HYDROFOIL R E S E A R C H P R O J E C T

F O R OFFICE O F N A V A L ‘ R E S E A R C H
NAVY DEPARTMENT, WASHINGTON, D.C.
C O N T R A C T NO. NONR-507(0(D)

HYDROFOIL H A N D B O O K

HYDRODYNAMIC CHARACTERISTICS
O F COMPONENTS

BATH IRON W O R K S C O R P .
BY GIBBS AND COX, INC.
NEW YORK 6, N.Y.
 FOREWORD
- -

THIS HYDROFOIL HANDBOOK HAS BEEN PREPARED

BY GIBBS & COX, INC. ACTING AS THE DESIGN
AGENT OF THE BATH IRON WORKS CORPORATION
UNDER OFFICE OF NAVAL RESEARCH CONTRACT
NONR-507(00). THE FOLLOWING WERE DIRECTLY
RESPONSIBLE FOR THE PREPARATION OF THE
TEXT OF VOLUME II:
Mr. W. H. Michel - Editor and Contributor
Mr. S. F. Hoerner - Contributor
Mr. L. W. Ward - Contributor
Mr. T. M. Buermann - Project Coordinator

ACKNOWLEDGEMENT
IN THE PREPARATION OF THIS HANDBOOK, USE HAS
BEEN MADE OF WORK PUBLISHED BY OTHERS. EVERY
ATTEMPT HAS BEEN MADE TO ACKNOWLEDGE: THIS
FACT BY SUITABLE NOTATIONS AND LIST OF
REFERENCES, THUS ANY OMISSIONS ARE INADVERTENT.
 HYDROFOIL HANDBOOK

VOLUME II

HYDRODYNAMIC CHARACTERISTICS OF COMPONENTS

INTRODUCTION

CHAPTER 1. Airfoil Principles

2. Submerged Foil Characteristics

3. Surface-Pfercing Foil Characteristics

4. Foil Parasite Drag

5. Flap Characteristics

6. Hydrofoil Wake

7. Strut Characteristics

8. Rudder Characteristics

9. Characterfstios of Nacelles and Small

Appendages
10. Skid Characteristics .

11. Hull Drag

12, The Influence of Cavitation

APPENDIX A. Calculation of the Lift and Drag

Characteristics of a Proposed 50
Ton Hydrofoil Craft
 HYDROFOIL HANDBOOK

VOLUME II

HYDRODYNAMIC CHARACTERISTICS OF COMPONENTS

INTRODUCTION

The hydrodynamic characteristics of the various components that

make up a hydrofoil craft must be known in order to prepare a design

or to analyze a given design or craft. It is intended in this volume

to present such information i n the form of concise engineering

formulations and methods, with some background material to give an

understanding of the underlying hydrodynamic relationships, the scope

and adequacy of available theory and data, and the general state of

knowledge.

Material was extracted from exfsting reports on hydrofoil theory,

analyses and experimental data and from pertinent established aero-

dynamic and marine information. Where sufficient reference material

was lacking, it was necessary to derive certain properties during the

preparation of this text, In other instances, where general relation-

ships could not reasonably be established, available data and suggested

 INTRODUCTION

means of analyses and derivation are given. In order to maintain this

work as a readily usable handbook for direct engineering use, hydro-

dynamic fundamentals and derivations are kept to a minimum, consistent

with clear exposition of the various relationships. FOP those

interested in reviewing the basic fundamentals or exploring further

into the various subjects, comprehensive references are given on each

subject.

In all cases, a conscientious effort was made to provide pertinent

engineering information on all components of hydrofoil configurations,

even though some of the proposed formulas may be tentative and even

conjectural; so that there would be a definite basis on which to pre-

pare a design and on which to make comparison when additional

information is forthcoming.

