A Fundamental Theory of Sailing and its application to the design of

a Hydrofoil Sail Craft

Stephen Bourn
October 2001

Abstract
This paper presents a general comprehensive but succinct theoretical
framework for analysing the forces acting on a sail craft and the resultant
sail craft performance. An innovative type of hydrofoil sail craft, whose
design was guided by the theory, is described. This type of craft should
have superior performance to all existing types of high performance sail
craft on all courses in most conditions. The theoretical analysis shows
that the system of forces acting on any sail craft at equilibrium can be
reduced to an equivalent system of three forces acting in a vertical plane.
The resultant forces represent the net aero, hydro and gravitational
forces. The geometrical relationships between these forces and the
air/water/craft velocity triangle in the horizontal plane leads to a
fundamental equation governing the limits of sail craft performance.
Consideration is given to the implications of the theory regarding the
necessary attributes of high performance sail craft in general. The
particular type of hydrofoil sail craft described in the paper would be
almost fully airborne when in use. A single inclined aerofoil and a single
submerged inclinable hydrofoil would generate the main aerodynamic and
hydro forces that would support and propel the craft.

List of symbols
AR aspect ratio, i.e. ratio of span squared to area
atm. unit of standard atmospheric pressure
cd profile or section drag coefficient
C coefficient
CG centre of gravity
CP centre of pressure
D drag, i.e. component of force parallel to flow
F force
Fn Froude number Fn = V S / g LWL
g acceleration due to gravity, g≈9.8 m/s2
kgf kilograms of force, 1 kgf≈9.8 newtons
L lift, i.e. component of force perpendicular to flow
LWL waterline length
P vertex of the velocity triangle
S area of foil
V velocity, speed
Vmg velocity made good, i.e. component of sail craft speed parallel to the
true wind
W displacement/weight, g ∆

1
 α (alpha) angle of attack or incidence
α0 zero lift angle
β (beta) apparent wind angle, i.e. angle between VS and VA or course with
respect to apparent wind
γ (gamma) course angle with respect to true wind
δ (delta) sail trim angle
∆ (Delta) displacement mass
ε (epsilon) drag angle, i.e. arctan(D/L)
θ (theta) elevation of CPA with respect to CPH
λ (lambda) 1. leeway angle
2. wavelength
ρ (rho) fluid density, ρA≈1.2 kg/m3, ρH≈1000 kg/m3
φ (phi) elevation, i.e. angle above the horizontal
ω (omega) relative position of CG

Indices
A air, aerodynamic, apparent wind
D drag, i.e. component of force parallel to flow
H water, hydrodynamic, hydrostatic
I horizontal (mnemonic - air water interface) component, projection
L lift, i.e. component of force perpendicular to flow
max maximum
min minimum
P pressure
S sail craft
T true wind
V vertical component

Mathematical signs and abbreviations

≈ approximately equal to
<< much less than
| such that, i.e. introduces condition

1 Introduction
This paper analyses the problem of the fundamental limitations on ultimate sail craft
performance. The general attributes required for high performance sail craft become
apparent from the analysis. Guided by the analysis a novel hydrofoil sail craft has
been designed, and is described.
A few well-understood basic principles are applied to the problem. The primary
principles come from simple Euclidean geometry and statics, that branch of
engineering mechanics that deals with rigid bodies at equilibrium under a system of
forces. Results from aerofoil theory, which is equally applicable to foils operating
submerged in water, are also used. Basic results from the hydrodynamic and
hydrostatic theory applicable to the motion of a hull on the water surface are also

2
needed. The principal initially used from aerofoil theory is the simple fact that lift, a
force component perpendicular to the direction of flow, can be generated.
An innovative approach is taken to the analysis of the complete system of forces
acting on a sail craft. The system of forces acting on a sail craft at equilibrium is
reduced to an equivalent system of three resultant forces with concurrent lines of
action lying in a vertical plane. The resultant forces represent the net aero, hydro and
gravitational forces. The geometrical relationships between these forces and the
air/water/craft velocity triangle in the horizontal plane leads to a fundamental
equation governing the limits of sail craft performance. The equation relates the
apparent wind angle to basic parameters associated with the net aero and hydro forces.
The approach taken in reducing the system of forces obviates the need to give
independent consideration to heeling, pitching and righting moments.
The theory is applicable to all sail craft. This includes displacement yachts,
dinghies, multihulls, sailboards and kite powered craft. The theory can be applied to
ice and land yachts, with appropriate substitutions for references to water. While the
theory may apply to all sail craft, the emphasis is on high performance, meaning high
speed relative to the true wind speed, and high absolute speed.
The hydrofoil sail craft described in the paper would be almost fully airborne
when in use. A single inclined aerofoil and a single submerged inclinable hydrofoil
would generate the main aerodynamic and hydro forces that would support and propel
the craft. The craft should have superior performance to all existing types of high
performance sail craft on all courses in most conditions.

