Proceedings of the 18th International HISWA Symposium on ‘Yacht Design and Yacht Construction’, HISWA, Edited by P.W. de Heer, ISBN: 90 370 0210 2, Amsterdam, The Netherlands

11

18th

International HIS WA Symposium on

"Yacht Design and Yacht Construction"

Amsterdam, 15 & 16 November 2004

PROCEEDINGS

Organized by HIS WA

- National Association of Watersport Industries in The Netherlands,

the International Trade Show of Marine Equipment METS 2002

and the Delft University of Technology

In co-operation with

EUROPE'S EXHIBITION FOR THE INTERNATIONAL LEISURE CRAFT TRADE AND

INDUSTRY

"Yacht Design and Yacht Construction"

Amsterdam, 15 & 16 November 2004

PROCEEDINGS

Edited by P.W. de Heer

November 2004

Organized by HIS WA National Association of Watersport Industries in The Netherlands,

The International Trade Show for Marine Equipment METS 2004

and the Delft University of Technology

Delft University of Technology Ship Hydromechanics Laboratory

CENDRIS DELFT

CIP-DATA KONINICLIJKE BLBLIOTHEEK, DEN HAAG

18th International Symposium on "Yacht Design and Yacht Construction": Proceedings of the 18th International Symposium on "Yacht Design and Yacht Construction", Amsterdam, 15 & 16 November

2004/P.W. de Heer (editor). Delft University of Technology, Ship Hydromechanics Laboratory, The

Netherlands.

ISBN: 90-370-0210-2

Subject headings: Yacht Design, Yacht Construction

Pone: +31 (0)15 2787019

TABLE OF CONTENTS

PROGRAMME

INTRODUCTION

BEST MAST: A NEW WAY TO DESIGN A RIG

Robert Janssen Msc, Centre of Lightweight Structures TUD-TNO, The Netherlands

THE RIG OF THE "UCA" FINITE ELEMENT ANALYSIS

Guenter Grabe, University of Applied Sciences Kiel, Yacht Research Unit, Germany

THE USE OF FIBRE OPTIC STRAIN MONITORING SYSTEMS IN THE DESIGN, TESTING

AND PERFROMANCE MONITORING OF THE NOVEL FREESTANDING DYNARIGS ON AN

87M SUPER YACHT BY PERINI NAY!, DESIGN BY G. DIJKSTRA

Damon Roberts, Director Insensys Ltd and Gerard Dijkstra, G. Dijkstra & Partners

THE APPLICATION OF SLENDER HULL TECHNOLOGY IN POWERED YACHTS AND

SMALL COMMERCIAL CRAFT

Nigel Irens Design, Tanners Yard House, St Lawrence Lane, Ashburton, Newton Abbot, UK

WATER WORLD OF HUMAN POWERED RACING BOATS

Leo de Vries, Delft Waterbike Technology (DWT), Centre for Mechanical and Maritime Structures

(CMC), TNO, The Netherland

"THE MALTESE FALCON: THE REALISATION"

Gerard Dijkstra, Naval Architect G. Dijkstra & Partners, NL, Tom Perkins, owner and project manager, USA, Perini Navi, Project team, IT, TK and Damon Roberts, Directeur Insensys Ltd, UK

CYCLOMAL PROPULSION

THE QUIET MANEUVERING PROPULSION FOR LARGE

MOTOR YACHTS

Marcel Flipse, Voith Turbo BV, Twello, The Netherlands, Dirk Jiirgens & Torsten Moltrecht,

Voith Turbo Marine GmbH & Co. KG, Heidenheim, Germany

A MATHEMATICAL MODEL FOR THE TACKING MANEUVER OF A SAILING YACHT

Erik Jan de Ridder, Jan Alexander Keuning & Kees Jan Vermeulen, Ship Hydromechanics Laboratory,

Delft University of Technology, The Netherlands

DESIGN FOR STABILITY AND FOR INSTABILITY FINDING THE RIGHT BALANCE FOR

SMALL CRAFT

Richard Birmingham, School of Marine Science and Technology, Newcastle University, UK

DESIGN CONSIDERATIONS FOR CANTING KEEL YACHTS

Andrew Claughton, Wolfson Unit MTIA University of Southampton & Clay Oliver, Yacht Research

International, UK

FINAL PROGRAMME

18th International HISWA Symposium

on Yacht Design and Yacht Construction

Monday 15 November and Tuesday 16 November 2004

Location: Amsterdam RAI Congress Centre, Room A

Monday 15 November 2004

08.30 09.30 hrs

Coffee/Tea/Registration

Moderator:

Jack Somer

10:15 - 10:30 hrs

Opening

Session 1

10.30 11.00 hrs

Robert Janssen, Centre of Lightweight

Structures, TUD-TNO, The

Netherlands

BEST MAST: A NEW WAY TO DESIGN A RIG

11:00 - 11:30 hrs

Coffee break

Session 2

11:30 - 12:00 hrs

Guenter Grabe, University of Applied Sciences Kiel, Yacht Research Unit,

Germany

THE RIG OF THE "UCA" FINITE ELEMENT ANALYSIS Session 3

12:00 - 12:30 hrs

12:30 14:30 hrs

Lunch break

Session 4 14:30 - 15:00 hrs

Damon Roberts, Director Insensys Ltd & Gerard Dijkstra, G. Dijkstra &

Partners

THE USE OF FIBRE OPTIC STRAIN MONITORING SYSTEMS IN THE

DESIGN, TESTING AND PERFORMANCE MONITORING OF THE NOVEL FREESTANDING DYNARIGS ON A 87M SUPER YACHT BY PERINI NA VI, DESIGN BY GERARD DYKSTRA

Nigel Irens, Tannes Yard House, St Lawrence Lane, Ashburton, Newton

Abbot, UK

THE APPLICATION OF SLENDER HULL TECHNOLOGY IN POWERED

YACHTS AND SMALL COMMERCIAL CRAFT

Session 5

15:00 - 15:30 hrs

Leo de Vries

WATER WORLD OF HUMAN POWERED RACING BOATS

15:30 16:00 hrs

Tea break

Session 6 16:00 - 16:30 firs Session 7 16:30 - 17:00 hrs Session 9 09:30 - 10:00 hrs

10:00 10:30 hrs

Session 10 10:30- 11:00 hrs 11.00 11.15 hrs

Closing

Gerard Dijkstra

MAL THESE FALCON: THE REALISATION

Marcel Flipse, Voith Turbo BV, Twello, The Netherlands and Dirk Jiirgens

& Torsten Moltrecht, Voith Turbo Marine GmbH & Co, Heidenheim,

Germany

CYCLOIDAL PROPULSION THE QUIET MANEUVERING PROPULSION

FOR LARGE MOTOR YACHTS

The VCR an Innovative Maneuvering Device for Yachts, Marcel Flipse

17:00 18:00 hrs

Welcome reception

drinks + fingerfood/snacks

Tuesday 16 November 2004

08.30 09.00 hrs

Coffee/tea

Session 8

09:00 - 09:30 hrs

Erik Jan de Ridder, Jan Alexander Keuning & Kees Jan Vermeulen, Ship

Hydromechanics

Laboratory,

Delft

University

of

Technology, The

Netherlands

A MATHEMATICAL MODEL FOR THE TACKING MANEUVER OF A

SAILING YACHT

Richard Birmingham, School of Marine Science and Technology, Newcastle

University, UK

DESIGN FOR STABILITY AND FOR INSTABILITY FINDING THE RIGHT

BALANCE FOR SMALL CRAFT

Coffee break

Andy Claughton, Wolfson Unit MTIA, University of Southampton

& Clay

Oliver, Yacht Research International, UK

DESIGN CONSIDERATIONS FOR CANTING KEEL YACHTS

INTRODUCTION

Once again, now for the 18th time, here in front of you lie the Proceedings of the International HISWA Symposium

on Yacht Design and Construction.

