12th International Symposium on
"Yacht Design and Yacht Construction"
V
Amsterdam, 12 and 13 November 1992
PROCEEDINGS
Organized by HISWA - National Association of Watersport Industries in theNetherlands and the RAI Exhibition Centre
and the Delft University of Technology at the occasion of the Marine Equipment Trade Show
W2
ura--""At'
TU Delft
University of Technology Ship Mybrobaohanica LaboratoryTable of contents
Introduction J. Gerritsma
Developments in Offshore Yacht Design 1
E. Dubois
Design and Construction of the America's Cup Yacht 12 "Challange Australia"
P. van Oossanen
Exhaust Cleaning Systems 50
J. Zijp, A. Hergart
The Effect of the European Directive on Recreational Boating 65
T. Nighy
Paint and the Environment 77
L. Bennison
Carbon Fibre Reinforced Spars and Masts 95
A.H.J. Nijhof
Sailing Yacht Performance Calm Water and in Waves 115
J. Gerritsma, J.A. Keuning, R. Onnink
DYNEEMA High Performance Fibres in Yachting: 150
Hulls, Sails and Ropes M.J.N. Jacobs
DEVELOPMENTS IN OFFSHORE YACHT DESIGN
Ed Dubois
Dubois Naval Architects Limited
Solent House Bath Road Lymington
Hampshire SO41 9RU
1. Evolution
The evolution of offshore yacht design over, say, the past thirty years
has been largely caused by three main factors:
Dramatic growth in the number of boats built - i.e. a market
explosion.
Improvement and development of materials from which vessels are built.
Rating rule changes.
Figure 1: UK Marine Industry Growth since 1965
1100 MILLION POUNDS 990 880 770 660 560 440 330 220 110 0 /965 1970 1975 [ WNW ...VIM impo.on. 1980 YEAR 1985 1990 1995 180 160 140 120 100 So so 20 0 /970 200 MILLION POUNDS 1975
SAILING BOATS POWER BOATS
1980 1985
YEAR
1990
The increase in the market size has led to competition to sell yachts of all types and at the offshore racing and cruising level this competition was centred primarily around success on the race course.
1995
1
There was once a maxim often heard in the 1960s and even the 1970s which was that the best form of cruising yacht was a good, perhaps retired,
racing yacht. While this adage seems ridiculous today there was
definitely good sense in its message 30 or even as recently as 15 years ago. This was because offshore racing yachts built to the RORC and CCA rules in the mid 1960s were strong and weatherly. Such a yacht would
look after its crew in the worst of weather even if badly handled.
Because they were seaworthy and quite fast, this type of vessel was good for cruising under sail and yacht builders keen to expand with the
growing market often looked towards offshore racing success when considering what type of design to adopt and offer to the public.
The introduction of glass fibre reinforced plastic as a boat-building material combined with increased wealth in Europe and the United States to allow the formation of companies producing many boats to exactly thE
same design with the use of moulds. The prices of such vessels were far less than labour intensive "one off" yachts which were all that was
available until the 1960s.
The pattern of copying successful offshore racing yachts for mass production was established and continued from the mid 1960s until about the mid 1980s when yachts produced by the International Offshore Rule (IOR) became so light, so tender and so short on internal volume that their worth as cruising yachts diminished to practically zero.
As a result of this change, the 1980s saw an enormous diversity of boat types come on to the market; some fast and sporty, some voluminous and comfortable, and some pretending to be both, but which, in fact, were
probably neither.
Meanwhile the number of boats being built to the IOR diminished
dramatically as no further use could be found for them after their
initial and very brief usefulness as racing craft.
The IOR was developed from the RORC and CCA rules in an attempt to allow international racing. It was a success in that it promoted such events
was starting from the premis of allowing all types of boat to compete, it was necessarily complicated and full of so-called loopholes. These loopholes allowed designers to come up with jumps in performance, some of which were accepted by the rulemakers and others which were one by one limited or closed off altogether. Hence one can think of a diversity of IOR yachts and their designers running free over the field of ocean racing, eventually to be penned into a small operational area by the sheepdogs of the rulemakers, at least in terms of the overall concept of
design.
Displacement/length ratio, sail area/displacement ratio and end shape were all controlled and design became almost stereotyped. Designers
therefore looked at so-called unrated areas in which to gain a speed advantage. Construction and rig design are largely unrated within the IOR and yachts became particularly expensive because large amounts of money were used to develop these two specifics. This had the combined effect of both reducing the weight of structure and rig and also
severely limiting the useful life of such craft. Hence IOR racing
quickly became the preserve of those who were willing to spend
relatively large amounts of money to compete at the top level. This keenness to win, seemingly at any price, also embraced the newly formed professional sailor - an expert who, since 1985, could be paid to sail the boat better than the owner who had to spend his time earning money
in a different way.
The sport of offshore racing at, say, Admiral's Cup level had therefore changed enormously from the late 1960s to the late 1980s. Those who were
not prepared to spend large amounts of money to win in IOR boats and who
still wanted to sail the yachts themselves looked to other types of boat, and offshore one-design racing (and cruising) has grown up in the place of the large IOR fleets seen in the seventies. However, part of the enjoyment and satisfaction that owners found in planning an IOR campaign was choosing the design and following its build. The selection
of the design and working alongside the designer, planning the layout of
the yacht and watching its creation at the shipyard often gave owners great pleasure and many designers have noticed that the establishment of IMS racing has been caused partly because of the vacuum left by the
demise of the IOR.
Figure 2: Race boat rating certificate statistics since 1985
2. The Current Position
So we now have offshore racing which is split into several distinct arenas:
IOR or Grand Prix racing where the owner is prepared to fund the building of such boats and the sailors who take part in it.
ii One-design offshore racing which is largely amateur and very much less expensive - Such boats normally have good resale value.
iii IMS racing which should, in theory, give fair time allowance racing for all types of yacht, be they production boats or
specialist one-off craft built to include the so-called
"cruiser-racer" facilities insisted upon by the rule.
Of course, the ideals of IMS are still unrealistic and much work needs to be done if its aims are ever to be achieved. However, there is a
market to fill its ideals.
Not mentioned so far has been the Channel Handicap System, developed mainly in England and France, and the new so-called Grand Prix rule which is to take the place of IOR in 1994/95.
The Channel Handicap System (CHS) is a subjective concept and not really a rating rule. However, it gives good racing to a large number of yachts.
7.666
X 111U1, 10.6
FOG
The Grand Prix rule is intended to take the place of IOR and is likely to be based on a Performance Prediction Programme (VPP) such as is used
for IMS. However, the intention is that boats will be designed to a given VPP so as not to exceed an envelope of performance, at certain
pre-determined level rating bands.
Because an enormous diversity of racing and racing boats currently exist, the Offshore Racing Council (ORC) has to make and implement major decisions to stabilise offshore racing and encourage owners to take part at the top level. It is thought that by stabilising and thinning down
the number of types, greater enthusiasm and more owners will be encouraged.
3. New Yacht Types
To help illustrate design type development, we can discuss the relative merits of a typical IOR boat designed today and its IMS counterpart.
Figure 3: Typical 1992 IMS 40 Racer/Cruiser
DUBOIS 5
To make a good comparison we can look at a One Tanner and a 40 foot IMS boat. The One Tonner is shepherded into a "pen" of displacement/length and sail area/displacement ratios which means that to be competitive against the existing fleet in all conditions the displacement and the length cannot vary very much at all from known limits without rendering the yacht vulnerable in a certain set of conditions. The other limiting
factor is stability. Stability, quite rightly, is rated using an inclination test.