Section outline
The paper is structured as follows. Section 2 introduces the velocity triangle
and the arc of constant apparent wind angle. The dependence of various performance
measures on apparent wind angle is presented. An expression is given for a velocity
ratio that changes and must be accommodated as course changes.
In Section 3 the horizontal components of the net aerodynamic and
hydrodynamic forces are examined and their geometric relationship to the velocity
triangle.
In the next section the full three-dimensional system of forces is considered.
These are reduced to three resultant forces, representing the net aero, hydro and
gravitational forces. The aero and hydro force to weight ratios are expressed as
functions of the force elevation angles. An equation relating the apparent wind angle
to the drag angles and the force elevation angles is derived. This is the fundamental
equation determining sail craft performance. The relatively low hull drag at low
Froude number is discussed. The basic equations for fluid dynamic lift and drag are
introduced and discussed in relation to the velocity ratio variation that accompanies
changes in course. Finally in this section the relevant literature is reviewed.
In Section 5 the general principles are considered regarding the relative
locations of the centre of gravity and the aerodynamic and hydro centres of pressure,
and the elevation angles of the resultant force lines of action. The attributes required
for high performance and for stable equilibrium are deduced. The design implications
of aspect ratio and induced drag are discussed. Finally the designs of a variety of
existing types of sail craft are reviewed.

3
 Finally in Section 6 the novel hydrofoil sail craft design is presented. Cavitation
is discussed. Alignment of the craft in response to the course, apparent wind and net
forces is shown. The resultant forces and lines of action in the vertical plane are
shown. The craft geometry is parameterised and an equation derived relating the
resultant force elevation angles. Representative values are assumed for the craft
geometry and drag angles, and a full analysis of relative craft performance is
undertaken. Finally specific performance measures are derived for a particular
example craft weight, aerofoil area and true wind.

2 The velocity triangle and the arc of constant apparent wind angle β
The velocity triangle showing the relationship between the true wind VT, the sail craft
speed VS and the apparent wind VA is depicted in Figure 1. The course γ with respect
to the true wind and the apparent wind angle β are also shown in the figure. The sail
craft speed made good Vmg is shown, and is defined to be positive in the upwind
direction and negative in the downwind direction.
The following property comes from Euclidean geometry.
The angle subtended at any point on a circular arc by its chord is
constant, and equal to the angle between the chord and the endpoint
tangents.
The property can be applied to the velocity triangle. For VT and β fixed, the locus of
the vertex P is a circular arc as shown in Figure 1. For the example portrayed in the
figure β <π/2, and this must be true for all craft sailing to windward or even on a beam
reach, and for all craft sailing faster than the wind, including craft sailing downwind
faster than the wind. In other cases it need not be true. The focus of this paper is on
high performance craft that can achieve β <<π/2 over a wide range of courses. Much
of the material presented is equally applicable for β≥π/2, and it is generally clear or
easily verifiable when this is the case.
The best speed made good upwind Vmg max is achieved on the course angle
γV that corresponds to P1 in Figure 1. The best speed made good downwind
mg max

Vmg min is achieved on the course corresponding to P5. Courses of practicable interest
range between these two extremes. The maximum craft speed VSmax occurs when the
vector VS passes through the centre of the circle and ends at P4. The remaining points
P2 and P3 correspond to V Amax and VA=VS respectively. Expressions for various
parameters are given for each of the five points in Table 1. Some parameters of
particular interest are plotted as functions of β in Figure 2. Small β is required for
high performance.

Velocity ratios on differing courses

As the course γ changes the relativity between VA and VS changes. At the
extremes

4
 DH
LIH

LIH
FIH
FIH
LIA
εIH
P1 FIA
P DA

DH β
γ DA LIA
Vmg VS εIA
εIH FIA
εIA
β
β P2
β

VA
O
VT β P3

P3
β
P4

P5

Figure 1 Velocities, forces and angles in the horizontal plane.

VA VS 1 + sin β
= =
VS V VA V 1 − sin β
mg max mg min

and the square of these ratios

VA2 VS 2 1 + sin β
= = . (1)
VS
2
VA
2
1 − sin β
Vmg Vmg
max min

5
 Course Boat speed Apparent wind Speed Boat speed
angle speed ratio made good
VS VA VA Vmg Remarks
γ VT VT VS VT

π β sin (π / 4 − β / 2 ) sin (π / 4 + β / 2 ) 1 + sin β 1 1 1

P1 + − Vmg max
4 2 sin β sin β 1 − sin β 2 sin β 2

π 1 1 1
V Amax
P2 0
2 tan β sin β cos β

π β 1 1 1 1 1
P3 + 1 − VS = V A
2 2 2 sin( β / 2) 2 sin( β / 2) 2

π 1 1
P4 +β cos β -1 VSmax
2 sin β tan β

3π β sin (π / 4 + β / 2 ) sin (π / 4 − β / 2 ) 1 − sin β 1 1 1

P5 + − + Vmgmin
4 2 sin β sin β 1 + sin β 2 sin β 2

Table 1 Relative speeds for selected course angles.