The Organizing Committee is proud that the HIS WA Symposium may be held now for the 18th time, meaning that it

is the oldest and the longest existing symposium in this field on the world. During the period of it's existence, i.e. since 1969, the HIS WA Symposium has always drawn a considerable amount of attention from people interested and active in this field, being it as designer, builder, user or researcher. As such it has drawn the attention of all who are active in the world of yachting and certainly has established the name of the Dutch designers and builders as leading in the field of Innovation, Research and Development. It has also shown that, certainly during this period, the liaison between "the theory" and "the practice" in the Netherlands is very strong, apparent fruitful and quite alive. To the benefit of both. The short "lines" and the short "distances" between all kind of interesting industries, research institutes and technical highly developed parties makes the field in the Netherlands very interesting and very innovative. In this field the Symposium has always played an important role. And will be!

The unique mix of parties that have always organized the symposium, i.e. the HISWA Association, the METS Marine Equipment Trade Show and the Shiphydromechanics Department of the Delft University of Technology, also shows this. For the second time the Symposium is organized in cooperation of the Royal Institute of Naval Architects of London, which strengthens it's role in the international arena.

To select the right topics of interest and to guard over the quality of the papers the Paper Committee has played an important role again. The members of this Committee perform their task out of the spotlights, but take it very seriously and they earn the full gratitude for their efforts by both the organizers and the delegates.

Finally I would like to express our gratitude to our sponsors. You can find their logos on the Proceedingscover.

Without their support the Symposium would be difficult to organize. In addition to this by connecting their names to the Symposium they underline and appreciate its importance.

I hope you will enjoy the material supplied and the gathering at the symposium itself I hope you will be able to meet a lot of interesting people, who are attending the symposium. This time and the times to come!

Jan Alexander Keuning

Chairman Symposium Organizing Committee

The 18th International HISWA Symposium

on Yacht Design and Construction

Organizing Committee

Jan Alexander Keuning

Chairman

Michael Steen hoff

Treasurer

Irene Drost

Member

Papers Committee

Jelle Gerritsma

Richard Birmingham

Gerry Dijkstra

Jan Alexander Keuning

Frans Maas

Michael Steen hoff

Hugo van Wieringen

Best Mast: a new way to design a rig

Robert Janssen Msc, Centre of Lightweight Structures TUD-TNO, Netherlands

Abstract

One of the most difficult tasks for a rig designer is to estimate the maximum loading condition for a rig. These loads determine the mast tube dimensions such as wall thickness and the stay diameters. The loads basically determine the total weight of a rig. The constant drive for better sailing performance pushes the design to the limits, even for cruising yachts. In combination with the growing use of composite materials for mast and rigging, this asks for a new way of rig design.

Best Mast is a generic rig design tool developed for the Dutch spar manufacturer Nirvana Spars® B.V. The tool consists of a new developed force prediction model, estimating the external forces acting on the rig during a specific sailing situation. Subsequently these forces are used in a finite element analyses to determine the structural behaviour of the rig. In several analyses steps the rig can be optimised. Due to the generic set up of the tool, different rig configurations can easily be compared.

This paper describes the development of the Best Mast design tool with special attention to the underlying load model and the finite element model.

Introduction

Nirvana Spars is a Dutch spar manufacturer specialized in building aluminium masts for cruising yachts from 60ft up to 160 ft. Apart from masts the company also manufactures carbon furling booms, poles and deck hatches. Almost all the yachts of the well known Dutch shipyard Jongert are equipped with Nirvana Spars masts. The market for cruising yachts asks more and more for better sailing performance and lighter yachts, but without reducing the luxury. This means a lot of weight saving is required on both yacht structure and rig. As a result complete carbon fibre rigs are slowly becoming the standard for super yachts. On the other hand, the quality and reliability needs to remain high while insurance companies are asking for certainties while they receive more and more claims for broken rigs.

A sailing yacht rig might seem a very simple structure but in reality it is not. It behaves in a very complex manner. The current design methods described in literature and the one used by Nirvana Spars are based on analytical approaches. In these methods various, relatively high and often also inexplicable safety

factors are used to take into account design uncertainties. By using more sophisticated models it is

possible to reduce the various safety factors. This results in either lighter or more reliable rigs. More knowledge is also necessary to make fully use of the benefits of new materials such as carbon fibre for the masts and aramid fibres for the standing rigging. An alternative for the current analytical approach is the use of a finite element analyses (FEA) program. These powerful tools are very useful to analyse the non linear behaviour of rigs, but their reliability heavily depends on the analyses method and the accuracy of the loading input. With respect to the loading of the rig from the sails there is not so much known.

The overall aim of Nirvana Spars is to design rigs in a more scientific way and specifically to be ableto

design high quality aluminium and carbon fibre masts. To achieve this, the company has set up the Best Mast project together with the Centre of Lightweight Structures TUD-TNO (CLS), MARIN, MSC Software Benelux B.V. and Van Oossanen & Associates B.V. The result is a tool based on the finite element analyses program MSC.Marc® Mentat® and a newly developed Force Prediction Program (FPP).

Finite element analyses programs are often very difficult to use due to the large number of features that are not all of interest for a structural engineer like a mast builder. To standardize the design process, MSC Software developed a user interface and a model generator procedure especially for the definition of a rig. The role of the Centre of Lightweight Structures was to develop a load analyses method to determine the loads for the finite element module. The resulting FPP consists of a Velocity Prediction Program module (VPP), mainly based on the International Measurement System (IMS) approach and a Rigging Load Program module (RLP). The VPP determines the forces generated by each sail individually for a certain load case. These forces are subsequently transformed by the RLP to forces acting on the rig. Figure 1 shows a diagram of the Best Mast tool.

Yacht & rigging data i<

Sail conditions

Geometric module Force Prediction Program VPP Module RLP Module

Rigging forces

Finite element analysis

LResuk

Design 3

Figure I: scheme of the Best Mast design tool

The Best Mast program is validated with results of full scale measurements performed on the 97 foot Jongert yacht "Flying Magic". MARIN equipped this yacht with a measurement system that collects 82 data signals full time and they analysed the collected data. Strain gauges were installed on the different panels of the four spreader aluminium rig as well as on the various transverse and longitudinal stays. The performance of the yacht and the dynamic motions are constantly registered a well. At the end of 2003 a series of measurements were performed to collect data for very specific predefined semi static sailing situations. During these trials the sail settings were registered as well. The yacht was also monitored during a transatlantic passage. The collected data gives insight in the dynamical aspects of the rig. After 2 years time the system is still working and is still collecting valuable data.

In this paper the development of the Best Mast tool is explained with focus on the FPP. The next chapter deals with the state of the art design method. This explains the relevancy of the definition of a new design tool. Subsequently the background of the FPP is explained and finally some of the possibilities with the Best Mast tool are shown.

At this stage the tool can evaluate the structural response on static forces from the sails. A future step is to develop a structural model for the response on the dynamic behaviour of the yacht. Until then safety factors are used derived from the comparison of the static results from the design tool with the dynamic and static measurements on the test yacht.

The Centre of Lightweight Structures is a cooperation between TNO, a contract research organisation, and the Faculty of Aerospace Engineering of the Delft university of Technology. The group is specialised in the field of lightweight structures and especially composite materials. They are active in structural design, implementation and optimisation of production processes, testing of materials and fundamental research. Apart from the marine industry they perform projects for the aerospace, automotive and civil engineering

industries.

2

State of the art

The function of a sailing yacht rig is to support the sails used to propel the yacht. To maximise the yachts stability and its sail carrying capacity, the rig should be as light as possible, with the centre of gravity as low as possible. At the same time the windage has to be minimized to reduce drag and the disturbance on

the airflow around the sails. On the other hand the rig should have a certain ability to deform in a

controllable manner, to trim the sails without losing the load carrying ability. These requirements make the design of a rig a very interesting challenge. In this chapter the design procedures as found in literature

([1], [2]) and as used by Nirvana Spars are explained. This will explain the need for a new design

approach.