The improvements that can be made in performance largely centre around construction and rig, as mentioned before. To this end, the weight of a structure must be reduced while still maintaining an overall stiffness and strength in compliance with ABS rules (part of the IOR rule).
High strength/high modulus materials such as carbon fibre are used and this has the effect of producing extremely high ballast ratios in IOR boats. The ballast, however, is not sited too low as this would produce too great a stability for all round competitive performance. The rating would go up significantly and this increase in rating would only be
reflected in extra speed in high wind conditions.
Figure 4: Comparison between 1991 IOR 40' and 1992 IMS 40'
PARAMETER IOR 40 IMS 40'
LOA 11.99m 12.05m LWL 9.44m 10.31m BMAX 3.76m 3.83m BWL 2.94m 2.77m Draft 2.30m 2.37m I 14.05m 15.00m 4.11m 4.35m 15.30m 15.50m 5.64m 5.70m Displacement 5.48 t 4.93 t Rig @ 10 142 kg-m 165 kg-m Weights
Hull, deck and structure 679 kg 1260 kg
Internal ballast 3178 kg
Hence, on a One Tonner, displacing say 5,400 kilograms, we have 4,000 kilograms of weight as ballast. However, 3,100 kilograms of this is
positioned in the bottom of the boat, not in the keel. The pitching period of such a boat is extremely low and the energy expended in pitching is therefore reduced over a boat where the structure weighs a great deal more. The cost of such a yacht is increased because of the high cost of materials necessary to achieve such a construction.
By contrast, an IMS boat is rated for all conditions, and high stability is only taken account of when such stability is useful. Therefore, there is no incentive to limit the righting moment. Equally, in recognition of the high cost of carbon fibre, for example, this material is not allowed under IMS rules. Hence the structure weight goes up considerably but so can righting moment because there is greatly reduced incentive to use
internal ballast.
Of course, pitching is still a big factor and the rule makers are currently working on ways to rate this. This would further reduce the incentive to use expensive building materials and would also, more
importantly perhaps, allow fully fitted cruising yachts to take part on more equal terms with the purpose-built IMS racers.
However, the VPP system of measurement under IMS, and the incentive to increase stability rather than limit it, means that a typical IMS 40 footer has the same overall length as a One Tonner but can be built using relatively inexpensive material to displace up to 500 kilograms less than a One Tonner, all on a waterline some one metre longer. Performance, both upwind and downwind, is therefore greatly enhanced,
along with handling and the seaworthiness.
The modern IMS type can therefore be thought of as a big improvement on the IOR type if speed, general handling and seaworthiness are to be the watchwords. The rule makers, it could be suggested therefore, must make every attempt to reduce the cost of such yachts and indeed the cost of running and maintaining these boats within the racing world.
4. Advances in Cruising Yachts
Moving on from the racing world, the changes in cruising yacht design have been no less apparent, but rather than diminishing the number of boats, as development in TOR yacht design has done for the racing world, the labour saving devices and speed improvements through technological development have seen the number of cruising yachts, and in particular
larger cruising yachts, increase enormously.
A general feeling of economic well-being in the mid/late 1980s combined with development in rig handling systems to allow large cruising yachts
to appear offering spacious and luxurious accommodation with good performance. Systems to limit the number of personnel required to operate such craft were developed, including furling headsails, furling
mainsails and powered winches.
New materials and new methods of using such materials were introduced into cruising boats from the racing boat field so that performance was
increased. Equally, the interior of such boats received far greater attention than had been the case previously. A new breed of designer became established the designer specialising purely on the interiors
-not a naval architect but a space planner and stylist.
The builders of such boats suddenly had to be expert in three more fields, other than the already established areas of construction,
conventional joinery and conventional machinery installation: they now had to cope with highly sophisticated interior construction, complicated
and sophisticated powered rig control and elaborate electronic navigation and communication systems.
Scope for development was enormous and so, by definition, was the scope for misplaced complexity and systems failure. The yacht designer, more than ever, had to decide on what was appropriate and what was likely to be inappropriate or even dangerous in a yacht which, after all, should be safe and as far as possible comfortable, in all conditions in the
open ocean.
The barrage of possibilities has enabled some wonderful concepts to be seen and the yacht designer has had a fascinating ten year period where he has been able to construct a vast array of yacht types.
However, the traditional tools of design development have not changed; they have been added to by the VPP systems but the test tank is still an
invaluable design aid, along with the wind tunnel.
Recently, our office conducted tank testing and wind tunnel work on a 42.5 metre craft to try and establish the best beam/length ratio
(tabulated below). Greater beam need not be a disadvantage to performance but too much beam, while promoting stability, can make for a
vessel which is not so sea-kindly.
Figure 5: 42.5m Fast Cruising Ketch Research Project
Optimum Upwind and Downwind VMG results
Carbon fibre, developed by the aerospace industry and greedily adopted by racing yacht designers and builders, has some wonderful properties and is currently being used in increasing amounts for the development of mast and rig construction for larger cruising yachts. These boats often have severe draft limitations for practical reasons and large weight savings can be achieved by using carbon fibre for the construction of
9
True Wind Speed Upwind Downwind Upwind Downwind
5.00 kts 3.06 -3.60 kts 2.95 -3.54 kts 7.50 kts 4.48 -5.36 kts 4.34 -5.27 kts 10.00 kts 5.59 -6.96 kts 5.52 -6.86 kts 15.00 kts 6.90 -9.69 kts 7.01 -9.59 kts 20.00 kts 7.46 -11.73 kts 7.70 -11.66 kts 25.00 kts 7.69 -13.21 kts 7.97 -13.17 kts 30.00 kts 7.76 -14.32 kts 8.06 -14.29 kts Hullform Data
Parameter Narrow Hullform Wide Hullform
LOA 42.50 m 42.50 m LWL 34.00 m 34.00 m BWL 7.59 m 8.08 m BMAX 8.16 m 9.08 m Hull Draft 1.94 m 1.88 m Displacement 227 t 227 t RM @ 1 deg. 7737 kg-m 10000 kg-m
masts and rigging. This area of development has probably caused the greatest improvement in performance in large cruising boats over the last 5 years. Not only is speed improved but heel angle is reduced and roll and pitch are greatly reduced in a seaway making for greater
comfort; a very important factor.
Figure 6: Typical rig weights for a 108 ft sloop : aluminium vrs carbon fibre
Construction Mast Tube Rig Weight (excl. boom) Boom weight
Aluminium 1188 kg 3179 kg 420 kg
Carbon fibre 650 kg 2706 kg 302 kg
Conclusions
Clearly, the world of yacht design and construction is diverse and
complicated - it is driven by imagination and a love of the sea rather
than by commercial forces - no one "needs" a yacht.
Equally, yacht design is by its physical nature a complicated science. Any boat, or as it has been called sometimes, any "water/air interface vehicle", is bound by a complicated set of physical laws. Because there is no commercial need for a yacht, most yacht design has limited funding and designers create their yachts with a limited knowledge of many
areas; hydro-dynamics, engineering, fashion, etc.
The one arena which has received a good deal of funding has been the America's Cup and these developments are passed down to "lesser" forms
of yacht design.
