfaces is a very important aspect of the modelling process that has to
be regarded throughout the whole process. This is of particular importance at the bottom
and freeboard. The fairing process constitutes a good practice of surface modelling, being a
requirement for having high quality surfaces (Henry P. Moreton, 1992). It consists in smoothing
the mesh through a proportional grid of control points that do not cross paths between each
other, except if strictly demanded by an irregular or unusual shape. The global trend of the
control points’ coordinates should also match the trend of the desired surface, except when
demanded by extreme shapes, that, for example, need trimming. The goal with this is to
obtain surfaces that are as smooth as possible in order to improve the results of subsequent
computations, i.e. to have computational results that are as smooth as possible.
There are no objective criteria to measure smoothness or fairness, and furthermore, it is highly
dependent on the application. Yet, a principle that should be followed in this particular ap-
plication of yacht design is that the surfaces should result as simple as possible without any
unnecessary details or oscillations. This means they should have the minimum number of con-
trol points that are not so restrictive as to inhibit the desired hull shape, but are just enough to
achieve it. They should allow for a similarly smooth and relatively easy further manipulation.
Adding to all of this, the aesthetics of the hull will also be greatly improved as a result. During
the design of the hull, fairness of several key lines and surfaces has always to be made. These
are explained in the following sections.
4.1.1.1 Keel line
The keel line extends over the length of the hull, and should assume a curvature close to that
depicted by the blue line in Figure 4.1.
As seen in the Figure 4.1, the curvature decreases to very small values near the stern. This
is important since a more convex curvature induces a low pressure that sucks the stern down,
therefore it increases resistance as the Froude number is increased.
17
4 Theory
Figure 4.1: Keel line curvature graph for final hull model
4.1.1.2 Waterline
The waterline extends for almost the full length of the boat. It should also be smooth all the
way along its length, to ensure there are no bumps on the hull surface and in order to minimise
the disturbance of the flow around the boat. The curvature graph of such line shall assume a
clear trend as well as absence of sign changes. Furthermore, the curvature will naturally increase
very significantly near the aft due to the geometry of the hull. This is evidenced in Figure 4.2.
Figure 4.2: Water line curvature graph for final hull model
4.1.1.3 Chine line
The same reasoning and principles mentioned for the keel and waterline apply for the chine line
as well. In this case, the line shall have no curvature along a certain percentage of the length,
this is from where the beam becomes constant. The chine line is presented in Figure 4.3.
18
4 Theory
Figure 4.3: Chine line curvature graph for final hull model
4.2 Computional Fluid Dynamics (CFD)
Practical CFD computations are carried out with the aim of predicting hull resistance with an
acceptable accuracy. Such accuracy is only achievable if turbulence models, wall functions, mesh
and evaluation times are appropriately considered. The tool used to carry out such evaluations
is STAR-CCM+ (CD-Adapco, 2017) and in the following sections theoretical support to the
solver set up will be presented.
4.2.1 Resistance
According to the theory of ship resistance initially developed by William Froude, the total
resistance of a hull proceeding at the water surface can be defined as the sum of a viscous
and a wave component. The former derives from the combined effect of the wetted surface,
the surface smoothness, the length and the speed, expressed in terms of tangential stresses
and viscous pressure resistance. The latter is instead caused by the distribution of pressure
developing around the hull when in motion and weighs on the total resistance, for a planing
boat with Froude number of 1.4, at most 44 %. For a containership with Froude number of
0.22 the wave components insists instead on at most 17.5 % of the total resistance (Larsson &
Raven, 2010). Such values for the distribution of the total resistance can be taken as maximum
and minimum references for the skiff under inspection in the project, which is computed for
boat speeds ranging between 3 and 15 kn.
As water flows along the hull, friction slows the water molecules and creates a layer that is
carried along with the hull itself. This initially thin layer, called the boundary layer, gradually
increases in thickness evolving from a laminar condition to a turbulent one and eventually breaks
into eddies near the stern. As the hull furrows through the water, the flow has to travel around
it. The local velocity of the water differs from the one characterising the undisturbed flow and
will be slowed at the bow and at the stern, but increased along the main part of the hull causing
an increase in friction.
Moving along the water surface, a hull will force fluid particles to move from their equilibrium
positions thus creating waves that radiate in two different patterns. The first, and most relevant,
19
4 Theory
is the transverse pattern. The second pattern is the divergent waves that fans out from the hull.
The resistance component generated by the energy removed from the wave system, which can
be found in the wake, is thus called wave breaking resistance. The remaining portion of wave
energy is instead radiated away from the hull and responsible for the so called wave pattern
resistance.
4.2.2 Governing Equations Solved
Practical CFD computations can almost always be considered satisfactory when details about
the time-averaged properties of the flow are obtained. At the heart of most these CFD ap-
proaches is the modelling of informations and effects of the turbulent fluctuations. The treat-
ment of such events can be carried out with procedures based on the Reynolds-Averaged-Navier-
Stokes (RANS) equations, which represents time-averaged equations of motion for fluid flow.
Assuming that a fluid behaves as a continuum, the Navier-Stokes equations allow to describe
fluid flows. However, these equations are inherently unsteady and nearly impossible to solve
unless multiple average solutions, solved numerically through discretization in space and possibly
time, at a series of time steps are used. As stated in Versteeg and Malalasekera (2007), the
decomposition of the Navier-Stokes equations, bringing to the definition of the RANS, is based
on the assumption that, introducing a set of unknowns called Reynolds Stresses, the time-
dependent turbulent velocity fluctuations can be separated from the mean flow velocity. Such
unknowns are then functions of the velocity fluctuations and require the implementation of a
turbulence model to solve them.
Unsteady RANS for transient flow is a straightforward variant of the RANS in which, while still
solving for the mean velocity separately from the turbulent velocity, a transient term is present
in the momentum equation and retained during computation. This means that turbulence time
scales are removed from the Navier-Stokes equations through the averaging procedure, while
larger time scales are resolved. The development of unsteady RANS methods within the ship
hydrodynamics evaluations unlocks extensions to applications for seakeeping and manoeuvring,
as presented in (Wilson, Carrica, & Stern, 2006) in addition to the already covered resistance
and propulsion evaluations. Nevertheless, the effects of turbulence on the mean flow have to be
described, since the time-averaging operation on the momentum equations discard all the detail
connected to the state of the flow in the instantaneous fluctuations.
CFD breaks a fluid domain into discrete cells to then solve in each one of them the conservation
laws. The accuracy of these simulations are strictly bonded to the grid generated.
4.2.3 Turbulence Model
Throughout the project both a normal k − ω turbulence model and the Shear Stress Transport
SST-Menter expansion are considered in place of the more commonly used k − because of the
high interest in the evaluation of close to wall features.
There are multiple reasons supporting such choice. According to (Versteeg & Malalasekera,
2007) the most important however can be found in the improved behaviour of k−ω when dealing
with wall effects, unconfined flows and curved boundary layers. The further implementation
of SST-Menter model allows to take into account properly the behaviour of the free-stream,
quality of the k − model, without falling into errors dictated by sensitivity as it would happen
to a normal k − ω. Therefore, the turbulence model chosen for the calculations is capable of
predicting, in a numerically stable way, both the near-wall and the far field behaviours of the
flow.
Another reason that guides the choice of the model is the under-prediction of separation provided
20
4 Theory
by k − , leading towards an overly optimistic, and therefore, non-conservative performance

prediction. On the other hand, SST-Menter results should provide higher accuracy when dealing
with onset and amount of flow separation under adverse pressure gradients and thus more
accurate in such prediction.
It should also be noted the fact that such turbulence model fails to take into account more subtle
interactions between mean flow and turbulent stresses when compared to the Reynolds Stress
Model (RSM) model. Versteeg and Malalasekera (2007) state in fact that two-equation turbu-
lence models are incapable of capturing the more subtle contributions introduced by anisotropy
of the normal stresses, as well as correctly predicting the effects on turbulence caused by body
forces and extra strains. The RSM is able to exactly incorporate such features, but the re-
quirement for several unknown turbulence processes to be modelled, together with a significant
increase in computational time and storage need, discourage its implementation.
4.2.4 Numerical Solution
The first step to process a CFD problem is the definition of the region of interest. Evaluation of
encountered resistance of a skiff has to deal with both air and water, therefore the modelling of
an eulerian multiphase computational domain is needed. Furthermore, as there is a change in
physical properties, problems may derive along the interface between the two fluids considered.
Subsequent mesh refinements are therefore needed in such region.
4.2.4.1 Computational Domain

The problem requires the definition of a multiphase computational domain where air and water
are considered as the fluids interfacing. Specific dimensions and boundary conditions of the
domain are needed in order to capture with satisfying accuracy the computations results. Only
half of the domain is modelled since symmetry properties of the problem allow to do so and the
computational time will decrease considerably. It has to be noted that all the results obtained
will refer to half of the hull, hence half of the total resistance if not otherwise stated. The
computational domain, presented in Figure 4.4, can then be defined as a rectangular box and
boundary conditions defined according to Table 4.1. The symmetry surface ABCD reflects the
symmetry of the problem, as water and air are considered to flow parallel to this boundary, with
no particle crossing through it. The two opposite surfaces ADHE and BCGF are respectively
given inlet and outlet boundary conditions; the outflow surface has to be placed sufficiently
downstream such that vortices are not yet present in the stream. The three remaining surfaces
are instead given wall boundary conditions.
The hull boundary is modelled by a wall law. To implement such condition, the skiff is positioned
in the domain with the centreline on the mirror plane with the bow facing the inflow boundary.
At least one ship-length L = 4.6 m has to be left clear between the bow and the inflow boundary.
The hull body is then subtracted from the domain itself and wall boundaries are defined at the
portion left empty as there is no velocity passing through these sides. The flow encountering a
wall will then have to travel around it, generating pressure distribution and a boundary layer
development that are the heart of the numerical computations. Finally, no boundary conditions
are applied to the free surface.
21
4 Theory
Figure 4.4: Computational domain setup
Table 4.1: Computational domain boundaries
Surface Corners Boundary
Top CDHG Wall

Bottom ABFE Wall
Side EFGH Wall
Symmetry ABCD Mirror
Inflow ADHE Inlet
Outflow BCGF Outlet
4.2.4.2 Turbulent Boundary Layer

According to Larsson and Raven (2010), a validated assumption is that the velocity profile
in the inner part of a boundary layer must be independent of the thickness of the layer and
dependent only on the shear stress at the wall, and the fluid density and dynamic viscosity.
Reasoning in terms of dimensions, such non-dimensional distance from the wall may be written
according to Equation (4.1).
U∗ y
y+ = (4.1)
ν
where y represents the distance to the nearest wall, ν the local kinematic viscosity of the fluid
and U∗ , defined according to Equation (4.2), and being τwall the wall shear stresses and ρ the
density of the fluid, represents the friction velocity at the nearest wall
τwall
r
U∗ = (4.2)
ρ
y + is thus used to define the proper size of a mesh for a specific flow pattern. The importance
of this element concerns wall functions and the definition of the velocity profile, divided in four
22
4 Theory
different regions according to the value of y + : viscous sublayer, buffer layer, logarithmic region
and wake region.
The first cell spacing, ∆S, on a flat-plate boundary is evaluated via the resolution of the following
algebraic system:
ρU∞ x
Rex = (4.3)
µ
0.026
Cf = 1/7
(4.4)
Rex
Cf ρU∞ 2
τwall = (4.5)
2
τwall
r
U∗ = (4.6)
ρ
y+µ
∆S = (4.7)
U∗ ρ
where Rex represents the Reynolds number at distance x from the stagnation point, U∞ the
freestream velocity and Cf skin friction coefficient.
Finally, knowing the Reynolds number, it is then possible to evaluate half of the total thickness
of the boundary layer, of the total thickness of the prism layer for the entire length of the hull,
according to the Equation (4.8). Such choice can be considered on the safe side an provide
results accurate enough since the velocity gradient in the outer part of the boundary layer is
very small.
0.185L
t= (4.8)
Re1/5
4.2.4.3 Volume of Fluid (VOF) Method

The Volume of Fluid (VOF) method belongs to the class of eulerian modelling techniques
characterised by the mesh, either stationary or moving in a prescribed manner, in order to
accomodate the evolving shape of the small contact area between two fluids. It is a simple
solver that allows for tracking the free surface using a volume fraction function for each fluid
to describe the location of the interface. Such fraction function is tracked for each cell of the
domain grid and assumes value 1 when the considered cell is full of water, zero when filled
with air. However, with this method the free surface is not sharply defined and a distribution
over the height of several cells is instead generated. Hence, in order to achieve accurate enough
results, local mesh refinements are needed.
4.2.4.4 Numerical Ventilation

An error that can occur in the cases where the flow is not properly resolved is numerical
ventilation. Such a computational error is encountered when two fluids are considered interfacing
in the domain and particles of one of them are considered trapped into the boundary layer and
transported into the domain of the second fluid. Hence, the net physical transition between
the two fluids is erroneously approximated as a gradual diffusion. An example of numerical
ventilation error can be appreciated in Figure 4.5. The hull presented was tested at a speed
of 15 kn and it can be seen how a wide portion of the submerged hull is covered by air (blue
portion) instead of water and only after a gradual transition does it reach the wet condition.
If not corrected, such error will affect the computations and provide a wrongly estimated lower
resistance.
23
4 Theory
Figure 4.5: Numerical ventilation error
4.2.5 Grid Description
Fundamentally, a mesh is defined valid if it respects the main concepts of classification and
topological compatibility. Further features defining a valid mesh must be the geometric trian-
gulation, the property of unique mappability, parametric intersection and geometrical similarity.
Once a valid mesh is generated, the attention has to move toward its quality, as a still valid
mesh can guide to entirely wrong results if the quality is too low or the physics models in use
are not properly assumed. Definition of an accurate mesh is then crucial to obtain valid results.
From such point of view, the given problem requires mesh refinements perpendicular to the wall
boundaries in order to properly capture the boundary layer and predict the frictional resistance.
4.2.5.1 Mesh Refinement

For fluid flow calculations, grid quality has to be optimised in order to best evaluate calculations
regarding convective and diffusive fluxes. To achieve such optimised grid, CD-Adapco (2017)
suggests to model the cells in a way that the line connecting the centroids of two neighbouring
cells would pass as orthogonal as possible to the face separating the cells and as closely as
possible to its centroid. Critical items to be considered are the first cell height, the total prism
layer height, and the number of layers.
It is good manner to initially generate an automatic valid volume mesh and later refine it. An
initial guess on the dimensions of the cells characterising the grid can be made referring to the
main dimension (length) of the geometry analysed and successively remesh the initial dimensions
of the cells and surfaces to improve the overall quality of the model. Directly connected to the
first cell height is the choice of a value for the element y + . Such choice is arbitrary, however,
specifications are needed on the first cell spacing and total number of layers in order to generate
a mesh that is accurate enough, with special regards toward the near-wall regions.
Once the first cell heights and the number of layers have been decided, the growth rate must
also be checked and adjusted in order to eventually modify the concentration of cells in the
direction orthogonal to the analysed geometry.
The mesh generated will result the same all over the geometry, final refinements are therefore
needed in order to consider the z-direction of the domain close to the waterline, in order to
resolve wave heights, and along the deck, hull and transom surfaces.
24
4 Theory
Unfortunately no universal rule exists when it comes to determining whether the total number
of cells forming a mesh is good enough. As mentioned earlier in this section, rule of thumbs can
be followed considering the characteristic dimension of the geometry analysed and derive from
it the dimensions for the first cell’s height. An other validation can be found when comparing
the generated mesh with one created for similar purposes and geometries
4.2.6 Aerodynamic Resistance
The computational domain in which the skiff is analyzed presents two different phases. It is
therefore necessary for completeness of investigation to evaluate the influence of different deck
shapes on the total resistance encountered by the skiff when sailing. In these evaluations, the
height of the freeboard and the geometry of the transom play major roles. Due to time limita-
tions such properties have not been properly analyzed with iterative investigations. Bethwaite
(2008) Higher Performance Sailing is instead taken as reference when shaping the transom.
According to Bethwaite,
"when wind flows away from a surface such as a vertical transom, the effect is like a filter
pump; the core of dead air behind the transom is dragged downwind by the viscous drag of
the surrounding moving air, the pressure decreases, and the pressure difference between the
undiminished pressure on the windward face and the reduced pressure on the leeward face shows
up as drag" (Bethwaite, 2008).
An inward sloped transom would therefore allow air to smoothly reach the water level without
generating sudden pressure differences. Additionally, the hull is shaped as a solid brick, hence
with no open top, such that the wind will only hit the windward end and sides of the brick and
gently tug the leeward faces backwards.
4.2.7 CFD Analysis Setup
Figure 4.4 shows a representation of the defined computational domain.

The turbulence model is chosen to be a k − ω model with SST implementation in order to
better capture the flow behaviour both near-wall both in free-stream.
It has been chosen to initially consider an automatically generated mesh and later refine it.
In order to accurately capture the viscous boundary layer a Prismatic Cell Mesher has been
chosen together with a Trimmed Mesher, which provides a robust method of producing a high-
quality grid, with a Medium Growth Rate, shown in Figure 4.6. All the size types have been
set to the value absolute in order to be able to iteratively modify each one independently of the
characteristic dimension of the geometry. Starting cell sizes have however been chosen to be
equal to 500 times smaller compared to the length of the hull. An initial value of 10 was chosen
for the Number of Layers. The first iterations however, showed that increasing such value to 14
provided a better quality of the grid without compromising the computational time.
Figure 4.6: Volume mesh growth
25
4 Theory
The interface between the two fluids is the most sensitive portion of the domain and easiest to
lead to numerical errors if not properly defined. A mesh refinement in such region is therefore
needed and defined between 15 cm above and 10 cm below the waterline
During the process, a value of y + = 30 is assumed. This value expresses the transition between
the buffer layer, where neither the linear law nor the log law are valid, and the logarithmic
region. The choice for a high y + wall treatment determines a logarithmic law for the near-wall
boundary layer and therefore a more accurate prediction of the behaviour of the flow in such
region. As a direct consequence, the chosen value eliminates numerical ventilation errors up to
speeds of 12 kn, while for higher speed the errors are considerably reduced but not completely
erased.
The first cell height, together with the total thickness of the boundary layer and therefore the
total number of cells generated, result different for each simulation. It is intuitive to under-
stand that, since cell height and total thickness of the prism layer depend on the flow velocity
and Reynolds number, their values will result different each time the ship speed is modified.
Reference values for these elements are presented in Table 4.2 while details of the grids at the
front and aft section can be appreciated in Figures 4.7a to 4.8b.
Table 4.2: Mesh properties vs ship speed
Speed [kn] First cell height [m] Total thickness prism layer [m] Number of cells
3 0.0006 0.038 4701421

5 0.00038 0.033 4704362
6 0.0003 0.033 4707762
7 0.00028 0.031 4734721
9 0.00022 0.031 4714780
12 0.00015 0.028 4670987
15 0.0001 0.027 4572475
Finally, an implicit unsteady solver is chosen with a time step of 0.01 seconds over a total time
span of 10 seconds. The solver is needed in order to assure the convergence of the computations.
It has been noticed that for the chosen time span the solutions can be assumed stable, oscillating
around a constant mean value with relatively small range. Therefore a longer evaluation time,
as was initially set, resulted not needed; moreover the computational time deriving from that
would drastically increase.
26
4 Theory
(a) Mesh detail at bow for 6 kn
(b) Mesh detail at bow for 12 kn
Figure 4.7: Mesh detail for the bow in the CFD-simulations
27
4 Theory
(a) Mesh detail at aft for 6 kn
(b) Mesh detail at aft for 12 kn
Figure 4.8: Mesh detail for the aft in the CFD-simulations
28
4 Theory
4.3 Finite Element Method (FEM)
The structural work within the project has a wide scope. It concerns the choice of construction
materials, finite element analysis, structural stiffening, construction method and more. The
process of reaching a final result is tedious, and depends on many variables. The following
sections describe the process of the structural work from the beginning to the current state.
4.3.1 Prestudy
Initially, a prestudy was performed. The study was divided into three parts where every part
was needed to produce a final hull. Firstly, material options for the hull structure was studied
with the aim of finding suitable materials for the hull plating, hull stiffening structure and deck.
Secondly, studies were performed with the intent of finding a suitable methodology to assess
the loads acting on the hull structure. These loads include the loads from the rig as well as
hydrodynamic loads acting on the hull. Lastly, studies on methodologies to assess the material
and load interaction in the hull structure using FEM-software were conducted.
4.3.2 Hull Components
In the prestudy, the hull structure was divided into three components: hull plating, internal
stiffening structure and deck plating. The reason for the division is to identify specific character-
istics for each component and thereby be able to adapt the material choice and design for each
component according to their characteristics, while still considering how the components will
interact with each other and how the material will affect this interaction. The material choice
and the design of all hull components is significantly limited in the 1001VELAcup regulations
that states: "the hull and racks should be made of wood or materials of vegetable and/or animal
origin, expressed in weight, not lower than 70 %". This rule and its effects on the design and
construction process will be covered in more detail further on in the report.
4.3.2.1 Hull Plating

The hull plating is the main structural component of the skiff. It makes up the closed volume
that provides buoyancy, as well as the main contribution of stiffness to the hull girder. This is
true especially in locations where the curvature of the hull is high such as the bow. The stiffness
of the hull girder is further increased by adding an internal stiffening structure. Common
practice when designing a lightweight hull structure for small sailing craft is to construct the
hull plating using sandwich laminates.
4.3.2.2 Internal Stiffening Structure

The internal structures of the hull, consisting of transverse bulkheads and longitudinal stringers,
contributes to the overall hull stiffness. The bulkheads and stringers are also the components
that transfer the local loads from rigging and mast through the chain plates and mast foot to
the hull. Using a sandwich structure similar to the one used in the hull plating is possible,
though another option to consider is to make use of ordinary marine plywood for the stiffening
structures. A complete FE-analysis comparing weight and strength of both options must be
performed to make a conclusive decision, although this is not part of the project scope.
4.3.2.3 Deck
The deck serves as the closing surface of the top part of the hull as well as contributing to overall
hull stiffness. It is also the part of the boat where components such as deck hardware, mast,
forestay, shrouds and mainsheet are attached to the hull. The material in the deck will therefore
29
4 Theory
have to be able to withstand large local point loads at the attachment points of rigging elements,
as well as the weight of the sailors during operation and the global bending moment. Making
use of the sandwich structure in the deck to achieve low weight and high bending stiffness is
desired. However to deal with the local point loads from the rigging, a sandwich structure in
the deck would possibly require reinforcement of the core in these areas, to be able to handle
the out of plane loading. FE-analysis is required to determine the design and location of these
reinforcements.
4.3.3 Centreboard and Rudder
The appendages, centreboard and rudder, are structurally designed with the same design phi-
losophy as the hull, which is to create a strong but light structure. The difference is that there
are no restrictions in construction materials for the centreboard and rudder, as opposed to the
hull’s limit of 70 % organic origin. This rule opens up new alternatives regarding the choice
of material. Since the structures of the appendages are supposed to be light and strong, a
sandwich structure may be favourable in this case as well. A material stronger than flax, such
as a composite laminate made from carbon or glass fibre, will most likely be used to achieve a
low weight and withstand the relatively high local loads.
The rudder is mainly affected by the hydrodynamic forces when travelling through the water,
whereas the centreboard needs to be dimensioned for the worst load case which is when the
crew is standing on the centreboard tip in order to tilt the skiff upright after it has capsized.
This load case will naturally cause a large bending moment at the base of the centreboard.
4.3.4 The Sandwich Structure
Early in the project, it was apparent that the final choice of construction material would stand
between marine plywood in a cold moulded configuration, and a sandwich structure using
organic constituents. Using plywood has the benefit of low cost and relatively easy building
process with a rigid cross-section suitable for attaching hardware relatively easy. It will however
make the hull heavier than necessary. A sandwich laminate using a lightweight core on the other
hand has the benefit of low total weight together with high flexibility regarding the hull shape.
The main reason for using a sandwich laminate is the ability to adapt the laminate layup
according to the stresses in a specific load case, as opposed to using an isotropic material for
example.
Based on these arguments, as well as the information found in the prestudy, it is concluded that
a sandwich structure would be the most beneficial, and also the most interesting to build. In
this section the procedure of choosing material for the core and the skins of the sandwich will
be described. The sandwich material consists of a weaker core covered by stiffer outer skins on
both sides of the core. The skins act as the load carrying element for global bending and axial
loads while the core is supposed to carry the major part of shearing loads as well as keeping the
two skins separated. The distance between the skins is a key element of a sandwich structure
since the distance increases the moment of inertia and strength of the sandwich section. The
principle is similar to that of a common I-beam.
The benefit of the sandwich structure is, as mentioned previously, its low weight to stiffness
ratio in the bending load configuration. The main drawback of the sandwich structure is its
limited ability to carry out of plane loads, especially compressive out of plane loads. Since the
sandwich structure consists of different materials in the skins and core, alternatives for both of
these had to be investigated.
30
4 Theory
Figure 4.9: Sandwich constituents
The R3 class rules limits the use of classical sandwich materials such as glass and carbon fibre
for the skins and divinycell foam or (most) honeycombs, as the material for the hull and deck.
This made it necessary to look for organic materials that are also readily available for purchase.
4.3.4.1 Sandwich Core

After researching alternatives for an organic material to be used in the core of the sandwich
structure three feasible options were found. The options were the following:
• Balsa wood,
• Cork, and
• Bio-based PLA (Polylactic acid) honeycomb.
The organic honeycomb made from bio-based PLA, could be discarded after researching the
possibility of finding a supplier of this kind of material. Only one company, Econcore, that
produces the PLA based honeycomb could be found. However upon contacting the company, it
was found that they had temporarily stopped their production but that the material could be
produced on demand. The cost of ordering this material produced on demand was considered
too high for this project. There would also be a risk that the sheets of honeycomb would be so
stiff that they would affect the design of the hull shape. Still the PLA based honeycomb is a
material with interesting properties that could be a viable alternative for future projects if the
material becomes more available.
Comparing the two remaining alternatives, the balsa wood has slightly better mechanical prop-
erties than the cork, especially in compression. Another important property of the core material
is its ability to absorb the epoxy used when laminating the skins of the sandwich onto the core.
To keep the weight of the material as low as possible, it is desired to keep epoxy absorption in
the core to a minimum. According to (Castegnaro et al., 2017) epoxy absorption is higher for
balsa wood than for cork. However this is only true if the balsa wood is placed with the grains
perpendicular to the laminates, so called end-grain. By placing the balsa grains parallel to the
laminates of the sandwich, it is assumed that epoxy absorption can be reduced drastically, since
the epoxy will not be applied to the end-grain surface in this configuration. Thus both lower
epoxy absorption and better mechanical performance of the balsa wood in the longitudinal
direction can be utilised.
31
4 Theory
The main drawback of placing the grains parallel to the skins of the sandwich is the loss of
ability of the sandwich to carry concentrated out of plane loads, e.g. concentrated rig loads or
crew stepping on the deck. This however mainly concerns the deck structure. It was reasoned
that, if necessary, the grains could be placed perpendicular on the deck or in certain areas of the
deck and keep the grains parallel to the longship direction in the rest of the hull. This idea was
later discarded because no supplier of end grain balsa could be found in Europe. To determine
whether the grains should be placed parallel or perpendicular on the deck, further analysis of
loads and strength would be required through FE-simulations. Through this reasoning it was
decided that the core material of the sandwich would be balsa wood. The properties of the
balsa wood used in calculations are presented in Table 4.3 (Newaz, Mayeed, & Rasul, 2016)
Table 4.3: Balsa wood properties
Elastic [MPa] EL ET EV GLT GT V GV L νLT νT V νV L

688 32.6 32.6 72.8 12.5 72.8 0.007 0.4797 0.007
Strength [MPa] SL ST SV SLT ST V SV L