A rig can be divided in a mast tube and standing rigging supporting the mast. The rigging consists of longitudinal and transverse stays. The whole structure is loaded in the following ways:

Distributed forces and point loads from the sails are acting on the mast and forestay.

Point loads are acting on the mast at the attachments of stays, spreaders, boom, pole and other equipment.

The dynamic behaviour of the yacht causes inertia forces.

The behaviour of a rig depends on all these loads that vary for the different sail conditions.

All rig design calculation procedures found in literature are more or less based on the Skene method. Also the current design procedure of Nirvana Spars is based on this method although a lot of experience is implemented in the form of additional coefficients. Starting point of the Skene method is the transverse stability of a yacht expressed in a righting moment. Figure 2 shows a typical stability curve of a yacht.

1.5.30,GZ,"

GZ 30.GZ,"

GZ,

3

Heel angle[Id weight

GZ

<

-buoyancy force

Figure 2: on the left a typical stability curve and on the right the forces on a yacht responsible for the heeling and

righting moment.

Based on the stability at a 300 heel angle the Skene method estimates the maximum compression force that can occur at the base of the mast with the following formula:

P

,

mast compression force [N]

A = weight displacement [kg]

1.5. GZ30. A- g 1.5- RM30. g ..- gravitational acceleration Irn/s2]

P =1.85- =1.85. b = chain plate width [m]

b/

b/

Gz34,. = righting moment arm at300heel [m]

/2

/2

RK, = righting moment300heel [Nm]

1.5 = coefficient taking into account heel angles greater than 30°

1.85 = coefficient for stays, sheeting and halyard loads

In general mast designers don't get to know the actual stability but only the one at a 10 heel angle. In such cases this value is multiplied by 30 assuming the first part of the stability curve to be more or less linear. As can be seen in Figure 2 this may heavily overestimate the righting moment. Apart from the factor for the extra loading for stays, sheets and halyard loads, the formula represents moment equilibrium in the transverse direction, as shown in Figure 2. The assumption is that the leeward shrouds are slack at this

The maximum compression force in the lower mast panel and the tensile force in the windward stay, both equal in magnitude, are used to determine the required stay dimensions and the panel bending stiffness (El), in both the transverse and longitudinal direction. The general method for the bending stiffness is to use the Euler buckling formula. A distinction must be made between the transverse and longitudinal direction due to the different support lengths.

k2 p L2

g2-E.I

Pc, Euler buckling load for the panel [N]

= E- I =

"

E modulus of elasticity in axial direction of the panel [Pa] k2 . L2 2 I moment of inertia in long, or transverse direction [ini

L length of the panel between the supports [m]

k support factor

An important remark is that the Euler buckling method is a linear representation of a non linear

phenomenon. The formula is a theoretical approach of the buckling or instability load of a compression column. It is only valid for ideal undisturbed structures under a pure compression force. The resulting bending stiffness El heavily depends on the type of support, expressed in the k factor. Figure 3 shows five

different support types for a column with the belonging k factor. The column carries a compression load up to a certain maximum, the bifurcation point. At that moment buckling theoretically occurs, see Figure 3 curve A. 4 Per ibtfurcationpotnt k=0 .5 k=[ k=2 k=0.7 k= 1 4 B & C

Figure 3: on the left five different types ofsupport with the belonging k factor. On the right the force to axial

displacement curve for a column under compression, theoretically (A) and actual (B &C).

In practise, like in case of a sailing yacht rig, this ideal situation never occurs. A distributed force of the mainsail or point loads from boom or stays make that a mast is never in a pure compression state. Right from the beginning there is a certain bending and an axial displacement as shown by curve B in Figure 3. As a result the actual bifurcation point will be below the theoretical value. Whether the structure is able to carry more load after that point, like curve C, depends on the post buckling properties of the column. This

so called global buckling does not automatically mean that the structure will collapse.

For the dimensioning of the rest of the rig, mast and windward rigging, it is

considered as a static

determined structure. The heeling moment at deck level is the result of heeling forces acting at the hinges

between panels and spreaders. Distributed forces as from the mainsail andforces acting between panels

need to be translated to forces acting at the hinges as shown in Figure 4. With equilibrium equations the transverse stay and panel forces can now be determined and so the required dimensions. The dimensioning of the longitudinal stays is also based on the transverse stability.

The Skene method is a rig design method solely based on

the transverse yacht stability and using

analytical formulas. In literature many variations on this method exist with the main differences in the various factors to take into account the modelling assumptions and effects as distributed forces from sails, halyard forces, and longitudinal forces from stays. The background of these factors is often not clear making optimization of the structure very difficult.

Figure 4: the forceofthe sails is translated to forces acting at the spreader heights, on the right a static determined mast structure

Due to this approach it is not possible to examine for example local buckling or the effects of variations in the pre loading, interactions due to swept back spreaders, fore and aft DI stays, jumpers, etc. It is also not possible to determine the general behaviour of a rig, for example bending and deformation under normal sailing situations. It requires a non linear analyses on a three dimensional model to do so.

These considerations lead to the development of a generic design tool with a finite element analyses program and a new developed load model to determine the sail loads acting on the mast.

Development of the Force Prediction Program module

To determine the sail loads on the rig, a load model is developed based on the performance of a yacht. In a static sailing situation the forces generated by the sails are in equilibrium with the hydrodynamic forces. These sail forces are transferred to the yacht at the connection points of sail and yacht or rig. The FPP module translates the sailing situation into rig loads.

The FPP consists of a Velocity Prediction Program module (VPP) mainly based on the IMS approach and a Rigging Load Program (RLP). The VPP predicts for a particular sailing situation the driving and heeling forces generated by each sail. The RLP translates these sail forces to forces acting on the rig. Input for the FPP is yacht, rig and sail data plus data for a particular sail situation. The result is a set of loads acting on the rig which is used as input for the finite element analyses program, see Figure 5.

//

_,

II / I / 1

I'

I I 1 I I , I I I 5

Figure 5: total forces generated per sail as determined by the VPP module are translated by the RLP module to forces acting on the rig.

The velocity prediction module

Common velocity prediction programs are used to predict the sail performance under various sailing situations. The Best Mast VPP predicts the driving and heeling force generated by each sail individually. The belonging speed of the yacht is not of interest for the analysis. Formulas derived from the Delft Systematic Yacht Hull Series (DSYHS) as available in literature ([3]) were used to develop the velocity prediction program from scratch.

41 \ I \ / \ _/ \ / I / I I / I 1 I I

Only four of the total six degrees of freedom are taken into account for the sake of simplicity, see Figure 6. F Vert' mike/

IF

Long" IF Trans" Displ + F_{buoyancy =}

Displ GZ + Fhl heeling arm = 0

Fay,ve + R0 = 0

Fhi Fheei = 0

Figure 6: four degreesoffreedom are solved in the Best Mast VPP. The figure on the right shows the longitudinal and transverse forces plus the vertical forces, the heeling and righting moment.

The total hydrodynamic drag is based on the viscous drag of hull, keel and rudder plus the upright and

heeled residuary drag for the hull keel combination plus the induced drag. For the rudder only the viscous

drag is taken into account, the induced drag is based only on the heeling force generated by the keel.

Effects of longitudinal trim and added drag for sailing through waves are not taken into account

in this

VPP module.

The hydrodynamic heeling force is assumed to be generated by only the keel and rudder, not by the hull.

The heeling force of the rudder is determined by the extended keel method ([4])with a rudder

angle of 2'

for true wind speeds up to 5 knots and 6' for wind speeds above 20 knots while sailing upwind. For

increasing apparent wind angles the rudder angle is reduced.