So the designer has many areas from which to amass information but much of it is limited and all of it is subject to his interpretation. Thankfully, at least for the likes of the design houses, the very
absence of true commercialism within the yacht design and construction
industry means that the small design houses can still exist and a personal relationship can be developed between client and designer. This
********
With regard to the future of racing, the flirtations with sponsorship at Admiral's Cup and level rating level during the late 1980s combined with professional sailors and yachts which cannot be re-sold effectively, amounted to a crash in the number of boats being built. I would like to propose that there may be a future in returning to the concept of the one-off yacht, built within size limits rather than to a level band, to encourage owners to express their individuality and enjoy the planning stage.
Perhaps handicap racing has a place at Admiral's Cup level alongside
level rating racing. The IMS style of yacht is, I believe, a success but
rather than banning certain materials, perhaps it should be restructured to allow only certain materials and building methods. This way the cost could be limited more easily. Equally, perhaps a divergence away from
the stereotyped windward/leeward and Olympic courses should be encouraged. Racing in The Solent, for example, with all its unfairness and peculiarities appealed to many - weather conditions will always mean that either smaller or larger boats might be favoured but is this really
so bad?
On the cruising front, recent economic restraint around the world has meant a retrenchment in yacht design type - we have seen a move towards more traditional craft being favoured by many but the development in
material type and systems means that yachts can offer both good performance and relative comfort.
One thing is certain; a yacht designer has many more possibilities with which to create and from which to build a yacht than ever before.
Acknowledgements: Royal Ocean Racing Club 'Channel Handicap 1992'
British Marine Industries Federation statistics
Paper to be Presented at the 12th International Symposium on "Yacht Design and Yacht Construction", The National Association of Watersport Industries (HISWA), Amsterdam, 12 and 13 November
1992.
by: Dr. Peter van Oossanen July 1992
DESIGN AND CONSTRUCTION OF THE AMERICA'S CUP YACHT "CHALLENGE AUSTRALIA"
DESIGN AND CONSTRUCTION OF THE AMERICA'S CUP YACHT "CHALLENGE AUSTRALIA"
by Dr. Peter van Oossanen Van Oossanen & Associates.
Australia: fax 61-2-4361327
Netherlands: fax 31-8370-24673
SUMMARY.
This paper describes the design and construction of "Challenge Australia", the Australian yacht that challenged for the
America's Cup in 1992. The races for this prestigeous sailing throphy in 1992 were the first in the new International America's
Cup Class, defined in 1988-1989. Special consideration in the
paper is given to the mathematical model that was used for the
investigation of optimum length, volume of displacement, sail
area, and other design parameters. The model tests carried out
for the enhancement and validation of this mathematical model are
described and specific results given and discussed. The final
design and construction of the canoe body and the chosen keel
configuration, and the performance thereof, is dealt with in detail. An account is also given of the performance of the yacht in the actual races for the America's Cup, together with an
explanation for not performing as well as expected.
1. INTRODUCTION.
From May 1989 through to February 1992, the author was head of the design team for tie Australian Challenge for the America's
Cup (ACAC). Together with David Lugg (structural design) and Neil
Loveland (design draughtsman), he was responsible for the design of "Challenge Australia", which participated in the races for the
America's Cup, off San Diego, early in 1992. This was the first
America's Cup in which the new International America's Cup Class (IACC) yacht was used.
The ACAC Syndicate had available a relatively small budget,
allowing for the design and construction of only one yacht. This
required that all of the research into the merits of different
design configurations had to be carried out by mathematical
modelling and model testing since comparative, side-by-side,
full-scale experimentation requires two or more yachts. It was
therefore decided to utilize 18 months of the available time for
research and design-oriented work, and to start building the
yacht no sooner than October 1990, leaving about 9 months for
the building programme and about 6 months for sailing,
perfor-mance evaluation, crew training, etc, before the races for the America's Cup began.
While it was realized that other teams competing for the 1992 America's Cup had significantly greater budgets and more
technical resources, it was also realized that if a novel canoe body, keel or rudder configuration could be found, leading to superior performance, it would be possible to bridge any
superiority in speed the other participants may have, caused by
having better sails, by better crew work, etc, resulting from having more facilities and better crew training possibilities. It
follows that the crux of the ACAC campaign plan lay in the
research, development and design area, this being another reason to reserve more than half of the available 32 months for this part of the campaign.
This paper describes the research and design work carried out, starting with the initial model test programme to generate a side force and resistance data base (Paragraph 4), the work done to improve the author's VPP used for determining the optimum length, displacement and sail area (Paragraph 5), the optimization study itself (Paragraph 6), the meteorology study for San Diego
(Paragraph 7), the second series of model tests to determine the optimum waterline beam and the best type of overhang forward and aft (Paragraph 8), the design of the final canoe body and
appendages (Paragraph 9), the final series of model tests
(Paragraph 10), and the structural design and construction of the yacht (Paragraph 11). Finally, in Paragraph 12, an explanation is given for the yacht's non-optimum performance in San Diego, followed by some final remarks and conclusions.
2. ADOPTED DESIGN METHODOLOGY.
An intricate plan was drawn up, composed of many mathematical modelling and model experimentation elements. This is shown in Figure 1, where the various work-elements are indicated in rectangular blocks.
An approach such as outlined in Figure 1 is essential in order to determine the canoe body and appendage configuration with the best performance. The central role herein of a mathematical model
(VP?) for the calculation of the speed of a specific design is necessary in order to investigate the desired value of design variables such as length, displacement, sail area, waterline
beam, canoe body draught, stability, longitudinal position of the centre of buoyancy, prismatic coefficient, size of keel and
rudder, etc. It is not possible to investigate all of these
design variables by model or full-scale testing, or a combination
thereof, due to the sheer magnitude of the number of variables,
while few, if any, yacht designers can make accurate judgements
about the required value of a large number of these variables on the basis of experience alone, particularly as the IACC yacht constitutes a new class.
On taking the inevitable decision to use a mathematical model, i.e. a computer program for the investigation of the main design
variables, it is important to possess a sound knowledge of the
strong and weak points of this mathematical model. In the case at
hand, the author's VPP was used which incorporates the results
and experience of many previous projects. Although this VPP has
been continuously updated and improved over the last years, it
was considered to be not accurate enough in certain areas. A
relatively large effort was made to improve the mathematical
modelling in those areas. Even so, the final decision with
respect to the required waterline beam, longitudinal position of
based on model testing rather than calculations. It was
con-siderad necessary, for the same reason, to validate the
perfor-mance of the final design through a full series of model tests,
before the building programme commenced.
EXPAND DELFT (DUT) DATA
BASE.
MODEL TESTS FOR 3 DIFFERENT BWL AND DIFFERENT FORWARD AND AFT OVERHANG DESIGNS.
MODEL TESTS WITH CHOSEN APPENDAGE CONFIGURATION.
TEST MODEL WITH FINAL KEEL, BULB AND RUDDER.
REFINE MATH MODEL IN VP? FOR: -side force; -induced drag;
wave resistance; -resistance due to heel
-wave-added resistance; -include moment
equil-ibrium in horizontal plane to calculate keel/rudder angles.
DETERMINE OPTIMUM BWL AND BEST FORWARD AND AFT OVERHANG SHAPE. OPTIMIZATION OF iE, CWP, CP, CX, AND LCB USING REFINED VP?. DEVELOPE FINAL LINES OF CANOE BODY. DESIGN FINAL KEEL AND RUDDER CONFIG. FINE-TUNE MATH MODEL FOR SIDE FORCE AND RESISTANCE OF KEEL, RUDDER AND BULB. DESIGN FINAL BULB. CHECK MODEL TEST RESULTS WITH VP?
PREDICTpNS.
CARRY OUT STRUCTURAL
DESIGN AND CALCULATIONS.