Tension 10.12 0.82 0.82 1.35 0.35 1.35
Compression 0.71 0.71 0.71 - - -
Density [kg/m3] 150
Where L denotes the property of the material in the fibre direction, T denotes the property
of the material across the fibres in the transverse direction and V the property of the material
across the fibres in the vertical direction. Combining these notations (LT, TV and VL) will give
the properties of the material in all three planes. These notations will be applied to describe the
properties of all non-isotropic materials in this project i.e flax epoxy laminate and balsa wood.
For compatibility with FE-software used in this project stress is denoted by S in this project.
4.3.4.2 Sandwich Skin

When studying previous boats built with sandwich structures in the 1001VELAcup it was
found that a common choice of material for the skins was a laminate made of flax fibre weave
and epoxy. This material seemed interesting, and some time was spent finding literature with
data and material characteristics. It was found that three main organic fibres were the flax,
jute and hemp fibres, where the flax fibres seemed to have the best mechanical properties for
the current application (Castegnaro et al., 2017). In order to limit the work spent on finding
alternative fibres, it was quickly decided that flax would be the fibre component of the sandwich
laminate. One major contribution to this decision was the fact that material testing of the skin
laminate would have to be made, a practical exercise that would be of interest for the whole
project group. However, shortly after choosing material another paper written in connection
to the 1001VELAcup containing all the material data needed for FEM simulations was found
(Pitarresi, Tumino, & Mancuso, 2015). This data was also given for different construction
methods, one of which was intended to be used for the construction of the boat covered in this
report. Therefore, the need to do material testing disappeared and the FEM simulations could
commence immediately. In Table 4.4 the properties of a flax epoxy laminate, manufactured in
a vacuum bagging process, are presented. The properties are obtained by contacting Bcomp
Ltd., the manufacturer of the flax fibre used in this project.
4.3.4.3 Sandwich Epoxy

The epoxy chosen to be used for the lamination is the FormuLITE 2501A resin, combined
with FormuLITE 2002B hardener. According to the manufacturer, these are high-performance
32
4 Theory
Table 4.4: Flax/epoxy laminate properties
Elastic [MPa] EL ET EV GLT GT V GV L νLT νT V νV L

31981 3239 3239 2552 - 2552 0.35 - 0.35
Strength [MPa] SL ST SV SLT ST V SV L

Tension 383 22 22 29 - 29
Compression - - - - - -
Density [kg/m3] 710

Area density [g/m2] 150
Thickness [mm] 0.21
Fibre ratio 0.51
bio-based products made from high amounts of renewable resources including Cashew Nutshell
Liquid (Cardolite, 2017). This specific property of the epoxy is of special interest to the project
due to the 1001VELAcup rules that demand a high level of bio-based construction components.
4.3.5 FEM Analysis Set Up
The finite element analysis (FEA) was carried out in the FE-software Abaqus CAE. The geome-
try to be analysed, i.e. the hull surfaces and stiffener geometry, was imported into the software
as a 2D shell structure. Since the hull structure can be considered to consist of thin plates, plane
stress can be assumed which allows for the FE-analysis to be performed on 2D-shells. After
importing the geometry, it was assigned material properties, based on the conclusions drawn
in Section 4.3.4. These properties, specifically the composite layup and balsa fibre directions,
were varied systematically to evaluate the most efficient construction in terms of strength versus
weight, with the goal of reaching the lightest possible structure while still being able to handle
the applied loads.
A mesh is then generated, shown in Figure 4.10, for the structure, where the goal is to min-
imise the number of distorted elements, with sufficient resolution to obtain reasonable results.
The mesh element geometry is of the quad-dominated type with a nominal element size of 30
millimetres. This size was chosen based on the shape of the internal structure, in order to cover
smaller details such as stiffener beams.
4.3.5.1 Element Type

A question raised before FE-simulations were performed was what type of element that should
be used in the simulations. The simulations can be performed with either shell or solid elements,
where the latter requires significantly more computation time and is significantly more complex
to model since it requires the skins and the core to be modelled separately and then assembled.
This also makes varying the number of plies in the layup very tedious. However the concern was
that the less complex shell elements would not capture shear stresses in the sandwich structure
in a sufficient way.
To determine the type of element to be used in the FE-simulation of the hull structure, tests
on simple plate geometries were performed. The shear stresses in the core of the sandwich were
compared for S4R shell elements and SC8R solid elements. According to (ABAQUS/Explicit,
2017) both of these are viable choices when modelling a sandwich structure. In a sandwich
structure, the core is supposed to carry the major part of the shearing load which is why the
core is chosen for comparison between shear stresses for the different element types in this case.
In this procedure, a plate with dimensions 1000 × 1000 mm is modelled once with shell elements
33
4 Theory
Figure 4.10: The meshing of the hull surface
and once with solid elements. The same composite layup, a sandwich structure with skins made
of flax/epoxy laminate and a core made of balsa wood, is assigned to both plates.
Laminate plate layup : [−45 0 0 90 c010 ]s
In this layup, c010 denotes the balsa core of thickness 10 mm with the fibres aligned to the
0-direction. The boundary condition simply supported along two edges and a uniform pressure
of p0 = 0.001 MPa, are applied in both models. Finally, the models are meshed with quadratic
elements with a size of approximately 30 × 30 mm. In the solid model, 3 elements are stacked
through the thickness. One element each for the skins and one element for the core.
Below, the results of the computations can be seen in Figure 4.11a for the S4R shell elements
and Figure 4.11b for the S8CR solid elements. Comparing the plots of the shear stresses in the
figures, it can be seen that similar stress distributions are achieved. A minor modelling mistake
caused the coordinate system in one of the models to be reversed, why one the plots is shown
upside down. As for the magnitudes of the shear stress, the following results are obtained for
the shell elements and the solid elements respectively.
S4R: SLT,max = 0.028 MPa SLT,min = −0.028 MPa
S8CR: SLT,max = 0.020 MPa SLT,min = −0.022 MPa
According to these results, using the shell elements will give larger shear stresses in the core
whilst the stress distribution will be roughly the same. It was therefore decided that S4R
elements are to be applied in the FE-simulations of the hull structure. The S4R elements
will give somewhat conservative results whilst significantly reducing modelling complexity and
computation time, therefore it is considered to be an economic and reliable option.
34
4 Theory
(a) Core with S4R (b) Core with SC8R
Figure 4.11: Shear stresses in core
4.3.5.2 Load Cases

To evaluate the strength of the hull structure, one needs to know the loads acting upon it. In
this project the loads are divided into two main groups: hydrostatic and hydrodynamic pressure
acting on the hull, and forces from the rig and its attachments to the deck.
The total hydrostatic and hydrodynamic pressure load acting on the hull below the waterline
is determined from the pressure distribution obtained from the flat water CFD simulation of
the hull at its maximum speed as predicted by VPP simulations Section 4.4. Since the pressure
distribution from the CFD simulations does not take any dynamic effects such as slamming into
account, the pressure calculated in the CFD simulations is multiplied with a dynamic factor
calculated from the ISO-standards for small craft design (ISO, 2015). In this case the dynamic
factor for multihull sailing craft was selected since the multihull is sailed on even keel in a similar
manner as the skiff. The dynamic factor is calculated using Equation (4.9).
3.086L2W L
kDY N = where 1.2 ≤ kDY N ≤ 2.4 (4.9)
∆0.66
With numerical values, length of waterline LW L = 4.6 m and estimated displacement of ∆ =

250 kg in Equation (4.9) the dynamic factor is calculated to kDY N = 1.7. The pressure obtained
from flat water CFD simulations is pf w = 0.0015 MPa, which is a representative mean value of
the pressure distribution on the hull. Equation (4.10) is used to calculate the dynamic pressure
acting on the hull.
pDY N = kDY N · pf w (4.10)
The final hull pressure, including dynamic factors, is calculated to be pDY N = 0.00255 MPa.
This should be interpreted as the sum of the hydrostatic and hydrodynamic pressures given
in the CFD simulations, multiplied with the dynamic coefficient kDY N . This means that the
final result includes the hydrostatic pressure and the hydrodynamic pressure, with the dynamic
coefficient also taken into account. This pressure is applied to the entire wetted surface of the
hull in the FE simulations. This can be considered as a conservative assumption since it can
be reasoned that hydrodynamic pressure from slamming and waves will not be acting on the
entire wetted surface at a given instance. On the other hand, from slamming, the craft could
take air at a wave crest, completely leaving the water surface, and thereafter land flat on the
water surface again. In that case, the hydrodynamic pressure would in fact act on the entire
wetted surface. Without further investigations, it is impossible to evaluate if such an event may
or may not occur. It was therefore decided to let the hydrodynamic pressure act on the entire
wetted surface and thereby, at least to some extent, design against such an event.
When performing FE-simulations, two load cases were considered. The first being the case
35
4 Theory
when the crew is hiking in the trapeze as in Figure 4.12b, and the second when the crew is
positioned on the rack without trapeze. The two load cases will hereafter be referred to as "crew
in trapeze" and "crew on racks" respectively. In both cases, the same hydrodynamic pressure,
as calculated in Equation (4.10), is acting on the hull bottom. All forces from the sails, the rig
and crew need to be transferred to the hull structure. The internal stiffening structure and hull
plating need to be designed with this in mind.
Rigging loads are calculated by assuming that the rigging forces and moments are in inter-
nal equilibrium at all times with one exception: the righting moment from the crew, that is
countered by heeling moment that derives from the wind force in the sails.
To determine the loads acting on the hull from the rig, expertise from external resources is
utilised. The mast manufacturer and supplier of rigging in this project, Seldén Mast, provided
estimated loads from the rig designs that were sent to them. The results from Seldén Mast rig
load calculations were then balanced to fulfil the above assumption concerning force equilibrium,
still with the above mentioned exception.
To accomplish this equilibrium, further simplifications are made. The first being that all forces
engage at the same z-coordinate, that is all forces engage on a flat deck. In reality, the deck will
have a slight camber giving slightly different vertical coordinates (z-coordinate) to the rigging
loads. In the load case crew on racks, reaction forces at the rack supports are assumed to be
evenly distributed between all four supports. To properly evaluate the reaction forces at the
racks, the position of the crew on the racks and more detailed specification on the rack design
need to be known, see Section 5.3.2. It was decided that evenly distributed reaction forces
would be a sufficient approximation at the moment and that a more careful evaluation of the
hull and rack interaction at the supports would take place when more was known about the
racks.
The two load cases are tabulated in Tables 4.5 and 4.6.
Table 4.5: Load case crew on racks
Crew on racks Coordinate [m] Force [N] Moment [Nm]

Component Point x y z Fx Fy Fz Mx My Mz
Forestay 1.70 0 0 416 -183 -1491 0 -2534 -311
Shrouds -0.60 -1.05 0 -205 -457 -2577 2705 1546 397
Rigging
Mast step 0 0 0 -211 608 4441 0 0 0
Main sheet -2.65 0 0 0 32 -373 0 989 -86
Fore 0 0 0 0 0 -1288 0 0 0
Rack inner supports
Aft -2.65 0 0 0 0 -1288 0 3414 0
Fore 0 -0.60 0 0 0 1288 -773 0 0
Rack outer supports
Aft -2.65 -0.60 0 0 0 1288 -773 -3414 0
P
0 0 0 1159 1 0
36
4 Theory
Table 4.6: Load case crew in trapeze
Crew in trapeze Coordinate [m] Force [N] Moment [Nm]

Component Point x y z Fx Fy Fz Mx My Mz
Forestay 1.70 0 0 819 -229 -2794 0 -4750 -389
Shrouds -0.60 -1.05 0 -10 -19 -91 96 55 22
Rigging Mast step 0 0 0 -146 455 5511 0 0 0
Main sheet -2.65 0 0 -467 193 -1065 0 2823 -511
Trapeze -1.20 -2.05 0 -197 -400 -1560 3198 1872 883
P
-1 0 1 3294 0 5
One should note that in the trapeze load case, see Table 4.6, the weight of the sailors is carried
by the trapeze itself which allows for the racks and the shrouds to be unloaded. This is the
ideal case. In reality, the racks will carry some of the crew’s weight since the crew needs to
transfer some of their body weight to the racks to be able to keep their balance in the trapeze.
In these calculations, this is neglected and the racks are assumed to be unloaded in the trapeze
load case.
As for the crew on racks load case, see Table 4.5, the opposite is true. The trapeze is unloaded
because the crew is disengaged from it and the racks and shrouds will have to carry the weight
of the crew.
Another point to note is that, in FE-simulations and other calculations, force components of the
trapeze load in Table 4.6 is to be added to the mast step force components, since the trapeze is
attached directly to the mast the load from the trapeze has to be countered by the mast step. As
an example the z-component of the mast step should be: Fz,mast step = 5511 + 1560 = 7071 N,
in the FE-simulations.
P
The sum of moments around the x-axis, Mx makes up the total righting moment, RM . From
Table 4.5 and 4.6 this is found as:
For the crew on racks load case: RMR = 1159 Nm

And for the crew in trapeze loadcase : RMT = 3294 Nm
Because the righting moment from the hull is negligible compared to the righting moment from
the crew due to the low water plane area moment of inertia and the small heeling angles the
craft operates with, it can be approximated with Equation (4.11).
RM
g = mc y cos θ (4.11)
In Equation (4.11), mc is the mass of the crew, y is lever arm from the centre line to the centre
of mass of the crew and θ is the angle from the water surface which the crew is hiking with.
Setting these to mc = 150 kg, θ = 0◦ and yR = 1.05 m and yT = 2.05 m for the racks and
trapeze loadcase respectively the following is obtained with Equation (4.11).
For the crew in racks load case: RM

g R = 1562 Nm
For the crew in trapeze load case: RM

g T = 3050 Nm
It can be seen that this approximation fairs quite well with the righting moments calculated
37
4 Theory
in Table 4.5 and 4.6. This indicates that the loads in the load cases are a sufficiently accurate
estimation of reality. It was decided to proceed with loads the crew in rack and crew in trapeze
load cases in the FE-simulations.
4.3.5.3 Boundary Conditions

If a boundary condition is introduced at a point in an FE-model subjected to external loading,
a reaction force is also introduced at that same point to fulfil Newtons 3rd law. In the case of
the skiff in this project, introducing a boundary condition on any point, edge or surface means
that a reaction force is also introduced at that location. This due to the fact that the load
cases calculated in Table 4.5 and 4.6 are not in equilibrium since the righting moment is not
countered by any heeling moment and the hydrostatic pressure is not countered by any reaction
force. If for example the transom was to be fixed and the trapeze load case to be applied
on the model, a reaction force and reaction moment would engage at the transom surface to
counter the righting moment and hydrostatic pressure. In reality, no reaction force or reaction
moment engages at the transom surface during operation therefore such a model would be a
poor approximation. Similarly, if any other geometry in the model is fixed or simply supported
the same argument can be reasoned. To get around this, inertia relief is introduced. Inertia relief
is used to balance external loading into equilibrium by applying rigid body accelerations to the
model (ABAQUS/Explicit, 2017). This means that righting moment and hydrostatic pressure
will be countered by rigid body accelerations instead of reaction forces and reaction moments.
Inertia relief gives a better approximation than introducing ordinary boundary conditions, thus
it was decided that inertia relief would be applied in the model.
4.3.5.4 Post Processing

Due to the numerous design options available, a systematic way of evaluating each option
is needed. A well designed hull structure requires sufficient strength to withstand both load
cases, sufficient stiffness to keep deflections low during operation and a low mass. Therefore
key parameters to be evaluated for each design iteration are stress magnitude and direction,
deflections and the mass of the model.
In the performed FE-simulations, perfect bonding between skin/core interface and between
the interfaces in the skin laminate is assumed. The deflection is thereby constant through
the thickness of the sandwich structure. In contrast to the stress, that varies with the fibre
directions through the thickness of the layers (Agarwal, Broutman, & Chandrashekhara, n.d.).
Because of the previous decision to model with shell elements, out-of-plane stress cannot be
captured by the solution. This means that no method exists to evaluate failure modes such
as delamination. A way of circumventing this problem is to create local solid models in areas
of the global model where large out of plane shearing is expected and from these local models
evaluate the risk of delamination. However, even though the ambition to create such models
existed from the beginning of the project, the time required to finalise the global model proved
this ambition to fall short.
4.4 Velocity Prediction Program (VPP)
A Velocity Prediction Program (VPP) is used to determine the performance of sailing boats by
using the properties of hull, rig, sails, and appendages. It calculates the boat speed for different
wind speeds and heading angles by solving force and momentum equations. The VPP used in
this project, originally created by Professor Kai Graf at the University of Applied Sciences in
Kiel, is modified to model the skiff’s properties.
38
4 Theory
4.4.1 Input and output
The input to the VPP is specified in an input file and defines the main properties of the skiff.
The input is divided into three different parts, the aerodynamic properties, the hydrodynamic
properties, and the geometric properties. The aerodynamic properties are defined as sail co-
efficients, used to determine the forces from the sails. The hydrodynamic properties are the
resistance curve of the hull for different speeds and the lift and drag curves for the centre-
board and rudder. The geometric properties are the main particulars of the hull, sails, and
appendages. The input file also specifies which wind speeds and which wind angles that should
be considered by the VPP.
The VPP calculates the boat speed, leeway angle, the lever arm of the crew (T CGcrew ), and
necessary depowering of the sails. The results are presented in a polar plot with the radius as
speed and the angle as the true wind angle. One curve per considered wind speed is plotted.
The calculated leeway angles, T CGcrew , and depowering of sails are presented in a table for all
considered wind speed and wind angles.
4.4.2 Solved equations
There are three equations considered in the VPP, a force equilibrium in the x-direction, which
is defined as the direction of motion. A force equilibrium in the y- direction, which is defined as
the horizontal direction perpendicular to the x- axis, the z-axis is defined upwards. The third
equation is a moment equilibrium around the x-axis. The equations are shown in Equation
(4.12). The forces are assumed to be in equilibrium which means that there are no accelerations
of the hull in the considered directions. It is assumed that the heeling angle can be set to zero,
which will result in the best performance of the skiff.
Fx = Fx,aero − Fx,hydro = 0
Fy = Fy,aero − Fy,hydro = 0 (4.12)
Mx = Fy,aero · (zCE + zCLR ) − mcrew · g · T CGcrew = 0
In Equation (4.12), Fx,aero and Fy,aero are the resultant forces from the sails in the x and y
direction respectively. These forces depends on the apparent wind speed (AWS), apparent wind
angle (AWA), the leeway angle (α), and the lift and drag coefficients of the sails. The lift and
drag coefficients depends on the type of sails and the AWA. The methods used to calculate the
forces from the sails are are explained further in Section 4.4.4.
39
4 Theory
𝐹𝑥,𝑎𝑒𝑟𝑜
𝐹𝑦,𝑎𝑒𝑟𝑜
𝑧𝐶𝐸
𝐹𝑥,ℎ𝑦𝑑𝑟𝑜
𝑧𝐶𝐿𝑅
𝑇𝐶𝐺𝑐𝑟𝑒𝑤 𝐹𝑦,ℎ𝑦𝑑𝑟𝑜
𝑚𝑐𝑟𝑒𝑤 · 𝑔
(a) Driving and resistance forces
(b) Heeling and righting forces
Figure 4.12: Simplification of forces acting on the skiff considered in the VPP
Figure 4.12a and 4.12b are simplified as the leeway angle is not taken into account, all lateral
forces in both Figure 4.12a and 4.12b therefore have a force component perpendicular to the
plane of the paper.
The forces Fx,hydro and Fy,hydro are the hydrodynamic fores on the skiff. Fx,hydro consists of
three parts, the resistance of the hull, the resistance of the centreboard and the resistance of the
rudder. The resistance of the hull depends on the speed and the shape of the hull, how this force
is obtained is explained in Section 4.2. The resistance of the centreboard and rudder depends
on the speed of the hull, the leeway angle, the area and aspect ratio, and the lift and drag
coefficients of the centreboard and rudder respectively. Since there is no speed or acceleration
in the y-direction, the forces in this direction will be caused by the speed of the skiff and the
leeway angle. Fy,hydro will have three components, caused by the hull, the centreboard, and the
rudder. The force due to the hull will be small compared to the other two. The forces on the
centreboard and rudder are explained more thoroughly in Section 4.4.3.
The moment equation in Equation (4.12) is calculated in the vertical centre of effort of the
appendages, (zCLR ). Since there is no force from the hull in the transverse direction, the
heeling moment will only be caused by the side force from the sails, and the righting moment
will be created by the crew hiking in the trapezes. zCE , is the vertical centre of effort of the
sails, zCLR is the vertical centre of effort of the appendages, mcrew is the mass of the crew,
g is the earth gravitational constant and T CGcrew is the transverse centre of gravity of the
crew. If the heeling moment is larger than the maximum righting moment created by the crew,
the VPP will start to depower the sails. Depowering of the sails can be done in two ways, by
twisting the sails and by flattening the sails. The VPP will start depowering by twisting the
40
4 Theory
sails. Twisting of the sails is implemented in the VPP as a factor that decreases the lift and
lowers the centre of effort. The twisting factor can lower the centre of effort by maximum 25%.
If the sails need to be depowered more, the VPP will start flattening the sails. Flattening of the
sails is implemented as a factor that effects the lift coefficients of the sails. It can vary between
1, where there is no influence on the lift, and 0, where there is no lift at all, This also decreases
the induced drag according to Equation (4.17).
4.4.3 Centreboard and Rudder
The purpose of the centreboard is to create a counter force to the side force generated by the
sails. This side force is created by the speed of the hull, creating an angle of attack for the flow
over the centreboard. The rudder’s main purpose is to create a moment that is used to turn the
boat when steering. The rudder also supports some of the side force and is thereby unloading
the centreboard. The forces on a foil, like a rudder or centreboard, are governed by Equations
(4.13).
1
FL = V 2 ρACL,3D
2 (4.13)
1 2
FD = V ρA(CD,2D + CDI )
2
The lift, FL and drag, FD , depend on the speed of the foil, V , the planform area, A, density of
the water, ρ, and the coefficients CL,3D , CD,2D , and CDI . These coefficients depend on several
factors, such as Reynolds number, wing section, and angle of attack.
4.4.3.1 Planform
When designing the planform for a centreboard and a rudder, the aspect ratio (AR), is one of
the most important parameters. The lift coefficient, CL,3D , and the induced drag coefficient,
CDI , are both affected by the aspect ratio according to Equation (4.16) and (4.17). The aspect
ratio is the ratio between the draft, TK , and the mean chord, C̄, and can be written as
TK TK 2
AR = = (4.14)
C̄ A
Where TK is the draft of the centreboard or rudder and C̄ is the mean chord.
C1 + C2
C̄ = (4.15)
2
When AR is increased the lift coefficient, CL,3D , is increased according to Equation (4.16),
where CL,2D,1◦ is the two dimensional lift coefficient for the wing section for a angle of attack
of one degree, and α is the leeway angle, approximately 2◦ to 3◦ in this case.
CL,2D,1◦
CL,3D = ·α (4.16)
2
1+
AR
The drag coefficient, CDI , is decreased with higher AR, as shown in Equation (4.17), but also
increased with higher lift coefficient.
CL,3D 2
CDI = (4.17)
π · AR
41
4 Theory
Figure 4.13: Description of planform
To maximise the performance, the CL,3D should be as large as possible and the CDI should be
as low as possible. In practice, this means that the draft of the centreboard should be as high
as possible, but as mentioned in Section 4.3.3, it is restricted due to structural reasons.
The tip shape of the centreboard and the rudder is not as important as the aspect ratio, but it
is still something that is investigated. A v-shaped tip is the best shape overall, in comparison
with a round, square and bulb tip (Larsson et al., 2014). By having a v-shaped tip, the AR will
be lowered by 19 %. Although the v-shaped tip has an AR that is 19% lower, the benefits of
the shape are bigger than the disadvantages (Larsson et al., 2014).
4.4.3.2 Wing Profiles

It is important to have as low profile drag as possible. One way to accomplish this, is to keep
the flow laminar around the profile for as long as possible. This means that the transition from
laminar to turbulent flow must be as far back as possible. By having the pressure minimum
far aft, the flow can be accelerated over a longer distance and the transition from laminar
to turbulent flow can be delayed. The location of the pressure minimum is governed by the
location of the maximum thickness, because of this, a wing section with the location of maximum
thickness far aft would be beneficial.
The main difference between the NACA four series which have a quite simple geometry, and
the six series which has a more complex geometry is that the six series profiles are designed to
have laminar flow along the wing section. Larsson et al. (2014) states that the NACA six series
is a good choice as the wing profile for a centreboard.
It is typical for the NACA 6 series to have low profile drag and more or less constant drag for
low angles of attack, but at a certain angle of attack, the drag increases quickly to a higher
level, from which it increases more moderate with the angle of attack. Such a profile are said
to have a drag bucket. This due to the shape of the graph created when plotting drag as a
function of angle of attack, see Figure 4.14. Comparing the NACA 4 digit series profile with
42
4 Theory
the 6 series profiles, both the 6 series profiles have a bucket shape while the 4 digit series have
a does not have that.
V ·L
Re = (4.18)
ν
The Reynolds number effects the lift and drag coefficients of a foil. The Reynolds number is
calculated according to Equation (4.18) and for a foil like a centreboard or rudder on a skiff
designed for speeds around 3.5m/s the Reynolds number will be between 5 · 105 and 106 . In
this range, the changes in properties can be considered small. The graphs presented in Figure
4.14, 4.15, and 4.16 are calculated for a Reynolds number of 8 · 105 using XFOIL 6.99.
10-3 Profile Drag vs angle of attack

10
NACA 63012
9.5 NACA 65012
NACA 0012
9
8.5
Drag coefficient C D
7.5
6.5
5.5
5
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Angle of attack [°]
Figure 4.14: Profile drag for different profiles
In Figure 4.14, all three profiles have a maximum thickness of 12 % of the cord length. The
effect of location of maximum thickness is important, and is visualised in Figure 4.14. Studying
the blue and red line, where the the red line has the location of maximum thickness more aft,
the resistance at small angles of attack is lower but the drag bucket is slightly more narrow and
steep than for the blue line.
The location of maximum thickness has effect on the shape of the drag bucket, although the
effect of changes in thickness are more significant than the effect of location. A large thickness
will cause a wider but but more shallow bucket than a thinner profile, which will have a more
narrow but deeper drag bucket, see Figure 4.15. A location of maximum thickness more aft
will cause a slight decrease in resistance for small angles of attack, although the bucket shape
will be more extreme so the drag at large angles of attack will be larger than for a section with
maximum thickness located more forward.
43
4 Theory
10-3 Profile Drag vs angle of attack

10
NACA 65018
9.5 NACA 65012
NACA 65010
9
8.5
Drag coefficient C D
8
7.5
6.5
5.5
5
0 1 2 3 4 5 6
Figure 4.15: Profile drag for different thicknesses
The purpose of the centreboard and rudder is to generate lift. The lift coefficients of symmetric
NACA profiles do not differ much from each other and the connection between lift and angle
of attack is highly linear for small angles of attack. The proportionality coefficient, CL,2D,1◦ ,
can therefore be approximated as the two dimensional lift coefficient for one degree, and can
approximately be stated as 0.1 for symmetric NACA profiles. Figure 4.16 shows that CL,2D,1◦
should be between 0.08 and 0.1 for all five wing sections, and that the relationship between lift
coefficient and angle of attack is linear. This means that for the design of the centreboard and
rudder, the lift will be approximately equal independent of the wing section while the drag of
the appendages to a large extent can be optimised by choosing the right wing section.
Lift vs angle of attack