These values are both plotted in Figure 2. Small values are beneficial. The reason
will be explained in Section 4. Small values occur when β is small.

3 Horizontal components of the net aerodynamic and hydrodynamic forces

The net aerodynamic and hydrodynamic horizontal force components FIA and FIH are
represented in Figure 1. The indices A and H signify aero and hydro respectively,
while I signifies horizontal. A mnemonic for I is the interface between air and water.
The force components FIA and FIH can be further separated into the horizontal
components of lift LIA and LIH and total drag DA and DH. The angles between the
horizontal components of force and lift are denoted εIA and εIH.
For equilibrium FIA and FIH must share a common line of action and be equal in
magnitude but opposite in direction. The vector sums are represented pictorially in
the inset to Figure 1.
From the superposition of the force components on the velocity triangle as in
Figure 1 it can readily be seen that
β = ε IA + ε IH . (2)
Small εIA and εIH are required for high performance. Note that εIA and εIH should not
be referred to as the aerodynamic and hydrodynamic drag angles. In accordance with
the conventional nomenclature used in fluid dynamics these titles are reserved for
related angles to be introduced in the next section. Figure 3 represents a conventional

6
5
VA2 VA
4 2
VS max VT VS Vmg max
VS Vmg max
3

1
Vmg max VT
0

1
Vmg min VT
2 VA
3 VS VT V
mg max
VT Vmg max
4

6
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90

β (degrees)

Figure 2 Relative speeds and speed ratios as functions of β.

sail craft with a generally vertical sail and fin keel or centreboard. It shows the
leeway angle λ, which is identical to the fin or centreboard angle of attack or
incidence αH, the sail trim angle δ and the sail incidence angle αA. Notice that
β = α A +δ +αH .
It is important to distinguish αA from εIA and αH from εIH.
Figure 3 also introduces a vector representing the apparent water flow VH with
respect to the reference frame of the craft. The vector VH is the reverse of VS.

4 The net aero, hydro and gravitational forces

Analysis of the totality of forces acting on a sail craft is assisted by the following
results from the sub branch of engineering science known as statics. Statics, in
contrast to dynamics, deals with the mechanics of rigid bodies at equilibrium under a
general force system.
Any system of forces can be reduced to an equivalent single force
vector and a collinear couple vector.

7
 αH=λ

VH=−VS
δ

αA
β

VA

Figure 3 Leeway, sail trim and incidence angles for conventional sail craft.

Couple is synonymous with moment or torque. This force-couple system is called a

wrench. It can be thought of as a push and twist along a single axis. A reference
is [1].
If three forces, acting on a rigid body, produce equilibrium, their
directions must lie in one plane; and must all meet in one point, or be
parallel [2].
Equilibrium exists when the vector sum of the forces is zero, and there is no net
moment. More succinctly, the lines of action must be coplanar; and must be
concurrent or parallel.
All of the forces acting on a sail craft belong to three distinct classes: aero,
hydro and gravitational. The resultant of the gravitational forces is a single force, the
weight W, acting vertically downwards through the centre of gravity CG. The
resultant of the aerodynamic forces is a single force FA acting through the
aerodynamic centre of pressure CPA, and possibly a residual moment. For most craft
it is reasonable to assume that the residual moment, if it exists at all, is negligible.
The hydro forces include hydrodynamic forces acting on hulls and foils, and
hydrostatic buoyancy. A major component of drag may be wave-making resistance.
The resultant of the hydro forces is a single force FH acting through the hydro centre
of pressure CPH, and possibly a residual moment, which similarly will be assumed to
be negligible or non-existent.

8
 F

LV φ

φL FI

εI LI
D

Figure 4 Net force, force components, and projection onto the

horizontal plane.