DEVELOPE WAVE-ADDED RESISTANCE MODULE COMPARE VP? PERFORMANCE OF 5 POSSIBLE KEEL/RUDDER CONCEPTS. CARRY OUT 30 FLOW CALCS FOR VARIOUS BULB SHAPES. STUDY OF FRAMING OPTIONS.
Fig. 1. Work chart used for the design of "Challenge
Australia". The work elements on the left are
ex-perimental in nature, while those on the right are
numerical/theoretical in nature. The work elements in the middle constitute the basic design path.
13
41111---DEVELOPE IACC RULE PROGRAM FOR CALCULATION OF ALLOWABLE L, DSPL AND SAIL AREA COMBINATIONS. OPTIMIZATION STUDY FOR L, DSPL AND SAIL AREA, USING REFINED VP? PROGRAM AND AMERICA'S CUP RACE SIMULATION PROGRAM. METEOROLOGY STUDY FOR SAN DIEGO TO SET DESIGN WIND SPEED. DEVELOPE PROGRAM FOR CALCULATION OF VP? INPUTS FOR SYSTEMATIC SERIES OF DESIGNS.
3. THE NEW AMERICA'S CUP DESIGN AND CONSTRUCTION RULE.
The International America's Cup Class (IACC) Rule, according to
which these yachts have to be designed. and built, came into being
in 1988-1989, after Dennis Conner and the catamaran "Stars and
Stripes" defeated the 40 metre K-Class yacht "New Zealand" in an
unfair match. All agreed that this should never happen again and it was decided to change the so-called "Deed of Gift" governing the America's Cup event, to no longer allow a challenger to
challenge for the Cup under the original terms thereof with
respect to the type of yacht, but only in a special, America's Cup Class yacht. An account of this, for future America's Cup events very important development, has been given by Kirkman (see
Reference 1), Chance (Reference 2) and Pedrick (Reference 3),
amongst others.
In a number of meetings in 1988 and 1989, between would-be
challengers and leading yacht designers, the overall parameters of the new class were defined to be as follows:
An overall length of about 23 metres;
The length-displacement ratio to be in-between present-day IOR maxi yachts and so-called ultra-light displacement
yachts (ULDB's), which was calculated to correspond to a
mass between 16,000 and 25,000 kg, depending on measured length and sail area;
To disallow hiking wings and unusual canoe body shapes which, in practice, was interpreted to mean that the canoe body is not to have any hollows except in way of appendages; A maximum beam of 5.5 metres;
A maximum draught to the bottom of the keel of 4.0 metres in the measurement condition;
A maximum of 2 rotatable appendages, the axis of rotation to be at a maximum of 45 degrees to the vertical;
A maximum height of the sloop rig of 32 metres, with the hoist of the genoa being 80% of that of the mainsail;
A maximum spinnaker area equal to 1.5 times the measured up-wind sail area;
Construction of hull and deck to be based on commercial grade carbon epoxy composites over a Nomex honeycomb core, cured at up to 95 degrees Celsius at 1 atmosphere of
pressure;
Construction of appendages and spars to be of higher grade carbon fibre, to be cured at up to 135 ,.,,grees Celsius in an autoclave at up to 5 atmospheres of pressure. Minimum skin thickness and weight of skins and core to be specified in 3 areas of the hull and deck.
The basic equation governing the value of measured length, measured sail area and displacement is based on the results of VPP predictions for about 500 designs. The formula derived by regression analysis, by the University of Southampton (see
Reference 4) is:
where L = Rated length in metres;
S = Rated sail area in square metres;
DSP = Rated displacement, equal to the mass in kg divided by 1025.
The rated length is given by the formula:
L = LM*(1+0.01*(LM-21.2)-8)+FP+DP+WP+BP
where LM = LBG+G
Here, LBG = Length of the hull between girth stations in
metres. The girth stations are the vertical,
transverse planes situated at the intersection of the waterplane 200 mm above the measurement
flotation waterplane with the stem and stern of the
hull, on the centre line;
G = Function of the girth and slope of the topsides at
the forward and aft girth stations, with a minimum value equal to 1.9 metres;
FP = Freeboard penalty, defined as 4 times the sum of
any freeboard deficiency at the forward, mid-length (i.e. 50% LBG), and aft girth stations, at which locations the minimum freeboards are 1.5, 1.25 and 1.2 metres in the measurement condition,
respec-tively;
DP = Draught penalty, defined as 4 times the excess
draught;
WP = Weight (i.e. mass) penalty, defined by formulae
for the case the mass is less than 16,000 kg or more than 25,000 kg;
BP = Beam penalty, equal to 4 times the excess beam over the maximum value, in metres.
The rated sail area is given by:
S = SM*(1+0.001*(SM-0.5-16.9)-8)
where SM = MSA+I*J/2
in which SM = Measured sail area in square metres;
NSA = Area of mainsail for which a formula is given based on the mainsail girths at 5 locations; I = Height of foretriangle in metres;
J = Base of foretriangle in metres.
The original Rule document counts some 45 pages, while the interpretations issued since May 1989, covering all aspects of design and construction, count that much again (see Reference 5).
4. FIRST SERIES OF MODEL TESTS: GENERATION OF DATA BASE FOR SIDE FORCE AND RESISTANCE.
The length-displacement ratio of the IACC yacht is significantly greater than that of the 12-Metre Class yachts they replace, IOR yachts, and other displacement-type yachts, with the exception of
ULDB's. Typical values are given in Table 1.
Table 1. Typical values of the length-displacement ratio of different types of yacht.
Since significant model testing has only been carried out for 12-Metre Class and IOR yachts, most designers of IACC yachts were suddenly confronted in 1989 with the problem of how to accurately determine the side force and resistance of these yachts without having to revert to the towing tank for each design studied, specifically with respect to the prediction of wave resistance and its dependence on canoe body characteristics. Although
existing VPP's will have provided a reasonable estimate for each design, an accurate determination of side force and resistance characteristics will generally not have been possible due to a lack of a suitable data base from which to derive formulae using regression analysis or appropriate relationships using some other method. This was also the case when the design work for "Chal-lenge Australia" was started.
To generate a data base on which to mathematically model the side force and resistance characteristics of IACC yachts accurately, a contract was entered into with the Delft University of Technology to expand their systematic series of yacht models, to cover the values of the length-displacement ratio, length-beam ratio, beam-draught ratio, prismatic coefficient, and longitudinal position of the centre of buoyancy, of interest. Half of the costs of these tests were paid for by ACAC, subject to the condition that the results of these tests remain confidential till after the 1992 America's Cup. The values of the design variables of the 12 models added to the systematic series are given in Table 2.
The standard keel and rudder of the Delft systematic series was used in these tests to render the results comparable to those of the earlier 28 models. Totally new polynomial expressions were derived for the residual resistance at discrete values of the Froude number, using all of the 39 models.
The results of the tests with these earlier models have been published by Gerritsma et al (see References 6 and 7). The
results of the added 12 models, called series III, will be
published by the Delft University of Technology as well (see Reference 8).
Type of yacht Value of LWL/VOLCB-(1/3)
12-Metre Class 4.6
IOR maxi yacht 6.0
IACC yacht 7.0
ULDB yacht 8.0
LWL = Length of waterline in racing condition
in metres;
Table 2. Values of design variables of the models added to the
Delft systematic series of models, on behalf of the
"Challenge Australia" project.