0.6
NACA 63009
NACA 65010
0.5 NACA 63012
NACA 65012
NACA 0012
0.4
Lift coefficient C L
0.3
0.2
0.1
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Figure 4.16: Lift for different profiles
44
4 Theory
4.4.4 Sail Set Up
(Larsson et al., 2014), the sails can be considered wings in the same way as as the appendages,
but with very low thickness. The presence of a jib in front of the main sail effects the flow over
the main sail and thus also the force on the sails. With the presence of a jib, the main sail
will be unloaded, compared to a case with no jib. The opposite is true for the jib. With the
presence of a mainsail, the jib will be more loaded than without a main sail. Overall, the force
from the sails are much larger with the presence of a jib.
4.4.4.1 Sail coefficients

The lift, drag, and induced drag of the sails are governed by Equation (4.13), the same equations
as the centreboard and the rudder. Using the coefficients governed by Equations (4.16) and
(4.17). However, the speed term V, should be the wind speed that the sails encounter, the
apparent wind speed (AWS). The AWS is calculated according to Equation (4.19).
q
AW S = (T W S · cos (T W A) + V · cos (α))2 + (T W S · sin (T W A) − V · sin (α))2
T W S · sin T W A − V · sin (α) (4.19)
AW A = arctan
T W S · cos (T W A)V · cos (α)
The forces from a sail can, as mentioned earlier, be calculated using classical wing theory,
according to Equation (4.13). The obtained forces, FL and FD , are orthogonal where FL is
perpendicular and FD is parallel to the wind acting on the sail, this is shown in Figure 4.17
where the x axis is the longitudinal direction of the skiff and the y axis is the transverse direction.
𝐹𝑦,𝐿
𝐴𝑊𝐴 + 𝛼
𝐹𝐿
𝑥 𝑦
𝐴𝑊𝐴 + 𝛼 𝐹𝑥,𝐿
𝐹𝐷
𝐴𝑊𝐴 + 𝛼
𝐹𝑥,𝐷 𝐹𝑦,𝐷
Figure 4.17: Forces acting on a sail
The resulting forces of FL and FD , in the x and y direction, will give Fx,aero and Fyaero , for
each sail in Equation (4.12), are obtained using the calculated FL , FD , and using the apparent
wind angle (AWA), calculated in Equation (4.19), and leeway angle α, according to Equation
(4.20).
Fx,aero = Fx,L − Fx,D = FL · sin (AW A + α) − FD · cos (AW A + α)

(4.20)
Fy,aero = Fy,L − Fy,D = FL · cos (AW A + α) − FD · sin (AW A + α)
45
4 Theory
The coefficients used for force calculations of the sails in this project are the coefficients used by
the Offshore Racing Congress (2016) (ORC). These sail coefficients are used in the ORC VPP
software to calculate handicaps for racing yachts. The ORC uses different sets of coefficients
depending on the type of sail. The sail coefficients depend only on the angle of attack of the
sails. The sail coefficient used in this project are displayed in Figure 4.18a and 4.18b.
CL vs CD vs
1.6 1.4
Main sail
Main sail
1.4 Jib
1.2 Jib
1.2
1
Drag coefficient C D0
1
Lift coefficient C L
0.8 0.8
0.6 0.6
0.4
0.4
0.2
0.2
0
-0.2 0
0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180
Angle of attack [°] Angle of attack [°]
(a) Lift coefficients (b) Drag coefficient
Figure 4.18: Lift and Drag coefficients as a function of the angle of attack α
4.4.4.2 Aspect Ratio

To reduce the induced drag in the sail, it is desirable, as for the rudder and centreboard, to have
as high aspect ratio as possible. Larsson et al. (2014), discuss the importance of the height of
the attachment point of the jib. Having the attachment point at the full mast length increases
the driving force compared to having it further down, while it does not have any effect on the
side force of the sail.
A limitation to an extremely high aspect ratio sail is that it will have very low cord length.
This means that the size of the mast section will be large compared to the sail section, which
also means that the influence from the mast on the flow around the sail will be large. A high
mast will also need a thicker profile for structural reasons which means that the influence off
the mast will be even more significant. The mast disturbance is introduced in the VPP as a
function, depending on the ratio between mast diameter and mean cord, effecting both the lift
and drag coefficients of the main sail. A higher mast also means that the centre of effort in the
sails will be moved upwards, this will cause a larger heeling moment than a shorter mast.
46
5 Design process
In this chapter, the design process of all the components of the skiff are covered. The knowledge
and theory behind this chapter is based on the previous Chapters 3 and 4.
5.1 Hull
In the design process of the hull, modelling of the geometry and CFD simulations are done
simultaneously in order to get updates for improvements and obtain an optimised hull for
desired speeds. When the design of the outer hull shape is decided, the internal structures are
designed. This process is presented in the following sections.
5.1.1 Hull Analysis
To obtain the best hull design, meaning a hull with a low resistance, all design changes have to
be compared. This is done using CFD-software. To obtain good results, the simulations need
to run in the same way. Therefore the same range of speeds for the hulls need to be evaluated
while maintaining same displacement. The trim of the hull needs to be set for each iteration.
All the simulations are run for just half of the hull, to save computational time. See Section 4.2
for further information and how to set up the simulations. Having just half of the hull analysed
makes it needed to correct the results for the total geometry.
5.1.1.1 Simulations Speeds

The speeds of the hull have to be the same in each simulation so they can be compared. To decide
what speeds will be run, inputs from the VPP are used. Here the resistance for the 49er© is used
with the same sail area as is allowed in this competition, 33 square metres(A. 1001VELAcup,
2017). This gave speeds according to Table 5.1, which will be considered in the simulations.
The speeds are chosen in knots but has to be implemented in metres per second.
From the VPP, the maximum speed the boat is most likely to reach is approximately 5-7 knots,
in approximately equal values for wind speeds. This wind speed is chosen according to the wind
statistics, see Section 3.1 for more information. The reason a wider range of speeds is analysed,
instead of considering just the one expected for boat speed is for the estimation of resistance.
To implement the resistance values into the VPP calculations for speeds that are not evaluated
through CFD simulations, an extrapolation method would be required. Analysis of speeds not
directly interesting for CFD evaluations make it instead possible for the VPP to obtain results
through interpolation of a known range of values (Wahab, 2017). This interpolation method
results more accurate than the extrapolation, and will therefore produce more accurate VPP
predictions.
5.1.1.2 Hull Trim

One peculiarity of the 49er© is that it can easily change the waterline just by relocating the
centre of gravity. This makes it possible to have a waterline good both for low speeds, when
trimming the hull forward, and for high speeds, when moving the centre of gravity toward the
aft. This feature is transferred to the hull designed since it is based on the 49er© hull. Therefore
the simulations will be run at different trims for different speeds. To have an idea of how to trim
the hull, the longitudinal position of the centre of gravity is scaled from a 49er© for each speed.
47
5 Design process
Table 5.1: The simulated speeds for each hull in the CFD-simulations converted from knots to
metres per second.
Speed [kn] Speed [m/s]
3 1.5433
6 3.087
9 4.630
12 6.173
15 7.7167
This data is derived from resistance tests carried out on a full scale 49er© in a towingtank test
(Persson, n.d.). The positions of the centre of gravity for the simulations are set according to
Table 5.2.
Table 5.2: Location of the centre of gravity (CoG) for each speed scaled from the 49er©.
Speed [kn] CoG from FP [mm]
1 1386.1
2 1718.2
3 1976.2
4 2172.7
5 2320.0
6 2428.9
7 2509.8
8 2572.0
9 2623.9
10 2673.2
11 2726.5
12 2789.7
13 2867.7
14 2964.6
15 3083.5
16 3226.8
17 3395.9
18 3591.3
19 3812.8
20 4059.1
The values in Table 5.2 are used in the first evaluation to find the optimal hull with the lowest
resistance.
5.1.2 Hull Concepts
To obtain a starting point for designing the hull, some concepts have to be analysed. Here the
two concepts that have been analysed are presented, the transonic hull© and the scaled 49er©
48
5 Design process
hull, respectively.
When creating the models of the two concepts, some properties need to kept constant. These
are so that the hull is not changed too much making comparisons through the CFD-simulations
invalid. These properties are,
• General hull shape
• Moment of inertia of the water-plane around the symmetry axis
• The displacement of 250 kg
The moment of inertia of the water-plane and the displacement are both taken from a 49er©
Rhino 3D model. Furthermore, a weight of 150 kg was assumed to simulate the weight of the
crew, and a weight of 25 kg for the estimated weight of the rig. These values are also the
references stated in the rules as they are the weights to be added in the real test in the tank
to simulate both the crew and the rig. Therefore, by assuming 250 kg of total weight, the
estimated weight of just the hull is left to be 75 kg.
5.1.2.1 Transonic Hull©

The first hull considered is the transonic hull©, whose main characteristics are limited by three
fundamental constraints. Firstly, all the hull curves below the waterline are of degree one
(i.e. straight lines) therefore the geometry is defined by the maximum draft and beam values.
Secondly, the stability is dominated by the moment of inertia of the water-plane area, IW P ,
since the difference of height between the centre of buoyancy and centre of gravity is similar to
the 49er©, leading to zB − zG ≈ (zB − zG )49er . Also, the displacement, ∆, considered is very
similar to the design displacement of the 49er© (between 250 − 260 kg), therefore the displaced
volume ∇ is also similar. Hence, the reason for which only considering the water-plane moment
of inertia can be a valid approximation to the one from the 49er© is evident by inspecting the
hydrostatic equation for the metacentric height, GM, Equation 5.1.
IW P
GM = + zB − z G (5.1)
∇
Therefore, the maximum beam is defined by making use of the first constraint, along with a
reference value for the water-plane moment of inertia. The used reference value is the moment
of inertia of the water-plane of the 49er© hull, assumed to have a satisfactory stability and
manoeuvrability. With this, the following equation arises by equalling the water-plane moment
of inertia of a triangular section with respect to the longitudinal axis (solved for the beam) to
the desired moment of inertia (from the 49er©), see Equation (5.2). From Equation (5.2) the
beam can be solved for, see Equation (5.3). The values that are used in these equations can be
seen in Table 5.3
L
· B 3 = I49er (5.2)
48
1
48 · I49er

3
B= (5.3)
L
Thirdly, the total displacement of the skiff assumes a first value decided based on previously
built skiffs, average weight of appendages, rig and assumed weight of the sailors (∆ = 250 kg).
Therefore, given the value for both the length, beam and the desired displacement, and con-
sidering the first constraint again, the draft is already fully determined through an iterative
process until the desired displacement is met. The final obtained value is then Tmax = 0.266 m.
49
5 Design process
Table 5.3: Beam fom moment of inertia
L [m] I49er [m4 ] B [m]
4.6 1.664 · 10−1 1.202
After computing these values, the obtained transonic hull© is depicted in Figure 5.1.
Figure 5.1: Transonic hull©
5.1.2.2 Scaled 49er© Hull

The other evaluated concept is the scaled 49er©, see Figure 5.2. As implied in the name, the
created model for the scaled 49er© has proportional dimensions to the 49er© skiff and is scaled
down to 4.60 m LOA according to the regulations of the competition. The main design properties
tried to be kept from the 49er© in the new scaled-down version, along with the reasons for doing
so are:
• Maximum beam far aft, to reduce resistance
• Flat bottom in the aft, to reduce resistance at higher speeds
• Concavity above the waterline, to reduce spray along the hull sides
• Slender and v-shaped bow, to reduce slamming and pierce waves with less interference
Similarly to the transonic hull©, the design displacement and the moment of inertia of the water
plane are also kept constant.
By having these conditions satisfied, the obtained hull is depicted in Figure 5.2.
50
5 Design process
Figure 5.2: Scaled 49er© hull
5.1.3 Hull Design Optimisation
First, a decision on which of the two concepts, among the transonic hull© and the scaled 49er©
hull, should be further investigated and optimised has to be taken. Hydrodynamic computations
provide as a result that the transonic hull© should be disregarded. This is because of its higher
resistances for all speeds when compared to the scaled 49er©, see Table 5.4 for the difference.
It must be noted the fact that the high resistance achieved by the transonic hull© is probably
due to its sharp edges that cause a large separation. Smoother edges between sides and bottom
would have probably determined a drop in terms of resistance. Such aspect has not however
been considered and the scaled 49er© hull was chosen for further optimisation.
Table 5.4: Resistance Comparison Between Transonic Hull© and Scaled 49er© Hull
Speed [kn] Transonic Hull© [N] Scaled 49er© [N]
3 15.32 8.29
6 79.81 60.71
10 261.79 140.60
15 688.697 337.80
The scaled 49er© hull is optimised adjusting its hull lines until an overall satisfactory shape is
obtained. This means that the overall dimensions and main curvature are set. When the major
hull lines are decided, the work continues optimising the hull with smaller changes. This means
that the hull is faired to get a smooth hull shape.
When the first model of the scaled 49er© is created, a flared freeboard characterises the hull.
This is a feature that is later removed to minimise the weight. Making the freeboard vertical
creates a clear transition between the bottom and the freeboard. This makes the water non-
slipping along the hull. As can be seen in Figures 5.3a and 5.3b the straight freeboard gives a
spray out from the chine, creating lift and decreasing the resistance. This feature is brought
within the following optimisation of the hull.
Unlike a displacement vessel, a racing skiff has to adapt and is affected by the conditions
surrounding it. Design optimisation for a single speed is therefore limiting in the performance
possibilities of the hull and a compromise between not-so-optimal solutions must be achieved.
It is verified for the present skiff, through its sailing conditions, that the most likely speeds to be
51
5 Design process
(a) Dry straight freeboard (b) Wet flared freeboard
Figure 5.3: Straight vs flared freeboard sprays differences
reached range between 3 and 12 kn, corresponding to Froude numbers between 0.23 and 0.91.
Wind statistics report guides toward a most probable ship speed around 6/7 kn and results for
such event are therefore be considered as more relevant in the decision process.
Main rule to be followed when positioning the maximum beam and draft is that both the
buttocks and the waterlines should be as straight as possible. Very straight aftbody is most
desired higher the speed is. This configuration helps reducing the low pressure caused by convex
lines that will otherwise lead to undesired trimming conditions by the stern and an increased
resistance. Similar higher resistance is experienced in case low pressure acts on the side of the
hull, causing wave troughs that also increase the resistance.
Geometrically, a hull for high speeds has to be designed positioning the maximum beam at the
stern and the maximum draft as far forward as possible. An optimum design for low speeds
would instead call for more curved lines such that the hull would taper off gradually towards the
aft. A compromise between the position of maximum beam and maximum draft has therefore
to be found.
5.1.3.1 Position of maximum beam

Through systematic variation, the maximum beam is placed at locations ranging between 70 %
and 97 % of LOA, see Table 5.5 for the variation scheme. Aft of the maximum beam the hull
will be parallel so the beam does not change after the maximum beam position. This is a feature
that does not come from the scaled 49er©, it is implemented to make the systematic changes
easier.
The result of such changes is that the further forward the maximum beam position is placed,
the more bluff but less wide the skiff hull will be. The initial hull has its maximum beam at a
position of 93 % of LOA, and does not have a parallel aft body. This has to be changed before
the systematic variation. After some simulations are run, a resistance curve can be plotted
against the position of the maximum beam, see Figure 5.4a to 5.4c.
The resistance curve is created from the computed data, marked with red stars, and there after
a two-degree polynomial curve, blue line, is fitted to better highlight the path of the analysed
data. This might not capture the path perfectly, but the resistance data has some scatter in it,
since it is solved by a numerical method, see Section 4.2 for theory behind the simulations.
Table 5.5: Positions for the maximum beam variation
% of LOA
xLOA 70 75 80 82 87 89 93 97
52
5 Design process
(a) Results for 3 kn
(b) Results for 6 kn
(c) Results for 9 kn
Figure 5.4: Max Beam position vs resistance. Red stars mark measured data; Blue line is the
fitted resistance curve
53
5 Design process
Aim of the computations is to determine a geometry that presents as low resistance as possible
and their results are plotted in Figures 5.4a to 5.4c. Main focus in the post-processing of the
results obtained is put on the range of speeds between 3 and 9 kn, while the higher speeds,
12 and 15 kn, are computed almost excusively in order to provide sufficient data to the VPP.
At speeds of 3 kn the optimal position of maximum beam is suggested to be placed between
80 and 90 % of LOA. Higher speeds suggest instead to move the position of maximum beam
toward the stern, reaching values in the range 85 − 95 % for speeds of 6 kn and even further for
9 kn. Furthermore, considering the fact that 6 kn of speed is the one most probably achieved
by the skiff, computations for speeds of 5 and 7 kn are also carried out. Choice for these two
speeds also gives the opportunity to carry out a validation comparison with other data. Such
comparison is presented in Section 5.1.4.1. The results of the computations for the full range
of speeds are presented in Table 5.6.
Table 5.6: Resistance vs beam position comparison
Speed [kn] 3 5 6 7 9 12 15
Position in %
of LOA for Resistance [N]
max beam
80 19.12 67.84 115.04 149.98 231.78 345.98 431.68
87 19.04 67.42 114.76 149.62 237.54 348.38 434.7
93 18.94 68.14 115.68 150.28 236.18 348.1 433.4
From Table 5.6 it is hard to evaluate which beam position to choose. Therefore all the values
have to be compared in some way. This is done by ranking all of the position based on resistance
for each speeds. There after the rankings are scaled by a factor depending on how probable
each boat speed is. The scale factor is chosen according to Table 5.7.
Table 5.7: Scale factors to compare which hull to choose for maximum beam position variation
Speed [kn] Scale Factor
3 0.16
5 0.23
6 0.23
7 0.23
9 0.15
12 0.00
15 0.00
The scale factors in Table 5.7 are set so they mostly privilege speeds around 7 kn, since this is
the design speed according to the VPP. Then 5 kn and 6 kn are speeds that also need to be
designed for. Therefore 5, 6 and 7 kn have the main influences on the final decision. Then the
rest is spread out equally between 3 and 9 kn since these speeds are equally probable for the
skiff to reach. For the higher speeds, 12 and 15 kn, it is not seen reasonable to take these into
consideration since the skiff is designed for speeds around 7 kn. To see the result for the scaled
ranking, see Table 5.8.
As can be seen in Table 5.8 the maximum beam position should be placed at 87 % of LOA from
the bow. After this position it should have a parallel aft body. This gives that aft of 4002 mm
54
5 Design process
the deck beam should be constant.

Table 5.8: Ranked and scaled comparison between the beam position to evaluate where to place
the maximum beam
Speed [kn] 3 5 6 7 9 12 15
Position in % Scale
SF · rank
P
of LOA for Factor 0.16 0.23 0.23 0.23 0.15 0 0
max beam (SF)
80 3 2 2 2 1 1 1 2.01
87 2 1 1 1 3 3 3 1.46
93 1 3 3 3 2 2 2 2.53
5.1.3.2 Position of maximum draft

After the variation of the beam is made, the same systematic variation is done to obtain the
position of the maximum draft. The starting position was placed of 22 % of the LOA from
the bow. Then the variation is made between 12 % and 32 % of the LOA. see Table 5.9 for
variation scheme. A similar plot for the resistance curve is made for this case but position of
the maximum draft instead of beam, see Figures 5.5a to 5.5c.
The resistance curve is created in the same way as for the analysis of the position of the maximum
beam, see Section 5.1.3.1. Which means by approximating the curve to a polynomial of degree
two. In the plots the red stars is the measured data and the blue line is the approximated
resistance curve.
Table 5.9: Positions for the maximum draft position variation
% of LOA
xLOA 12 17 22 27 32
By analysing Figures 5.5a to 5.5c, which are for speeds of 3, 6 and 9 kn, an interval for which
position that should be chosen is derived. For speeds of 12 kn and 15 kn see Appendix 8.2.5.
For 6 kn the preferred position of maximum draft is somewhere between 16 − 20 % of LOA,
according to the curve. For low speeds of 3 kn it is around 28 − 32 % and for 9 kn it is almost
the same as for 6 kn, around 18 − 22 %. One thing to take in mind when doing this analysis
is that this is for the curve which is interpolated between the measured data. Therefore Table
5.10 is created to compare all the measured data whit each other. However, this table is just
for the normal simulated speeds, without 5 and 7 kn.
Since Table 5.10 just shows the resistance values, a new table is created to compare the ranking
between the hulls with a scale factor. The scales factors can be seen in Table 5.11. The factors
is chosen according to that the 6 kn is the one that is closest to the design speed of 7 kn why
it has a quite big scale factor. For 3 kn and 9 kn there are still speeds that should be taken
into account, why they have smaller values. Again the scale factors for higher speeds are set to
zero, since such will not be reached.
According to Table 5.11 the scale factors are added to the ranking of comparison of the resistance
for draft position. Table 5.12 is created to give the results.
According to Table 5.12 the maximum draft should be positioned at 27% of the LOA from the
bow. This means that the maximum draft should be placed 1242 mm aft of the bow.
55
5 Design process
Table 5.10: Resistance vs draft position comparison
Speed [kn] 3 6 9 12 15
Position in %
of LOA for Resistance [N]
max draft
12 20.3 114.42 238.1 350.76 429.96
17 19.66 113.74 242.88 363.1 428.02
22 19.04 114.76 237.54 348.38 434.7
37 18.74 114.52 237.72 347.36 440.08
32 18.8 116.12 243.58 292.66 404.24
Table 5.11: Scale factors to compare which hull to choose for draft variation
Speed [kn] Scale Factor
3 0.25
6 0.50
9 0.25
12 0.00
15 0.00
Table 5.12: Ranked and scaled comparison between the beam position to evaluate where to
place the maximum draft
Speed [kn] 3 6 9 12 15
Position in % Scale
SF · rank
P
of LOA for Factor 0.25 0.50 0.25 0 0
max beam (SF)
12 5 2 3 3 2 3.0
17 4 1 4 5 5 2.5
22 3 5 2 2 3 3.75
27 1 3 1 1 4 2.0
32 2 4 5 4 1 3.75
56
5 Design process
(a) Results for 3 kn
(b) Results for 6 kn
(c) Results for 9 kn
Figure 5.5: Max Draft position vs resistance. Red stars mark measured data; Blue line is the
fitted resistance curve.
57
5 Design process
5.1.4 Final Hull Geometry and Hydrodynamics
From the optimisation of the hull design, the shape of the canoe body is determined. By this its
main dimensions are set from an optimal point of view. All the main dimensions are presented
in Table 5.13. Drawings for the final design of the hull shape are detailed in Appendix 8.2.5.
Table 5.13: Final hull dimensions for the skiff
Bmax Position of max beam Tmax Position of max draft

LOA [m]
canoe-body [m] [%LOA] canoe-body [m] [%LOA]
4.6 1.202 87 0.17 27
For the final canoe body the resistance is presented in Table 5.14. Also a resistance curve is
created, see Figure 5.6.
Table 5.14: Final drag along x-axis for the canoe body
Speed [kn] Resistance [N]
3 19.52
5 69.62
6 114.22
7 151.32
9 233.80
12 348.65
15 431.56
Figure 5.6: Resistance curve for final canoe-body
58
5 Design process
The resistance curve will further on be used in the VPP to get how the hull will perform in
different conditions, see Section 5.2.1.1 for performance according to the VPP.
From the CFD-simulations the wave pattern of the canoes body is simulated. This is not
taken into account when doing the optimisation since only the total resistance is taken into
consideration. However, the generated waves are one component in the resistance, so it will in
someway be taken into account but not specifically be optimised for. The final wave pattern
for 6 kn can be seen in Figure 5.7.
Figure 5.7: Wave pattern generation at 6 kn speed
5.1.4.1 Validation of Results
Being the first year that Chalmers University of Technology is taking part in a project of
this kind, no direct reference hulls nor computational results were available. Validation of the
results obtained comes from the comparison with the results presented by Mancuso, Pitarresi,
and Tumino (2017). In this paper, results for the CFD evaluations carried out on LED3 are
presented. The skiff chosen as reference was built with the same purpose of the one treated in
this text. Another fact in favour of this choice is that the reference skiff graduated champion of
the first Midwinter Indoor Race in February 2017. Table 5.15 presents a comparison between
the obtained computational results for resistance, for three different speeds and no appendages
attached, by LED3 and the ongoing project, CTH1. This comparison is not carried out on the
final design, it is made for when the position of the maximum beam is decided to see that the
results is reasonable.
Most importantly the comparison validates the modelled setup for the solver. Furthermore,
the results obtained can be considered quite satisfying. With exception made for the speed of
5knots, the hull here under discussion is shown to perform better than last year’s best skiff.
59
5 Design process
Table 5.15: Drag along x-axis for hull LED3
Speed [kn] Resistance LED3 [N] Resistance CTH1 [N]
5 67.9 69.62
6 115.1 114.22
7 168.2 151.32
5.1.5 Deck
When designing the deck, several parameters have to be considered. Such parameters are ease
of movement for sailors on the deck, aerodynamics, water on deck and stiffness given by deck
shape. Three different designs are discussed, one with a v-shaped deck downwards, a straight
deck and one with a slightly upwards cambered deck.
For the v-shaped deck the main feature with this design is that is make the movement for the
sailors easily. The angle should follow the racks in this case to make get this easy movement.
The disadvantage for this design is that there are risk for collecting water in the middle of the
V, as well as the structure will be weaker by this solution. By this reason, it will not be chosen.
Next design concept is the straight deck. This is some mix between the two other options and
does not give any specific advantage or disadvantage. This is why this is not chosen since both
of the other concept has bigger advantages that can be used.
The last design is the slightly upwards cambered deck. This has it advantages in giving stiffness
to the structural performance, see Section 5.1.6.2 for more information about it. Also the feature
that the water not will be collected on the hull is something that is preferred as well as there
are no big problems with the movement. One problem that can occur further on in the project
when designing the deck layout is that there is no place under the freeboard to collect thing in.
Like the gennaker sock. This does not seem as a big problem and with all advantages for this
design this is chosen.
5.1.6 Structural Hull Design
The structural design of the hull can be divided into three steps. Design of the hull plating
sandwich, design of the deck sandwich and design of the internal stiffening structure. In this
section the design of each of these components will be explained. Structural analysis is performed
in the FEM-software Abaqus. Details on the set up of the FE-model, elements used, boundary
conditions applied and load cases considered are found in Section 4.3.5.
5.1.6.1 Design of Hull Plating Sandwich

In this section, the process of designing the sandwich for the hull plating is described. Design
of the sandwich include determining core thickness and design of the skin laminates.
To begin, the core thickness is determined. This is done by choosing an arbitrary layup sequence
for the skin laminates. It is not of great importance which layup is chosen at this stage, only
that it is a reasonable layup and that it is held constant while the core thickness is varied.
In this case, the following layup sequence for the sandwich was chosen, where x is the core
thickness to be varied. See Section 4.3.5.1 for details on layup notation.
[−60 60 c0x ]S
60
5 Design process
By varying the core thickness in the FE-model while applying the trapeze load case described
in Section 4.3.5, the influence of core thickness on deflection and stress levels in the model could
be evaluated. The evaluation is presented in Table 5.16.
Table 5.16: Maximum stress for varying core thickness
Stress [MPa] Deflection [mm]