The aerodynamic forces are based on the lift and drag coefficients, CI and Cdp, as used by the IMS VPP. Based on the series of 9 lift and profile drag coefficients, continuous curves are created as a function of the apparent wind angle for mainsail, jib and spinnaker, see Figure 7. In the Best Mast VPP four different sail

types are distinguished: mainsail, jib, spinnaker and gennaker. The curves

for a gennaker are created by

manipulating the spinnaker curves.

-CI ob Cdp 017

111.MOWNWMIIMMIE

spmneker Up spinnaker

IIIMMILMEMERIIMMEIMMIEM

-CI gennakar C4070*

IMIMNIENIM1101111

MMINSNII

11111MINIMIIMIL,

a.

INIIIMIIMIIIMII

_WEIM11111

NN

INNINEINIIIMINM 111MIIMMI NuMbk

MI

ill MEM

=MN

mo.im

mum'

Mil

IMOINIMMINIEN

EININENII

MI

NINNINNINMINIMENEI

INIIIIINNIL_

MEOPM111111311111111

IIIIIIMI ME

INIMINIMMIMWOMIIII

..., id ...

, ,

...,

IIMIMIIMIIMINIIMMIIIIIMIIIIMIII--6 74,4 GZ 0 10 20 30 40 50 90 70 SO 90 100 110 120 130 140 /50 160 170 180 AVVA

Figure 7: the Cl and Cdp curves for mainsail, jib, spinnakerand gennaker based on the IMS coefficients.

The EMS values assume an optimum setting regarding sail combination and sail trim (profile shape, trim

angle and twist). By introducing reduction factors the VPP can

also be used for single sail settings. The

heeling arm per sail, the vertical distance between the centre of lateral resistance (CLR) and the

1.8 1.6 111 1.5 'a 1, 1 2 11 1.0 07 08 05 0.4 0.3 at 0.1 0.0

aerodynamic centre of effort (Con is based on the relative positions of the CoEa per sail as shown in Table 1. For the mainsail it is a fraction of the luff length with respect to the boom and for the other sails a

fraction of the I height above the deck.

Table I: the position of the CoE° for the different sail types.

Mainsails 0.40 P

Jib or Staysail 0.39 'top-jib

Gennaker 0.60 'top-g

Spinnaker 0.60 I_-top-sp/

The measurements performed on the "Flying Magic" yacht were used to validate the results of the VPP as boat speed and heel angle. For this particular yacht the VPP slightly overestimates the performance. The set up of the VPP is such that it is possible to use more accurate data for a specific yacht if available from tank and wind tunnel tests.

The RLP module

The total driving and heeling force acting at the centre of effort of a sail is the result of a pressure

difference between the windward and leeward side of the sail. The sails can also generate vertical forces;

however these are not taken into account. In a state of equilibrium the three resulting forces are

counteracted by forces from the rig and sheets acting on the sail. The Rigging Load Program determines these forces for the different sail types ([5]). Starting point is the fact that sails are made from cloth that can only transfer tensile forces. Just like a rope the orientation indicates the direction of the force. Due to the different supports the RLP uses different methods for fixed sails like mainsail and jib and for the free flying sails as spinnaker and gennaker.

Mainsail and jib

In the RLP these sails are divided in triangles running from the clew to the luff of the sail as shown in Figure 8. Each triangle carries part of the total generated force; this is assumed to occur as a distributed force along the diagonal running from the clew to the luff of the sail. This distributed force is in a state of equilibrium with the reaction forces acting in the directions of the diagonal at the clew and leech. The angles are different for each diagonal and are a function of the shape and twist of the sail.

clew

7

Figure 8: division of the mainsail in triangles. The force on each triangle is transferred to the rig by a diagonal.

At the luff the sail can slide along either the mast or forestay. The components of the diagonal forces in the leech direction are transferred to the top of the sail and are counteracted by the halyard. For a jib a small part will be transferred to the tack of the sail. The force components perpendicular to the mast and forestay result in a distributed force along mast or forestay. At the clew the forces of the diagonalsare

summed to components in the three main directions acting at the outboard end of the boom, or on the jib sheet. These forces are counteracted by the outhaul and clew tie down for the mainsail or by the jib sheet.

While sailing upwind the driving force of the main and jib is transferred to the yacht by the outhaul and

the jib sheet. The forces at the mast and forestay tend to heel the yacht and pull it backwards. At higher

apparent wind angles the driving force of the mainsail is transferred more and more by the sheet. The

mainsail and jib are assumed not to generate a vertical force so the vertical components of the halyard and sheet counteract each other.

The mainsail shape is rather constant for the different sail angles. Only the angle of the boom relative to the yacht centre line needs to be defined as a function of the apparent wind angle. However the shape of the jib changes with sheet tension and track position. For the jib an entrance angle of the luff leech, with

respect to the yacht centre line, is assumed as a function of the apparent wind angle. Together with the

geometry of the sail, the jib track position and the twist define the sail shape and so the required angles of the diagonals.

Figure 9 shows the decomposition of the resulting force on the diagonal in the vertical plane, to the forces acting on sheet and forestay.

Bow Apparent wind direction Fsail -Mest Chtqn plate Clew Sheet 'Centre tine-', "Streetrt. block

stem, _directionApparent vonxi

Fsail Chia plate Sheet couimuz. -Sheet-it-4_61.94 4 stent

-Figure 9: the left figure shows a horizontal cross section of a jib at clew height in upwind sailing condition with the various forces. The figure on the right shows a high reaching situation.

The vertical component of the mainsail tack force acting at the outboard end of the boom is mainly

counteracted by the mainsheet when sailing upwind and by the yang when sailing lower courses, see

Figure 10. Due to its short length and forward position the yang can cause high loads on the lower part

of

the mast. The RLP uses a yang factor to take into account the distribution of the vertical force between the

sheet and the yang. This so called yang factor is an additional trim facility and needs to be defined

for each sailing situation.

Figure 10: the vertical component

of

the tack force ofthe mainsail is counteracted by the sheet and the yang. Using the yang causes high extra loads on the lower partofthe mast.

The result of the RLP for the mainsail is a distributed force along the mast in both the driving and heeling

direction, a compression force on the mast due to the halyard and forces acting at the boom and yang

attachments. For the jib the result is a distributed force along the forestay and also a mast compression

force due to the halyard. Additionally the jib sheet force is determined. Fsheet vertical

Free flying sails

The spinnaker and gennaker are controlled by two sheets plus the halyard. The tack of the spinnaker is

kept in position by the pole; especially at reaching angles the resulting pole force at the lower part of the

mast can become very high. A gennaker can be set with the tack directly running to the bow or to a

bowsprit or pole. Figure 11 shows for a gennaker the decomposition of the resulting sail force plus the

sheet and halyard forces in the driving, heeling and vertical direction.

In a static state all forces on the sail are in equilibrium. This requires at every apparent wind angle a

certain relation between the resulting sail force and the pointing direction of sheets and halyard which is

dictated by the shape of the sail. In the RLP calculation procedure the sheet and halyard loads are

determined by adapting the shape of the sails, within certain limits, till equilibrium is reached. The shape of the sail is defined by:

the entrance angle of the luff with respect to the apparent wind angle. the angle of the boom to the yacht centre line.

the geometry of the sail.

the position of the sheeting block.

As starting point for the calculation a base transition is defined for both the entrance and boom angle, as a function of the apparent wind angle.

Figure 11: on the left and in the middle the resulting force generated by a gennaker and the counteracting forces

of

the sheets and halyard. On the right a cross sectionofa spinnaker at pole height.

For both the spinnaker and gennaker the following results are generated: a driving, heeling and

compression force acting at the mast. When the pole is used the driving and heeling force on the mast at pole height are also determined. Additionally the sheet, tack and guy forces plus the compression force in the pole are determined.