Model No. LWL/VOLCB-(1/3) LWL/BWL BWL/TC CP LCB
29 30 31 32 33 34 35 36 37 38 39 10.87 0.549 -4.4 7.07 0.549 -4.4 15.82 0.549 -4.4 10.86 0.551 -2.1 10.87 0.545 -6.6 10.37 0.520 -4.4 11.47 0.579 -4.4 10.16 0.550 -4.3 9.45 0.551 -4.5 19.32 0.549 -4.4 6.96 0.549 -4.4
BWL = Maximum beam on the waterline; TC = Maximum draught of the canoe body; CF = Prismatic coefficient;
LCB = Longitudinal position of the centre of buoyancy in % of LWL, relative to the mid-length (50% LWL) location. A minus sign indicates that LCB is situated behind the
mid-length location.
5. REFINEMENT OF VELOCITY PREDICTION PROGRAM.
One of two basically different methods can be adopted in a
mathematical model for the calculation of the speed of a sailing craft. The first and most frequently adopted method is to
determine the thrust and side force developed by the wind on the sails and then to calculate at what boat speed, heel and leeway angles these aerodynamic forces make equilibrium with the
hydrodynamic forces. Here, 3 equations are available to find these 3 unknowns, viz: force equilibrium in the horizontal plane, in the direction of the boat speed vector and in the direction at 90 degrees thereto (the hydrodynamic lift direction), while
the third equation involves the heel angle and moment equilibrium of all forces in the vertical, transverse plane.
The second method, less often adhered to, is to assume a specific value for the boat speed and the leeway angle, and to calculate the hydrodynamic side force and resistance in a straight-forward manner. Knowing the boat speed and the resultant hydrodynamic force, the calculation of the resultant aerodynamic force on the sails, for equilibrium, is a relatively simple matter. This, in turn, permits the calculation of the apparent and true wind
speeds and angles. On performing these calculations for a series of leeway angles from 0 degrees to, say, 12 degrees, all the points-of-sail ranging from running square before the wind, to sailing high-on-the-wind, are covered. An interpolation method
then yields boat speed, heel angle, leeway angle, etc, at any
required value of the true or apparent wind speed and angle.
19 7.5 4.0 6.5 4.0 8.5 4.0 7.5 4.0 7.5 4.0 7.5 4.0 7.5 4.0 7.5 4.0 7.5 4.0 7.5 3.0 7.5 5.0
This second method has been adopted by the author in his VPP. Apart from the mathematically more simple process for finding the
conditions for aerodynamic-hydrodynamic equilibrium, it places
the hydrodynamic properties of the canoe body and its appendages
in a more central role then does the first method. This results
in a favourable situation with respect to mathematical modelling
since much more is known about hull hydrodynamics then is known
about rig and sail aerodynamics, by virtue of the fact that
towing tank testing has yielded a wealth of information while, by
comparison, little is known about the forces on the individual
sails. The various modules in the program are as follows:
Input Part
Input of values for all geometric "design" parameters;
Input of values for density of air and water, viscosity of
air and water;
Input of matrix of boat speed and leeway angles for which calculations have to be done;
Input values for the rotation angle of rotatable appendages
(such as trim tab and rudder) if these are not to be
calculated by the program, which has the option of utilizing the condition that the momemts of all forces in the horizon-tal plane must also be zero;
Hydrodynamic Part
Provide first estimate of heel angle;
Calculate side force of canoe body and all appendages; Calculate induced drag of canoe body and all appendages; Calculate skin friction and pressure drag of canoe body and all appendages;
Calculate wave resistance of canoe body and of any appendage piercing the water surface;
calculate resistance due to heel; Calculate stability (GZ curve);
Calculate centres of all hydrodynamic forces on canoe body and appendages and calculate resultant force and its point of application (in x, y and z coordinates);
Calculate or input the height of the vertical centre of the resultant force on the sails;
Calculate the heeling moment;
Calculate the effect of crew weight and position on GZ curve; Calculate new heel angle by equating heeling moment to
righting moment;
If heel angle is greater then the value at which sails are "reefed" then limit heel angle to this value and adjust the height of the vertical centre of the sail force or shorten
foot length of sails;
Repeat steps 6 through 17 until heel angle value has converged;
Aerodynamic Part
Estimate a first value for the apparent wind angle and make a first estimate of what sails are being used;
Calculate the lift and drag forces on each sail (CLs and CDs), based on the assumed apparent wind angle;
Calculate the resultant aerodynamic force and associated drag angle equal to ARCTAN(CDs/CLs);
21
the resultant aerodynamic force direction opposite to the
resultant hydrodynamic force direction (the sum of the
hydrodynamic and aerodynamic drag angles is equal to the
apparent wind angle plus the leeway angle);
Determine which sails are being used, based on the calculated
apparent wind angle value (spinnaker versus jib or genoa, for example);
Repeat steps 20 through 23 until the value of the apparent wind angle has converged;
Calculate the apparent wind speed by equating the resultant aerodynamic force equal to the resultant hydrodynamic force; Calculate true wind speed and angle from the apparent wind speed and angle and the boat speed;
Correct the calculated true wind speed and apparent wind speed and angle for a standard wind gradient, to derive the values valid for the top of the mast where the wind sensing gear is fitted;
Added Resistance due to Waves Part
If true wind is forward of abeam calculate added resistance due to waves on providing either inputs for wave height and
period, or fetch and duration of wind over fetch.
Repeat steps 12 through 27 until wind direction and speed no
longer change, that is until the calculated added resistance
due to waves has converged. Interpolation Part
Repeat steps 5 through 27 for all boat speeds and leeway angles prescibed (usually about 30 boat speeds and about 50
leeway angles per boat speed), filling a set of
two-dimensional arrays for such important quantities as boat
speed, leeway angle, heel angle, total side force, total
resistance, apparent wind speed and angle, true wind speed
and angle, etc;
Interpolate to find the required performance for
previously-input values of true wind speed and angle; Output Part
Output of performance in tables and graphs.
The author's VPP, has 3 modes of operation. The first of these
is a detailed output mode, which gives the results of all of the
hydrodynamic and aerodynamic calculations for each speed and
leeway angle combination considered. This mode is utilized when
detailed design problems need to be addressed and not the
overall performance of the vessel. Because of the vast amount of
output generated in this mode, only a few boat speed-leeway angle
combinations can be addressed in one run of the program. The
second mode of operation gives one line of output for each boat
speed-leeway angle combination, printing the calculated values of
the main parameters, such as heel angle, sail thrust and sail
heel forces (equal to the total resistance and hydrodynamic side
force respectively), apparent and true wind speeds and angles,
and the speed-made-good to windward. The third mode of operation only outputs the results of the interpolated values for boat
speed, speed-made-good to windward, and the true and apparent
wind angles relative to the track of the yacht through the water,
the polar performance diagram is based. This program mode writes the polar performance data to an output file, which is read by the program that draws the polar diagram. To facilitate the editing of input files (over 100 variables have to be input for any configuration considered), special input and input file editing programs have been prepared.