Core thickness [mm] SLT,max SL,max SL,min ST,max ST,min U3,max
12 0.244 39.5 -87.1 5.6 -11.0 4.04
10 0.245 41.7 -89.2 6.0 -11.2 4.54
8 0.245 41.7 -89.2 6.0 -11.2 5.21
5 0.260 43.1 -87.2 7.2 -12.0 6.71
In Table 5.16, the stress global maximum is shown for each stress component where index L
denotes the fibre direction and T is the transverse direction relative to the fibres. Index LT
denotes the shear stress. Maximum deflection for the third (vertical) direction recorded at the
mast step, is presented as well. From these values it can be concluded that core thickness
will not have a significant influence on stress levels in the sandwich structure. It is seen that
the core thickness will be the governing factor for the deflection of the structure. Thus it is
concluded that increasing core thickness will primarily add stiffness to the structure. It should
be mentioned that when these values were recorded the design of stiffening structures and deck
laminate layup were still at the conceptual level meaning the numbers are not representative for
the stress levels or deflection of the final design. The design of the internal stiffening structure
is covered in detail in Section 5.1.6.3.
A previous FE-analysis conducted on a similar boat participating in the 1001VELAcup showed
a deflection of 10 mm, recorded at the mast step (Mancuso et al., 2017). The boat in question
utilises cork as the core material, a flax/epoxy laminate in the skins and marine plywood as
the material of the internal stiffening structure. The core thickness is not mentioned. Another
boat that previously participated in the 1001VELAcup was built with an end-grain balsa core
thickness of 6.5 mm covered with flax/epoxy laminate (Castegnaro et al., 2017). In both
these cases, the structural design, performance of material and load cases considered in design
may vary from this project. However, the numbers mentioned in these reference studies are
considered to be appropriate to serve as guidelines when choosing core thickness in this project.
The core thickness chosen in this project is 8 mm. Increasing the thickness to 10 mm will
result in an increase of 20 % of the core weight while only reducing the maximum defection
with roughly 13 %. Since the deflections seen in Table 5.16 are already low compared to the
reference studies it was reasoned that, if necessary, reducing the deflection could be done in a
more weight efficient manner further on in the design work. Core thicknesses below 8 mm are
deemed to fragile to handle in the building process intended for this project, where long thin
strips of balsa is cut and attached to a male template. More details on the building process are
found in Section 7. It should be pointed out that the method intended in this project is different
than the one described in (Castegnaro et al., 2017) where the authors employed vacuum bag
moulding in a female mould with 6.5 mm end grain balsa in the core of the sandwich.
To decide which layup to use in the laminate of the skins, a total of 20 layup options are
investigated. The results of this investigation are presented in Table 5.17. In the investigation,
5 nodes of interest are chosen in the FE-model. The locations of these are shown in Figure 5.8.
For each layup tested the deflection of these nodes is measured. Stress levels are overall low for
each layup. The effects of varying layups on the stress distribution and stress magnitudes in
the hull were not analysed. The philosophy behind this is similar to when the thickness of the
core was decided. A thorough analysis of stress is instead performed in Section 5.1.6.4.
61
5 Design process
Figure 5.8: Location of hull nodes investigated
The trapeze load case is considered the worst case scenario for the hull plating, since it induces
the largest deflection and stress magnitudes, which is also why it is the only load case analysed
in this section. A core thickness of 8 mm was already decided earlier in this section. The
thickness of each ply is 0.21 mm according to flax/epoxy laminate properties in Table 4.4.
The nodes that are investigated are chosen based on if they are located at a critical location
or if the magnitude of deflection in a certain area is large. Deflection at the location of the
fore stay and the mast step (nodes #1236 & #1182) in the vertical direction are two of the
largest deflections on the hull. It is important that the hull layup is stiff enough to keep these
deflections under control. Furthermore, node #7896 belongs to the hull plate where bulkheads
are located furthest apart from each other in the conceptual design of the stiffening structure.
Lastly, node #3216 belongs to the aft most hull plate. In the keel, this plate has the largest
lateral deflection. At node #498 the largest deflection in the transverse direction occurs. The
hull plating collapses inwards at this location. This node is chosen to be examined to evaluate
the effect of hull layup on the transverse stiffness of the hull.
In Table 5.17 the layup chosen for the hull plating in this project is layup no. 17. The deck
plating is designed to give as large stiffness as possible to the keel plates while adding as little
weight as possible. In Table 5.17 the two most promising layups are layup no. 17 and 18.
Observing the difference in deflection of node #7896 for layup no. 17 (U3,no17 = −0.25mm) &
layup no. 18 (U3,no18 = −0.49mm), the choice of layup no. 17 over layup no. 18 is motivated
by the philosophy that hull plating should contribute more to stiffness in the vertical direction
at the keel than in the transverse direction at the side plating. The transverse deflection of node
#498 is U2,no17 = 1.91mm for layup no. 17 and U2,no18 = 1.27mm for layup no. 18. Side plating
stiffness will to a large extent be given later in design by the deck and bulkheads. Further, it
should be mentioned that the possibility to add a bulkhead at the location of node #498 to
reduce the transverse deflection at this point is possible, see design of stiffening structure sec
5.1.6.3. Other differences in deflection between layup no. 17 & 18 are considered to be so small
that they are negligible. Adding more plies improves the stiffness of the hull plating while at the
same time adding on to the total weight. In layup no. 17 four plies are stacked on the outside
laminate. Four plies is assumed to be the maximum number of plies that can be handled during
construction. In the inside laminate, three plies are stacked. There is also a possibility to add
an extra 90 degree ply on the inside of the hull to improve the hulls transverse stiffness if found
necessary in later analysis.
62
Table 5.17: Variation of the sandwich skin laminates
Outside ply no. 8 mm Inside ply no. U3,i [mm] U2,max [mm] Mass [kg]
0 1 2 3 Core 4 5 6 7 #1236 #1182 #7896 #3216 #498 Of the plates
Layup no.
1 - 0 60 -60 0 -60 60 0 - 4.75 -2.56 -0.53 2.33 1.40 10.30
2 - 0 30 -30 0 -30 30 0 - 4.71 -2.79 -0.22 2.38 2.11 10.30
3 - 0 45 -45 0 -45 45 0 - 4.69 -2.68 -0.40 2.34 1.71 10.30
4 0 0 60 -60 0 -60 60 0 - 4.27 -2.54 -0.48 2.14 1.35 11.25
5 0 0 30 -30 0 -30 30 0 - 4.32 -2.70 -0.24 2.21 1.95 11.25
6 - 0 0 90 0 90 0 0 - 4.83 -2.78 -0.38 2.38 1.50 10.30
7 - 0 30 90 0 90 30 0 - 4.78 -2.72 -0.53 2.35 1.37 10.30
8 0 0 60 -60 0 -60 60 0 0 3.91 -2.45 -0.45 1.99 1.29 12.21
9 - -60 60 0 0 0 60 -60 - 4.73 -2.63 -0.54 2.32 1.33 10.30
10 - 60 -60 0 0 0 -60 60 - 4.73 -2.63 -0.54 2.32 1.33 10.30
11 0 60 -60 0 0 0 -60 60 - 4.25 -2.53 -0.49 2.13 2.13 11.25
12 0 30 -30 0 0 0 -30 30 - 4.31 -2.70 -0.25 2.21 1.92 11.25
13 0 60 -60 0 0 0 -60 60 0 3.89 -2.44 -0.45 1.98 1.24 12.21
14 0 30 -30 0 0 0 -30 30 0 3.98 -2.61 -0.26 2.07 1.79 12.21
15 - 30 -30 0 0 0 -30 30 - 4.70 -2.79 -0.23 2.37 2.08 10.30
16 0 30 -30 0 0 0 -30 30 - 4.31 -2.70 -0.25 2.21 1.92 11.25
17 30 -30 0 0 0 0 -30 30 - 4.30 -2.69 -0.25 2.21 1.91 11.25
18 60 -60 0 0 0 0 -60 60 - 4.25 -2.52 -0.49 2.13 1.27 11.25
19 30 60 0 0 0 0 -60 60 - 4.50 -2.68 -0.54 2.24 1.52 11.25
20 0 0 30 60 0 0 60 30 - 4.52 -2.69 -0.53 2.25 1.57 11.25
Node no. Description

#1236 Forestay
5 Design process
#1182 Mast step
#7896 Largest plate in keel
#3216 Aftmost plate
#498 Side bow plate
63
5 Design process
5.1.6.2 Design of Deck Sandwich

Design of the deck sandwich is performed in a similar manner as the design of the hull plating
sandwich, only in this process the core thickness is varied together with with skin laminates,
why the procedure for the deck is different from the procedure for the hull is explained in Section
5.1.6.5. In Tables 5.18 to 5.20, the variation of the core thickness and laminate layups can be
seen. In Figure 5.9, the locations of deck nodes investigated are shown. Again, nodes at the
location of the mast step (#6309) and forestay (#1236) are chosen since these are two critical
locations with large deflections. Node #13108 is located at the midpoint of the largest plate in
the deck, enough stiffness in the vertical direction (3rd direction) is required here to allow for
crew members to operate on the deck without structural failure. The main sheet is predicted
in this design stage to be located somewhere around node #15769. The large load from the
main sheet will require high stiffness in this area. Node #2341 where the shroud placement
is predicted, belongs to the largest deflection in the transverse direction (2nd direction). This
node is chosen to evaluate the decks transverse stiffness.
Figure 5.9: Location of deck nodes investigated
In this project layup no. 3 in Table 5.18 is chosen. This means that a core thickness of 8
mm in the core of the deck is used. The gain in stiffness with core thicknesses above 8 mm is
considered too small in relation to the weight added to viable. Decreasing the core thickness
from 8 mm is not considered, since 8 mm is the minimum thickness assumed to be manageable
during construction. A maximum of four plies in each laminate is considered, for the same
reason.
The deck is subjected to the largest risk for concentrated chock loads, such as tools being
dropped on the deck, during rigging. Four plies are placed in the outside laminate to provide
as much protection against concentrated shock loads as possible. On the inside laminate three
plies are stacked (layer 5,6 and 7) and the eighth layer is reserved if it is found in later design
work that additional plies are needed. Also a fifth layer on the outside is reserved for this reason.
It should be noted that if a fifth layer on the outside is added it could reduce the possibility to
laminate the deck properly, since as already stated four layers are assumed to be the maximum
manageable amount of layers during lamination.
64
Table 5.18: Deck layup analysis with 8 mm core thickness
Outside 8 mm Inside U3,i mm U2,max mm Mass [kg]

0 1 2 3 4 Core 5 6 7 8 #1236 #6309 #13108 #15769 #2341 Of deck
No
1 - - 45 -45 0 0 0 -45 45 - 4.55 -2.71 -1.88 2.08 -0.38 6.10
2 - - 90 0 0 0 0 0 90 - 4.34 -2.68 -1.91 2.00 -0.31 6.10
3 - 90 0 0 0 0 0 0 90 - 4.17 -2.61 -1.83 1.92 -0.31 6.66
4 - 90 90 0 0 0 0 0 90 - 4.30 -2.62 -1.87 1.94 -0.31 6.66
5 - - 90 45 0 0 0 -45 90 - 4.60 -2.70 -1.93 2.05 -0.32 6.10
6 - - 60 -60 0 0 0 -60 60 - 4.67 -2.71 -1.93 2.08 -0.34 6.10
7 - - 30 -30 0 0 0 -30 30 - 4.34 -2.73 -1.84 2.05 -0.39 6.10
8 - - 45 -45 0 90 0 -45 45 - 4.78 -2.75 -1.94 2.15 -0.36 6.10
9 - - 90 0 0 90 0 0 90 - 4.53 -2.71 -1.97 2.07 -0.31 6.10
10 - - 90 45 0 90 0 -45 90 - 4.91 -2.77 -2.02 2.17 -0.31 6.10
11 - - 60 -60 0 90 0 -60 60 - 5.00 -2.78 -2.03 2.21 -0.34 6.10
12 - - 30 -30 0 90 0 -30 30 - 4.46 -2.72 -1.87 2.06 -0.36 6.10

#1236 Forestay
#6309 Mast step
#13108 Crew step area
#15769 Main sheet
#2341 Shroud
5 Design process
65
66
5 Design process

0 1 2 3 4 Core 5 6 7 8 #1236 #6309 #13108 #15769 #2341 Of deck
No
13 - - 45 -45 0 0 0 -45 45 - 4.50 -2.66 -1.84 2.04 -0.37 6.77
14 - - 90 0 0 0 0 0 90 - 4.30 -2.63 -1.87 1.97 -0.30 6.77
15 - 90 0 0 0 0 0 0 90 - 4.14 -2.56 -1.79 1.89 -0.30 7.34
16 - 90 90 0 0 0 0 0 90 - 4.27 -2.57 -1.83 1.91 -0.30 7.34
17 - - 90 45 0 0 0 -45 90 - 4.56 -2.65 -1.89 2.01 -0.31 6.77
18 - - 60 -60 0 0 0 -60 60 - 4.61 -2.65 -1.89 2.04 -0.33 6.77
19 - - 30 -30 0 0 0 -30 30 - 4.30 -2.68 -1.80 2.02 -0.38 6.77
20 - - 45 -45 0 90 0 -45 45 - 4.78 -2.70 -1.92 2.12 -0.35 6.77
21 - - 90 0 0 90 0 0 90 - 4.54 -2.66 -1.94 2.04 -0.30 6.77
22 - - 90 45 0 90 0 -45 90 - 4.92 -2.72 -2.00 2.15 -0.30 6.77
23 - - 60 -60 0 90 0 -60 60 - 5.01 -2.72 -2.01 2.18 -0.32 6.77
24 - - 30 -30 0 90 0 -30 30 - 4.46 -2.72 -1.84 2.03 -0.35 6.77

#1236 Forestay
#6309 Mast step
#15769 Main sheet
#2341 Shroud

0 1 2 3 4 Core 5 6 7 8 #1236 #6309 #13108 #15769 #2341 Of deck
No
25 - - 45 -45 0 0 0 -45 45 - 4.45 -2.61 -1.80 2.01 -0.36 7.45
26 - - 90 0 0 0 0 0 90 - 4.27 -2.57 -1.82 1.94 -0.30 7.45
27 - 90 0 0 0 0 0 0 90 - 4.12 -2.51 -1.75 1.87 -0.29 8.02
28 - 90 90 0 0 0 0 0 90 - 4.23 -2.52 -1.78 1.88 -0.29 8.02
29 - - 90 45 0 0 0 -45 90 - 4.51 -2.60 -1.85 1.98 -0.30 7.45
30 - - 60 -60 0 0 0 -60 60 - 4.56 -2.60 -1.85 2.01 -0.32 7.45
31 - - 30 -30 0 0 0 -30 30 - 4.27 -2.63 -1.77 2.00 -0.37 7.45
32 - - 45 -45 0 90 0 -45 45 - 4.78 -2.65 -1.90 2.09 -0.34 7.45
33 - - 90 0 0 90 0 0 90 - 4.54 -2.61 -1.91 2.02 -0.29 7.45
34 - - 90 45 0 90 0 -45 90 - 4.93 -2.67 -1.98 2.13 -0.29 7.45
35 - - 60 -60 0 90 0 -60 60 - 5.02 -2.67 -1.99 2.16 -0.31 7.45
36 - - 30 -30 0 90 0 -30 30 - 4.46 -2.61 -1.82 2.00 -0.34 7.45

#1236 Forestay
#6309 Mast step
#15769 Main sheet
#2341 Shroud
5 Design process
67
5 Design process
5.1.6.3 Design of Internal Stiffening Structure

The main purpose of the internal stiffening structure, from now on referred to as the stiffeners,
is to provide structural support for the hull plating, ensuring that the global deflections are
minimised. This task demands a well thought through design to render a good result. The
design process of the stiffeners has been iterative, and is based partly on engineering guesses but
also the results obtained from the finite element simulations. The stiffeners consist of two main
component groups. One is the longitudinal supports that include one large stringer running
along the length of the boat, and the centreboard casing. The transverse supports consist
of several bulkheads. The geometry of the stiffening components have gone through several
changes during the duration of the design process. The changes concern both the number of
stiffening elements as well as their geometry. The goal was always to achieve a structure that is
as light as possible while still keeping the global structure adequately stiff enough. The overall
goal demanded the stiffeners to be efficient, in the sense that they are placed and designed
where they are needed the most.
The final result of the stiffener design is shown in Figure 5.10. The general design philosophy, as
already mentioned, is to keep the stiffening structure light. This is achieved by cutting holes in
the stiffening components, only keeping material where it is needed. The longitudinal stringer
is reinforced with diagonal elements in the forward part in order to counteract the bending
moment from the mast and forestay. The transverse bulkhead underneath the mast step is
made solid since this is the point where the concentrated loads are the largest. Also note the
diagonal bulkheads in the same position. These bulkheads are angled to account for the loads
originating from the racks mounted on the deck. The intersection of the stringer, the diagonal
bulkheads and the solid transverse bulkhead in this one point will create a very concentrated
structure that will be sufficient to counteract the compressive forces from the mast step. The
four other bulkheads will provide transverse stiffness to the hull as well as compressive forces
acting on the deck from crew in operation. All internal corners of the stiffening structures have
been modelled with a radius in order to decrease the stress concentrations. The radius was
varied until an acceptable stress distribution was reached in the FE-analysis. The outer upper
corners of stringer and bulkheads have been cut diagonally to facilitate attachment of deck.
Figure 5.10: Overview of internal stiffening structure
A possibility that is also implemented in the final design is to add flanges on the edge of the
holes, as shown in Figure 5.11a where a bulkhead with a flange around the edge of the hole is
shown to compare with Figure 5.11b where a bulkhead without flange is shown. This flange is
mainly a result of the building process, but has been shown to contribute to the overall hull
stiffness in FE-simulations.
68
5 Design process
(a) Bulkhead with flange (b) Bulkhead without flange
Figure 5.11: Comparison of stiffener design
The flange would be achieved by wrapping the laminates on the outside and inside of the core
around the edge of the stiffener core template. Between the two laminate skins fibres will be
placed along the edge of the hole for additional reinforcement. This is visualised in Figure 5.12,
where the green lines symbolise the laminate on the inside and outside of the core, that is shown
in grey. The fibres running along the edge of the hole are shown as orange cross-hairs.
Figure 5.12: Schematic flange layup design of stiffener
To evaluate the flange contribution to stiffness versus its contribution to weight, two simple
models are compared in Abaqus. The models are shown in Figure 5.13a and 5.13b. Both consist
of a plate with dimensions 600 × 600 mm which is the approximate space between bulkheads.
In both cases the plates are modelled with a web of height 60 mm. Both models have identical
laminate layup, load, boundary conditions and mesh applied. The load is the hydrodynamic
pressure in Equation (4.10) acting on the entire plate surface. Both plates are simply supported
along two edges and clamped along the other two. S4R shell elements with a approximate size
of 10 × 10 mm is applied in both models. The only difference is that the model in Figure 5.13b
is modelled with a flange of width 15 mm on the web.
69
5 Design process
(a) Without flange (b) With flange
Figure 5.13: Comparison of model designs
The layup of the plates is layup no. 17 in Table 5.17. The web is modelled with the following
layup, where 0◦ corresponds to the long edge of the web.
[45 − 45 90 c908 ]S
c908 indicates that the fibres of the core are parallel to the short edge of the web and that it
has the thickness 8 mm. In Figure 5.13b, again with the long edge of the web as a reference.
The flange is modelled with the following laminate layup.
[45 − 45 90 0]S
The first three plies are extensions of the web laminates and the fourth is the fibres that are
supposed to give the flange its stiffness. The deflection plots in Figures 5.14a and 5.14b are
achieved with FE-simulation.
(a) Without flange (b) With flange
Figure 5.14: Comparison of model deflection
In Table 5.21 below, deflection and mass data are compared between the two models.
Table 5.21: Comparison between models with and without flange.
Maximum deflection [mm] Mass [kg]

U1 U2 U3 mtot
Flange ∼0 0.122 0.491 0.71
No flange ∼0 0.204 0.728 0.70
∆ ∼ 0% 40% 33% -1%
It can be seen that vertical deflection increases with 40 % when the flange is removed and that
transverse deflection is increased with 33 % while the weight only reduces with roughly 1 %. The
flange adds some complexity to the structure and requires extra work in the building process.
70
5 Design process
Still the benefits of adding this laminate flange is considered large enough for the additional
construction time to be invested in them. It was decided that the bulkheads and stringers will
be built with a flange along the edges of the holes with a width of around 25 mm. The width
may vary on some stiffening components since 25 mm of flange width will not fit at all locations
inside the hull.
As for the final layup of the stiffeners, only one was tested at first. The following layup was the
one that was analysed.
[45 − 45 0]S
This layup was chosen based on the wish to keep the fibres of at least one ply more or less
perpendicular to the hull surface around the whole circumference of the stiffeners in order to
make use of the fibres’ strength along their length. The layup turned out to give good results
already in the first FE-analysis. Since the building process was about to start, no further
investigations were made.
5.1.6.4 Evaluation of Hull Structure Design

When a draft design of hull, deck and internal stiffening structure is established, a final eval-
uation of deflection and stress is performed to verify the design. If intolerable levels of stress
or deflections are found the design of the internal stiffening structure is changed. This stage
mainly concerns increasing radii of corners, slightly modifying the positions of the bulkheads
and other minor geometrical modifications of the internal stiffening structure. The intention
of this process is to systematically improve the design towards lower stress levels and deflec-
tion. This stage can be seen as the fine tuning of the design towards its final form. Since
these changes are many and minor presenting them in a structured way is hard and could cause
confusion. Therefore, this section will focus on how the evaluation of stress and deflection is
performed, rather than the many and minor modifications done to the stiffening structure. In
Figure 5.15, the deflection of the hull subjected to the trapeze load case is shown. The largest
deflections occur at the bow due to the force from the forestay, pulling the bow upwards causing
a global bending of the hull. The force from the mast compression at the mast step causes large
deflections in this area. This is visualised more clearly in Figure 5.16.
Figure 5.15: Deformations magnified 50:1 for hull-with-trapeze load case
Figure 5.16 shows how the internal stiffening structure deforms under the trapeze load case.
The figure clearly shows how the mast step causes the stringer and how (from the bow) 2nd,
71
5 Design process
3rd and 4th bulkheads below it to deform a lot more than the other bulkheads. Except for the
transom which is deformed by the main sheet force.
Figure 5.16: Deformation of hull magnified 50:1 for hull-with-trapeze load case. Hull plating
and deck are hidden
Figure 5.17 shows how the elements around the node associated to the forestay force are dis-
torted. The distortion is possibly caused by a local buckling mode in this area, due to the
forestay force being associated to a single node. When evaluating the deflection at this point
this large distortion has to be taken into mind. Choosing a node belonging to an element just
aft of the elements heavily distorted by the forestay force, will yield a more realistic deflection.
The same reasoning can be applied when evaluating deflection at other points where a large
force is associated to a single node, such as the locations of the mast step and the main sheet.
Figure 5.17: Distortion of elements at forestay magnified 50:1 for hull-with-trapeze load case
Considering the stress levels, they are analysed by plotting: stress along fibre direction SL ,
stress across fibre direction ST and the shear stress SLT , through the thickness of the sandwich.
An example of how this is performed is shown in Figures 5.18 to 5.22.
In Figure 5.18 the element of interest is marked with a red arrow. The element is situated
in close proximity to the mast step force. An element not directly associated with the single
node in which the mast step force is applied is chosen, with the same reasoning as above. The
72
5 Design process
Figure 5.18: Bulkhead element for which stress is plotted in Figure (5.19)
element in question is positioned in a critical area. Since the stress is high at the 2nd (from the
bow) bulkhead due to the mast step it is an example of an element where special attention is
required.
Figure 5.19 shows stress through the thickness of the sandwich. The stress varies in the sandwich
skin through the plies of the laminate as the fibre direction changes, giving the characteristic
stress-thickness graph of a laminate. Stress appears to stay more or less constant through the
thickness of the core and the plies, which was not expected.
Figure 5.19: State of stress in bulkhead element shown in Figure (5.18)
When comparing the results to already solved examples in (ABAQUS/Explicit, 2017), where
analysis of sandwich structure in a sail yacht is performed, it is shown that strain varies through
the core and that the strain stays constant through the thickness of the plies, but varies between
plies as the fibre direction changes. To verify that the same behaviour is achieved in this project
a plot of the strain through the thickness of the sandwich is produced, shown in Figure 5.20
below.
The figure shows that the strain varies through the thickness of the core. This complies with the
example analysis found in (ABAQUS/Explicit, 2017). In Figure 5.21 another element considered
critical is marked with a red arrow. This element belongs to the hull and is located directly
below the area of the mast step.
73
5 Design process
Figure 5.20: Strain plotted for bulkhead element in Figure 5.18
Figure 5.21: Hull element for which stress is plotted in Figure (5.22)
Figure 5.22 shows the stress plotted through the thickness of the sandwich for the element in
Figure 5.21. It can be concluded that the behaviour noted previously in Figure 5.19 is repeated
for the core. As for the plies the stress varies through the thickness of each ply. An explanation
could be that the element in question is subjected to more pronounced bending in contrast to
the bulkhead element that is subjected to compression from the mast step.
Similar analysis are performed on several elements throughout the model where considered
necessary, making sure that stress levels are at an acceptable level. One has to take into consid-
eration that certain areas, such as the mast step and the forestay, will be reinforced to be able to
handle the loads in these areas. The stress and deflections presented here are calculated without
these extra reinforcements, and therefore, may not represent the final structure exactly. How-
ever, the bulkheads and stringers have to be defined before the extra reinforcement is designed.
After analysing several elements in a similar manner as previously shown, and modifying the
internal stiffening structure, the final design of the internal stiffening structure is decided. 2D
drawings of the structural hull design is produced and the building process can be initiated,
even though the design work of extra reinforcements still remains.
74
5 Design process
Figure 5.22: State of stress in hull element shown in Figure (5.21)
5.1.6.5 Conclusions on Hull structure Design

In this project the design of the hull plating sandwich is performed first, followed by the design
of the deck sandwich and lastly the design of the internal stiffening structure was performed.
In hindsight, a potentially better methodology could be to perform the design of the stiffening
structure before performing the design of the sandwich structures for hull plating and deck. That
way the procedure of working with conceptual design of the stiffening structure when designing
hull and deck structures could be avoided. This is assumed to be a better methodology since
stiffening structure influences hull and deck in a larger extent than the hull and deck influences
the stiffening structure from a structural point of view.
The reason this project did not employ this order has to do with time frame and planning. By
the time the load cases had been established and a working FE-model was created, templates
used in the building process had to be ordered for fabrication. To be able to order these
templates the core thickness of the sandwich had to be decided to get the correct offset on the
templates. The hull structural design process in this project starts at the core thickness of the
hull plating sandwich. This also explains why core thickness in the hull is not varied together
with laminate layups as was done with the deck in Section 5.1.6.2
5.2 Sail and Rig
When designing the sail and rig, there are many different aspects that must be considered.
According to the class rules (A. 1001VELAcup, 2017), there are no limitations of either mast
height or draft, although the total sail area cannot be larger than 33 m2 and the mast and boom
must be made of aluminium. Both parts, the sails and the rig, are design with professional help
from the companies Seldén and North Sail. As a basis to start from, the 49er© distributions
concerning both sections are used and customised. The adjustments made are stated in the
following sections.
5.2.1 Sail Set Up
There are several different factors that effect the performance of the sails. The most important
of these are described by section 4.4.4. Several of these interfere with each other in a complex
way and the design process of the sail setup, therefore, consists of systematic variations where
75
5 Design process
one parameter is altered at a time in order to find a value for this parameter that results in
the best performance. The performance for a change in parameter value is evaluated with the
VPP.
To design a high performance sail setup it is important to consider in which weather condition
it will operate. The wind statistics in Table 3.1 are therefore crucial for the design of the rig.
It will be designed to perform optimally at the most likely average wind speed. The expected
wind speed differs slightly between Windguru (3.86 m/s) and ISPRA (3.36 m/s). The design
wind speed will therefore be chosen to 3.5 m/s as an intermediate value.
There are also several types of sail setups that can be considered. In this project, four different
concepts have been considered and evaluated. To choose the main particulars of the sail setup,
as well as the best sail concept, the performance of different setups are evaluated with the VPP.
Furthermore, the sail shape is adapted from the 49er©, see Figure 3.4. It provides more sail
area in the upper part, gaining more lift (wind is stronger) and managing the twist of the sail is
easier. These advantages apply on a rectangular sail shape but on a triangular. Another point
of having the 49ers© shape of the main sail is the fitting of the area into the rig dimensions.
A triangular sail needs a higher mast to fit in the same main sail area while keeping the same
boom length.
5.2.1.1 Sail Concepts