It was felt that the hydrodynamic characteristics should be

presented without qualifications as to what combination of components

provide the best hydrofoil configuration. There is a wide variety of

configurations, each of which may have some particular advantage under

particular requirements; and to attempt to introduce such operational

factors would complicate any presentation of basic information,

For similar reasons, only the "steady-state" characteristics of

the components are considered, Investigations of dynamic stability,

behavior in waves, turning characteristics, etc. should be undertaken
 INTRODUCTION

only after a complete configuration has been chosen tentatively on the

basis of specified functional requirements,

Although treatment of the "best" foil size, öptimum" number of

struts, "most effective" configurations, etc, has therefore not been

included in this volume, some practical limitations as to size, speed,

geometry, etc. were considered so that effort could be concentrated on

deriving more exact properties in the range of applications considered

to be most frequently employed. These limitations are based on

evaluations of existing craft and analyses of general design studies,

such as are indicated in Volume I, In most instances, it is stated in

this text where such limitations are used and the procedure for obtain-

ing information for cases beyond such limits is: shown or inferred,

The material contained herein is presented under chapter headings,

as indicated in the table of contents, with the chapters so sequenced

as to fall into three main groupings, Chapters 1-6 contain the basic

hydrodynamic characteristics of the foils, including the effects of

the various other influences on the foils. Chapters 7-11 deal with

the characteristics of struts, hulls and the various other appurtenances

in a configuration, Finally, Chapter 12 indicates the influence of

cavitation on the characteristics of the various components. Cavitation

is separately treated to stress its importance,, to show the ranges of

craft speed and size in which it is a factor and those in which it can
 INTRODUCTION

be neglected, and possibly of greatest importance at the present time,

to indicate the dearth of necessary information for predicting the

hydrodynamic effects of cavitation on a foil configuration with

sufficient accuracy,

Design examples have been distributed throughout the text to give

ready indication of how to apply some of the important formulas and

methods,, An appendix showing the detail calculations of the lift and

drag of a chosen design is also included to serve as a summary of the

progosed material and its application,

Since the basic principles (and geometry) of hydrofoils are the

same as those of airfoils, most hydrofoil properties have been derived

from airfoil theory and data, with airfoil nomenclature being generally

adopted. Chapter 1, an introduction to airfoil principles, has there-

fore been included to familiarize the reader with this subject, which

is necessary to the proper understanding of hydrofoil hydrodynamics.

On the other hand, the hydrodynamics of hulls have not been com-

prehensively presented in the text. Information on hull drag is readily

available from many published sources, most of which are known to the

naval architect, and the detailed coverage of such information in this

work would be needless duplication, In this particular case, a short

discussion herein with references to the more basic works has been

considered sufficient.
 INTRODUCTION

The criterion for presenting information in this volume has been,

therefore, to elaborate on subjects that are new to the naval architect,

but to abridge the treatment of more familiar subjects which are amply

treated in general marine texts.

 CHAPTER I. AIRFOIL PRINCIPLES

1. Introduction

2. Airfoil Notation

3. Lifting-Line Airfoil Theory

4. Modifications to Lifting-Line Theory

5. The Influence of Fluid Boundaries

6. Pitching Moment Characteristics

7. Airfoil Drag

8. Airfoil Data

'The essential principles of foil sections and wings are presented

as the basic material used in developing hydrofoil characteristics.

Fundamental airfoil theory is outlined, including those factors that

are most applicable to hydrofoils. Viscous drag considerations are

given, and the availability of airfoil data .for design use is pointed

out *
 AIRFOIL PRINCIPLES

1. Introduction

Basically, a hydrofoil is an airfoil operating in water.

Aerodynamic foil principles can therefore be uti:Lized in developing

the hydrodynamic characteristics of a hydrofoil.

In some instances, such as where the foil operates at large

distances below the water surface, established airfoil theory (in

incompressible flow) and pertinent airfoil data can be used directly

for determining the hydrodynamic characteristics of the foil. In

more practical cases, the airfoil methods must be corrected by suit-

able factors to take account of the surface effects, supporting struts,

high speeds (cavitation) and other factors associated with practical

hydrofoil configurations.