User interface for the FPP

When using the Best Mast tool the first step is the definition of the rig geometry in the dedicated user

interface built by using MSC.Marc® Mentat®, see Figure 1. When the model is generated the next step is

the definition of various load cases to be analyzed. Part of

_{the required information for the FPP follows}

from the geometry but additional information needs to be defined in a special FPP user interface. The

following input groups are distinguished:

Hull and appendage data required for the VPP module. Rig data required for the RLP module.

Sails data required for both the VPP and RLP module. Load case data required for both the VPP and RLP module.

In the FPP several load cases can be defined, the result of each can subsequently be used as input for the FEA module. A load case is defined by a sailing situation with a certain wind angle and wind speed plus a combination of sails. Figure 12 shows the required input in the FPP user interface.

Ant Pint .111._mc data Nmontlan Snot, Foot A.m7. SW Lontent CMIEEIMIN. AMAndaGe iNniNN SW CoNn *OP* Name

Fkmo New own. 77 15 Mama Venn

Nam It,ona N won.

TWA 77F

IWO

-5...--- Wei Arowetert t17 Nam 1 ... Z.I 1 Danr ...1. I,or z_l

.

JitcorAwatont 1,min :LI 1

ele.S Urea 7

Mee limm tail 11101e zi 1 Pryor. pl.laMot r

Pm.

-OM

1WA 746 Sal Nontmatmo

MOW NI SIN

LaminaeI Re.. I TWA I TWS Sal combintlin

nmeNmm AtrowlI 77 15 MonsalN,Nlow

New /FIN. MeNm NontetA tramen

OFT-1n. Nnt wtNt tomot1 10 STV-- W

km Hann. 1.9M2

ted.mmo % WON

Amman nod NVM APP.. mod Non

Mg nee 32811;-7. N N OnN De.fflorle non I No. NI tenI I0 1

Figure 12: the load case definition window of the FPP user interface, depending on the mode, design or stiffness a safety factor or a forestay sag has to be defined

For each load case distinction is made between the load to be used for rig strength design and rig stiffness evaluation. For rig strength design the worst case rig load should be applied. Therefore a load safety factor is used on the load predicted by the FPP to account for uncertainties. For rig stiffness evaluation the FPP

loads without safety factors are used to evaluate the rig deformation under nominal sailing situations.

During the design phase this enables the designer to judge for example the mast bend, the effect of the

applied transverse pre load or the loads on the stays useful for fatigue analyses. It can also be used to

evaluate the loads during a rig failure.

For a stiffness evaluation it is necessary to define a sag factor representing the maximum sag of the

forestay as a percentage of its length. In normal sailing situations the aim is always to

minimize the sag

because of the negative influence on the performance. This is achieved by tightening the backstay orthe

runner resulting in extra loads on and deformation of the rig. In the Best Mast tool this can be simulated

Figure 13 shows the typical result of the FPP, in this case for the yacht "Flying Magic" when

sailing

upwind in 15 knots of wind. The VPP estimated a boat speed of 11 knots at a heel angle

of 17°. The

windows at the right show the results of the RLP. For both mainsail and jib the distributed forces along the mast and forestay are given in both the driving and heeling direction. Negative values indicate a force to windward or backwards. EM=2111IMPF -11:1110 sho., Si tom. Lk. Iowa 0,0.1,0 1w- N Ham,. 156ArF3-8 Lmtat.na /972M1S N Fad.191-W1V- N Now 19711r Nang ITiiiThSi N lialeffr=10. G111.'11'" ion FO111 N 7.01.9 toms I N LoNN,N1N., mmtNm tSOCnos N Fthap N

FNAwy teN E.07

forte

Figure 13: the results of the FPP for the yacht "Flying Magic" while sailing upwind. The window at the left shows the results of

the VPP module while the two windows at the right show the resultingforces for both mainsail and jib.

]

SAG I N

Best Mast tool to show the required

are free to choose. Stays can be selected

the mast

as fore and aft

a jumper. The first result after the definition of

centre of gravity.

U 3)3)03

350011

act_cAloflatad

user interface showing the generi,

,c1 up of the Be,t tool. METE OF SPREADERS ..1 17304 yed_170 22 'FE, rod.,3 AREA rc,150 Lc. rc4_48 11 46E4 rod_91 14 4004 x,Dd_48 04 AREA rod_60 IS 42220 red_tiO 35 AREA ES kii, ,, 11104 20 1.0221 11 113000110 NaytecRod 2: 062E022: HaytecRed 1> 03-00112 Naateacd -,1 HATFPI, NeNtee.:.R. Vl 141700I12 Itavee,0..1 20 MATEPI, Havte.c.nd 71 137E0I2L HavtecRod 32 ,033E0022 Na,,tecNad 35 vA-75331 ,,-0011.010-. FORCE SECTION 1 ,177 0 2021110 FORCE SEETION 2 .323 10:145 0 -FMK'S SECT, 0 3 FORCE SECT- 0 4 06133 95;145 165713 264653 :111133 BOOT DATA (ASS CT YACHT 14,03 ooneuu NAY AZ UWE 347306

HEEL ANGIE AT MAX GZ

55 037001 RN AT 1 11011EE HEEL 20315 OUCCO

IOW

calculatedalthCaAX

After the definition of the geometry, the FPP module is started and the rigging loads are determined for a

specific sailing condition. The finite element model is built using shell elements for the mast tube, rod

elements for the transverse rigging and beam elements for the longitudinal rigging. The non linear analysis starts with the application of the transverse pre load followed by a pre bend check. If necessary the length

of both fore stay and backstay is altered to achieve the required pre bend. Next the external forces are

added and the load case is analysed.

The general behaviour of the rig can be monitored by monitoring the different load application steps.

Figure 15 shows the geometric model and the deformed structure after the pre tension is applied. Also the final deformed structure is shown where the sag of the fore stay is clearly visible.

htsi>, Me> 11111...4101

/

/

//

/ /

/ /

Tann 7 000...(1. -7 10.407 -1 ,0...000 -:322..000 -1 727..009 OM7X.7

/

r1,00 10.01 100. 0.1001.01

Figure 15: on the left the geometric model, in the middle the model under only pre tension and the final deformed structure on the right.

By examining the vertical displacement of the top of the mast as a function of the load steps global

buckling can be monitored. The analyses also results in the stays forces and the stresses

in the different directions of mast tube, see Figure 16. The later can be used to examine local stress concentrations and the

risk for local mast wall buckling. Based on the results of several load cases the

designer may decide to

change some rig parameters and perform the analysis again.

140000 0000.. 1.0.0 LOO,

Figure 16 stresses in the mast as determined by the Best Mast tool, on the left the top of the mast and on the right the mast up till the first spreader.

The Best Mast tool is validated with the measurements obtained on the yacht "Flying Magic". As

explained in the previous chapter the results of the VPP are compared to the measured boat speed and heel

angle for various sailing situations. The finite element model is validated with the measurements

performed during an inclination test of the yacht and the data from the pre load situation. For the semi static load cases the measurements are split in a static and a dynamic part. The static loads are used to validate the external forces as predicted by the load model. The long term measurements are used together with the dynamic part of the short term measurements to derive dynamic safety factors to be used for the different sailing situations.

At this stage the tool is capable to evaluate the structural response on static forces from the sails. A future step is to implement a model for the mast response on the dynamic behaviour of a yacht. Probably slamming and longitudinal decelerations do have a significant effect on the response where transverse linear and rotational accelerations are less dominant. Until then dynamic safety factors are used which are derived from full scale measurements in combination with the design tool.

Conclusions

In this paper the development of a new design tool for Nirvana Spars is discussed. Current design tools, as until recently used by Nirvana Spars, do not allow for further optimisation of sailing yacht rigs. This is due

to the fact that they are based on a simplified statically determined model of the rig and the use of

relatively high and often not traceable safety factors to take into account several design uncertainties. Lowering these factors may result in particular cases in unsafe rigs.