For either fluid, air or water, the force F may be separated into the components
lift L and drag D, as shown in Figure 4, which are normal and parallel to the direction
of flow, respectively. The lift L can be further separated into vertical lift LV and
horizontal lift LI. The horizontal force component FI is also shown in the figure. The
angle ε between force F and lift L will be referred to as the aerodynamic or
hydrodynamic drag angle, as appropriate, notwithstanding that in the hydro case there
may be a significant hydrostatic contribution to the vertical lift LV. The angle εI
between the horizontal force and lift components FI and LI is the projection of the
drag angle ε onto the horizontal. The angle φ is the elevation of force F with respect
to the horizontal. In the figure φ is depicted as positive, but it could be zero or
negative. The angle φL is the elevation of lift L with respect to the horizontal. This
too may assume positive, zero or negative values. If a single foil generated the force
F, then the angle φL would be the roll angle of the foil’s lateral axis measured from the
vertical. The angular relationships may be expressed algebraically by
sin ε = sin ε I cos φ , and
sin φ = sin φ L cos ε .
Add the indices A or H to all of the above values to denote the specific aero or
hydro quantity. The angular relationships are
sin ε A = sin ε IA cosφ A , (3)
sin ε H = sin ε IH cos φ H , (4)
sin φ A = sin φ LA cos ε A , and (5)

9
 FIA

LVA FA

φA

FH
LVH

φH
FIH

Figure 5 Vector sum of net forces.

sin φ H = sin φ LH cos ε H . (6)

To picture the totality of forces acting on a sail craft visualise aero and hydro
versions of Figure 4 built up on the foundations in Figure 1, together with the weight
W acting vertically downwards through P. Note that
LVA + LVH = W .
The vector sum of the resultant forces in the vertical plane in which they act is
represented in Figure 5. The relationships can be expressed algebraically by
FA cos φ H
= , and (7)
W sin(φ A + φ H )
FH cos φ A
= . (8)
W sin(φ A + φ H )

The fundamental equation

Using (3) and (4) to substitute for εIA and εIH in equation (2) yields

sin ε A sin ε H
β = arcsin + arcsin , (9)
cos φ A cos φ H

which relates the apparent wind angle β to the drag angles εA and εH, and to the
elevation angles φA and φH. This is the fundamental equation governing sail craft
performance. High performance requires small β. From the equation it is clear that
this requires small drag angles and small elevation angles, but with rapidly

10
diminishing returns as φA and φH tend to zero. Small elevation angles result from
large aerodynamic and hydro force to weight ratios FA/W and FH/W. More succinctly,
the equation gives analytical rigour to the intuitive notion that high performance
requires a high force to weight ratio and low drag. The equation also provides the
major reason why sail craft apparent wind angles are so much greater than the glide
angles achieved by sailplanes, a question raised by Bethwaite [3].
The theory presented above applies to all sail craft. The theory is equally
applicable to land and ice yachts if references to hydro forces are replaced with the
forces generated by the wheels or runners respectively. For conventional craft the sail
is generally vertical, and so φA≈0, however φH is generally much greater than zero.
The Froude number is the non-dimensional value
VS
Fn =
g LWL
where g is the acceleration due to gravity and LWL is the waterline length of the craft.
The speed of deep-water waves of wavelength λ is
gλ
.
2π

When Fn = 1 / 2π the boat speed corresponds to the speed of waves with

wavelength λ= LWL. For a conventional displacement craft the drag increases
dramatically as Fn approaches 1 / 2π . At lower craft speed drag is very low and εH
is typically very small, whereas φH is very large, resulting in a moderate value for εIH.
Kite powered craft and sailboards generally have inclined aerofoils, which provide
some of the vertical lift, so φA>0. On a conventional craft which is heeled the sail
generates some negative vertical lift, and so φA<0, which is counter productive.

Accommodating velocity ratio variation

The lift L and drag D components of the force generated by a foil may be
expressed as functions of the fluid density ρ, velocity V, foil area S, and the
coefficients of lift CL and drag CD by
1
L= ρ V 2 S C L , and (10)
2
1
D= ρ V 2 S CD .
2
Over some useful working range of foil incidence angle α the coefficient of lift varies
approximately linearly with respect to α
C L ≈ 2π (α − α 0 ) ,
where α0 is the incidence angle at which CL=0. The coefficient of drag is a more
complex function of α, and also depends on aspect ratio AR, section profile and
Reynolds number Re. It suffices to say that CD is much less than CL over the useful
working range of α. Incidentally, the presence of the fluid density ρ in the above

11
equations explains why the aerofoils are so much larger than the hydrofoils. Recall
from Section 2 that as the course angle γ varies so too does the relative speed of the
air and water flow over the respective foils. In order to maintain equilibrium it will be
necessary to vary one of the foil areas SA or SH, or one of the coefficients of lift CLA or
CLH, or a combination of these, to compensate for the relative change in V A 2 and
2
V H . The expression (1) gives a guide to the relative magnitude of the compensation
required at the course extremes. Small values of β have the advantage of reducing the
variation required to foil areas or coefficients of lift as the course γ changes.
The performance polar of a craft for which the apparent wind angle β remained
constant over all courses γ would be a circular arc as shown in Figure 1. However for
real craft the performance polar will not be perfectly circular due to changes in the
drag angles εA and εH, and the elevation angles φA and φH, as adjustments occur to
compensate for the changes in VS and VA. The most dramatic change is the wave
making drag increase for a displacement hull as Fn approaches 1 / 2π . Detailed
analysis of the forces generated by displacement hulls and planing hulls is given
in [4].