Regarding the racing time, more than two thirds of it is spend on the upwind course and one
third going downwind. This observation sets the focus of sail set up optimisation on the upwind
sail area. The aim of the VPP investigations is finding the sail concept providing the highest
mean VMG, with focus on the upwind VMG, in design wind speed without the appearance of
twist or flattening the sail. These boundaries make it possible to be able to sail in higher wind
speed which might occur in gusts or even in several hours over the day.
Therefore, the different ratios and dimensions of the 49er©, see Table (3.2.2), are used to get
a design basis within the limited dimensions of the A. 1001VELAcup (2017). Different ratios,
especially different main sail to jib ratios (50/50, 60/40 and 80/20) are investigated. The
conclusion is an adjusted 49er© total upwind sail area ratio (70/30) is chosen, since unrealistic
rig dimensions and worse mean VMG values are the results from above stated upwind sail
ratios. Thus, the connections of upwind sail area and centreboard area, centreboard area and
centreboard draft, rudder area and centreboard area as well as rudder draft and rudder area
of the 49er© are set fix in the VPP input. Hence the sail area change in the input file of the
VPP goes along with all ratios resulting in a simple smart optimisation process. Furthermore,
a second focus lies on the mast height which is adjustable in the input of the VPP as well. The
mast height opens the opportunity of applying higher ARs. Nevertheless, the limitation of it
is done by using common sense. The higher the mast, the more weight, the more difficult to
handle the skiff. However, the sail area has to fit into the rig dimensions as well, which limits
the mast height regarding a lower bound. The Table 5.22 shows the different sail concepts, its
upwind sail area and reasonable mast height range, which are investigated within the designing
process.
The Classic Concept implies a main sail, a jib and a gennaker, which is hoisted downwind. The
reason to investigate upwind sail areas from 18 m2 to 24 m2 is to evaluate the opponents sail
setup, see Table (3.4). For the VPP calculations the ORC aerodynamic coefficients are used for
the main sail, jib as well as the gennaker and the hydrodynamic coefficients are imported from
the CFD results, see Section (4.4.1). Investigating different total upwind sail areas combined
with different mast heights show that the mean VMG increases with the mast height and the
upwind sail area as stated in Table 5.23.
76
5 Design process
Table 5.22: Different concepts
Concept Upwind sail area [m2 ] mast height [m]
Classic 18-24 8-10

Two Sails 33 8-10
Three Sails 33 8-10
Jib 0 18-24 (33) 9,5
The larger the sail area, the higher the lift force resulting in a high driving force which is
connected to a high VMG. Furthermore, the greater the aspect ratio (see Section 4.4.4.2) of
the sail, the higher the lift force, which ends in an increase of the driving force as well. These
connections lead to the investigation of the Two sails Concept with different mast heights. An
upwind sail area of 33 m2 (23 m2 main sail area, 10 m2 jib) is analysed.
The results, described in Table 5.23, in design wind speed show that the produced lift force is
too high to be handled by the righting moment of two sailors. Twist and flattening of the sails
occur. Also, changing the main sail shape to a triangular by lowering the luff height of the main
sail for 25 % in the input file to reduce the CE or adjusting the mast height for the same reason
does not result in a competitive VMG without twist or flattening. Reshaping the main sail also
involves an extension of the boom length to still ensure the fitting of the main sail into the rig.
Digging deeper into using the total sail area upwind, investigation of so called ’stay sails’ are
performed. Stay sails are additional small sails between the jib and the main sail. They can
enhance the performance of a skiff significantly. The driving force is increased by up to 12, 5 %
depending on the AWA (D.J. Le Pelly, 2008). The idea is to use this advantage downwind and
take down the stay sail when going up wind, still keeping a comparably large upwind sail area
to handle. For the first VPP analysis of this concept a reasonable small mast height of 8 m
is connected to a 14 m2 main sail, 13 m2 jib and 6 m2 stay sail. The calculations of the stay
sail and the jib use the ORC jib aerodynamic lift and drag coefficient, see Section 4.4.4.1. The
result, described in the Table 5.23, in design wind speed show a small mean VMG of 2.27 m/s
as well as the occurrence of twist and flattening. Besides these facts, taking down the stay sail
for the upwind course turns out to be difficult regarding forestay handling, rig tensions, taking
down system and storage of the sail.
The results, see Table 5.23, show that for the design wind speed the Classic Sail Concept is the
best suitable. In almost all investigated setups neither twist nor flattening occur. The maximum
VMG of the Classic Sail Concept is 2.38 m/s at a mast height of 10 m. This value is considered
as competitive compared to the two other concepts. Regarding the twist and flattening of the
sail the values are not acceptable. Lowering the mast height to reduce the heeling moment,
though results in a non competitive VMG. Analysing the results of the Three Sails Concept
neither the VMG nor the flattening and twist turn out to be competitive values, even with a
low mast height of 8 m. Therefore, the Classic Concept is investigated further.
To narrow the sail setup selection down, the mast height of 8 m is crossed out because of the
high difference in the VMG values. Furthermore, two factors are taken into account: Firstly,
the weight of the mast section, and secondly, the highest possible draft of the appendages. The
adjustment of the centreboard draft is done by choosing a higher AR.
(1.4m)2
AR = = 7.69.
0.255m2
77
5 Design process
Table 5.23: Systematic variations within different Sail Concepts
Upwind sail area [m2 ] Mast height [m] mean VMG [m/s] at 3.5 m/s TWS TWIST / FLAT
Two Sails Concept

33 8 2,29 0 / 0,83
33 triangular 10 2,38 0,095 / 0,77
33 10 2,32 0,121 / 0,71
Three Sails Concept
27 8 2,27 0,031 / 0,95
Classic Concept
18 8 2,20 0/0
18 9 2,26 0/0
18 10 2,30 0/0
20 8 2.24 0/0
20 9 2.30 0/0
20 10 2.35 0/0
22 8 2.26 0/0
22 9 2.34 0/0
22 10 2.39 0,031 / 0
24 8 2.28 0/0
24 9 2.35 0,033 / 0
24 10 2.39 0,116 / 0,98
The centreboard area of 0.255 m2 is calculated by using the 49er© centreboard area to upwind
sail area ratio, see Equation 3.1, with an upwind area of 18 m2 . The centreboard draft 1.4 m
results from reasonable hand calculations. Elaborating a smaller mast height variation step of
0.5 m within the investigation process leads to a preferable selection of the sail setup. The
results of the Classic Concept with the higher AR is shown in the Table 5.24.
The highest mean VMG is achieved with 22 m2 upwind sail area and a mast height of 10 m while
35.6% (0.089) twist occur. Taking into account the first factor, Per Wretlind from Seldén Mast
states that the cross section has to be change though the weight per meter of mast increases
by 0.05 kg/m when using a higher mast then 9.5 m. This increase has to be compensated by a
higher mean VMG. The difference between the highest mean VMG in total and the maximum
mean VMG regarding a mast height of 9.5 m is 0.02 m/s. Additionally, the occurring twist is
lowered by the half to 16.4% (0.041). The higher weight as well as the higher twist results in
huge disadvantage regarding the mean VMG compared to the increase of it by 0.02 m/s. The
final sail set up is set to a 9.5 m mast height combined to a 22 m2 upwind sail area. This is split
into a 16 m2 main sail and a 6 m2 jib. For finalising the setups performance result, the rudder
area is adapted as derived in Section 5.3.1.1. The following polar plot show the performance of
the chosen setup in different wind speeds.
The last concept refers to the usage of the remaining sail area of the final sail setup. The classic
downwind sail is called gennaker which is usually a light huge sail, which is hoisted sailing
downwind. A new invention is called Code 0, which is a huge reaching sail designed for offshore
racing. The third alternative is a way of combination of both sails. Called Jib 0, it is a sail
which can be hoisted and taken down like the gennaker, but has the advantage of being able to
be used upwind as well. The following polar plot shows the performance of the Jib 0.
78
5 Design process
Table 5.24: Systematic variations with an centreboard AR = 7.69
Upwind sail area [m2 ] Mast height [m] mean VMG at 3.5 m/s TWS TWIST / FLAT
Classic Concept
18 9 2.29 0/0
18 9.5 2.31 0/0
18 10 2.33 0/0
20 9 2.30 0/0
20 9.5 2.36 0/0
20 10 2.38 0/0
22 9 2.36 0/0
22 9.5 2.39 0.041 / 0
22 10 2.41 0.089 / 0
24 9 2.36 0.077 / 0.978
24 9.5 2.38 0.103 / 0.956
24 10 2.39 0/0
79
5 Design process
Formula Sailing VPP V17 22m2 cb adapted 9,50m AR

7
,69 Polar Plot
TWA (deg)
0
10
7
20 TWS / Sailset
2
30 2.5
6 3
3.5
40
4
4.5
5 5
50 SS1
SS2
4
BS (m/s)
60
3.1418
3.0066
2.8558
3 2.6665
2.2843 70
1.9434
2
1.6442
80
1
0 90
100
-1.3508
-1.5504
-1.8085
-2.1104 110
-2.5106
-3.0484
120
-3.5814
130
140
150
160
170
180
27-Oct-2017 13:26:39
Figure 5.23: Final sail area and mast height dimensions
80
5 Design process
Formula Sailing VPP V17 22 m2 cb adapted 9,50m AR 7,69 Jib0 Polar Plot
TWA (deg)
0
10
7
20 TWS / Sailset
2
30 2.5
6 3
3.5
40 4
4.5
5 5
50 SS1
SS2
4
BS (m/s)
60
2.964
2.867 2.8725
3 2.7003
2.4844
2.1485 70
2 1.7734
80
1
0 90
100
-1.3502
-1.5497
-1.8085
-2.1086 110
-2.5069
-3.0437
120
-3.5768
130
140
150
160
170
180
03-Dec-2017 16:10:47
Figure 5.24: Final sail area and mast height dimensions using the Jib 0
81
5 Design process
The polar plot shows, that the Jib 0 can not be used in wind speeds higher than 4.0 m/s without
loosing mean VMG. However, up to 4.0 m/s wind speed the mean VMG increases compared
to the mean VMG in Figure 5.23. Until a wind speed of 3 m/s the increase of mean VMG is
0.2 m/s. The detailed wind statistics (3.2) show an increase of the mean wind speed over the
racing hours (11h to 17h), starting at ∼ 3.7 m/s, ending at ∼ 4.0 m/s.
5.2.1.2 Final Sail Concept
After the investigation of different sail setup concepts combined different mast heights and its
possibility of implementation, the final decision is a 22 m2 upwind sail area connected to a 9.5 m
high mast, see Figure 5.25.
Figure 5.25: Final sail plan
82
5 Design process
The sail area is divided into a 16 m2 main sail, a 6 m2 jib and a 11 m2 Jib 0. Regarding the
sheeting, the main sail is sheeted on a rail in the aft of the hull, the jib is sheeted on a rail
shortly in front of the mast and the Jib 0 is sheeted out on the racks extension. The sail shape
is determined by the rig dimensions disregarding the main sail. It is shaped in a rectangular
way having as much sail area as possible in the upper part. Furthermore, this shape makes it
easier to twist the main sail.
5.2.2 Rig
The rig of a sailing dinghy is a complex part to design and optimise. This is the reason the R3
class rules allow outsourcing the rig design to an external company (A. 1001VELAcup, 2017).
For the rig concept of this skiff, the expertise comes from Per Wretlind at Seldén Mast who
takes care of the design.
There are two rig solutions that could be an option for this skiff. One is the traditional rig,
similar to the 49er© rig, and the second one is a diamond rig, used mainly on catamarans.
These two concepts are based on different ways to handle the loads that the rig will be exposed
to. The traditional rig is well used and a lot of the already existing skiffs, which are underlying
these class rules, have something similar. This also shows that this type of rig works in reality.
The diamond rig is not been used for mono hulls as first chose of design. By implementing this,
some new ideas are brought into the design of the skiffs for 1001VELAcup.
One feature of the mast, which will be unique, is the mast height. A mast height of 9.5 m is
chosen based on the calculations from the VPP, see more information about this decision in
Section 5.2.1.1. Since the mast is relatively high and needs to be built in aluminium, there is a
risk of a heavy rig construction, as well as a high centre of gravity.
The rig has to take care of the loads that are created from the sails, where mainly the side
forces are important. For this, shrouds and stays are attached between the mast and the hull
to make sure that the rig do not break or fall down. Because of this, the shrouds have a force
pulling upwards. This force has its origin in the attachment point of the shrouds, the chain
plates. Since the force is pulling upwards and the rig downwards, compression loads will occur.
5.2.2.1 Traditional 49er© rig

The current rig that is used in the 49er© class is a conventional single spreader setup with top
mast supporting jumper spreaders, which carries a large main sail on a fixed mast with a boom
along the foot of the main sail. It also carries a smaller head sail, a jib, attached to the forestay.
The standing rig is comprised of the mast and boom, which are stabilised in all directions by
two or four shrouds, a forestay and a backstay. The fore- and backstay serves the purpose
of providing the mast with the necessary longitudinal stability, while the shrouds stabilise the
mast in the transverse direction. The shrouds are usually deflected backwards from the mast
using a set of one or more spreaders in order to provide more longitudinal stability to the mast
as well.
By applying this rig design on the skiff the hull needs to be stiffened up due to the compressing
loads that is transferred from the rig into the hull. Since these forces will occur and the hull
needs to be stiffened up due to this, some extra weight will be added to the hull.
In the case, where this design is applied on the skiff, the mast will both be heavier and higher
than the 49er’s© mast. The rig will have a weight of about 25 kg, which is approximated from
the weight of the 49er© rig of carbon fibre, that has a weight of 15 kg. By having both a higher
and a heavier mast the skiff will be more unstable. This is something the sailor has to adapt
when sailing.
83
5 Design process
In this design there is a need to have a kicker that holds the sail down and maintain the correct
shape of the sail. With this solution, the mainsheet just needs to change the attacking angle
for the sail. To be able to hold down the boom, a kick is needed. This kicker is placed on top
of the boom, why it is called gnav (vang, American word for kick, spelled backwards). When
having this gnav, it needs an extra pair of shrouds that will be placed closed to the height of
the boom, attached to the mast. This is because the gnav will contribute with bending moment
at the mast and there will be a risk of buckling at this point.
The possibility for movement in front of the mast will be limited by having the regular top
shrouds to the deck together with the lower shroud. This makes it harder for the sailors to
move in front of the mast, which can be needed in low wind speeds to change the CoG.
This kind of rig is a proven and common solution, being used for the majority of small sailing
dinghies. Therefore this is an alternative that is investigated.
5.2.2.2 Diamond Rig

The design of the diamond rig looks as it sounds, the shrouds create a diamond shape around
the mast. This rig is also able to carry the main sail as well as a jib and a Jib 0. For the
mainsail it will have a boom that holds the sail out to get its shape. To stabilise everything,
shrouds arranged in a diamond shape around the mast together with at least one pair of shroud
out to the deck, will be used. These will make sure that the rig is stable in the transverse
direction. For the longitudinal direction it will be stabilised with the help of the forestay, and
some solution to be sure it do not fall forward, like an backstay.
The diamond shaped shrouds around the mast are only attached to the mast and not to the
deck. With this solution, the diamond shrouds will take care of the compression loads that
occurs from the sail and by that, the hull does not need to be extra stiffened. To achieve the
diamond shape, the spreaders will keep the shrouds out. The spreaders will at the same time be
swept backwards, which will make the diamond shrouds to contribute to longitudinal stiffness
as well.
The outer shrouds to the deck will partly serve to achieve the longitudinal stiffness, but when
the sailors are in the trapezes they will instead act as the shrouds that hold the mast straight.
In the case the sailors are not in the trapezes the outer shrouds will be the ones that take the
load and make sure that the mast will not fall.
To solve the problem with the support in the longitudinal direction, the mainsheet together
with the backwards swept shrouds or the trapezes will make sure that the rig does not fall
forward. To make sure that the rig will have some kind of longitudinal support, the mainsheet
can not be released too much in the case it needs to change sheeting angle of the sail. To solve
this, the point touching at the hull needs a rail where the sheeting point can be transported on,
to be able to change the sheeting angle.
This solution with a rail in the aft makes it possible to sail without having a kick, since the
sheet holds the boom down because it is connected to the most outer part of the sail. By this,
the mainsheet acts as an backstay, main sheet, and a kicker.
Since the rig is of the type free standing, it allows it to rotate around its own axis. This is a
feature that is brought into the design. By doing this the interference between the mast and
the sails is reduced since there is a smoother transition from the mast to the sail. As well as
that the mast section will be in an attack angle against the wind and will contribute with lift
force.
By having the diamond rig, the mast can be seen as an individual truss structure with the
mast and shrouds. This setup can therefore deliver a mast which can be both lighter and more
84
5 Design process
slender than the traditional 49er© rig. The weight will approximately be around 15 kg for this
concept. One problem that can occur by having this less robust mast is that there are risks for
large deflections in the top of the mast, mainly from the Jib 0 but also from the main sail in
upwind sailing. This will then affect the sail shape both for the downwind sail as well as for the
main sail.
This design is mainly used on multihulls where it is beneficial to have a free standing rig as this
type. The reason for this is that the hull take less compression loads, therefore this solution
is good for a multihull since then the hulls will not be pushed upwards due to the vertical
component from the compression force. By applying this mast design on the skiff there will be
less forces into the hull, why this is an alternative.
5.2.2.3 Bowsprit
For the design of the bowsprit, the main factors that have to be considered are those that
influence its dimensions. Mainly, it should be long enough so that there is at least 1 m distance
between the jib and the jib 0. This condition is depicted below, in Figure 5.26:
Figure 5.26: Highlighted are the Jib 0 (red) and Jib (green)
As seen in Figure 5.26, the design of the bowsprit is not independent, being directly connected
with racks. Its length is fixed to a total of 1.6 m in total, and 1.1 m outside from the bow.
Worth noting is that the bowsprit is also highly dependent on the sail plan and position of the
mast, thus meaning that the design has also to be adapted to both of these elements.
The attachment of the bowsprit to the hull is done by a tube through which it passes, which is
then tapered with fibers to the hull, as illustrated in Figure 5.27.
Figure 5.27: Schematics of the attachment of the bowsprit to the hull
85
5 Design process
In order to obtain a preliminary design of the bowsprit, several assumptions have to be made
regarding both it’s shape and structure, as well as theoretical assumptions in order to simplify
hand calculations. Firstly, regarding its shape, it should be a conical section. This is motivated
by the fact that this solution ensures that the bowsprit doesn’t progress forward of a certain
point (in this case set to be when 1.1 m of the bowsprit structure is hanging forward of the
forward-most point of the bow). This assumption is illustrated in Figure 5.28 below:
(a) Retracted bowsprit (b) Fully extended bowsprit
Figure 5.28: Different levels of extension of the bowsprit, with the fixed conical section high-
lighted in red
As observable, by increasing the diameter of the bowsprit aftwards and making the opening
through which it passes (highlighted by the red lines) having the same dimensions than those
of the thicker part of it, it cannot physically pass.
In order to obtain a first estimate of the bowsprit’s diameters at both ends, as well as it’s
thickness, basic beam theory calculations are performed. The assumptions needed to compute
are:
• The material properties considered are those of carbon-fiber
• Even though it is not, the material is considered isotropic for the sake of simplicity on the
calculations
• The diameter variation is linear between both ends of the bowsprit
• A stayed option is considered, where the bowsprit is stayed with a dyneema rope at half
its length
• Elementary case to be considered to represent the situation is a beam fixed at one end
with a point load applied at the other (free) end
• The load to which the bowsprit is subjected to is represented by a local load (of magnitude
P) at its tip and inclined of an angle α = 73.58◦ on the x-direction. This represents the
force that the jib 0 exerts locally
The values used for the computations are summarized in Tables 5.25 and 5.26,
Table 5.25: Constants assumed for subsequent computations
LBS [m] PBS [N ] ρ [kg/m3 ] σu [M P a] s.f. [−]

1.1 600 · sin(73.58) 1600 570 1.5
Table 5.26: Variables used for subsequent computations
rBS [cm] tBS [mm] IxBS [m4 ]

∈ [2, 3] ∈ [2, 4] 3 ·t
(π · rBS BS )
where LBS is the length of the bowsprit sticking out of the forward-most point of the bow,
86
5 Design process
PBS is the point load a the tip of the bowsprit, ρ is the density of the material composing the
bowsprit and σu is the ultimate stress of the material. The safety factor is represented by s.f..
Regarding the variables, rBS represents the radius of the bowsprit, tBS represents the thickness
and IxBS the moment of inertia of the cross section.
These factors are then inserted in Equation (5.4) in order to compute the resulting compressing
stress at all points of the bowsprit, in an iterative process.
PBS · LBS
σBS = · s.f. (5.4)
2rBS · IxBS
Finally, these values are compared with the ultimate stress considered in Table 5.25. The failure
criterion chosen is a very simplistic one, it being that there’s failure if the computed stress is
larger than the ultimate stress for the material (criterion formulated by Equation 5.5).
σBS > σu (5.5)
The results from computations applying this procedure for both the non-stayed and stayed (at
0.5 · LBS ) cases is summarized in Table 5.27, for a bowsprit with a diameter going from 60 mm
at the base to 40 mm at the tip, and with a thickness of 2 mm.
Table 5.27: Results from the Matlab computations
σBS [M P a] σu [M P a] Failure
non-stayed 191 570 No
stayed 96 570 No
So as easily verified by reading Table 5.27, the stress values on the component are well below
the tolerated ultimate values, therefore meaning that the bowsprit could be in principle thinner.
Yet, a thinner bowsprit isn’t considered for three reasons. First, these values for its diameter
are already among the smallest manufactured by some companies. Second, it should have
enough inner space for the attachments to slide through. Third, as this is a coarse approach,
since in reality a composite is not isotropic, being constituted by several plies of fibers oriented
in different directions, the results are not completely trustworthy, requiring further detailed
analysis. This future analysis is further discussed in chapter 8.
Another concept considered for the bowsprit was the rotating bowsprit. This concept consists
on a pivoted bowsprit, that enables the Jib 0 to be at the desired shape for different angles of
attack, by rotating the desired amount. The concept is depicted in Figures 5.29a and 5.29b.
The bowsprit is pivoted around an axis, and the hull has to have a cut at the bow in order
to allow the rotation. Through it, the attachments of the Jib 0 pass just like in the regular
bowsprit. It can be seen that the staying of such a bowsprit can’t just be a dyneema rope due to
the fact that it rotates. Also, two levers can be seen, whose function is to serve as attachments
points for the ropes that would pull them. Due to the distance they would have had to be
pulled, two horns would have to stick out of the deck, on top of which two wheels would be
mounted in order to direct the rope. Such horns were never modelled given the fact that this
concept was abandoned at a fairly early stage.
This concept ended up being discarded. The main reason driving this decision is the fact that
the sail plan consists on three sails. Therefore with the rotating bowsprit the distance between
the Jib 0 and the jib would become small enough for it to be rendered inefficient.
87
5 Design process
(a) Rotating bowsprit-hull interaction (b) Pivot detail on top of the stiffeners
Figure 5.29: Rotating bowsprit concept model
5.2.2.4 Final rig design

For the final rig design a diamond rig will be used. This is because with this solution the rig
will absorb some of the compression forces by itself. This will make the hull lighter since it not
needs to be stiffened up for these loads. Also by this solution, the rig itself could be lighter.
Since the rig is made in aluminium the weight will be higher, saving weight is one important
part when choosing the rig design.
Since this concept allows the rig to be a rotating rig, this has to be investigated as well. To solve
this, the boom will be locked in rotation around the z-axis, z-direction upwards parallel to the
mast. By locking this when the boom is released with the sail, it will rotate the mast. This will
be possible since it uses the generated force in the sails to rotate the mast. To stop the rotation
of the mast, the rail for the mainsheet will be used. Since the rotation of the mast depends on
the sheeting angle, the sheeting point will be moved onto the rail, and the maximum possible
rotation will thereby be determined by the end of the rail.
The height of the mast will be 9.5 m, this is calculated using the VPP, see Section 5.2.1.2. The
boom will be between the mast and the transom of the deck. Compared to the sail, this means
that the boom will go more aftwards than the sail. This to be able to have the the sheeting
point on the the rail without a to big angle between the sheeting point on the deck and the
boom.
Trapezes is as well added to the rig to help the sailors to create more righting moment. These
is also contributing with stability to the rig when they are in use. When they are in use they
will remove the forces in the shroud and the compression forces comes from the sailors.
To achieve more transverse stability to the rig, the shrouds will be attached out at the racks.
This puts them further away from the centre line and gives therefore more stability. They will
also be placed 1000 mm aft of the mast to also contribute with longitudinal stability to the
mast. For the final rig design, see Figure 5.30. For drawings over the rig, see Appendix 8.2.5.
To complement the mast and the boom, a bowsprit is stacked out from the bow. This will
push the Jib 0 forward and extend the gap between the jib and Jib 0. In this way the sail
performance is improved.
The bowsprit will be one long piece made out of carbon fibre and will be retractable to be able
to not have the whole bowsprit out when it is not in use. In the pulled out mode 1.1 m will be
stacked out in front of the bow. With a total length of 1.6 m it will leave 0.5 m inside the hull
when it is pulled out and nothing when it is retracted. The diameter is 40 mm at the tip and
60 mm at the base. See Figure 5.28 for the design of the bowsprit.
The dimensions of the bowsprit is just made by simple beam theory and therefore it needs to
be analysed by some FEM-software to make sure it does not break. This will tho be a future
work. Also, the attachment to the hull needs to be decided later on.
88
Rig data no. M2211: FORMULA SAILING 2018 9.5M II
Customer: CHALMERS UNIVERSITY

Sales manager: Per Wretlind
Date created: 17-11-27
Latest revision: 17-11-30
Not approved 5 Design process
For quotation only.
Specifications may be changed in actual production
Figure 5.30: Final rig design, with the diamond rig

Copyright © 1996-2017 Selden Mast AB [email protected] printed 2017-12-06 11:18:49
5.3 Appendages and Racks
As mentioned earlier there are no limitations on the appendages designs according to the rules.
As stated in Section 4.4.3, the induced drag of a wing, like a centreboard or rudder, decreases
with increased aspect ratio, which means that a high aspect ratio is desirable for the design of
both centreboard and rudder.
5.3.1 Centreboard and rudder
The design process of the centreboard and the rudder started by deciding the draft of the
centreboard in the VPP. The choice centreboard draft has to be well balanced together with
the mast height and sail area providing enough side force and keeping high aspect ratio without
the need of depower the sails.
5.3.1.1 Planform
When the draft is decided, the planform of the appendages is designed. Two different planforms
are designed, where both of the designs are based on theory for keelboat designs (Larsson et
al., 2014) together with expertise from experienced skiff designers.
One of the planforms is designed with a straight chord line at 25 % of the total chord length
from the front, keeping the same ratio at all points of the chord. With a straight chord line,
the leading edge and the trailing edge need to have elliptical shapes in order to have a elliptical
89
5 Design process
side force distribution over the centreboard and rudder. An elliptical side force distribution is
desirable since the CD will in other case be higher and CL will be lower (Larsson et al., 2014).
The other design is a trapezoidal form, based on the taper ratio, that is decided dependent on
the sweeping angle that is being used. With a sweeping of 0◦ the taper ratio is recommended
to be 0.4 (Larsson et al., 2014). The taper ratio is the ratio between the upper and the lower
chord length of the wing.
C2
Taper ratio ≡ T R = (5.6)
C1
The approximate design of the two planforms is showed in Figure 5.31a and 5.31b.
(a) Straight quarter chord line (b) trapezoidal shape
Figure 5.31: Planforms comparison
The same planform design is used for the rudder as the centreboard. As stated earlier the
rudder planform area is scaled to 50 % of the centreboard planform area.
5.3.1.2 Wing Profiles