To be able to design lighter and reliable rigs in aluminium and carbon, a more sophisticated model for the structural behaviour of a rig is needed. Only a non linear finite element analyses can provide a prediction of this behaviour. This requires a reliable input of the external forces acting on the structure. The loads determined by the current design tools are not detailed and reliable enough for this purpose.

For Nirvana Spars a new design tool, Best Mast, has been developed. The tool consists of the finite element analysis program MSC.Marc® Mentat® and a sophisticated load model. Due to specially developed user interfaces for the input of mast and yacht data the result is a very generic rig design tool. The load model is based on a VPP and a force translation routine. The VPP determines the total driving and heeling force as generated by each sail individually. These are subsequently translated to forces acting on the rig like a distributed force on mast and fore stay, halyard loads and boom and pole loads.

At this stage the Best Mast tool can predict the response of the structure on static forces from the sails. A next step is to develop a structural model for the response on the dynamic behaviour of the yacht. Until then dynamic safety factors are used derived with full scale measurements in combination with the design

tool.

References

Claughton, "Sailing yacht design: theory", Addison Wesley Longman Limited, Harlow, 1998 Larsson, L. & Eliasson R.E., "Principles of Yacht Design", Adlard Coles Nautical, London, 1997 Keuning, J.A. & Sonnenberg, U.B., "Approximation of the hydrodynamic forces on a sailing yacht based on the 'Delft Systematic Yacht Hull Series", 15th International symposium on "yacht design and yacht construction-. Amsterdam, 1998

Keuning, J.A. & Vermeulen, K.J., "The yaw balance of sailing yachts upright and heeled", International journal of small craft technology, RINA transactions 2003 part BI, London, 2003 Janssen, R.J., "Comparison of different rig configurations for an Open 60", Delft university of Technology, Delft, 2001

The Rig of the "UCA" - Finite Element Analysis

University of Applied Sciences Kiel Yacht Research Unit Prof. Dr.-Ing. Guenter Grabe

Guenter.Graber&FH-Kielde

1. Introduction

Results of global Finite Element Analysis (FEA) computations of the "UCA" rig are presented in this

report. They are performed within the scope of a research project from the German Ministry of Education

and Research (BMBF) called: "Development of Procedures for the Dimensioning of Rigs from Modern

Sailing Yachts". The aim of the computations is to simulate the behaviour of the rig in three different

conditions:

dock tune, pretension of the standing rigging sailing close hauled with up to 30 degrees of heel sailing on a broad reach with spinnaker.

The maxi racer "UCA" was designed by judeUvrolijk & co Yacht Design & Engineering in Bremerhaven.

The 85'-racer is a yacht designed for racing the DCNA-Challenge in June 2003 and IMS-racing

_{in the}

Mediterranean.

The "UCA" was built by the Shipyard Knieriem in Kiel and launched after only 7 month of work in the

The rig was produced by Nordic Mast in Aabenraa, Denmark according to the sail plan from judel/vrolijk

and a spread sheet supplied by Applied Engineering Services (AES) in Auckland, New Zealand. Nordic

Mast developed the detailed engineering for the carbon rig.

The rig was certified by the Germanische Lloyd in Hamburg [ I ].

2. Description of the "UCA" rig and the FEA models

The "UCA" has a top-fractional rig like the rig of a Volvo Ocean racer of the race in the years 2001/2002. Differences are the five spreaders instead of four and that the spreaders are not in line but are raked aft 22 degrees. The -UCA" rig is larger than a Volvo Ocean Racer rig. The main measurements are:

The rig of the "UCA" is very slim. The staying base of the vertical shrouds is 2.334 m. The relation of the length I divided by the staying base is 13.2. The rig is made out of high tensile and high modulus carbon.

The modulus of elasticity for the mast tube is 110 kN/mm2. The standing rigging is out of rod. Figure 1

shows the sail plan.

Two FEA models are built up. The first FEA model is for the case of dock tune and sailing close hauled.

The second FEA model is for sailing on a broad reach (apparent wind angle (AWA) = 135').

The

geometries of the finite element models are illustrated in Figures 2 and 3.

The global FEA models of the rig are constructed with beam elements for the mast tube, the spreaders, the

boom and the spinnaker pole. Nonlinear link elements with the ability to fall slack are chosen for most of

the standing rigging. Only the forestay is made out of a beam element to make visible

the sag of the

forestay. Bearing elements are used for the goose neck of the boom and the spinnaker pole. The hull of the "UCA" is not modelled. But a spring at the chain plate of the forestay simulates the hull stiffness.

2

Figure 2 FEA model for dock tune and sailing close hauled

I = 30.700 m

P = 32.250 m main sail = 240 m2

J = 9.500m

E= 12.000m

100 % genoa = spinnaker = 140 m2 600 m'

Figure 3 FEA model for sailing on a broad reach, AWA: 135°

The computations for the load cases dock tune and sailing close hauled

_{consider large displacements}

(geometrical nonlinearity). The computations for the load case downwind are performed with small

3. Dock Tune

The rig is pre tensioned in the FEA by initial strains of the standing rigging. The size of the initial strains is computed according to a chosen percentage of the breaking strength of the different rods. Table 1 shows the pretensioning of the single shrouds and the forestay.

Table 1 Pretensioning of the standing rigging

The rig of the "UCA" is pre tensioned with a hydraulic mast jack. The typical procedure for adjusting the pretensions in the rigging is performed in six steps:

the cap shrouds and the forestay are tensioned, all diagonals are slack additional the Dl's

additional the D2's additional the D3's additional the D4's additional the D5's

For every step the hydraulic pressure is controlled and the turnbuckles are adjusted until the computed mast compressions are achieved. Figure 4 shows the tension and the compression forces and also the mast bending curve of the rig after step 6. The forces in Newton [N] are made visible with contours and colours. The width of the contour is proportional to the force. The mast bending curve is shown with vectors in meter [m]. The length of the vectors and the colour correspond with the deformationin three dimensions. The length of the vectors is in an enlarged scale to make the small deformations visible. The mast bending curve is made plain by a black line.

Table 2 lists mast foot compression forces for the steps 1, 2 and 6. In the second column are the values computed with FEA. Chris Mitchell from Applied Engineering Services (AES) in Auckland, New Zealand computed also the compression forces in the mast foot (column 3). He used a spread sheet. The values from the FEA and from AES spread sheet are compared with each other in the third and fourth columns.

-4

N

D Length Breaking strength Chosen % of breaking

strength Pretension force Initial strain

Shortening of shrouds and forestay D1 16,76 7,408 267813 15 40172 0,00104 7,71 D2 12,70 5,868 169713 14 23760 0,00107 6,29 D3 12,70 6,005 169713 14 23760 0,00107 6,44 D4 12,70 5,556 169713 14 23760 0,00107 5,95 D5 14,27 4,990 213858 14 29940 0,00107 5,34 D6 12,70 4,432 169713 29 49217 0,00222 9,84 V1 22,23 7,475 513063 29 148788 0,00219 16,37 V2 19,51 5,831 401229 29 116356 0,00222 12,97 V3 17,91 6,121 338445 29 98149 0,00223 13,63 V4 16,76 5,804 267813 29 77666 0,00201 11,68 V5 12,70 5,274 169713 29 49217 0,00222 11,71 Forestay 17,91 31,982 338445 15 50767 0,00115 36,83 mm m N N mm

Table 2 Mast foot compression forces .007797 .015594 .023392 .031189 .038986 .096783 .05458 .062378 .070175 393891 291824 239807 -187790 135772 -83755 -31738 20279 72296 124314 [N [ml

FEA AES Difference % difference

Cap's only 126915 160678 -33763 -26.6

Cap's, DI 212554 225061 -12507 -5.9

Cap's, all D's 343841 339308 4533 1.3

The compression forces are about the same for the final step. But the mast compressions for the first step differ with 26.6 % very much. The reason for this is that the spread sheet computation can't take into account large displacements. All diagonals are not tensioned in the first step. The mast bends forward caused by the compression forces of the spreaders. The mast compression is smaller when the mast bends than when the mast is a straight line. The mast bending is reduced when the Dl's are tensioned in the second step. In the last step the mast bending is even smaller and the mast foot compressions from FEA and AES are about the same. The comparison shows the importance of applying large deformation effects in FEA for rigs.