6. OPTIMIZATION STUDY FOR LENGTH, DISPLACEMENT AND SAIL AREA. 6.1. Outline of Optimization Study.
With the refined VPP, a systematic study was carried out to find that combination of length, displacement and sail area, allowed by the America's Cup rating formula, yielding the fastest time around the America's Cup race course. For this purpose a computer program was written, called "ACRULE", for the calculation of all of the variables stipulated by the IACC Rule once the length between girths (LBG) and the displacement mass (W) are input. These variables were treated as the main, independent variables. With the Rule parameters calculated in this way for each design
(described more fully in Paragraph 6.2.1), a weight and stability calculation was then carried out and the remaining VPP inputs determined, using another computer program, called "ACDESIGN", specially written for this task as well (see Paragraph 6.2.2). For each design, the speeds calculated by the VPP on each of the
legs of the America's Cup course were tabled in LOTUS 123 for different true wind speeds and a set of races were simulated in
the computer in which the wind speed remained constant or was allowed to fluctuate according to specific, prepared scenarios
(see Paragraph 6.3). Investigation of the results clearly
revealed the sensitivity of length, displacement and sail area on the race results and it was relatively simple to determine the optimum values of these 3 main design variables, once the
applicable wind speed scenario had been determined.
6.2. Generation of Systematic Series of Designs. 6.2.1. IACC Rule Requirements for Each Design.
The ACRULE program, specially prepared for this systematic study,
requires inputs for the girths forward and aft, topside angles
forward and aft, freeboards, and the maximum beam and maximum
draught, in addition to LBG and W. The program then calculates
all of the dependent design values, including the sail area of
the mainsail, genoa and spinnaker, assuming maximum hoist values
and using previously-determined data on the required type of
mainsail roach, in the form of (non-dimensional) mainsail girths,
and the ratio of genoa area to total sail area. In the program, a
maximum and minimum value of the upwind sail area is adhered to,
to obtain practical designs. If the input values for LBG and W
give rise to too large or small a sail area, the program allows modification of one or both of these inputs to obtain a design with a previously determined maximum or minimum sail area.
LBG in metres
Fig. 2. The design space allowed by the IACC Rule is considerably larger then the area in which little or no penalties are
incurred. ter 'are SM dl ter re th0 1% t salt area)
205
I 23The initial matrix of designs and the associated selection of
LBG, W and S, was based on wanting to cover the entire design space, not only the design space without penalties. The entire design space is shown in Figure 2, from which follows that heavy
(length) penalties, leading to substantial reductions in sail
area, arise when LBG is less than about 18.3 m and greater than
about 20.3 m. Likewise, substantial reductions in sail area
arise when W is less than 16,000 kg or greater than 25,000 kg. In
this figure the adopted maximum and minimum rated sail area (S)
is that value for which the corresponding measured sail area (SM)
is no more than about 1% smaller than S. It was thought that
larger or smaller values of S would result in too much of a sail area penalty. It should be remarked, however, that during the America's Cup races it became apparent that in the light
conditions prevailing off San Diego, a large sail area was
extremely important and this, above all else, was being adhered
to by most of the designers, often leading to a trade-off with
length and displacement in such a way that the difference
Table 3. Values of the length between girth stations (LBG) and
displacement mass (W) used in the initial part of the
optimization study.
(prior to Round Robin 2), for "Challenge Australia", after her
rated length was reduced by 0.73 m, was S=349.7 m-2 and SM=339.6 m-2, i.e. a difference of about 2.9%.
The initial matrix of LBG and W values, for which the optimiza-tion study was carried out, is given in Table 3. For each length up to 5 displacement and (corresponding) sail area values were
chosen, thereby covering a large part of the permissable W and S
values for that length. A small value of the displacement W corresponds with a small value of S. while for a large mass a large value of S is possible. On increasing W to over 25,000 kg, however, the value of S decreases again because of the heavy W penalty that then comes into play. For the wind conditions prevailing off San Diego it was not considered worthwhile to
investigate the performance of designs in this latter regime in
detail.
Further combinations were studied in the region of best perfor-mance, after the associated LBG and W values thereof had been ascertained.
The program prints a complete formula calculation, with penalties and sail measurements, in the format adhered to in the IACC Rule documentation. An example of part of the output of this program
is given in Table 4.
6.2.2. VPP Inputs for Each Design.
For each design, the required inputs for the VPP calculations was generated in a systematic way using a program that was specially written for this task, called "ACDESIGN". This program uses the
output values of ACRULE, in addition to pre-determined values
for the ratio of various design variables, the mass of the
various components of the structure and equipment on board, the vertical centre of gravity thereof, etc.
The ACDESIGN program starts with the calculation of the design
variables for the "racing" condition of the yacht as opposed to
the variables valid for the "measurement" condition, calculated
by the ACRULE program. Inputs have to be provided concerning the Length between girths
(LBG) in metres Mass of displacement (W) in kg 18.2 15748, 16000, 19802, 21171 and 26320 18.6 15741, 16000, 18564, 21126 and 26337 19.0 15888, 16000, 19028, 22055 and 25998 19.2 19.4 19.8 20.2 20.6 15964, 16000, 16214, 17019, 18049, 22986, 19255, 19627, 20538, 21524 23993 22544 and 23040 and 24057 and and 25000 and 25000 25824 25652 25307
mass and centre of mass of the crew, additional gear and sails, etc, that are on board during racing, and not included in the displacement W. Inputs then have to be provided for the beam on the racing-trim waterline, the prismatic coefficient, waterplane area coefficient, maximum transverse area coefficient, and the volume, in % of the total displacement volume, of the combined appendages. With these inputs, together with the values for LBG
25
Table 4. Example of part of output of program for the calculation of IACC Rule parameters, for which values of LBG, W, freeboards, girths, topside angles, maximum draught, etc, have to be input.
Input value of overall length in metres; 24.123
Input value of length of aft overhang in metres; 2.033
Calculated length of forward overhang in metres; 1.890
Input value of forward topsides angle in degrees; 20.000
Input value of aft topsides angle in degrees; 45.000
Input value of forward girth (FG) in metres; 2.750
Input value of aft girth (AG) in metres; 3.933
Calculated value of forward beam correction (FBC); -0.116
Calculated value of aft beam correction (ABC); 0.000
Calculated value of forward girth correction (FGC); 0.300
Calculated value of aft girth correction (AGC); 1.600
Calculated value of girth component of L in in (G); 1.900
Input value of freeboard at forward length mark in m; 1.500
Input value of freeboard at mid-LBG location in m; 1.250
Input value of freeboard at aft length mark in m; 1.200
Calculated value of freeboard penalty (FP); 0.000
Input value of draught to bottom of keel in m (D); 4.000
Calculated value of draught penalty (DP); 0.000
Input value of maximum beam in metres (B); 5.000
Calculated value of beam penalty (BP); 0.000
- Input value of length between girth stations (LBG); 20.200
Input value of mass of displacement in kg (W); 25000.000
Calculated value of mass penalty (WP); 0.000
Calculated value of rated displacement in m-3 (DSP); 24.390
Calculated value of measured length in metres (LM); 22.100
Calculated value of rated length in metres (L); 22.195
Calculated value of rated sail area in m-2 (S); 324.615
- Calculated value of measured sail area in m-2 (SM); 323.885
Assumed height of mast datum band above sheer; 0.500
Assumed height of boom above datum (BAD); 1.500
Assumed height of upper mast band above datum; 32.000
- Input value of clew offset of mainsail in metres (CO); 0.000
Input value of ratio of genoa area to total sail area; 0.306
- Calculated mainsail girth along boom in metres (E5); 10.253
Calculated value of E4 mainsail girth (E4/E5=0.931); 9.544
Calculated value of E3 mainsail girth (E3/E5=0.813); 8.332
Calculated value of E2 mainsail girth (E2/E5=0.585); 6.001
- Calculated value of
El
mainsail girth(E1/E5=0.082);
0.837
-
Calculated value of measured mainsail area in m-2; 224.835Calculated value of measured foretriangle area in m-2; 99.050
Assumed value of hoist of genoa in metres (I); 25.600
- Calculated value of foretriangle foot in metres (J); 7.738
Calculated spinnaker pole length in metres; 10.447
and W, the calculation of the total volume of displacement and canoe body volume of displacement is carried out. The overall length and beam, waterline length, length of forward and aft overhang, etc, is then determined using a set of constants for
the ratios LOA/LBG, LWL/LBG, BMAX/BWL (with BMAX limited to a
maximum of 5.5 metres), etc. These were determined from a
"parent" design on which all other designs considered in this optimization study were based. The value of the canoe body draught and the maximum draught in the racing condition follow directly from the inputs supplied after some mathematical
manipulation.