Review of the literature

The full coherent method presented in this paper for the analysis of the total
forces acting on a sail craft and the resultant speed performance is novel. However
many individual parts of the complete theory have been appreciated in isolation for
many years. Lanchester [5] in 1907 pointed out that
“the minimum angle at which the boat can shape its course relatively
to the wind is the sum of the under and above water gliding angles.”
This is reproduced in Marchaj [6] along with its algebraic representation equivalent to
equation (2), however in this paper the distinction is made that the angles to be
summed are not referred to as the drag angles, but rather the projections onto the
horizontal of the drag angles. Barkla [7] in 1971 pointed out that the polar diagram of
ice yacht speed would be a circular arc, similar to that shown in Figure 1, but
consideration was restricted to craft with zero runner friction and so εH=0 and β =εA.
This too is reproduced in Marchaj [6]. Marchaj presents expressions similar to those
appearing in Table 1 for Vmg max , VSmax and Vmg min , and the corresponding course
angles, but again restricted to the context of ice yachts with zero runner friction, and
so the independent variable is εA rather than β. Bethwaite [3] in 1993, apparently
independently of the ice yacht examples, gives examples of polar diagrams showing
circular arcs for constant apparent wind β. However Bethwaite does not mention the
decomposition of β into aerodynamic and hydrodynamic contributions as given by
equation (2). Perry [8] in a 1998 report describing a hydrofoil sail craft design refers
to
“concurrent lines of action for the gravity force, sail force and the
windward hydrofoil force.”

12
5 Location and direction of the net forces
At the level of abstraction required by the above theory, the description of a sail craft
comprises the relative locations of the centre of gravity CG, the aerodynamic centre of
pressure CPA and the hydro centre of pressure CPH, and the elevations of the
aerodynamic lift φLA and the hydro lift φLH. The locations and elevations may be
variable so that feasible solutions can be obtained which satisfy the conditions for
equilibrium over a range of VT and γ. The hull and foils must be capable of generating
the required forces.
The weight W acts vertically downwards so the requirement for concurrent lines
of action can be simplified. The aerodynamic and hydro force lines of action must
intersect on the vertical line through the centre of gravity CG. Clearly there must be
some vertical component of separation between CPA and CPH. It follows that at least
one of CG, CPA and CPH must be horizontally separated from the rest. Recall that
high performance requires small elevation angles. The exceptions are ice and land
yachts, and displacement watercraft with small Froude number, all of which incur
extremely small drag penalties for vertical lift. Small elevation angles require that the
horizontal separation be large relative to the unavoidable vertical separation.
The aspect ratio of a foil AR is defined to be the ratio of the span squared to the
area, which for a rectangular foil reduces to the span to chord ratio. The coefficient of
drag for a foil may be decomposed as follows
CL2
C D = cd + ,
π AR
where cd is the profile or section drag coefficient and varies with α. The second term
is the induced drag. Low induced drag calls for high aspect ratio, and therefore large
vertical separation between CPA and CPH. The largest contributor is of course the sail
or aerofoil span. The inescapable conclusion is that while the virtue of a ship being
tall is defended, it should also be relatively wide. An Orwellian motto for high speed,
high performance sailing could be “tall ships good, wide ships better”.
It is generally preferable that neither elevation φA nor φH should be negative, in
fact the sum of the projected drag angles would generally be less if φA and φH were
similar. If multiple independent aerofoils were used, then parallel lines of action
would be more efficient. The same principle holds for a combination of hydrofoils
and planing surfaces. Note that this principle is not satisfied by a typical hull and
vertical foil combination.

Stability of the equilibrium

Stability of the equilibrium is assisted if CPA is to leeward of CPH. In this sense
stability refers to the general orientation of the craft, not its resistance to heeling. For
example consider the instability of a long rod when two compressive forces in
equilibrium are applied at opposite ends. The slightest misalignment would be
immediately amplified. Compare this with the stability when two tensile forces are
applied at opposite ends. Stability is also assisted if CG lies below the line from CPA
to CPH. Natural stability reduces the need for the addition of stabilising features to
the craft with their inevitable associated drag. If a foil is vertical, then the line of
action is horizontal. It follows that the intersection of the line of action with the

13
vertical line through CG is unaffected by the lateral position of CPA. However
stability of the equilibrium is affected by the lateral position of CPA.
The combination of the requirements that the elevations φA and φH are non
negative, CPA is to leeward of CPH, and CG lies below the CPA to CPH line, can only
be satisfied if CG lies laterally between CPA and CPH.