4. Sailing close hauled

Sailing close hauled is for a rig with a good aft rigging like on the "UCA" the critical load case. Most of the computations are performed at a standard heeling angle of 30 degrees. Results are deformations like bending curves of the mast and forces, bending moments and torque in the rig. Also stresses are computed. In addition to the standard 30° heeling angle other heeling angles are also computed. The heeling angle was increased from 0° (dock tune) in steps of 50 as high as possible.

All computations are performed considering large displacements (geometrical non linearity).

4.1 Heeling at 30 degrees

A load model for rigs sailing upwind and downwind is developed in the BMBF research project [2, 3] on the base of real size measurements and a FEA of the rig of the sailing research yacht "DYNA" from the Technical University of Berlin. The load model is applied on the Rig of the "UCA".

The loading of the rig for the close hauled case is simulated on the base of the righting moment of the "UCA" at 30 degrees of heel. The apparent wind angle (AWA) is defined to be 30 degrees. The size of the righting moment at 30 degrees of heel is 46,900 kgm. With full main and 100 % genoa the necessary wind speed to heel the "UCA" 30 degrees is 13.3 m/s or 26 knots. The load model for the close hauled load case gives single forces and uniform loads. The single forces act in the three directions of a Cartesian coordinate system at different points like sheave axles of the sails, the boom tip and the goose neck. The uniform loads act on the forestay and the mast tube.

Figure 5 shows an overview like for the dock tune case about the mast bending curve, the tension and the compression forces for a heel of 30 degrees. The largest deformation is in the mast top with 717 mm. In the case of dock tune it was only 70 mm. The sag of the forestay is 156 mm (137 mm aft and 74 mm leeward). That is a very small value. It is only 0.5 % of the forestay length. The tension in the forestay is 141090 N (14,4 t). The mast compression force rises from 343841 N (dock tune) to 498876 N or about 50 t. The leeward rigging is slack above the third spreader. V3, V4 and the diagonals D4, D5, D6 are without tension.

More deformations are visualised in figures 6, 7 and 8. All deformations are in scale with the rig

dimensions. The mast bending in figure 6 looking from the side is visible but small. Figure 7 shows a view looking from above the mast top down to the deck. The whole mast is twisted a little bit clock wise. Figure

8 shows the sag of the forestay.

Figure 9 shows torque in the rig. The torque rises stepwise from the mast top down to the mast foot. Every spreader and every diagonal on port and starboard side change the torque. The spreaders push and the

diagonals pull. Above the third spreader the leeward rigging is slack. There is only small torque.

Bending moments for the mast can be seen in figure 10. The largest bending moments are at the height of the forestay connection to the mast. The leech of the main sails pulls aft. The backstay is not tensioned. The step in the bending moment is caused by the tension force of the forestay pulling at the lever from the neutral axis from the mast to the forestay tang and the halyard pulling at the lever of the halyard sheave.

Figure 13 shows von Mises stress in the mast tube and the mainboom. The von Mises stress is computed

for a homogeneous and isotropic material. Carbon is strongly inhomogeneous and anisotropic. The

-6-computed von Mises stress values can only give an idea of the distribution and the size of the real stress in the carbon.

The largest stress computed after the formula of von Mises is at the backside of the mast at the height of the forestay tang. At that point the bending stress and the compression stress are added. The von Mises stress is there up to 200 N/rnm2. [m] .079738 159477 .239215 .318953

.!

.398692 .97843 .558169 .6379071 .717695 [N] 498876 -427769 356662 -285554 214447 143340 -72232 -1125 69982. 141090

8

Figure 6 Deformed shape looking from the side, mast bending, close hauled, AWA 300,heel 30°

0 .079738 .159477 .239215 .318953 .398692 .47843 .558169 .637907 .717645 [m]

Figure 7 Deformed shape looking down from the top to the deck, sag of forestay and distortion of mast

Figure 8 Deformed shape, sag of forestay, close hauled, AWA 300, heel 30° [N m] -4361 3809 -3256 2704 2151 -1598 -1046 I -493.26 59.331g 611.92111

wawrisim

Noma._miff

TAN01

EWA

WAPA,

Figure 10 bending moments, close hauled, AWA 30°, heel 30°

Figure 11 Stress von Mises, close hauled, AWA 30°, heel 30° 10

[Nm]

76707 56161 35615 -15069 5477 26023 46569 67115 87661 108207 [N/m2] .222E+08 .444E+08 .667E+08 .889E+08 .111E+09 .133E+09 156E+09

.178E+091

.200E+09

4.2 Heeling from 5 to 32 degrees

Figure 12 shows tension and compression forces for a range of heeling angles between 0° and 300. The heeling angle 0° belongs to the dock tune. The forces of the sails rise with the heeling angles and the corresponding righting moments and wind speeds. The windward shrouds are tensioned more and the leeward shrouds become less tensioned. At 15° heel the leeward shroud D5 starts to fall slack. A larger pretension in the D5 would keep the tension in the D5 up to higher heeling angles. It would also reduce the

bending of the mast in panel 5. The rising mast bending reduces the pretension in the shrouds for _the

spreaders are swept aft. At 30° heel V3, V4, D4, D5 and D6 fall slack. The heeling angle for the falling slack of the leeward shrouds depends also on the tension force of the runner. The tension force of the runner is adjusted by initial strain. It is 15.1 kN at the heeling angle of 30°. The runner pulls the rig aft. That reduces the pretension in the shrouds still more. Table 3 shows important data for the different _heeling angles up to 32°. The righting moment is assumed to rise approximately linear with the heeling angle.

Table 3 Data for a range of heeling angles from 00 to 32 °

All computations are performed with the consideration of large displacements_{(geometrical nonlinearity).} The computations don't converge any more when heeling angles are larger than 32° of heel.

The forces of the sails are computed in the load model depending on the righting moment. The computation of the sail loads in the load model assume, that both sails also the main sail are still full standing in spite of the rising mast bending. In reality this will not be the case. The mainsail forces will become smaller and the fore sail forces will become larger. This results in less bending of the mast and more tensioned leeward

shrouds. That stabilises the rig again and the rig will keep standing up to higher heeling angles.

heeling angle

Righting moment

AWS force in mast

panel 1 force in forestay displacement in mast top shrouds falling slack 0 0 0.0 343.8 33.3 70 ---5 7,817 5.4 393.4 104.6 370 ---10 _15,633 7.7 407.8 111.7 418 ---15 23,450 9.4 423.4 118.9 472 D5 20 31,267 10.9 441.3 126.2 531 D5 25 39,083 12.2 460.6 133.5 595 D5 30 46,900 13.3 498.8 140.9 717 V3, V4, D4, D5, D6 32 50,027 13.7 533.1 143.6 823 V3, V4, D4, D5, D7 ° kgm m/s kN kN mm

500000 0° 360000 220000 -80000 60000' 200000 [N] 12 20° 25° 30°

Figure 12 Tension and compression forces, dock tune heel 0°,

close hauled, heel from 5° to 30°

5. Sailing on a broad reach

The downwind load case is not as critical as the close hauled load case for the "UCA" mast has a good aft rigging. The FEA computations for the downwind load case consider small displacements only. It was tried to take into account large displacements but there are still numerical instabilities and the computations do not always converge.