The calculation of the mass and vertical centre of mass is carried out on the basis of formulae for each component of the structure, equipment and crew on board. These are based on a detailed study of the carbon structure of hull and deck, spars and rigging, sails, winches, etc. The amount of lead ballast is determined by simply subtracting the mass determined in this way from the total mass of displacement. A calculation is then
carried out to determine the amount of ballast carried in the bulb at the bottom of the fin keel. The previously-supplied input specifying the total volume of the appendages is used for
this purpose. If all of the ballast cannot be housed in the bulb, the vertical position of the centre of mass of the lead that has to be fitted elsewhere has to be input. The keel fin and rudder geometry are kept the same. Only the bulb size is varied in accordance with wanting to house most of the ballast therein. An example of the output of ACDESIGN is given in Table 5.
6.3. Some Typical Results.
The results of the VPP calculations revealed that length and sail area are extremely important and that an increase in displace-ment, to gain more length and sail area, is worthwhile. In fact, it became clear that the maximum displacement without penalty needs to be adopted in order to maximize performance in every wind speed over 8 knots. This is revealed in Figure 3 where the time around the America's Cup race course in decimal hours is set
out against the length between girth stations in metres for
different true wind speeds at the mast-head. The curves shown, for different wind speeds, seem to be rather insensitive with LBG but in actual fact the opposite is true. The differences between
the various designs add up to several minutes over the course, except for 8 knots of true wind speed, for which the overall performance is almost independent of LBG for the 4 designs here considered. The results of Figure 3 are for a maximum value of
the displacement W, without penalty, and a constant waterline beam of 4 metres.
Many more results of this type were studied. All of these pointed
to the same conclusion, viz; that in more than 8 knots of true
wind the optimum performance is obtained for the maximum LBG length (around 20.2 metres) and the maximum displacement of 25,000 kg, without penalties. At 8 knots of true wind speed or
less, however, optimum performance favours a smaller length and
Table 5. Example of part of output of the program used to generate a systematic series of designs for the (VPP)
study of the optimum length, displacement and sail
area.
- Code number of design 20225
LBG length in metres; 20.200
- Mass W in measurement condition in kg; 25000.000
- Length of waterline in racing trim flotation in m; 19.131
Overall length in metres; 24.123
- Maximum beam of waterline in racing trim flotation; 4.000
Overall, maximum beam in metres; 5.000
- Total volume of displacement in m-3; 26.349
- Volume of displacement of canoe body in m-3; 23.910
Maximum draught to bottom of bulb in racing trim; 4.038
Maximum draught of canoe body in racing trim in m; 0.907
- Average chord length of keel in m; 1.500
Height of keel fin in metres; 2.431
- Average chord length of rudder in metres; 0.700
Height of rudder in metres; 2.500
Maximum height of bulb in metres; 0.700
- Maximum width of bulb in metres; 1.000
Maximum length of bulb in metres; 5.946
Volume of bulb in m-3; 1.847
Mass of total yacht in racing trim in kg; 27056.207
Vertical centre of mass of yacht above racing
trim waterline in metres; -2.13
ITEM WEIGHT in kg VCG WEIGHT IN %
7. METEOROLOGY STUDY TO DETERMINE DESIGN WIND SPEED.
To ascertain the wind conditions on the proposed race site off
San Diego, one of Australia's leading marine meteorologists was
requested to carry out a wind study. His report (see Reference 9) revealed that 90 percent of the measured wind speed values, between 12.00 and 17.00 hours, at Lindbergh Airport in San Diego
and nearby Mission Bay, in March, April and May, were between 7
and 14 knots, with an average value between 10 and 11 knots. Wind speeds on the actual race course were predicted to be between 6
27
Hull and deck; 2074.09 0.45 7.67
Deck fittings; 928.04 1.40 3.43
Spars; 901.90 13.02 3.33
Keel fin; 342.13 -1.88 1.26
Rudder and steering; 59.25 -1.25 0.22
Sails; 303.11 8.34 1.12
Running rigging; 113.10 1.28 0.42
Batteries, elect.; 150.00 1.28 0.55
Crew; 1440.00 1.78 5.32
Sails, etc, below; 500.00 0.00 1.85
Ballast; 20244.59 -3.76 74.83
other ballast; 0.00 0.00 0.00
and 13 knots. These values were at 10 metres above sea level and by allowing for a standard wind gradient with increasing height above sea level, these values were interpreted to be between 8
and 15 knots, with an average value in-between 11 and 12 knots,
at the mast-head. With this result it was concluded that the "small boat" option was not a viable one.
8. SECOND SERIES OF MODEL TESTS: DETERMINATION OF WATERLINE BEAM, LCB, AND TYPE OF FORWARD AND AFT OVERHANG.
At this stage linesplans were designed for the manufacture and
testing of 5 models at the Delft University of Technology. These tests were aimed at addressing 3 problem areas which could not be satisfactorily dealt with in the VPP study. These were: the
determination of the optimum waterline beam (BWL), the
determina-tion of the optimum posidetermina-tion of the longitudinal centre of
buoyancy (LCB), and the type of forward and aft overhang. In each case the VPP was considered either not accurate enough to
pin-3.85 3.65 3.45 3.25 3.05 2.85 2.65 19 19.2 19.4 19.6 19.8 20 20.2 20.4 LBG In metres
Fig. 3. Calculated time around the initially-proposed America's Cup race course (21.9 nautical miles) for designs with maximum displacement (W), without penalty. The initial VPP study revealed that in over 8 knots of true wind the optimum LBG was 20.2 metres and the optimum W was 25,000 kg.
Time around
Amert's Cup
Race ourse In hou s
29
point the desired value (in the case of BWL and LCB), or the VPP
could not address the problem adequately al all (type of
overhang).
The design of these 5 models was kept as similar as possible in order to only vary the parameters that had to be studied. The first 3 models, models A, B and C, had varying waterline beam
(3.5, 4.0 and 4.5 metres) and an LCB position corresponding to a
distance of 3.5% of LWL behind the mid-LWL location. A modest
12-Metre type of bow overhang was adopted, similar to that of the
mini-America's Cup day-sailers designed by the author's team
early in the project. A drawing of this mini AC design, two of which were built for the purpose of evaluating mainsail and spinnaker shapes, the best ratio of genoa area to total sail area, etc, is given in Figure 4. These yachts are 40% of a full
size, "mid-Rule", America's Cup yacht. They were later used by
the crew to gain more match-racing experience.
The fourth model, model D, was similar to model B except for a
more aft position of the longitudinal centre of buoyancy (5.7% of
LWL behind the mid-LWL location). Model E was similar to model B, except for the forward and aft overhang. A raked IOR type of stem was adopted while the overhang aft was eliminated and a
skiff-type stern was adopted with a narrow waterline, somewhat similar
to the design of the final New Zealand IACC yacht. A stern of
this type is allowed by the IACC Rule and gives rise to an
important increase in waterline length. The main particulars of these 5 models are given in Table 6.