The selection of wing sections will largely be based on the leeway angle of the skiff, as shown in
Figure 4.16. The lift of the foils are not dependent on the profile, but the drag of the profiles
are, see Figure 4.14. This means that a suitable profile would be one that has low drag in the
range of angles of attack where the foil is acting. If a profile can be selected so that the angles
of attack are within the drag bucket, and that the drag bucket is as narrow as possible this
would mean minimal drag, which is visualised in Figure 4.15.
The leeway angle is an output from the VPP. The leeway angles at optimum VMG for different
true wind speeds are visualised in table 5.28. Sailing upwind, the leeway angle will vary between
around 2.9◦ and 3.2◦ and sailing downwind the leeway angle will be smaller than that.
Since the largest leeway angle is approximately 3.2◦ , the centreboard profile must have low drag
up to this angle. In Figure 4.15, the NACA 65010 drag bucket is too narrow, the NACA 65018
90
5 Design process
Table 5.28: Leeway angle for different TWS at VMG
TWS [m/s] TWA [◦ ] Leeway angle [◦ ] TWA [◦ ] Leeway angle [◦ ]
2 40 2.94 130 0.72

2.5 40 3.14 130 0.72
3 45 2.92 170 0.19
3.5 40 3.21 125 0.89
4 40 3.09 120 1.07
4.5 40 2.91 120 1.07
drag bucket is wide enough but has more drag than the NACA 65012 profile, which also has a
wide enough drag bucket. Compared to other 12% thickness profiles, see Figure 4.14, the four
digit series profile has lower drag at small angles but not at angles close to 3◦ . The NACA
63012 and NACA 65012 have similar properties but the NACA 65012 seems to have slightly
better properties than the NACA 65012 series profile. According to Larsson et al. (2014) the
NACA 6 series is a good option in general for centreboards and rudders, since the NACA 65012
series seems to have the best properties in this case, it is selected for usage for the centreboard.
Larsson et al. (2014)
The leeway angle can not be used as the angle of attack for the rudder, the influence of the
centreboard redirects the flow and the angle of attack will be smaller than for the centreboard.
However, the rudder are not fixed to the hull but are used to steer the skiff. This means that
the angle of attack for the rudder will vary more than for the centreboard, to account for this,
the NACA 63012 profile is used rather than the NACA 65012 due to the wider drag bucket.
5.3.2 Racks
The design of the racks is based on the reference hulls in Section 3.2.4. Since most of the skiff
is based on the 49er©, but slightly modified the racks will be so too. This is justified because
it is shown that the 49er’s© design works, and that the crew will possibly be 49er© sailors so
they will be familiar with the design.
One feature brought from the 49er© is the height over the still water surface, since it works well
according to 49er© sailors. One aspect to take into account is that the forward part should be
high enough to not hit the water in case of either trim or heel. Since the racks are wider on
the 49er© than allowed for the skiff, the height of the racks will be at the same height as the
49er’s© at the same corresponding width. One feature of the 49er© that is not brought to these
racks is the solidness of the racks. This because it gives unnecessary increase in weight to the
boat.
From the rig design, see Section 5.2.2.4, the attachment points for the shrouds are designed on
the racks. This configuration adds an extra force component to the racks. This force is pulling
both forward and upward, reason why the racks need to be designed such loading condition.
For the planform of the racks, inspiration is taken from the 14 Footer and existing skiffs com-
peting in 1001VELAcup. This means that they are be built with tubes, so that it is possible to
place them as far out as possible. The maximum distance from the centreline is be 1050 mm
since this is half of the allowed beam(A. 1001VELAcup, 2017). For the length, the racks needs
to be inside the LOA. A request from the 49er© sailors was to place the racks as far back as
possible, why the end of the racks are at the same distance as the transom of the hull. The
length of the racks is decided based on both sail performance and structural performance.
91
5 Design process
The position of where to attach the racks to the hull needs to be decided to design the concept.
For this there are several options, on the deck, through the deck or freeboard and, at the side
of a bulkhead e.g. If they in some way will go through the hull there will be a hole that has to
be water tightened. Another problem is that the racks will be fitted there permanent. By this
it will be hard to replace them. Therefor the supports is attached on the deck where there is a
bulkhead underneath. This to be sure that the forces acting on the racks will be taken cared of
and transferred to the hull.
For the dimensioning of the racks all the forces will be calculated with values according to Table
5.29. Where the sailors estimated gravity force is set to S ≈ 2000 N . The shroud force,PY , is
set according to values in Section 4.3.5.2.
Table 5.29: Forces to dimension against
Force Value [N ]
S 2000
PY 2600
The material that is used is aluminium EN-AW 6082 T6, with material data according to Table
5.30.
Table 5.30: Material data for Aluminium EN-AW 6082 T6 taken from Alutrade (2017b) ans
Alutrade (2017a)
Material data Value
Young’s modulus, E 70 GP a
Yield stress, σu 290 M P a
Density, ρ 2.71 g/cm3
5.3.2.1 Aft Support

The desired location of the attachment point of the support is on a bulkhead to make it easier
to attach and also to be able to take the forces that will be applied on the rack. Therefore, the
attachment point was set to the most aft bulkhead since it is desired to have the racks as far aft
as possible. It will be attached at the centreline on the longitudinal stiffener. There will also
be an attachment point at the gunwale. Between these to attachment point the support will be
horizontal to do not have curvature in to directions. The design of the attachments point will
be done after this project ends.
In an early stage one idea of having rotating racks was investigated. This idea build on having
one attachment point at the centre line where it is able to rotate around. The two support arms
for the racks are then attached to each other with a stay and a low angle between them. When
the crew is moving from one side to another, the racks will be pushed down by the crew at the
windward side and the leeward racks will be lifted up. This concept will not be implemented,
but could definitely be investigated in the future. The reason for not implementing it is because
it will be outside the maximum beam, but one option to solve this is presented in Section 8.2.4.
For dimensioning the racks simple beam theory is applied to get both the reaction forces onto
the hull and the stresses and forces in the support. In this case, load case is chosen as seen in
Figure 5.32. Where R1 is the reaction force acting on the centreline, and R2 is the reaction
force at the gunwale. The overhanging part a is the one from the gunwale to the rack. P is
92
5 Design process
the load applied, in this case the sailor standing out at the end. This case is not totally perfect
since it just is the support that is analysed and not the whole structure, as well as that the
reaction force at the centre line should be a fixed. This case was not found in elementary cases
why the most similar was used to get some initial guess.
Figure 5.32: Analysed load case for the aft support of the rack
The reaction forces R1 and R2 according to Equations (5.7) and (5.8). These are decided to
get the moment, Equation 5.9), in the support so the stress could be analysed. From this the
diameter and thickness could be set.
Pa
R1 = (5.7)
l
P
R2 = (l + a) (5.8)
l
Mmax = P a(at R2 ) (5.9)
Mmax r
σmax = (5.10)
Ix
Ix = πr3 t (5.11)
This is done by iterating over a different diameters,D, and thicknesses,t, by making sure that
they will pass the yield stress of the material, see Table 5.30. By using the maximum moment
occurred in the rack, Mmax , the maximum stress could be calculated by help of the moment of
inertia, Equation (5.10) respectively Equation (5.11).
With forces according to Table 5.29 and a safety factor, SF = 1.5, the dimensions for the rack
will be according to Table 5.31. These dimensions is corrected from calculations so there will
be a manufacturer that can provide the profile.
Table 5.31: Dimensions for the aft support.
Variable Value [mm]
D 55
t 2
L 1100
For the attachment to the hull there will be need of reinforcement due to the reaction forces.
For this support the reaction forces will be R1 = 1.33 kN and R2 = 3.33 kN
93
5 Design process
5.3.2.2 Bow Support

Since the rig design is constructed so the lower shrouds is attached out on the racks this will
give forces pulling the rack upward and inward to the hull. To be able to handle this forces the
racks supports has to be designed to be able to take the loads. The forces will contribute with
both bending and compression. For the location of where the loads is acting, see Section 5.2.2.4
for the rig design. The load will act 1000 mm aft of the mast out on the racks.
To be able to handle the compressing loads the support will be placed from the mast and
800 mm aft of the mast. It would have been preferred to have it to 1000 mm but then material
would have been needed to be added. Underneath the support there will be a bulkhead to stiff
up the hull. This is done to be able the take care of the compressing loads from the racks, for
more information see Section 5.1.6.3. This give the bow support an angle backwards from the
mast.
To get the racks in the correct height they have to have some kind of angle and the options
is similar to the one for the aft support. This will be taken care of when designing the bow
support. What also needs to take into consideration is that there will be a lot of ropes that
will come from the forward part of the hull. This need in some way pass the support. Also the
sailors need to be able to move smoothly in front of the support if that is needed in low wind.
To be able tho get the sheets passing the support, the support will be one straight piece with
angle increasing height. This is the geometry that will be dimensioned. This will be of a circular
cross-section.
For the bow support the dimensioning is using the same load case as for the aft support, see
Figure 5.32. Therefore the equations for solving this will be the same. The force from the
sailors is applied at the tip of the support out on the rack. Quite close to this point, 200 mm
aft of the point the shrouds will be attached. This is discarded and will act as a safety instead,
why the safety factor is set to SF = 1.0 in this case. This will give the dimensions according
to Table 5.32.
Table 5.32: Dimensions for the bow support
Variable Value [mm]
D 65
t 2
L 1250
For this support the reaction forces that the reinforcement need to handle will be R1 = 2.15 kN
and R2 = 4.14 kN .
5.3.2.3 Rack beam

The last part of the racks is the rack beam that runs between the two supports. The length
of the beam should go as long aftwards as possible, according to 49er© sailor. This gives that
the racks should go all way back to the aft, behind the aft support and create a overhang of
250 mm.
The length in front of the forward support will mainly be decided by where the sheeting point
for the Jib 0 will be attached onto the rack. This will for sure be in front of the mast, for more
information about the reasoning about where to place it see Section 5.2.1.2.
The rack beam will be analysed in two different parts to see which one that is the most critical.
The analysed cases are sailors between the supports, see Figure 5.33a, and sailors in front of
94
5 Design process
the forward support, see Figure 5.33b.
(a) Loadcase for (b) loadcae 14
Figure 5.33: Planforms comparison
First analysed case is the sailors between the supports. The forces that is taken into account
is the sailors weight downward and the shroud pulling upward. This gives reaction forces is
calculated by Equation (5.12) and (5.13).
P1 (l − a) + P2 b
R1 = (5.12)
l
P1 a + P2 (l − b)
R2 = (5.13)
l
(
R1 a if R1 < P1
Mmax = (5.14)
Rb if R2 < P2
The reaction forces is calculated to get the maximum moment, Equation (5.14), and get the
stress, Equation (5.10). By varying the diameter, D, and thickness, t, a design could be set that
both passes the yield stress as well as having lowest weight as possible. This gives dimensions
between the supports according to Table 5.33
Table 5.33: Dimensions for the beam rack between the support.
Variable Value [mm]
D 55
t 2
L 1110
To see which case that is the dimensioning the other load case is evaluated. For this the sailor
will be placed att to most forward point which from a start point will be 1700 mm in front of
the bow support. This to achieve the best possible sheeting angle for the Jib 0.
The load case analysed in this case can be seen in Figure 5.33b. Where l = 1700 mm and
a = 0 mm. This since the sailors will be places at the tip. The equations that is solved for the
dimensioning are presented as Equations (5.15) and (5.16).
R=P (5.15)
Mmax = P b (at f ixed end) (5.16)
From the maximum moment, Mmax , the maximum occurring stress σu , is calculated for different
diameters, D, as well for the thickness, t. This gives dimensions that is not possible to have
since they are so much bigger than the part between the supports needs. This is calculated
with a safety factor of SF = 1.5
95
5 Design process
Therefor the beam dimensions between the supports is used for this section as well. Therefor
it had to be decided at wich position both of the sailors could be placed on. For this load case
according to Figure 5.32 is used. In this the reaction forces is comes from the supports, why
this not is conservative. By moving the sailors forward the will be allowed to move 500 mm
in front of the forward support. But since the racks needs the extra part for achieving good
sheeting angle they will in total be 1200 mm.
5.3.2.4 Final Rack Design

The final rack design is presented in Figure 5.34 with dimensions according to Table 5.34. There
can though be some problem when the will be built. There can be some problem with mounting
different diameters together. One solution is to change the beam to the biggest diameter that
is needed for the supports. Though, the bow support has some safety factor that has not been
taken into account when designing, it is the shroud force that is pulling upwards.
Furthermore, since the racks are up to date designed as a singular piece, a FEM-analysis is
needed to be carried out to analyse the whole structure. More about this is Section 8.1.4.
Table 5.34: Dimensions of all the parts of the rack
Part Dimension Value [mm]
L 1100
Aft support D 55
t 2
L 1250
Bow support D 65
t 2
L 2300
Rack Beam D 55
t 2
Figure 5.34: Analysed load case for the aft support of the rack
96
6 Balancing of Final Design
After designing the single parts related to the hull, Section 5.1, sail and rig, Section 5.2 and
the appendages, Section 5.3, a few details have to be considered while setting up a general
arrangement, see Appendix 8.2.5, of the skiff. Figure 6.1 shows the final GA which is used as a
basis for the building preparation.
Figure 6.1: General arrangement drawings of the skiff
97
6 Balancing of Final Design
The stiffeners skeleton, as displayed in Section 5.1.6.3, is the result of the balancing of the
different elements dimensioned during the project. The second transverse bulkhead from the
bow is in fact placed right below the mast foot at a distance of 1.70 m aft from the bow. Such
distance is scaled, for the current skiff, from considerations made on the mast positioning on a
49er©. Therefore, knowing the stability and good handling properties of the 49er©, the current
skiff is expected to present similar properties. Right in front of the mast is placed the rail for
the self-tacking jib. Hence, this second bulkhead is directly asked to provide support to the
forces transferred from the rail to the hull.
Moving toward the stern, bulkheads are strategically distributed and designed to not only struc-
turally support the hull itself, but also provide additional support to the racks, the appendages
and the sheeting points. In order to provide the best support to the forces transmitted to the
hull by the racks, bulkheads are therefore positioned right below their attachment portions.
Hence, the diagonal bulkheads spread from the mast foot and follow exactly the direction of the
front attachment of the racks. One innovative feature of this skiff is that the front part of the
racks is used to place the shroud attachment. With this solution, the shrouds are placed further
out and will decrease the compression load of the rig and stabilise it, since the side forces will
be greater. To provide additional stability in longitudinal direction, upper dyneema shrouds,
starting at the top area of the mast and reaching the lower shroud attachment on the racks, are
added.
The most aft bulkhead provides instead support to the back attachment of the racks. This
bulkhead is also given the task of supporting the main sheet, which is positioned in order to
provide the crew with as much clear space as possible. Right beyond the last bulkhead the
rudder is positioned. Opposed to more common solutions, the rudder is located under the hull.
The reason for such solution is given by the consequent elimination of interference with the
water surface. This provides a better inflow on the rudder blade resulting in more lift and less
resistance.
Once the mast and the rudder are set into position, considerations can be made on the place-
ment of the centreboard. Such is critical in order to not compromise the capability of steering
manoeuvres, as the higher the distance between rudder and centreboard the higher the lever.
Hence, higher lever produces a higher steering moment. Furthermore, the lever is constrained
by the lead, which describes the distance between the centreboard and the CoE of the sail area.
Analysing reference skiffs and dinghies the lead is set to 8 %, see Table 3.5.
Finally, local additional reinforcements are placed in correspondence of all the high load attach-
ment points.
98
7 Building process
In this chapter the building process is explained in detail. The parts that are covered are the
designing and assembling of the templates, followed by the planking of hull and deck, as well
as their lamination and of the stiffening structure.
7.1 Templates
The first step in the building process is to design the templates from the final hull shape. One
set of templates is designed for the bottom and freeboard while another is designed for the
deck. The templates are designed in Rhino 3D using the final hull design as a reference. Each
of the bulkhead templates is placed within a distance of 230 mm from the the neighbouring
template(s). The smaller the distance between them, the more accurate will the shape of the
hull be, therefore if a bigger distance would have been used, the precision when it comes to
shape of the hull would be lost when planking. To obtain the desired shape of the aft part of
the hull, one more template is placed in the aft, making the template model 230 mm longer
than the actual hull.
When designing the templates, all joints have a gap of 0.15 mm as a tolerance in order to make
the assembly easier. Furthermore, all the inner corners have to be made with so called ’dog
bones’, which have a diameter of 6.2 mm. This is due to the cutting process, where sharp inner
corners can not be made due to the milling bit being circular, see Figure 7.1.
Figure 7.1: Dog bones in one of the deck’s template
When the design of the templates is done, they are ordered in Medium Density Fibre boards
(MDF) that are 16 mm thick. The MDF boards are later assembled into the full scale template
plug using epoxy to glue them together. A model of the final design of the templates and a
picture of the assembly can be seen in Figures 7.2a, 7.2b, 7.3a and 7.3b.
Furthermore, templates for the bulkheads and stringers need to be ordered. These components
are to be mounted and cut over these templates. This process is described in the following
section dealing with planking.
99
7 Building process
(a) Model of hull templates (b) Model of deck templates
Figure 7.2: Templates models in Rhino 3D
(a) Templates for the hull (b) Templates for the deck
Figure 7.3: Template assemblies used for planking
7.2 Planking
The entire hull structure of the skiff will be built in a composite sandwich. The core will be
built using balsa wood. The following sections will describe the process of building the balsa
core for the different parts of the hull.
7.2.1 Planking the hull and deck
The planking process starts once the templates are assembled. Balsa-planks with 8.8 mm
thickness to leave a margin for sanding, 45 mm width and 2.5 m in length are used. The
planks are placed on top of the templates and bent along their curvature in order to achieve
the desired shape. At the joints and sides of the planks, fish glue is used to glue the planks
together. During the curing process of the fish glue, the planks are held in place by screwing
them to the templates. To prevent the screws from damaging the planks and also to distribute
an even pressure around the screw, small blocks of divinycell are placed between the planks and
the screws. This can be seen in Figure 7.4.
100
7 Building process
Figure 7.4: Planking with divinycell blocks
When the planking is done and the fish glue is cured all the screws and divinycell blocks are
removed and the hull is faired using sandpaper and a planer. This process is important in order
to get the surface as smooth as possible when starting the laminating process. The faired hull
core can be seen in Figure 7.5.
Figure 7.5: Faired hull
7.2.2 Planking the internal structure
In order to build the bulkheads and stringers different pieces of the balsa core are cut and
arranged so they cover the entire bulkhead template. See Figure 7.6 for an example where the
largest stringer template is covered with balsa pieces that will constitute its core.
Figure 7.6: Bulkhead template completely covered with balsa for the core
101
7 Building process
The pieces of balsa that are then glued together and fixed with blocks of divinycell like the
hull. Once the glue sets and the divinycell blocks can be removed, the balsa is aligned with the
template underneath and clamped together being ready for trimming. The contour is then cut
into the balsa by making use of a router machine. An example of half a bulkhead core after
cutting is presented in Figure 7.7, where the shape was followed and the edges were filleted by
the router.
Figure 7.7: Finished half-bulkhead core
The need for filleting the edges arises from the concern of having a smooth fibre transition
through those edges when laminating later. The lamination process is covered in the following
section.
7.3 Lamination
The general process of laminating the components of the skiff is hand layup of the fibre and
epoxy, followed by vacuum bagging of the laminate during curing. Specific details in the process
are described in the following sections.
7.3.1 Laminating the hull and deck
The lamination of the hull uses the finished planked and faired core as the starting point. The
two skins of the hull will be laminated in two turns, starting with the outer skin. When the
outside of the hull is sanded and faired, it is detached and removed from the templates. A sheet
of plastic is then spread out over the templates, and the hull core is put back into place again.
This sheet of plastic will create an air-tight seal on the inside of the hull, which allows for the
usage of the vacuum bagging method. This method uses vacuum to create a well saturated fibre
layup with evenly spread out epoxy resin. It also allows for some extraction of excess resin.
With the inside of the hull sealed by plastic, it is ready for the outer skin to be laid up. The
likely method laying up will be to prepare all plies and saturating them with epoxy resin on
a flat surface in advance. The wetted layup is placed between two sheets of plastic to make
handling easier. The layup is then carried and placed on top of the hull, making sure it lies in
contact with the hull in all places. Naturally, the two sheets of plastic used during the layup
are removed after this step.
A new sheet of plastic is then used to cover the hull. An air tight seal is created between the
two sheets that now surrounds the hull, only leaving a hole where a vacuum pump is connected.
When the pump is turned on it will force the two sheets to press down on the wet laminate,
spreading the epoxy resin evenly across the hull and through the thickness of the fibres. This
102
7 Building process
setup is then left to cure in a heated atmosphere. The heat is needed for the epoxy to reach
its maximum strength properties. After the epoxy has set, the vacuum bag is removed and the
inside of the hull is ready to be sanded, faired and laminated in the same fashion. The deck
is laminated on a fabricated female mould using the same method as the internal structure, of
which the method is described in Section 7.3.2.
7.3.2 Laminating the internal structure
The components of the internal stiffening structure, the stiffeners, are laminated in a slightly
less cumbersome way than the larger hull and deck. The stiffeners are also laminated using
the vacuum bagging method, but their size allows for the whole component to be laminated
in one step. A set of templates that have the same contours as the stiffeners are milled in an
MDF-based board which has very smooth surfaces in another material to help in releasing the
hardened epoxy. These templates are placed on a table. The epoxy resin is then applied on
the template and covered with the correct layup of fibres. The balsa core, that has been cut
to correct dimensions prior to the lamination, is placed on top of the fibres on the template.
The other skin is then placed on top of the core. All edges of the core are covered with fibres
which are shaped as a flange, mentioned in Section 5.1.6.3, around the edge of the template
underneath. The component and laminate is then covered with a sheet of plastic, taped air tight
to the table, and then connected to a vacuum pump. The setup is finally left under vacuum
until the epoxy cures. The concept is visualised in Figure 7.8.
Figure 7.8: Layup procedure of stiffener component.
7.4 Further Building
At the time of writing, several construction details are left to build. The building process
of these parts differ from the techniques described in previous sections. These new building
processes will be described briefly in the following sections.
7.4.1 Rudder and Centreboard
The rudder and centreboard will be manufactured using a different kind of building technique
compared to the hull structure. This is because of the higher demands on the appendage
surfaces due to the desired properties of the water flow during operation. One of the best
methods to obtain a smooth surface is to manufacture the appendages in halves, using CNC-
milled aluminium female moulds. The halves are then joined around a honeycomb or foam core
to make the finished part. The material used for the appendages will most likely be carbon
fibre due to its high strength to weight ratio. As mentioned before, the updated R3 class rules
103
7 Building process
allow this choice of material since there are no restrictions regarding material content for the
appendages.
7.4.2 Racks
Since the racks can be built in aluminium as long as they comply with the rule of 70 % of natural
material they are designed to be build in aluminium tubes(A. 1001VELAcup, 2017). Since the
racks contain three parts of different shape they have to be fitted together. The method of
joining the aluminium parts is still to be determined. The most likely solution will be a mix
of welding and blind riveting. The finished racks will be attached to the hull using epoxy and
flax fibres and possibly rope where applicable. To get the shape of the aft support, see Section
5.3.2.1, the tube will be bent.
7.4.3 Bowsprit
The bowsprit is allowed to be made out of carbon fibre, and since it assumes a conical shape,
as mentioned in Section 5.2.2.3, the method for building shall be roll wrapping. It consists
on wrapping sheet of pre-impregnated fibres (pre-preg) onto a mandrel, according to desired
layers and angles. Then it is pressurised and heat cured to solidify. After solidifying the shape
is extracted from the mandrel. This process is generally used for small to medium diameter
tubes, that is from around 6 mm to 30 mm diameter, up to 2.5 m length, (west composites,
n.d.).
104
8 Future Work
Due to the limited time available for the project, some work and investigations will inevitably
have to take place further on in time. The following chapter will cover both work that needs
to be done within the current project, as well as investigations that lie outside of the project
scope but may be interesting to look at in future projects.
8.1 Remaining Work
Although the main part of the project is finished, there is still some design work left to be
done in order to be able to produce a complete skiff. This work has to be done parallel to the
remaining building work, mentioned in Section 7.4. The remaining design work is covered in
the following sections.
8.1.1 Sails
The plan form of the sail is set from the VPP program, see Section 5.2.1.2. Since the design for
the sails are not done yet, some concept for handling of them is not decided. Therefore, some
of the future work is presented at this section.
8.1.1.1 Jib 0 and jib handling