Review of sail craft types

Some example classes of craft will now be considered. On a winged skiff CPA
remains vertically above CPH and CG moves a variable amount to windward. For a
large multihull CPA remains vertically above CG and CPH moves a variable amount to
leeward. On a sailboard CG and CPA move slightly to windward of CPH, and φA>0,
but the interconnection of components prevents the achievement of small φH. For kite
boards the kite may be a long way to leeward, but the kite and in particular the lines
attaching it to the sailor, do not constitute a rigid body. Analysis could be conducted
by considering the force exerted by the kite lines at their point of attachment to the
board rider. An advantage is that a relatively high aspect ratio aerofoil is generating a
force at the attachment point just a few feet above the water surface, however small
φH cannot be achieved.

6 A high performance hydrofoil sail craft

The author has designed a novel type of hydrofoil sail craft, which should be capable
of sailing at very high speeds and small apparent wind angles. It is fully described in
the patent application [9]. A perspective drawing taken from that publication is
reproduced at Figure 6. The numbered parts are described in [9].
The design goal was a craft of minimum necessary complexity that was capable
of high performance over a wide range of true wind speed VT and course γ. This
capability was required on both tacks and in unsheltered waters.
All of the general principles for high performance and stability of equilibrium
were considered and generally accommodated. The design choice process led to
selection of the following general features. The locations for CG, CPA and CPH are
fixed. In use a single aerofoil and a single submerged hydrofoil generate the main
aerodynamic and hydro forces, respectively. Only the hydrofoil, together with its
supporting struts and stabilisers, remains submerged, the rest of the craft being
airborne. The main aerofoil and hydrofoil have fixed areas SA and SH.
The craft must be able to accommodate a range of projected drag angles εIA and
εIH, and it must be capable of providing a range of force elevation angles φA and φH,
and coefficients of lift CLA and CLH. This can be achieved as shown in Figure 6 by
providing three degrees of freedom in the gimbal assemblies supporting the aerofoil
and hydrofoil. Yaw rotation about the vertical axes maintains alignment with the air
and water flow respectively, keeping the foil lateral axes transverse to the respective
flows. Roll rotation about the horizontal axes controls the lift elevation angles φLA
and φLH. Finally pitch rotation about the foil lateral axes changes the incidence angles
αA and αH, thereby controlling the coefficients of lift CLA and CLH.

14
 Figure 6 Perspective drawing of a hydrofoil sail craft [9].

In addition the hull is free to rotate, about a vertical axis, with respect to the
main beam connecting the aerofoil and hydrofoil assemblies. Yaw rotation combined
with rolling φL through π/2, that is rolling the foil to be horizontal and then beyond on
the other side, allows the craft to sail on either tack. Furthermore the foils can be
thick and asymmetric. This is necessary to provide a wide range in the coefficient of
lift while maintaining low drag.
Since the ratio of foil areas SA to SH is fixed, it is desirable that the FA to FH ratio
remains reasonably constant as the force magnitudes change. This is achieved if CG
is close to midway laterally between CPA and CPH. Fortunately this is not
inconsistent with other requirements.
Recall that each foil has three degrees of freedom associated with it. Yaw can
be controlled automatically by provision of fins to maintain alignment with the fluid
flow. The pilot must control roll and pitch. This could be achieved by provision of a
joystick for each foil. Steering is achieved by generating transient net lateral forces
by coordinated adjustment of the aerofoil and hydrofoil. The foil assemblies should
be reasonably well balanced about their axes, both with respect to fluid dynamic

15
pressure and inertial mass. The wind velocity gradient must be accommodated to
fully achieve balance.
Further discussion and detail regarding the design and variations may be found
in the patent application. The topics discussed include control and operation of the
craft, including take off from rest and changing tack, stability, the use of stabilisers
and elevators, and choice of foil section shapes and properties.
Use of a single submerged hydrofoil is virtually mandated by the stated design
goals and the principles governing high performance. There are additional benefits.
Hydrofoils have superior lift to drag ratios compared with planing surfaces. A
submerged hydrofoil avoids the problems associated with the rough state of the water
surface. It also avoids ventilation that can affect surface piercing foils. A
disadvantage is the drag associated with the supporting struts. Ventilation may
further increase drag on the struts.