Table 6. Main particulars of models A through E, used for
determining the optimum waterline beam (models A, B, and C), to check the viability of positioning LCB further aft (model D), and to determine the effect of maximizing
the length of the design waterline by adopting an
IOR-type stem and a skiff-type stern.
Entity Model A Model B Model C Model D Model E
LOA (m) 24.619 24.398 24.430 23.996 22.246 BOA (m) 4.501 4.977 5.500 4.830 4.746 LWL (m) 18.974 18.798 18.660 18.860 19.892 BWL (m) 3.500 4.000 4.500 4.000 4.000 TC (m) 1.013 0.905 0.800 0.863 0.820 VOLCB (m-3) 23.867 23.867 23.850 23.870 23.871 LCD (%LWL) -3.526 -3.681 -3.838 -5.669 -3.690 - CP 0.536 0.537 0.539 0.530 0.535 CB 0.355 0.351 0.355 0.367 0.366 CWP 0.672 0.670 0.674 0.676 0.661
where LOA = Length over all; BOA = Beam over all;
LWL = Length on the design waterline;
BWL = Maximum beam on the design waterline; TC = Maximum draught of the canoe body; VOLCB = Displacement volume of the canoe body;
LCB = Position of the longitudinal centre of buoyancy; CP = Prismatic coefficient of canoe body;
CB = Block coefficient of canoe body;
From the values presented in Table 6, it follows that these models did not constitute a true systematic series since
variations occur in more than one variable at a time. Although
the models A, B, C and D could have been derived from one parent model by transformation techniques, this was not done because Of having to fulfil various IACC Rule requirements with respect to girths, slope of topsides, slope of the buttocks aft of the aft girth station, etc. This lead to the decision to design each
model separately (for which the hull design system "MACSURF" was
used), leading to the variations identified.
Fig. 4. Drawing of the mini-America's Cup day-sailers, designed
by the author's team, two of which were built for
31
One and the same keel and rudder configuration was used on each of the 5 models. A basic design was made for this purpose
consisting
of a fixed fin with an average chord length of 1.61 inand a height of 2.62 m. A simple axi-symmetric bulb with a length
of 3.41 m and a maximum diameter of 0.7 in was fitted to the
bottom of this fin. This
bulb,
filled with lead, was sufficientto obtain the stability of the full-size yacht in the test tank.
The results of the upright resistance tests for the models A, B
and C, as a
function
of Froude number, are given in Figure 5.Noticeable differences in the resistance values occur in the important speed range between the Froude number values of 0.35 and 0.45, and at relatively very high speeds, in excess of the Froude number value of 0.525. In the lower of these two speed
ranges, model A is not much better than model B, but both are
better than model C. In the higher speed range the differences are greater, favouring the more slender canoe body.
30
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Fig. 5. Comparison of the upright resistance of models A, B and
C, each with a different waterline beam. The values are
For the important Froude number value of 0.35, being
approximate-ly the maximum speed attainable on-the-wind, the resistance
associated with heel and side force of Models A, B and C is shown
in Figures 6a, 6b and 6c, for 10, 20 and 30 degrees of heel,
respectively. This resistance increase is given as a function of
the side force-squared. The fact that the lines shown are more or
less straight, indicates that the resistance increase with
increasing side force is mainly composed of induced resistance.
0.6
.5
0.4 0.3 0.2 0.1 Resist e due heel side In kN so le of eel - 10 degree Model C 100 Model 13 150 4 2 1 0 A Resistance ue to heel Eind s for In kN 200 400 2 1.5 1 0.5 0 200 250 0 100 200 300 400 500 600 700Mg of h
I e ees Mod C Angl Resistance heel and el In kN IsC2 600 800 1000 120* 1400 B of heel - ko degrees Side force-squared In KN-2Figs. 6a, 6b and 6c.
Resistance due to heel and side force, for heel angles of 10, 20
and 30 degrees, for a Froude number = 0.35, for models A, B and
C. All values are for the full-scale.
Although the keel and rudder of these 3 models are the same, the canoe body influence on the development of side force is
different, primarily due to the differences in the BWL/TC ratio.
The keel of model C has the smallest effective aspect ratio and
hence the greatest induced resistance at equal side force values. This can be seen from the relatively steep slope of the resis-tance versus side force-squared line for model C in Figure 6c, valid for 30 degrees of heel. (At other heel angles this is less noticeable.). Model C has the largest BWL/TC ratio.
From analyses of this type for a range of speeds, and from
additional VPP considerations involving the differences in (form) stability between the 3 designs, the conclusion was drawn that
the LWL/BWL ratio should not be smaller than 4.5, from which the
required BWL value was determined.
The upright resistance of model D, compared to model B, is shown
in Figure 7. Although the scale of this figure is insufficient to
properly determine the exact differences, it can be seen that in the Froude number range around 0.35, model D is marginally
better. At higher speeds, however, this is not the case. The more slender forebody of model D, associated with the shift of LCD
further aft, does not lead to sufficient reduction in wave
resistance to offset the increase in viscous resistance caused by
the fuller aft-body. Accordingly, a moderate LCD position was
adhered to.
The results of the tests with the model possessing the long
waterline (model E), realized by adopting an IOR-type stem and a
skiff-type transom, was disappointing. The upright resistance was
greater than for the otherwise equivalent model B over the entire speed range. The analyses thereof revealed that the lack of
overhang aft, and the associated lack of buoyancy due to the
adopted narrow waterline aft, lead to appreciable trim down by
the stern and greater wave resistance because of this, even
though the waterline length was greater. The increased wetted area of this model was the cause of a greater resistance at low
speeds.
Although the performance of model E could have been improved by 33
The resistance at zero side force is theoretically that due to
heel alone. From these figures it follows that model A is better than model B, while model B is better than model C. The
resis-tance due to heel for the 3 models, as can be deduced from these
figures, is given in Table 7.
Table 7. Approximate full-scale values for the resistance due to
heel alone for models A, B and C in kN, as determined
from model tests.
A 0.01 0.22 0.40
B 0.04 0.45 0.53
C 0.08 0.50 0.46
Model identification 10 degrees 20 degrees 30 degrees
adopting a greater waterplane area aft to reduce the encountered trim problem, this would have increased the wetted area and, thus, the resistance in the low-speed region. As demonstrated by the final New Zealand yacht, such a design is probably only
feasible if moderate Rule dimensions are adhered to and not the
maximum dimensions.
Further VPP analyses revealed that for the design wind speed of between 11 and 12 knots (see Paragraph 7), the average boat
speed over the America's Cup course (as proposed in 1989) is
ap-proximately 10.2 knots. That is equivalent to a Froude number value of about 0.38. At this speed more than half of the
resistance is associated with wave-making and the decision was
therefore taken to maximize the design waterline length by adopting a raked IOR stem rather than a 12-Metre type overhang. The skiff-type stern was not adopted. As it turned out (see
Paragraph 12), the wind conditions off San Diego were found to be significantly lighter then predicted, leading to importantly
lower average boat speeds. If this had been realized during the design phase, a 12-Metre type of overhang would have been adopted leading to a lower wetted area and better performance in the
lighter conditions. 30 25 20 15 10 5 Reels In kN nee Froude number Model 0 0.1 0.2 0.3
04
05
06
0.7Fig. 7 Upright resistance of model D in comparison to that of
model B. Movement aft of LCB was found to only yield a (small) advantage near a Froude number of 0.35. At
higher speeds the increase in viscous resistance is