The choice of using Jib 0 as a down wind sail that possibly can be used upwind sail as well is
preferred. In the case the wind is too strong, the Jib 0 is just used on the down wind legs. For
this, the Jib 0 has to be stored on the deck when going upwind. The system for this has to be
decided. One idea for this is to furl it around the stay and then take it down onto the deck.
The same problem will occur if the Jib 0 is used in upwind. In this case the jib and the Jib
0 will come to close to each other why the jib has to be taken down. Furling is one option to
store the jib for the upwind leg. For making the handling of the jib easy it will be kept on a self
tacking rail. By this the position will be close to the mast, and out to the freeboard. Another
investigation is done on the shape of the jib which might be adjusted to a fat head jib, making
it easier to trim. Both the systems, for the jib and Jib 0, are decided before the skiff will go
out on sailing.
To be able to use the Jib 0 both upwind and downwind the sheeting point has to be able to be
moved around. The solution for this is not decided and has to be designed. Two options that
should be investigated are the extend of the rail for the self-tacking jib so the sheeting point for
the Jib 0 also will be attached on this and a flying sheeting point which will be fixed by sailing
upwind
8.1.2 Local Reinforcements
Structural calculations that remain are those regarding local reinforcements of the hull where
the basic construction may not be sufficient. One such point on the hull is the mast step where
the mast rests on the deck. This is one of the highest loaded parts of the hull and needs to be
reinforced accordingly. As mentioned in Section 5.1.6.4 the internal structure of the hull will
most likely be able to support this load. However, the balsa core of the deck in this point will
most likely be compressed beyond acceptable levels, which is why it will have to be replaced by
105
8 Future Work
a stiffer material. The current options are to either use a stiffer wood core, or simply making a
small area just underneath the mast step out of solid laminate. The decision is still to be made.
Another high load point of the hull is the one where the rudder is attached to the hull. This point
is positioned slightly aft of the transom, which means that some sort of supporting structure
has to be designed in that position.
There are still some uncertainties concerning the attachment of the standing rigging as well as
the attachment of the racks to the hull. These attachment points are also yet to be designed
and dimensioned. The design process of these attachment points will take inspiration from
existing solutions on high-end sail racing boats and adapting these to the material options that
are available, as well as the geometry of the hull.
8.1.3 Structure of Centreboard and Rudder
The centreboard and rudder need to be dimensioned and analysed using the FE-software Abaqus
CAE just like the rest of the structures. A big difference is the separate rules regarding the
material content. A late change in the R3 class rules now allows appendages to be fully made
from carbon fibre composites, in addition to the bowsprit which already allowed the usage of
carbon fibre. This change in the rules will allow for these parts to be designed at a lower
weight than what would have been possible otherwise. Naturally, this will affect the structural
calculations that have been carried out up to this point, as well as coming calculations.
8.1.4 FE-Analysis of Racks
Regarding the racks, some further structural work needs to be done. This work concerns the
attachment of the racks to the hull, as well as FE-analysing the racks themselves. For the FE-
analysis the whole rack structure should be investigated since the preliminary design looks at
each member separately, and only using basic beam theory in hand calculations. The analysed
load cases should be the following.
1. Sailors not on the racks
• Pulling force from the down wind sail and,
• Shroud force pulling upwards and inwards
2. Sailors as far forward as allowed, in trapeze
• Sailors at a point 500 mm in front of the forward support,
3. Sailors as far backward as allowed, in trapeze
• Sailors at a point 250 mm back of the aft support,
4. Sailor at the mid of the racks,
• Sailors placed at 600 mm aft of the forward support,
106
8 Future Work
8.1.5 Structure of Bowsprit
Given that the bowsprit’s structural characteristics do not affect any of the other different
components of the skiff, only its main dimensions are needed during the design process. For
this reason, a FEM analysis needs to be conducted as future work, in order to access the desired
characteristics, such as the fibres orientation, number of plies and their respective thicknesses.
8.1.6 Deck Layout
Designing and building a skiff implements the point to make it run as well. After designing,
building and joining all the individual parts an operating system on the deck is installed. The
main sail is sheeted with a block on a rail at the stem connected to a block on the end of the
boom. This sheeting replaces the kicker and can be adjusted via the main sheet. A rope which
goes out to both sides of the racks is used to move the block on the rail veering the main sail.
The trapezes are fixed with stretching ropes on the racks. Next to the mast and still reachable
from the racks, the sheet of the selftacking jib is mounted and spread to both sides. The jib is
sheeted over a block sliding on a rail, which is installed closely in front of the mast. Next to
jib sheet, the system to hoist and take down the Jib 0 is installed. This system is connected to
the retractable bowsprit and the Jib 0 storage making sure that the bowsprit is retracted when
the Jib 0 is stored in the hose and extended when the Jib 0 is set. The Jib 0 is sheeted on the
top part of the racks extension to enable the best wind attacking angle possible. Furthermore,
the Jib 0 sheet is connected to a floating sheeting point between racks and hull for being able
to use the Jib 0 upwind with a perfect sheeting angle as well.
To make sure that all these system is working together, all the system has to be implemented in
the GA. This to make sure that the ropes do not cross each other as well as the do not interfere
with each other in a bad way. Therefore it has to be more investigated in the future.
8.2 Future Investigations
Many concepts have been mentioned during the project, but not all have been investigated.
There are several reasons for this. Some concepts did not seem plausible, and many would
take too much time from designing a proper hull. The fact that these concepts have not been
investigated does not mean they are not interesting in the project context, which is why they
are mentioned in this section for future reference.
8.2.1 Hull Fluid Dynamics
During the process, a proper CFD analysis has not been carried out as far as it regards the
shaping of deck, freeboard height, and transom. Bethwaite (2008) has been used as a reference
when alteration of the transom have been computed. Moreover, given the satisfying results
obtained with the first sloped transom, no further analyses has been implemented. The alter-
ations that saw a lowered freeboard have been on the same track. Finalised the optimisation
of the hull in terms of submerged portion, the freeboard has been cut lower. The interaction
between the lowered freeboard and the sloped transom determined a further improvement in
the performances of the skiff, however no systematic variation of these two geometries and their
interaction has been carried out. It might therefore be worthwhile investigate such combination
of solutions in order to achieve an optimal ratio between the two components.
The introduction of the cambered shape for the deck has been considered in terms of ergonomics
advantages for the sailors, in order to avoid in total the presence of water on board and to
107
8 Future Work
improve structural properties of the complete geometry. The contribution to total dynamic
resistance has however not been computed, therefore such shape is probably not optimal and a
further step in terms of CFD computations has definitely to investigate such solution.
Furthermore a study on boat-tailing the skiff should also be conducted, as it might prove
beneficial from a resistance point of view.
Moreover a global resistance prediction, considering the hull complete with appendages, should
be carried out. Time limitations and set up complexity have however prevent the implementa-
tion of such computations and therefore hydrodynamic optimisation under such point of view.
Different CoG, therefore different trim angles, have been assigned to specific speeds. It has not
been evaluated however a variation of the trimming conditions within a fixed speed. Future
investigations can focus on such aspect and optimise the configuration of the skiff while sailing.
8.2.2 Wing Sails
The usage of wing sails on the skiff instead of a traditional rig and sails is allowed in the R3
class rules. A wing sail can be made as a hard wing consisting of two or more hinged vertical
elements, or as a soft wing where the wing surfaces are made from regular sail cloth. The
common trait for the two alternatives is the lack of standing rigging, and the variable camber of
the wing. A wing sail is more efficient than a traditional rig and sail and is affected by induced
drag to a much lower extent due to the more efficient square or elliptical planform, as opposed
to the triangular planform of an ordinary sail (Nielsen, 2014).
The main obstacle when using a wing sail in the design however, is that the R3 class rules
state that this kind of sail must be designed and built by the project group. Aside from the
high difficulty of building an advantage bringing wind sail, a considerable amount of workload
is added. Furthermore, a certain amount of water hours are needed to perfect the handling of
this sail. This may not be a problem if an existing hull is retrofitted with a wing sail.
8.2.3 Foiling
It has been proven and shown in practice that making a boat sail with its hull completely
emerged (to foil), only resting on a set of hydrofoils attached to the lower end of the ap-
pendages, is entirely possible. Recently, even an optimist dinghy could be made to foil with
some modification (Andersson et al., 2017). This opens up the future possibility of designing a
foiling skiff, since the design of the appendages is completely unregulated in the R3 class rules.
If such a design could be made successful, it would give a great advantage to the team.
The idea of designing a foiling skiff in this project was dropped very soon in the initial stages
since it would add a considerable amount of work to an already work intensive project. It
was decided instead to design a more traditional skiff in order to focus fully on the hull de-
sign. Adding foils would most likely need the hull to be altered in order to provide optimal
hydrodynamic performance.
The possibility to use foils could however be interesting in future projects where more time
can be spent on evaluating new alternatives to the design. There is already a bachelor thesis
available at Chalmers University of Technology where the option of adding foils to the current
skiff will be evaluated (Larsson, 2017).
8.2.4 Rotating Racks
As mentioned in Section 5.3.2.1 one opportunity is to implement a rotating system if the racks.
This is crossed out due to the fact that it was not possible at that stage in the project to have
108
8 Future Work
rotating forward support as well. Cause in the case it was just the aft section that was rotated
it gave a to wide beam. If it is possible to get the forward support rotating this should be one
thing to investigate, cause then it is possible to earn some lever arm for the sailors. This will
help to create a higher righting moment which gives the possibility to have more sail area.
At the same time the system can be built in to rotate the mast to the windward as well. This
will give even better sail performance since then the projected sail area will be larger than it
would have been in the case if the rig has some heeling angle.
8.2.5 Rotating Bowsprit
Furthermore, the rotating bowsprit concept mentioned in Section 5.2.2.3 is of great interest,
as having the sails at their desired shape increases their efficiency. Therefore a study on the
combination of such a bowsprit with a two sails plan is of potential interest to be carried out
in the future.
109
8 Future Work
110
References
1001VELAcup. (2017). Istruzioni di regata. Retrieved 2017-12-03, from http://www
.1001velacup.eu/images/documenti/EVENTO-2017/IDR%201001%20VELA%202017.pdf
1001VELAcup, A. (2017). Norme speciali di classe r3. Retrieved 2017-12-6, from http://
www.1001velacup.eu/regolamento/regolamento-di-classe.html
ABAQUS/Explicit. (2017). Abaqus documentation [Computer software manual]. Vélizy Villa-
coublay, France..
Adam Persson. (2017). bowspritcompetitor. ([Picture; September 29, 2017])
Agarwal, D., Bhagwan, Broutman, J., Lawrence, & Chandrashekhara, K. (n.d.). Analysis and
performance of fiber composites. Wiley.
Alutrade. (2017a). Fysikaliska värden. Retrieved from https://www.alutrade.se/
fysikaliska-varden/ ([Online; accessed November 29, 2017])
Alutrade. (2017b). Hållfasthetsvärden, extruderat. Retrieved from https://www.alutrade.se/
hallfasthetsvarden-extruderat/ ([Online; accessed November 29, 2017])
Andersson, A., Barreng, A., Bohnsack, E., Lundin, L., Sahlberg, R., Werner, E., . . . McVeagh,
J. (2017). The foiling optimist. In The proceedings of the 4th international conference on
innovation in high performance sailing yachts, lorient, france, 28-30 june 2017. (p. 19-30).
Bethwaite, F. (2008). Higher performance sailing: Faster handling techniques. Bloomsbury
USA.
Bethwaite, F. (2010). High performance sailing: Faster racing techniques. Bloomsbury USA.
Calderon, A. A., & Maskew, B. (2015). Transonic Hull: Theory, Validation, Breakthroughs
and Applications to Ships.
Cardolite. (2017, December). Formulite. Retrieved from https://www.cardolite.com/
formulite
Castegnaro, S., Gomiero, C., Battisti, C., Poli, M., Basile, M., Barucco, P., . . . Lazzaretto, A.
(2017). A bio-composite racing sailboat: Materials selection, design, manufacturing and
sailing. Ocean Engineering, 133 (February), 142–150.
CD-Adapco. (2017). Star-ccm+ (12.04.11 ed.) [Computer software manual].
CDIO. (2017). Cdio. Retrieved 2017-12-06, from http://www.cdio.org/
Creative, F. L. (2012). 49er full realistic model - january 2012. Retrieved from http://
www.9eronline.com/library/ ([Online; accessed October 12, 2017])
DeSantis, M. (2016). Progetto di una imbarcazione classe r3 per la regata 1001velacup.
D.J. Le Pelly, R. F., L. Kjellberg. (2008, dec). The effects of staysails on yacht performance.
In 3rd high performance yacht design conference (p. 247-256).
Eike Jacobs. (2017). 49erbowsprit. ([Picture;, November 6, 2017])
Henry P. Moreton, C. H. S. (1992, July). Functional optimization for fair surface design.
Computer Graphics.
Imeche. (2017). Formula student. Retrieved 2017-12-07, from http://www.imeche.org/
events/formula-student
ISO. (2015). NWIP, Small craft - Scantlings - Part 7: Multihulls (Tech. Rep.). Author.
Kaupp, R. (2007). A 49er, fast sailboat. Retrieved from https://commons.wikimedia.org/
wiki/File:49er_sailboat.jpg ([Online; accessed December 6, 2017])
Larsson, L. (2017). Formula sailing – en flygande kappseglingsjolle? Re-
trieved from http://www.chalmers.se/sv/institutioner/m2/utbildning/
kandidatutbildningar/PublishingImages/FORMULA%20SAILING%20%e2%80%93%20En%
20flygande%20kappseglingsjolle.pdf ([Online; accessed December 7th, 2017])
111
References
Larsson, L., Eliasson, R., & Orych, M. (2014). Principles of yacht design. Bloomsbury Pub-
lishing Plc.
Larsson, L., & Raven, H. (2010). Ship resistance and flow. Society of Naval Architects and
Marine Engineers.
Mancuso, A., Pitarresi, G., & Tumino, D. (2017). Using FEM simulation to predict struc-
tural performances of a sailing dinghy. International Journal on Interactive Design and
Manufacturing, 1–12. doi: 10.1007/s12008-017-0427-7
McNeel, R. (2017). Rhino 5 features. Retrieved 2017-12-02, from https://www.rhino3d.com/
features
Newaz, G., Mayeed, M., & Rasul, A. (2016). Characterization of balsa wood mechanical prop-
erties required for continuum damage mechanics analysis. Proceedings of the Institution
of Mechanical Engineers Part L: Journal of Materials: Design and Applications, 230 (1),
206-218.
Nielsen, P. (2014). Have wingsails gone mainstream? Retrieved 2017-12-03, from https://
www.sailmagazine.com/diy/have-wingsails-gone-mainstream
Offshore Racing Congress, . (2016). Orc vpp documentation.
Persson, A. (n.d.). Resistance from towingtank test for a 49er.
Persson, A. (2017). 1001 VELA Cup existing skiff rack.
Pitarresi, G., Tumino, D., & Mancuso, A. (2015). Thermo-mechanical behaviour of flax-fibre
reinforced epoxy laminates for industrial applications. Materials, 8 (11), 7371–7388.
R.M.N. (2017). Velocita’ e direzione del vento. Retrieved 2017-10-03, from
http://www.mareografico.it/?session=0S1584794367D886587866680P&syslng=
ita&sysmen=-1&sysind=-1&syssub=-1&sysfnt=0&code=STAZ&idst=12
Versteeg, H., & Malalasekera, W. (2007). An introduction to computational fluid dynamics:
The finite volume method. Pearson Education Limited.
Wahab, M. A. (2017). Interpolation and extrapolation.
west composites, R. (n.d.). Roll wrapping. Retrieved from https://www.rockwestcomposites
.com/custom/processes/roll-wrapping ([Online, accessed December 07, 2017])
Wikipedia. (2010). 49er. Retrieved from https://commons.wikimedia.org/wiki/File:49er
_skiff.svg ([Online; accessed December 2, 2017])
Wilson, R., Carrica, P., & Stern, F. (2006, June). Unsteady rans method for ship motions with
application to roll for a surface combatant. Computers Fluids, 35 (5), 501-524.
Windguru. (2017). Archive wind statistic. Retrieved 2017-10-03, from https://www.windguru
.cz/archive.php?id_spot=11912&id_model=3
WorldSailing. (2017). World sailing classification. Retrieved 2017-12-07, from http://members
.sailing.org/classification/
112
Appendix A - Drawings
1. CFS 2017/2018 - 0 - General Arrangement
2. CFS 2017/2018 - 001 - Hull Drawing
3. CFS 2017/2018 - 102 - Deck Drawing
4. Seldén Mast - Rig Drawing
I
Rig data no. M2211: FORMULA SAILING 2018 9.5M II
Customer: CHALMERS UNIVERSITY

Sales manager: Per Wretlind
Date created: 17-11-27
1000
Latest revision: 17-11-30
Not approved
For quotation only.
CAP
Specifications may be changed in actual production 8720
General rig description D3
8410
Double diamonds with lower shrouds FS
Conventional Untapered ø3
11,2°
Rig dimensions 1/4"
Forestay height FH: 8 550 mm
D3
Main sail luff length P: 8 700 mm ø3
Boom height above deck BH: 800 mm CAP
2850
15,8°
Main sail foot length E: 2 340 mm ø3
6,5°
Main boom sheet pos S: 2 595 mm 1/4"
Chainplates dist. from cl mast
Lateral Long
Cap shrouds: 1050 1000 mm (43,6°)
Lower shrouds: 1050 1000 mm (43,6°)
Forestay: 1700 mm
800 P35 Linked
Deck above chainplates: 0 mm
Deck above waterline: 246 mm
Craft data and righting moment
CATAMARAN
Length: 4 600 mm
Beam: 2 100 mm RD2
10,7°
C-C Hulls: 2 100 mm
2850
M24
Displacement: 310 Kg LWR
Equipped boat ø3
Dimensioning RM: 3,2 kNm 10,0°
1/4"
Spar sections
Mast: C069
Boom: B087 No kicker
Square mainsail roach
2:1 purchase for main halyard 800
P35
Masthead gennaker and code zero OK
2:1 purchase system for code zero halyard
Furlex
Forestay: X
Notes (17 lines)
2850
Rig setup for Chalmers University Formula sailing RD1/V2

competition 2018. ø3
17,3°
1/4"
Total length: 9550
(800+8700+50)
Total sail area with gennaker app. 33m2
Monohull with max length 4600mm, max beam 2100mm RD1

400
2x75kg crew in trapeze at cap shroud fitting 8720
Rotating mast, main boom to be fixed to the mast in Diamond angle: 17°
rotating direction.
Main boom Spin poole S038 main sheet fitting on mast

S1=2500 Sheet lead block position on boom to be set.
(section requirement may be changed)
Panel 1 = 3250, Panel 2 = 2900
Copyright © 1996-2017 Selden Mast AB [email protected] printed 2017-12-06 11:18:07

Appendix A - Drawings
VI
Appendix B - Plots
B.1 Max beam position vs Resistance
(a) Max beam position vs Resistance at 12 kn (b) Max beam position vs Resistance at 15 kn
Figure B.1: Max beam position vs resistance. Red stars mark measured data; Blue line is the
B.2 Max draft position vs Resistance
(a) Max draft position vs Resistance at 12 kn (b) Max draft position vs Resistance at 15 kn
Figure B.2: Max draft position vs resistance. Red stars mark measured data; Blue line is the
VII
B.2 Max draft position vs Resistance
VIII
Appendix C - Rules
IX
1001VELACUP
CHALLENGE R3
Special Rules Class R3
2017-2020
!!! This is a translation of the original text : If there is a conflict between languages the Italian text will take
precedence. !!!
Premise
This Rules aims to promote a regatta’s program for sailboats designed and manifactured inside universities,
under a common teaching program and in accordance with a Class Rule R3 which sets out the features.
Premessa
Con il presente regolamento si intende promuovere un programma di regate tra barche a vela progettate e realizzate all’interno
di strutture universitarie, nel quadro di un programma didattico comune e nel rispetto di un Regolamento di classe R3 che ne
stabilisce le caratteristiche.
Aims
Aims of the universities participating in the initiative is to organize within their internal structures (in thesis
courses, workshops, laboratory activities, extracurricular activityes, etc.) annual teaching activity aimed at
the design and implementation of sailboats 15 feet long in accordance with the specifications and
requirements of the Class Rule R3.
Obiettivi
Obiettivo degli Atenei che aderiscono all’iniziativa è quello di organizzare all’interno delle loro strutture (nell’ambito di corsi o
tesi di laurea, workshop, attività di laboratorio, extracurriculari ecc.) un’attività didattica annuale finalizzata alla progettazione
e realizzazione di una deriva da 15’ che risponda alle caratteristiche e ai requisiti previsti dal Regolamento di classe R3.
Purpose
The purpose of the initiative is to promote an annual or bi-annual, meeting between the universityes that
agree the 1001VELAcup aims.
The calendar and the program of events will be defined by the Technical Committee provided for in
Regulation R3 class, after hearing the proposals of the University who won the previous edition and the
other participating universities.
Finalità
Scopo dell’iniziativa è quello di promuovere, con cadenza annuale o biennale, un incontro tra le Università che aderiscono
all’iniziativa “Mille e una vela cup”.
Il calendario e il programma delle manifestazioni verranno definiti dalla Commissione Tecnica prevista dal Regolamento di
classe R3, sentite le proposte dell’Ateneo che si è aggiudicato la precedente edizione e degli altri Atenei partecipanti.
Generalities
The boats partecipating to the program 1001VELAcup must comply with these "Special Rules, be designed
and constructed, in accordance with the Rules of Class R3, and be governed by teams composed of
Universities students.
Generalità
Le imbarcazioni che partecipano al programma “Mille e una vela cup” devono rispettare le presenti “Norme speciali di Classe
R3”, essere progettate e realizzate, secondo quanto stabilito dal Regolamento di Classe R3, ed essere governate da equipaggi
formati da studenti iscritti presso qualsiasi Ateneo.
1) Elegibility for the “1001VELAcup's Throphy

Crew:
a) The crews must be composed by students enrolled in participating Universities
b) All crew shall not hold a Group 3 Classification (World Sailing Sailor Code) (Annex “D”).
c) In regattas with multiple consecutive trials, the crew (min. 2 persons) is fixed for the duration of the event
d) Students older than 30 are not admitted.
Ammissione alla regata per il "Trofeo 1001VELAcup"
Equipaggi
a) Gli equipaggi che partecipano al programma “Mille e una vela cup” devono essere costituiti da studenti iscritti ad un ateneo
partecipante (cfr. Allegato D).
b) Tutto l'equipaggio non deve essere in possesso di una Classificazione di Gruppo 3 (World Sailing Sailor Code) (cfr. Allegato
D).
c) In una manifestazione a più prove consecutive la composizione dell’equipaggio (min. 2 persone) è fissa per tutta la durata
della manifestazione.
d) non sono ammessi studenti di età superiore a 30 anni.
Boats:
e) Are allowed in the " Mille e una vela Cup " program, only boats that have been designed and built by
students enrolled in degree courses of participant university.
f) The boats will be identified by a sail number issued by 1001VelaCup.
g) hulls and keel must to be built by students in a university's laboratory, also under teachers guide. If a
university will use an external structure, the boatyard must release a declaration that attests that the
personnel has not directly been involved in the construction or has furnished consultations.
h) equipment, rigging and sails, except for wing sails structures, can came from external supplies of the
university
Imbarcazioni
e) Sono ammesse al programma “Mille e una vela Cup” unicamente imbarcazioni che siano state progettate e costruite dagli
studenti iscritti a corsi di laurea di un ateneo partecipante.
f) Le barche saranno identificate da un numero velico rilasciato da 1001VelaCup.
g) Lo scafo e le appendici devono essere realizzati, presso un laboratorio di ateneo,dagli studenti, sotto la guida di docenti scelti
dall’Ateneo. Qualora l'ateneo decidesse di utilizzare una struttura esterna, il titolare del cantiere dovrà rilasciare una
dichiarazione che attesti che il personale non sia stato direttamente coinvolto nella costruzione o abbia fornito consulenze.
h) L’attrezzatura, l’armo e le vele, ad esclusione delle strutture delle wing sails, possono provenire da forniture esterne
all’Ateneo partecipante.
Participants
i) Each university participating in the race can bring up to 3 boats.
l) For each boat must be associated univocally a crew, each crew member can be associated only to a single
boat
m) If more universities will participate as consortium, there must be designate the representing university
Partecipanti
i) Ogni ateneo partecipante può portare in gara un massimo di 3 imbarcazioni.
l) Ad ogni imbarcazione deve essere associato univocamente un equipaggio, ogni membro di equipaggio può essere associato ad
una sola imbarcazione.
m) Nel caso più atenei decidano di partecipare in consorzio, deve essere nominato l'ateneo rappresentante.
2) Regatta rules
The regattas will be run in real time under the following rules:
a) The World Sailing Racing Rules of Sailing 2017 – 2020 using the “Low Point System”;
b) The notice of Race and the Regatta's Instructions released for the Event ;
c) The Special Rules R3 Class;
d) the regattas will be run inside the following wind speed limits: min. 3 m/sec – Max 10 m/sec. The regatta
Committee is free to evaluate the conditions of security even in that range.
2) Regole di regata
Le regate si correranno in tempo reale applicando:
a) Il “Regolamento di Regata World Sailing” in vigore con le norme integrative emanate dalla F.I.V. Sarà applicato il
“punteggio minimo”;
b) Il Bando emanato per la manifestazione e le Istruzioni di Regata conseguenti;
c) Le “Norme speciali di Classe R3”
d) Le regate saranno disputate all’interno dei seguenti limiti di vento: min 3 m/sec - Max 10 m/sec. A discrezione del comitato di
regata resta in ogni caso la valutazione delle condizioni di sicurezza della manifestazione anche nel range di vento definito.
3) "Paolo Padova" Trophy

Premise:
The Trophy has been created in 2013 and is dedicated to Paolo Padova, a meaning figure in sharing, with
his students, the 1001VELAcup's principles and philosophy.
Boats
a) will be applied the same rules of the 1001VELAcup trophy
Crews
b) The crews must be composed by a teacher and a students enrolled in participating Universities.
3) Trofeo "Paolo Padova"

Premessa:
Il trofeo, istituito nel 2013, è dedicato a Paolo Padova che con il suo agire e per il ruolo che ricopriva è stato una figura
emblematica dei principi condivisi da 1001VELAcup.
Imbarcazioni e Partecipanti
a) Valgono le stesse regole del Trofeo 1001VELAcup
Equipaggi
b) Gli equipaggi che partecipano al Trofeo “Paolo Padova” devono essere costituiti da un docente strutturato nell'ateneo di
rappresentanza ed uno studente iscritto ad un ateneo partecipante.
R3 class rules
Introduction
These class rules was compiled with the intent of making the universities designing and building sailing boats limited
in dimensions and materials, with easy technologies and limited costs as well as be easily transportable on land and
manoeuvrable at sea, putting universities on the same level even with different available skills and equipment.
Objectives
The following rules aims to make regattas with boats, realized inside the Athenaeum, having similar characteristics
and letting a great range of design possibilities
Regolamento di classe R3
Premessa
Il presente regolamento è stato redatto con l’intento di far realizzare delle imbarcazioni a vela di limitate dimensioni, con
tecnologie accessibili e costi contenuti, facilmente trasportabili via terra e manovrabili in acqua, mettendo sullo stesso piano
laboratori didattici universitari con capacità e attrezzature differenti, anche attraverso l’introduzione di limitazioni sulla scelta
dei materiali.
Obiettivi
Il presente regolamento si propone di far regatare insieme barche con le stesse caratteristiche, realizzate all’interno degli Atenei,
lasciando ampio margine per la loro progettazione.
1- Dimensions
a) Maximum length overall: 4.60 meters.
b) Maximum beam overall: 2.10 meters.
c) For the measurements of the dimensions given at point a) and b) a maximum tolerance admitted is 15mm.
d) The ruder and its mounting system are excluded in the overall length measurements.
e) Bowsprits are allowed, only if used for flying sails off, and are excluded from the overall length
measurement.
1- Dimensioni
a) Lunghezza massima fuori tutto: 4,60 m.;
b) Larghezza massima fuori tutto: 2,10 m;
c) Per la misurazione delle dimensioni definite ai punti a) e b) è prevista una tolleranza max di 15 mm.
d) Il timone e il suo sistema di aggancio all’imbarcazione sono esclusi dalla misura della lunghezza massima fuori tutto.
e) E’ ammesso l’uso di un bompresso o di un tangone, escluso dalle dimensioni massime, solo se necessario alla manovra delle
vele.
2- Materials and construction