Cavitation
The pressure distribution over a foil may be expressed by
1
ρ V 2 CP
2
where CP is the coefficient of pressure and varies over the surface. The pressure
distribution changes as the incidence angle α changes. When the magnitude of the
pressure drop at some point on a hydrofoil surpasses the ambient fluid pressure,
cavitation occurs, and performance is impaired. For example water at VH=28 kn has
dynamic pressure
1
ρ H V H 2 ≈ 1 atm .
2
Cavitation could be a performance limitation, and foil shape for high speed should be
carefully selected to control CPH. This requirement limits the foil thickness and
reduces the maximum value and range of CLH. Although control of CLH is limited,
equilibrium can still be achieved by adjusting CLA, φLA and φLH. Ultimately further
increases in speed will require the adoption of super cavitating hydrofoils, but these
have markedly inferior lift to drag ratios.

Force geometry
Figure 7 is a plan view of a hydrofoil sail craft. Superimposed on it are vectors
representing the water flow and apparent wind. Also superimposed are the horizontal
force components. For simplicity the forces are shown acting at the aerofoil and
hydrofoil gimbal centres. This is equivalent to assuming negligible drag on the main
connecting beam and hull. Finite drag would slightly offset the craft total
aerodynamic centre of pressure. The inset diagram shows the velocity triangle. The
velocities and forces shown in Figure 7 are identical to those in Figure 1. Notice that
the main connecting beam is aligned with the forces and lies in the resultant force
plane, under the assumptions of negligible beam and hull drag. A bonus of this
design layout is that the main structural component, the beam, tends to be in tension,
rather than compression. This allows it to be a comparatively lighter structure,
although there may be bending moments that it must withstand.

16
 γ VS
Vmg
εIH
εIA
β

VA
VT

LIH

FIH VA
VH=-VS
εIH

DH

εIH

DA
εIA LIA
FIA
εIA

Figure 7 Plan view of craft with forces and velocities overlaid.

Figure 8 shows an offset frontal view of a hydrofoil sail craft. In fact it is a

view normal to the resultant force plane. The waterline is represented in the figure,
and all but the lower part of the hydrofoil assembly is airborne. The resultant forces
and their lines of action are superimposed. The resultant forces and their lines of
action are reproduced without the craft but with the force components shown, and the
elevation angles φA and φH marked. To complete the force analysis the craft must be
characterised by two further parameters. These are the elevation θ of the aerodynamic
centre of pressure CPA with respect to the hydrodynamic centre of pressure CPH, and
the lateral position of the centre of gravity CG expressed as a proportion ω of the
lateral distance from CPH to CPA. These parameters are shown in the figure. Also

17
 FA

CPA

LVA
CG FA

FH CPA φA
CPH FIA

CG
FIA
LVH
φH FH
θ
FIH CPH
FA
LVA W W

FH ωb (1-ω) b
LVH
FIH b

Figure 8 Net forces in their vertical plane.

shown in Figure 8 is a pictorial representation of the vector sums of the forces and
their components.
The condition of concurrency of the force lines of action is equivalent to the
algebraic constraint
(1 − ω ) tan φ A = ω tan φ H + tan θ . (11)
This determines the relationship between the elevation angles φA and φH. The ranges
of φA and φH are limited by
π
θ ≤ φA ≤ − ε A , and (12)
2
π
−θ ≤ φH ≤ −εH .
2
The elevations φA and φH are related by (11), and so the upper bounds are determined
by whichever is the most restrictive of these two conditions. Recall from the previous
section that high performance requires a craft to be relatively wide. For the hydrofoil
craft described in this section that is equivalent to requiring small θ. This is
confirmed by (12), which states that φA is bounded below by θ.

18
Performance analysis
A hypothetical example will now be given to predict the performance that could
reasonably be expected in practice. Modest parameter values that should be
achievable are assumed. Let θ =30°, ω =0.45 and ε A = ε H = 7.5° . As shown in
Figure 9, φA can be plotted as a function of φH, using equation (11) and the assumed
values for θ and ω. Next the necessary corresponding force to weight ratios FA/W and
FH/W can be plotted using (7) and (8). Also the resultant projected drag angles εIA
and εIH and the required lift elevation angles φLA and φLH can be plotted using (3), (4),
(5) and (6), and the assumed drag angles εA and εH. Following this, the apparent wind
angle β = ε IA + ε IH can be plotted. Finally the expressions dependent on β in Table 1
for relative speeds on various courses can be evaluated. In Figure 9 the relative
speeds VS max VT and Vmg max VT , and their corresponding course angles, have been
plotted.
Now suppose the example craft has a gross weight of 175 kg, so W=175 kgf,
and is operating in true wind VT =10 kn. If the craft could sail with apparent wind β
=20° then from Figures 2 and 9 it can be seen that a maximum speed VS max ≈30 kn
would be possible, and the best speed made good upwind would be Vmg max ≈10 kn on
course γ ≈55°. The elevations φA ≈50° and φH ≈15° are required to achieve β =20°.
From Figure 9 it can be seen that this would require FA slightly greater than
W=175 kgf. On course γ ≈55°, with β =20°, the apparent wind would be VA ≈25 kn.
Suppose the craft has an aerofoil area SA =15 m2. Assuming a coefficient of lift
CL=1.2 and applying (10) gives sufficient lift LA ≈180 kgf.
In winds too light for the craft to become airborne, there are niches in which
existing craft may have an advantage. For displacement mode sailing, longer craft
have a Froude number advantage. Craft that can deploy massive light air rigs may be
able to achieve a force to weight ratio advantage.