The "UCA" has different gennakers up to 599 m' but no symmetrical spinnaker. The developed load model

for downwind sailing is based on the symmetrical spinnaker of the "DYNA". In the computations a

configuration of full main sail and a fictive symmetrical spinnaker with 600 m2 is considered. The chosen apparent wind angle (AWA) is 135 degrees (broad reach). The loading of the rig for the downwind case is

simulated on the base of the defined apparent wind angle and a chosen apparent wind speed (AWS). It is

always difficult to define a sound AWS for downwind sailing. The chosen AWS is 10 m/s or about 20

knots. 20 knots AWS is quite much for the 600 m2 spinnaker. The load model for rigs sailing downwind

with a spinnaker is explained in detail in [3].

Figure 13 shows the deformed shape, figure 14 the deformations, tension and compression forces. The

maximal mast deformation is

with 272 mm much smaller than sailing close hauled. Most of the

deformations are to leeward and not aft. The main boom and the spinnaker pole move upwards a little bit.

The compression force in the mast reaches 382 kN. The tension force on the windward VI is 172 kN. No

part of the standing rigging is falling slack.

Torque in the mast tube is shown in figure 15. The boom and the pole distort the mast anti clock wise. The rotating direction of the torque changes at the goose neck. The maximal torque is with 7387 Nm larger than sailing close hauled (4361 Nm). The torque will be still bigger considering large displacements.

Figure 16 visualises von Mises stress. The von Mises stress is smaller when sailing on a broad reach than when sailing close hauled. Hot spots are on the front side of the mast. Here the spinnaker pole pushes into

the mast. That leads to compression stress that is super positioned with the mast compression

_{stress. The}

stress level is with 50 N/mm2 smaller than sailing close hauled (200 N/mm").

All computations are performed for an AWS of 10 m/s. The wind speed can rise in a sudden gust by a factor of 1.5 or more. That raises the sail forces by a factor of 2.25 or more. The rig will be loaded much more in the case of a sudden gust than computed here.

14 [M] 0 .030169 060338 .090507 120677 150846 .181015f .211184 .241353' .271522 [N] 381716 320147 258578 -197009 -135440 7 387 2 12304'3 49266 110835

I

1724032

Figure 15 torque, broad reach, AWA 135°, AWS 10 m/s

Figure 16 von Mises stress, broad reach, AWA 135°, AWS 10 m/s

[Nm] -7387 6244 5102 -3959 2817 1674 -531.864 610.636 1753 2896 [N/M2] .556E+07 .111E+08 .167E+08 .222E+08 278E+08 .333E+08 .389E+08 .444E+081, .500E+0811

6. Conclusions

A finite element analysis is performed for the rig of the "UCA". Three steady state load cases are computed: Dock tune

Sailing close hauled

- Sailing on a broad reach.

All three computed load cases are steady state load cases. There are also unsteady loads. E.g. inertia forces caused by movements in a sea way like pitching when sailing close hauled enhance the steady state loading of the rig. Also peak loads occurring when a spinnaker after having lost the wind fills itself up again with a "bang" are neglected in the performed computations.

A tall rig like that of the "UCA" is a very complex structure. A lot of effects can be considered today applying FEA for rigs. But there are still some imponderables in known - and also until today - unknown effects. The human factor in tuning the rig and the sails depending on wind and waves plays an important part. FEA is only an endeavour to simulate the real world of

the rig of the "UCA".

7. References

[ 1] Germanischer Lloyd: Rules for Classification and Construction, Ship Technology, Special Equipment, Guidelines for Design and Construction of Large Modern Yacht Rigs, 2002

[ 2 1 Grabe, G., The Rig of the Research Yacht "DYNA" Measurement of Forces and FEA, HP-Yacht,

Auckland, 2002

[ 3 ] Grabe, G., Downwind Load Model for Rigs of modern Sailing Yachts for Use in FEA, 16th Chesapeake Sailing Yacht Symposium, Annapolis, Maryland, 2003

16-The use of fibre optic strain monitoring systems in the design, testing and

performance

monitoring of the novel freestanding Dynarigs on an 87m

SuperYacht by Perini Navi, design by G Dijkstra.

By

Damon Roberts, Director Insensys Ltd

Gerard Dijkstra, G Dijkstra and Partners

SUMMARY

The use of novel fibre-optic based strain monitoring systems opens new opportunities for

structural and performance measurement, particularly in composites where the fine fibre-optic

cable networks can be embedded within the structure to provide ongoing, absolute and repeatable

data to an extent not previously attainable.

The potential future possibilities are demonstrated by reference to this 87m yacht project. Three

freestanding 57m tall carbon fibre masts, each supporting cross yards and sails are currently

in-build. The sails furl into mast cavities (creating further structural demands!). This is the largest

yacht built with freestanding spars. To minimise the risk structural models have been tested

including two full size yards and mast to represent one' panel' and a sixth size model of the

mast. The models incorporate the fibre-optic strain sensing system the data from which has been

used in the design process

Data from the rigs on board will provide the operator with real load against design limits and

background data for design corroboration and ongoing monitoring. Driving and heel force

figures from each spar will be presented in real time.

Figure 1. Sail plan.

8,1 CLIPPER VACI, A SF. -orr.

dl

1

INTRODUCTION

1.1

VESSEL

This 87m vessel is currently in build by Perini Navi. It is a three masted sailing clipper

ship with Naval Architecture by G Dijkstra and Partners and interior and exterior design

by Ken Freivokh. The masts are freestanding and are based on the Dyna Rig concept. The

rig engineering and rig build management is by Insensys

1.2

DYNARIG

The Dynarig owes its origin to work done in the sixties by Mr W. ProIls, at the time he

believed the system could provide additional propulsion for ships. The Dyna Rig is

effectively a square rig. The mast is freestanding and the yards are connected rigidly to

the mast, in this case each mast supports six yards. The yards, unlike a conventional

square rigger, have built in camber of 12%. The sails set between the yards in such a way

that when deployed there are no gaps to the sail plan enabling each spar's sail plan to

work as a single sail. The sails, when not deployed, furl into the mast. The sail is trimmed

to the wind direction by rotating the mast .As there is no rigging the yards have no

restriction on rotation and this taken together with the curved (shaped) yards, low

windage and effective single piece sail combine to give the rig improved aerodynamic

efficiency compared to a traditional square rigger

1.3

TEST PROGRAM

The development of this novel rig for such a large yacht takes the structural design into

areas where there is no historic experience. Empirically derived standards are not

available for this design. It is thus essential to carry out extensive modelling and testing

to minimise the development risks In order to maximise the benefit of such testing

extensive use is made of fibre optic strain sensing systems embedded into the composite

structure. The data from these sensors provides the basis for

comprehensive real time

analysis of the response of the test structures and will be used to corroborate the design

in

the final as built structure and can provide useful performance limits.

2

STRUCTURAL ISSUES

Lightweight high strength modern composites permit the building of large freestanding

rigs to reasonable weight and good fatigue resistance. We have previously successfully

applied novel design rules to develop the successful large freestanding rigs including the

Aero Rig which has been proven in Round the World voyages on yachts of up to 95ft in

length (currently an AeroRigged sloop of 160 ft length is in build). The Aero Rig design

engineering was acclaimed when awarded finalist status in

the prestigious MacRobert

award by the Royal Academy of Engineering in 1999. However the extrapolation of the

design approach to a yacht of 87m (290ft) length overall requires special design and

engineering consideration

2.1

MAST

The Masts are approximately 57m in height above the bottom bearing. The Dyna rig

concept calls for an elongated section (to

reduce the drag) and this needs to be

symmetrical as the rig can be tacked to allow flow in both directions, The mast rotates

about deck and heel bearings. It is weight effective and practical for the section to be

circular at these points. The principle structural demand for a freestanding rig relates to

the bending moment requirement brought about by the application of the driving and drag

forces through the sails at the mast attachment points. This is maximum at the deck