a) For the purposes of educational training in these rules, the hull and racks should be made of wood or
materials of vegetable and/or animal origin (also called “natural”) , expressed in weight, not lower that 70%.
The glue within natural materials can be calculated as part of natural composite only if it's not used as
structural material.
b) The use of hiking wings (or racks) is allowed if i's in accordance with the dimensional and material
composition limits.
c) The following materials are banned: aramid fibres, carbon fibres, titanium.
d) Carbon fibre can only be used for the bowsprit and the appendices.
e) Spectra (polyethylene) can be used for rigging.
f) The sails cannot be made from Kevlar, spectra, carbon or other high modulus fibres and they must have a
statement from the sail maker.
g) The mast and boom must be made from extruded aluminium or, in exclusion of the matrix material, in
natural fibers.
2- Materiali e costruzione
a) Per gli scopi didattico formativi che si pone il presente regolamento, l’insieme scafo-coperta-terrazze dovrà prevedere un
contenuto in legno o materiali di origine vegetale e/o animale (in seguito detti anche “naturali”), espresso in peso, non inferiore
al 70%. Il collante può rientrare nel calcolo dei materiali naturali solo se necessario all' incollaggio tra questi e non sia
utilizzato come materiale strutturale.
b ) E’ ammesso l’uso di terrazze nel rispetto dei limiti dimensionali e costruttivi definiti.
c ) Ad eccezione di quanto specificatamente consentito, nella realizzazione di scafo, coperta, terrazze appendici, albero e boma
non sono ammessi i seguenti materiali: fibre aramidiche, fibre di carbonio, titanio.
d) E' ammesso l'utilizzo di fibre ad alto modulo per la realizzazione del solo bompresso e delle appendici.
e) E' ammesso l'uso di polietilene per il sartiame.
f) Le vele non possono essere realizzate in fibre aramidiche, carbonio o altre fibre ad alto modulo e dovranno essere
accompagnate da una dichiarazione del velaio riportante le misure di stazza come descritte negli allegati C e C1.
g) L’albero e il boma devono essere realizzati con estruso in alluminio o, ad esclusione della matrice, in fibre di origine vegetale
e/o animale.
3- Appendices
a) The number of daggerboards and foils is free, the shapes are not restricted in design.
b) It is allowed the use of Kevlar, spectra, carbon or other high modulus fibres for daggerboards and foils.
3- Appendici
a) Il numero delle appendici è libero, le sagome e superfici sono libere.
b) Per le appendici sono ammessi i seguenti materiali : fibre aramidiche, fibre di carbonio, titanio o altra fibra ad alto modulo.
4- Rig
a) The height of the mast is not restricted.
b) Trapezes are allowed.
4- Armo
a) L’altezza dell’albero è libera.
b) Sono ammessi trapezi.
5- Sail plan
a) The sail plan will have a maximum total surface area of thirty-three square meters.The calculation of the
areas will be carried out in accordance with Annex C for laminate sails and according to what is indicated in
annex C1 for wing sail type; by wingsail is meant any sail which, when used, has a distance between
windward surface and leeward surface is bigger that the thickness of the sheet.
b) No kites are allowed.
c) In a regatta with more manches, only one set of sails are allowed.
d) Each university can present a measured sail set signed by an official measurer of the sail federation from
the university's country, the measures will be taken as description in annex C and C1
5- Piano velico
a) Il piano velico avrà una superficie massima complessiva pari a trentatre mq. Il calcolo delle superfici sarà effettuato secondo
quanto riportato nell’allegato C per le vele in laminato ed in base a quanto riportato nell’allegato C1 per le vele di tipo alare
(wing sail); si intende per wing sail o vela alare qualunque vela che, in esercizio, presenti una distanza tra superficie al vento e
superficie sottovento superiore allo spessore proprio della lamina.
b) Non sono ammessi armi tipo kite o aquilone
c) In una manifestazione a più prove consecutive è ammesso un solo gioco di vele
d) Qualunque ateneo può presentare le vele già stazzate da uno stazzatore, autorizzato della federazione velica dello satato di
appartenenza, seguendo le indicazioni dgli allegati C e C1.
6- Hull
a) The hull must not have concavities in cross sections below the waterline.
b) It must be a monohull; multihulls are not allowed.
c) The hull must not have asymmetrical cross-sections.
d) The hull must be able to provide sufficient buoyancy to the safety of the vessel and its crew, of at least 80
litres; in the form of foam, expanded material, air bags and/or inspectable watertight compartments. The
caps of the inspection holes will be minimum 2 an have a diameter of 12 cm min. They must be located in
the anterior and posterior part of the boat to provide a complete inspection.
e) The dagger-board and rudder must be securely fixed to the hull; the use of a quick release shackle with
elastic is allowed for each appendix.
6- Scafo
a) Lo scafo non deve presentare concavità nelle sezioni trasversali al di sotto della linea di galleggiamento.
b) Lo scafo deve essere unico; non sono ammessi multiscafi.
c) Lo scafo non deve presentare sezioni trasversali asimmetriche.
d) Lo scafo deve poter garantire una riserva di galleggiamento sufficiente alla sicurezza dell’imbarcazione e del suo equipaggio,
pari ad almeno 80 lt., sotto forma di schiuma, espanso, sacchi d’aria, compartimenti stagni ispezionabili. I tappi di ispezione
dovranno essere minimo 2, avere diametro minimo di 12 cm ed essere posizionati uno nella zona posteriore e uno nella zona
anteriore della coperta e comunque secondo logica in modo da permettere una adeguata ispezione.
e) E’ obbligatorio assicurare la deriva e il timone allo scafo; l’uso di un grillo a sgancio rapido con elastici è permesso per ogni
appendice.
7- Conformity and book of rules

a) The boats must be in possession of a self-certification of the University represented showing compliance
with this Regulation in accordance with the models in Annexes A and B.
b) The boats could be subject to compliance testing.
c) If a boat is damaged, repair or replacement of parts is permitted, as long as it complies with the class
rules.
d)If it is not possible to correct any discrepancies, they will be remedied with the installation of an on board
corrector weights from 5 kg, in number established by T.C., placed at a distance less than 50 cm from the
transom.
e) Protests regarding measurement will be accepted until two hour before the start signal of the first race,
except for infraction happened during the race/s
f) 15 days before the start of the regatta/s each entry must present (also in numeric format) the following
documents:
f1) Annex A “Self Certification of Compliance”
f2) Annex B “Weight Calculations”
f3) Draws of the forms, waterlines and construction plan
f4) Detailed technical description about the construction
f5) Technical sheets of the used materials
f6) photos and/or clips about the construction process to support the points: f1,f2,f3 and f4.
f7)the annex D “Crew Sheet”
7- Conformità al regolamento
a) Le barche dovranno essere in possesso di una autocertificazione dell’Ateneo rappresentato attestante la conformità al
presente regolamento secondo i modelli allegati A e B.
b) Le imbarcazioni potranno essere sottoposte a verifiche di conformità.
c) Qualora una barca venga danneggiata è ammessa la riparazione o la sostituzione delle parti a seguito della quale si
verificherà nuovamente il rispetto delle regole di stazza.
d) Qualora non sia possibile correggere eventuali difformità, queste saranno sanate con l’istallazione a bordo di pesi correttori
da Kg 5, in numero stabilito dalla C.T., posti a distanza inferiore a 50 cm dallo specchio di poppa.
e) Le proteste di stazza saranno ammesse entro le 2 ore dall'inizio delle regate o serie di regate, salvo infrazioni commesse nel
corso dell'evento.
f) Entro 15 giorni dall'inizio delle regate dovranno essere presentati i seguenti documenti:
f 1) Scheda di adesione all’evento “1001VELAcup”
f 2) Allegato A “Dichiarazione di conformità” con allegata dichiarazione del velaio;
f 3) Allegato B “Calcolo dell’esponente del peso”;
f 4) Disegni delle linee d’acqua e del piano di costruzione;
f 5) Descrizione dettagliata della tecnica di costruzione adottata;
f 6) Schede tecniche dei materiali usati;
f 7) Foto e/o filmati di tutte le fasi della costruzione sufficienti a supportare quanto riportato nelle voci 2, 3, 4 ed 5
f 8) Allegato D “Scheda equipaggi” che comprende : categoria di appartenenza degli atleti (sono esclusi gli atleti facenti parte
del gruppo A – World Sailing), certificato di iscrizione all’ateneo per l’anno in corso (allegato alla scheda), tessera della
federazione velica del paese di appartenenza valida per l’anno in corso.
8- Technical committee
a) The technical committee is composed by four members named by 1001VelaCup and by the
representatives appointed, time to time, from the first three universities classified in the regatta for the
“Trofeo 1001VELAcup”.
8- Commissione Tecnica
a) La Commissione Tecnica è costituita da quattro membri nominati da 1001VelaCup e dai rappresentanti nominati di volta in
volta dai primi tre Atenei classificatisi nella Regata per l’assegnazione del “Trofeo 1001VELAcup”.
Emendments
a) This Regulation will be valid until October 2016.
b) This regulation could be supplemented and / or modified by the Technical Committee as a result of any
proposals by the universities concerned.
c) Proposed amendments must be submitted to the Technical Committee within 7 days of being awarded the
"Trofeo 1001VELAcup."
d) The Technical Committee will review the proposed amendments within 30 days from the date of award
of the "Trophy 1001VELAcup."
Emendamenti
a) Il presente regolamento è valido fino a November 2020.
b) Il presente regolamento potrà essere integrato e/o modificato, dalla Commissione Tecnica a seguito di eventuali proposte da
parte degli Atenei interessati.
c) Le proposte di emendamento dovranno essere sottoposte alla Commissione Tecnica entro 7 giorni dall’assegnazione del
“Trofeo 1001VELAcup”.
d) La Commissione Tecnica provvederà a esaminare le proposte di emendamento entro 30 giorni dal termine della assegnazione
del “Trofeo 1001VELAcup”.
Interpretation
If a university requires clarification on the interpretation of some of these rules only the interpretation
expressed by the Technical Committee will apply.
The official language of these class rules is Italian, all interpretations will refer to the Italian language
meanings
Interpretazione
Qualora un Ateneo richiedesse chiarimenti sull’interpretazione di alcune delle presenti regole sarà valida solo l’interpretazione
espressa dalla Commissione Tecnica.
Roma, 16 Novembre 2017

Signed
the president of the Technical Committee
Massimo Paperini
ANNEXES
TO R3 CLASS RULES
ALLEGATI
AL REGOLAMENTO DI CLASSE R3
ANNEX A
CLASS RULES R3
SELF CERTIFICATION OF COMPLIANCE (pt. 7A)
Boat's name : ……………………………………………………………………………...
Sail Number: (3 letters for the nationality) ….......... - (sail number)……………………..
Construction year : ……………………….
University: …………………………… Faculty : ……………………………….

Address : ………………………………………………………………………………
Telephone : +…… ………….…………….... Fax: +........ ………………………
E-mail:……………………
HULL: Rule. R3
LH (Leinght overall max) 1a (max mt. 4.60) : mt.…………………………

B (beam overall max) 1b (max mt. 2.10) : mt.…………………………
PIANO VELICO Rule. R3
SA (total sail surface) 5a (max mq. 33,00) : mq……………………

In attachment the certification of the sailmaker.
Norme Speciali
Wing Sail 1d) □NO □YES.
I declare to have read the Special Rules of R3 Class, I certificate the exactness of the information released in
this declaration. I accept to make the boat available for verifications.
Date : ………………… stamp of the athenaeum and signature of the person responsible :
………………………………...
____________________________________________________________________________________
Pb (corrector weights) rule 7d □NO □YES → : kg. ………………………..

.
notes:
..................................................................................................................................................................
at .............................. date: ..../..../........
the measurer the president of T.C.
.................................................. ..................................................
Annex B – WEIGHTS CALCULATIONS
Brief technical description about the construction …………………………………………………………………….
……………………………………………………………………………............................................................................................
................................................................................................................................................................................................................
................................................................................................................................................................................................................
....................................................................................................................................................................................................
Note: for “Natural Material” is intended Wood or vegetal/animal materials
HULL
Hull is considered being the the ensemble of surfaces for planking, topside, underside, stern surface.
Natural Materials Specific Weight (kg/m3) Surface (m2) Weight (kg)

1
2
3
4
5
Total
Others Materials Specific Weight (kg/m o kg/m ) 2 3
Quantity (m o m ) 2 3
Weight (kg)
1
2
3
4
5
Total
STRUCTURES
Structures are the ensemble of parts such as frames, keelson, stingers, and reinforcements. The
reinforcements for rigging settings must be calculated in the “Rigging” section.
Natural Materials Specific Weight (kg/m3) Quantity (m2 o m3) Weight (kg)
1
2
3
4
5
Total
Others Materials Specific Weight (kg/m2 o kg/m3) Quantity (m2 o m3) Weight (kg)
1
2
3
4
5
Total
SOVRASTRUCTURES (only in natural fibres builded)
For sovrastructures is intended l' ensamble of the planking surfaces. Structural elements must be
cocalculated in the “Structure” section.
Natural Materials Specific Weight (kg/m3) Surface (m2) Weight (kg)
1
2
3
4
5
Total
Others Materials Specific Weight (kg/m o kg/m )
2 3
Quantity (m o m )
2 3
Weight (kg)
1
2
3
4
5
Total
RACKS (only if maiden in “non-natural” fibers)

This section must be filled only in case the racks (or “wings”) are maiden in materials different than “natural
fibers” . It's necessary to calculate also the local renforcements and fastenings.
Material Specific Weight (kg/m2 o kg/m3) Quantity (m2 o m3) Weight (kg)
1
2
3
4
5
Total
2 3
Quantity (m o m )
2 3
Weight (kg)
1
2
3
4
5
Total
RIGGING
To calculate only the elements that are directly attached to the hull. In this sections iare to be calculated also
the reinforcements for the rigging .
Fixed rigging Quantity Weight (kg)
1
2
3
4
5
Totale
2 3
Quantity (m o m3)
2
Weight (kg)
1
2
3
4
5
Totale
Natural Materials Specific Weight (kg/m3) Quantity (m2 o m3) Weight (kg)
1
2
3
Totale
Total Weight
Natural Materials Weight
Other Materials Weight
Percentage of Natural Material
Date : …………………
stamp of the athenaeum and signature of the person responsible :
………………………………...
Student's Responsable (must be a student) : ……………………………….

ANNEX C
DRAWINGS AND INSTRUCTIONS

Measurement of sails
The fabric must be dry, the tension applied for measurements must be sufficient to eliminate all the folds of the fabric.
Measurements must be effectuated spreading the sails to the ground.
Il tessuto deve essere asciutto e la tensione esercitata su di esso durante le misurazioni deve essere tale da eliminare tutte le
pieghe. La misurazione va effettuata con le vele stese a terra.
Randa/Mainsail
MSA = (LI/2 x (E5+E4)/2) + (LI/4 x (E4+E3)/2) + (LI/8 x (E3+ E2)/2)

+ (LI/8 x (E2+ E1)/2) + FTA
In cui :
a) E5 > E4 > E3 > E2 > E1 ;

b) LI is the distance, in straight line, between head point and tack; both
identified as descripted below è la distanza in linea retta tra il punto di
drizza e il punto di mura, ambedue rilevati secondo il sistema di
identificazione sotto riportato;
c) FTA = E5 x CO / 2 ;
d) FTA is to be considered only if the case that the clew is not coincident
with the endpoint of E5; CO is the distance between the measurement
point E5 and the clew point. FTA si calcola solo nel caso in cui il conto
di scotta non coincida con l'estremo della misura E5 sulla balumina. CO
è pari alla distanza dalla linea di misura di E5 al punto di scotta.
e) The rounded excess of fabric under the line between tack point and clew
point must not exceed 8% of the mainsail's base. La rotondità della base
al di sotto della linea che unisce il punto di mura al punto di scotta non
deve eccedere l’ 8% della base della randa ;
f) E1 is coincident with the width of the head measured perpendiculary to
the luff by the head point coincide con la larghezza della penna presa
perpendicolare all’inferitura, passando per il punto di drizza ;
g) E2 is coincident with the width of the mainsail taken at 1/8 of the luff
perpendicularment to the line between between the head point and the tack point coincide con la larghezza della randa
in corrispondenza a 1/8 della lunghezza dell’inferitura.
h) E3 is coincident with the width of the mainsail measured on the perpendicular to the line between the head point and the
tack point at 1 / 4 of the line itself ¼ oincide con la larghezza della randa in corrispondenza a 1/4 della retta che
congiunge il punto di drizza al punto di mura, in direzione perpendicolare a questa retta ;
i) E4 is coincident with the width of the mainsail measured perpendicularment to the line between the head point and the
tack point at 1 / 2 of the line itself coincide con la larghezza della randa in corrispondenza di 1/2 della retta che
congiunge il punto di drizza al punto di mura, in direzione perpendicolare a questa retta ;
j) E5 is coincident with the width of the mainsail measured perpendicularment to the line between the head point and the
tack point at the level of the tack point coincide con la larghezza della randa all’altezza del punto di mura, rilevata
lungo la direzione che si trova seguendo le istruzioni riportate di seguito, e comunque perpendicolarmente alla LI.
k) For the identification of the points on the leach refer to the following instructions ( the same procedure is for the
gennaker) Per quanto riguarda l’identificazione dei punti di misura alle estremità delle vele far riferimento alle figure di
seguito riportate ( valide anche per le misurazioni di fiocco e gennaker ) .
l) On the mast will be installed tree measurement stripes , two on the mast and one on the boom. These will be black
colored and be 25 mm. high. The higher edge of the lower stripe on he mast will be at the same high of the top face of
the boom. The sail will never be tacked beyond that line high Sull’albero e sul boma dovranno essere installate tre
strisce di misurazione, due sull’albero e una sul boma, di colore nero e di altezza pari a 25mm oltre le quali la vela non
potrà essere bordata. Il filo superiore della banda inferiore di P dovrà essere posizionato in corrispondenza della faccia
superiore del boma.
Measurement method / Metodo di misurazione :

1) Overlap the head point on the tack point and pull the luff . Sovrapporre il punto di drizza sul punto mi mura e tendere
bene l’inferitura.
2) The crease identifies two points, the first on the luff and one on the leech. Mark these points
La piega ottenuta identifica due punti, uno sulla balumina e uno sull’inferitura. Segnare questi due punti.
3) Measure the distance (E4) between these points and mark the trace of this line nearby the luff. Misurare la distanza
(E4) tra questi due punti e segnare la traccia della fettuccia in prossimità dell’inferitura.
4) Unfold the sail and pull the luff, trace the line between tack and head. The line will intercepts the taken trace of the
measure E4 in the point O1. Stendere nuovamente la vela, tesare l’inferitura e congiungere con la fettuccia anch’essa
ben tesata il punto di drizza al punto di mura. La fettuccia intersecherà la traccia della misura di E4 precedentemente
presa in un punto O1.
5) Overlap the head point on the O1 point and pull the luff . Sovrapporre il punto di drizza al punto O1 e tendere bene
l’inferitura.
6) The crease identifies two points, the first on the luff and one on the leech. Mark these points .La piega ottenuta
identifica due punti uno sulla balumina e uno sull’inferitura. Segnare questi due punti.
7) Measure the distance (E3) between these points and mark the trace of this line nearby the luff. Misurare la distanza
(E3) tra questi due punti e segnare la traccia della fettuccia in prossimità dell’inferitura.
8) Unfold the sail and pull the luff, trace the line between tack and head. The line will intercepts the taken trace of the
measure E3 in the point O2, mark this point. Stendere nuovamente la vela, tesare l’inferitura e congiungere con la
fettuccia anch’essa ben tesata il punto di drizza al punto di mura. La fettuccia intersecherà la traccia della misura di E3
precedentemente presa in un punto O2.
9) Overlap the head point on the O2 point and pull the luff . Sovrapporre il punto di drizza al punto O2 e tendere bene
l’inferitura.
10) The crease identifies two points, the first on the luff and one on the leech. Mark these points. La piega ottenuta
identifica due punti uno sulla balumina e uno sull’inferitura. Segnare questi due punti.
11) Measure the distance (E2) between these points. Misurare la distanza (E2) tra questi due punti.
12) Overlap the tack point on the O1 and mark the position of the E4 leech point on the overlapped fabric. Sovrapporre
infine il punto di mura sul punto O1 e riportare la posizione dell’estremo di E4 dalla balumina al lembo sovrapposto.
13) Spread the sail, the extention, on the leech, of the line between the last point ant the tack point gives the position of the
point where to measure the E5 distance. Ristesa la vela, congiungendo il punto identificato con il punto di mura e
proseguendo fino alla balumina si ottiene la posizione del punto per la misura di E5.
__________________________________________________________________________________________
Fiocco
F = 0.125 ×JL×(2×LP+3×JGM+2×JGU)
a) JL is the leinght of the luff. Unfold the sail

and pull the luff, measure the distance between
head and tack. JL coincide con la lunghezza
dell’inferitura. Si stende la vela in modo da
evitare la formazione di pieghe e si misura la
distanza tra il punto di drizza e il punto di
mura;
b) LP is the width of the sail, on the
perpendicular of the luff, trough the clew
coincide con la larghezza sulla
perpendicolare all’inferitura passante per il
punto di scotta. Si misura rilevando la minima
distanza tra il punto di scotta e l’inferitura;
c) JGM : overlap the head to the clew. The
crease identifies the midpoint of the leech.
Mark this point and measure the minimum distance between this and the luff.
sovrapporre il punto di drizza sul punto di scotta. La piega che si viene a formare
identifica la metà della balumina. Segnare questo punto e ristendere la vela in modo da poter misurare la minima
distanza che intercorre tra tale punto e l’inferitura della vela.
d) JGU : overlap the head to the point of JGM on the leech and mark the point on the leech in correspondence of the
crease. Unfold the sail and measure the minimum distance between the point and the luff. piegare la vela fino a
sovrapporre il punto di drizza al punto precedentemente trovato e segnare il punto sulla balumina in corrispondenza
della piega. Stesa di nuovo la vela misurare la minima distanza tra questo punto e l’inferitura.
Gennaker
S = (SLU+SLE)/2×((SF+(4×SHW))/5)×0.83
a) SLU : il the leinght of the luff. Pull the luff and measure
the distance between head and tack. coincide con la
lunghezza dell’inferitura. Tendere l’inferitura e misurare
la distanza tra il punto di drizza e il punto di mura.
b) SLE : is the leinght of the leech. Pull the luff and measure
the distance between head and clew. coincide con la
lunghezza della balumina. Tendere la balumina e
misurare la distanza tra il punto di drizza e il punto di
scotta.
c) SF : pull the the lower edge of the gennaker and
tmeasure the distance between tack and clew. SF :
tendere il bordo inferiore del gennaker e misurare la
distanza tra il punto di scotta e il punto di mura.
d) SHW : overlap the head on the tack and measure, on the
luff, the point in correspondence of the crease that
identify the half of the luff. Overlap the head point on the
clew point and measure as above the midpoint of the
leech. Unfold the sail and measure the minimum distance,
on the sail, between the midpoint of the luff and the
midpoint on the leech. portare il punto di drizza sul punto
di mura e rilevare, sull’inferitura, il punto in
corrispondenza della piega che coincide con la metà
dell’inferitura stessa. Sovrapporre il punto di drizza al
punto di scotta e rilevare, in maniera analoga alla
precedente, la metà della balumina in corrispondenza
della piega. Stendere la vela in corrispondenza di questi due punti e misurare la minima distanza che intercorre tra il
punto di mezzo della balumina e il punto di mezzo dell’inferitura.
ANNEX C1
DRAWINGS AND INSTRUCTIONS

Measurements on wing sails
Definition
For “Wing Sails” is intended sails where the distance between the upper-wind surface and the lower-wind
surface is higher than the own thickness of the fabric or lamination even for a limitate portion of the chord
line
Measurements
The measurement of the wing sail will consider the larger of the surfaces of the wing eventually adding the
half of the lateral surface of the wing-mast but if this is inserted inside the wing sail
Definizione
Per vele alari si intendono vele la cui distanza tra la faccia sopravvento e quella sottovento sia superiore,
in esercizio, anche se limitatamente, ad una porzione della corda, allo spessore proprio della lamina che le
compone.
Misurazione
Per la misura della superficie di vele alari si considererà la maggiore delle due facce della vela, compresa
la semisuperficie laterale dell’albero, ad esclusione dei casi in cui l'albero sia inserito all'interno della
vela.
The Wing Sails will be measured in working configuration with the devices, such as flaps or slats, if
present, in extended position.
Le vele di tipo alare verranno stazzate nella configurazione di esercizio, con eventuali ipersostentatori di
bordo di attacco (slat), ipersostentatori di bordo d’uscita (flap) e alettoni in posizione di massima apertura.
individuation of the side to be measured for simple and multi sectioned profile
the measurement points will be indicated by the measurer or by the T.C.

The calculation of the surface will be developed by decomposition of the surface in trapezes or triangles
where possible
In case it's present at last one curved edge on the leading edge and/or the trailing edge, will be adopted the
Cavalieri-Simpson method on min. 5 partitions
I punti di misura verranno definiti a discrezione dello stazzatore o della C.T.

Il calcolo della superficie sarà effettuato per scomposizione in trapezi e/o triangoli dove possibile.
Nel caso sia presente anche un solo bordo curvo si adotterà la formula di Cavalieri-Simpson su un minimo
di 5 partizioni
Some examples: Si riportano a seguito alcuni esempi:

Annex D – CREW SHEET
MILLE E UNA VELA CUP
Place ___________________________ date _______
NOME IMBARCAZIONE – BOAT NAME ______________________________________________
Numero velico / Sail number _______
EQUIPAGGIO – CREW
Group (World Federation's
Nome / Name Cognome / Surname Ruolo / Role Sail) inscription N°
1
2
Trofeo “Paolo PADOVA” / “Paolo PADOVA” TROPHY

EQUIPAGGIO – CREW (Regata docente-studente / student-teacher Regatta)
Nome / Name Cognome / Surname Ruolo / Role
Docente/
Teacher
Studente/
Student
Data / Date FIRMA / SIGNATURE
________________ ________________________
Gommone Accreditato/ Assistance Boat
___________________________________________ TEL. ___________________________
Annexed, at this sheet, the registration of the students for the current Academic Year
Certificate of conformity
Place ........................................ date …./..../20........
Sailmaker :
Address :
Tel.:
Mail.
Web:
Certification of conformity for the sail set in accordance of the R3 class rules
I declare that the material used for the realisation of the sail set for the boat:
…......................... sail number: …............. DO NOT includes Kevlar, spectra, carbon or
other high modulus fibres.
I declare that the sail set has been measured in accordance with the “Annex C and C1” of the
R3 Class Rules
Sheet sails
MAIN SAIL JIB
LI JL
E1 LP
E2 JGL
E3 JGU
E4 GENNAKER
E5 SLU
C0 SLE
SF
SHW
Wing sails
H1 X1
H2 X2
H3 X3
H4 X4
H5 X5
Mr/Ms
….............…......................................
(stamp and signature)

Chalmers Formula Sailing

Uploaded by

Copyright:

Available Formats

Chalmers Formula Sailing

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chalmers Formula Sailing

Uploaded by

Copyright:

Available Formats

Chalmers Formula Sailing

High Performance Skiff

Naval Architecture and Ocean Engineering International Master Programme

Department of Mechanics and Maritime Sciences

Department of Mechanics and Maritime Sciences

Keywords: skiff, dinghy, sailing, 1001VELAcup, Chalmers University of Technology, CDIO

To Professor Kai Graf for supplying his VPP.

To Chalmers University of Technology.

CAD Simon Granli

CFD Tommaso Acerbi

FEM Eric Eriksson

VPP Rasmus Andersson

Type High Performance Racing Skiff

CDIO - Conceive, Design, Implement and Operate

Variable Unit Description

A [m2 ] Planform area

6 Balancing of Final Design 97

8 Future Work 105

Appendix B - Plots VII

2.1 Map of Mondello, Sicily . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

3.1 Weibull probability distributions . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

4.1 Keel line curvature graph for final hull model . . . . . . . . . . . . . . . . . . . . 18

5.1 Transonic hull© . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

5.15 Deformations magnified 50:1 for hull-with-trapeze load case . . . . . . . . . . . . 71

6.1 General arrangement drawings of the skiff . . . . . . . . . . . . . . . . . . . . . . 97

7.1 Dog bones in one of the deck’s template . . . . . . . . . . . . . . . . . . . . . . . 99

4.1 Computational domain boundaries . . . . . . . . . . . . . . . . . . . . . . . . . . 22

2.1 Class Rules

As a proper sport competition, the 1001VELAcup is subjected to a regulation class specifically

No limits are put on draft, displacement nor the freeboard.

2.2 The Race

Figure 2.1: Map of Mondello, Sicily

3.1 Wind Statistics

Windguru [m/s] ISPRA [m/s]

Mean Value 3.8438 3.3716

(a) ISPRA data (b) Windguru data

Figure 3.1: Weibull probability distributions

(a) ISPRA data (b) Windguru data

Figure 3.2: Weibull probability density functions

Time step 8-11 [m/s] 11-14 [m/s] 14-17 [m/s]

Mean Value 3.5011 3.7382 4.0253

(c) 14-17 am time step data

3.2 References for Design

3.2.1.1 49er© Hull

Figure 3.4: Model of a 49er©, (Wikipedia, 2010)

3.2.1.2 Transonic Hull ©

(a) Scaled 49er©-based hull (b) Transonic hull©

Figure 3.5: Reference hull concepts

3.2.2 Rig and Sails

(b) Previous competitor in 1001VELAcup, pic-

Figure 3.6: Bowsprit reference concepts

Table 3.3: Important dimensions of the 49er©

Main Sail 16.1 m2

Table 3.4: Important dimensions of Federica (DeSantis, 2016)

Main Sail 13.794 m2

3.2.3.1 Rudder and Centreboard

centreboard area 0.3 m2

Boat class Rudder Area [% of centreboard area] Lead [%]

Figure 3.7: Racks reference concepts

3.3 Overall Design Features

by k − , leading towards an overly optimistic, and therefore, non-conservative performance