7 Conclusion
It has been shown that the system of forces acting on any sail craft at equilibrium can
be reduced to an equivalent system of three forces representing the net aerodynamic,
hydro and gravitational forces. The resultant force lines of action lie in a vertical
plane and are concurrent. These are sufficient conditions for equilibrium, it is not
necessary to give separate consideration to heeling, pitching and righting moments.
There is a direct geometrical relationship between the resultant forces and the velocity
triangle. This relationship leads to the fundamental equation (9), which seems to be
new, relating the apparent wind angle β to the drag angles εA and εH and the force
elevation angles φA and φH. This single equation encapsulates the factors controlling
performance, namely the aerodynamic and hydro lift to drag and force to weight
ratios. Note that equation (2) relates the apparent wind angle β to the projected drag
angles, not the drag angles themselves.

19
 180
180

170

160

150

140

130

120
φ BHltmax )) γV
S max
110

) 100
degrees
ε H, θ , λ ) 90

80
A, ε H, θ , λ )
70

ε A, ε H, θ , λ ) 60 γV
mg max
2
50

) φLA
Hltmax
40 φA
φ BHltmax )) 30

20 β
εIA
10
εIH
0

10

φLH
20

− 30 30
30 20 10 0 10 20 30 40 50 60 70 80 90
− 30 φ BH φ BHltmax φ BHφHltmax φ φ φ BHβ lt90 φ BHβ lt90 φ BHφHltmax φ BHltmax
BHltmax(degrees) 90
, , , H , , , ,
5 deg deg deg deg deg deg deg deg
5

4
,λ )) VS max VT
)
BHltmax , φ BHltmax ) 3

λ , φ BHltmax )) FA W
2
1 FH W
−
,θ,λ )) 2
1 Vmg max VT
0 0
30 20 10 0 10 20 30 40 50 60 70 80 90
− 30 φ BHβ lt90 φφH (degrees)
BHltmax φ BHltmax φ BHβ lt90 90
, , ,
deg deg deg deg

Figure 9 Angles, velocities, forces as functions of φH .

20
 Equation (1), which seems to be new, indicates the extremes to be encountered
in the VA 2 / VS 2 ratio as course γ changes. Sail craft must be able to accommodate
these extremes. A benefit of decreasing apparent wind angle β is a corresponding
reduction in the variation of this ratio.
The achievement of high performance and stable equilibrium imposes certain
general design requirements on the relative locations of the centre of gravity CG, the
aerodynamic centre of pressure CPA and the hydro centre of pressure CPH, and the
elevations of the aerodynamic lift φLA and the hydro lift φLH.
A hydrofoil sail craft has been designed with minimal necessary complexity that
generally accommodates all of the attributes required for high performance and
stability of equilibrium. The craft is designed to operate on both tacks in unsheltered
waters over a wide range of true wind speed VT and course γ. The craft should have
superior performance to all existing types of high performance sail craft on all courses
in most conditions. The feasibility is demonstrated by example calculations showing
exceptional performance.

References
1. Beer, F P and Johnston, E R, Vector Mechanics for Engineers – Statics, McGraw-
Hill, Fourth Edition 1984.
2. Thompson, Sir William (Lord Kelvin) and Tait, P G, Principles of Mechanics and
Dynamics, first published as Treatise on Natural Philosophy by Cambridge
University Press in 1879, last revised 1912, Dover Publications unabridged
and unaltered edition 1962.
3. Bethwaite, F, High Performance Sailing, International Marine, 1993.
4. Larsson, L and Eliasson, R E, Principles of Yacht Design, Adlard Coles Nautical,
London, Second Edition, 2000.
5. Lanchester, F W, Aerodynamics, Vol 1 p 431, A Constable and Co, London, 1907.
6. Marchaj, C A, Aero-hydrodynamics of sailing, Adlard Coles Nautical, London,
Second Edition 1988, reprinted 1993.
7. Barkla, H M, Physics of sailing (Encyclopaedic Dictionary of Physics -
Supplementary Vol 4) Pergamon Press, 1971.
8. Perry, J, AYRS Newsletter, p 7, October 1998.
9. Bourn, Hydrofoil Sail Craft, PCT international patent application, priority date
2 November 1998, publication number WO 00/26083, 11 May 2000.

21