In any case, aerodynamic foil principles form the basis for the

development of hydrodynamic characteristics of h;ydrofoils. Therefore,

this chapter is presented to familiarize the reader with the airfoil

notation employed, the underlying airfoil theory that is most

applixable to hydrofoils, and the availability of pertinent airfoil

data for use in deriving hydrofoil characteristics,

 AIRFOIL PRINCIPLES

2. Airfoil Notation

Foil Section Geometry

Figure 1.1 shows the typical airfoil section notation.. The

chord C of the foil is the distance from the leading edge to'the

trailing edge, and the angle of attack OC iti the angle between the
chord line and the direction of advance, as indicated.

MAXIMUM

GEOMETRY OF AN AIRFOIL SECTION

a- SECTION WITH THICKNESS, t
b- THE SECTION'S "SKELETON". THE MEAN LINE
(CAMBER LINE) HAS A MAXIMUM CAMBER,
AT A DISTANCE X FROM THE LEADING EDGE

FIGURE 1.1

The mean line of the section is called the camber line, the camber
being considered the maximum separation between this line and the

chord line. For simple circular arc sections, the mean line has

symmetrical camber with the maximum separation and maximum foil

thickness at the mid-chord position, but for typical airfoil

II - 1.3
 AIRFOIL PRINCIPLES

sections the shape of the camber line and the chordwise location

of maximum camber and maximum thickness vary, depending on the type

of section emplcyed. A description of modern airfoil section

shapes employed by NACA is given by Abbott1j2.

Wing Geometry

Figure 1.2 shows the typical

wing geometry and notation. Dim-

ensions are characteristically

taken about the quarter-chord

line (1/4 of the chord from the

leading edge of the foil). Thus,

the span is the projected distance

between the tips of the quarter-
chord line, and the sweep and

dihedral angles are measured I FRONTAL VIEW

from the projected quarter-chord

IW I N G G E O M E T R Y
line in the horizontal and
transverse planes respectively
FIGURE 1.2
as indicated. Two important

"parameters" in dealing with

wings are2

-.-------.-. I ._-I __--_ _^-” . .

 AIRFOIL PRINCIPLES

the plan form area, 5 - the projected area of the wing

in a horizontal plane

the aspect ratio, A - the ratfo of span to mean chord,

or b’,s

The difference b8tW38n rVdisplac8ment bodies"

and '"lifting surfaces " in the selection of
reference areas should be noted here. For
ships and similar bodies the total wetted
area is usually chosen as the reference area,
whereas for wings the projected area of one
side of the wing is taken, i

Force Notation

The force system acting on a foil (or any body) may oonveniently
_- be referred to the components along and about the three co-ordinate

axes of the body, These are identified, for the foil, as%

Lift - force in the vertical direction

Drag Force - force in the fore and aft direction

Side Force - force in the transverse direction

Pitching Moment - about the transverse axis

Rolling Moment - about the fore and aft axis

Yawing Moment - about the vertical axis

 AIRFOIL PRINCIPLES

Considering the forces in

the plane of symmetry for

stra:Lght forward'flight, the

forces on the foil section

are shown in Figure 1.3.

(The transverse axis is taken
,
to be at the ltaerodynamic

center", which is discussed

F IlGURE 1.3
below.) *

In many applications, pressures and forces in fluid flow are

proportional to the "dynamic pressure1t

$ = p/2v2 (lb/.ft*) (1.1)

where p - mass density of the fluid (#sec?/fth)

V = speed of advance (ft/sec)

The mass density of the fluid varies slightly with temperature.

At the standard 59°F (l!?'C),

P - 1.94 (#sec2/ft 4 ) for fresh water

p = 1.99 (#sec*/ft 4 ) for salt water

Thus, using these units, p/z may be considered equal to 1.0

and s=: \/' in water.
-

--.-.,
II - 1.6
 AIRFOIL PRINCIPLES

What is the dynamic pressure in sea water at 30 knots?

V = 1,69 x 30 = SO07 ft/sec

p = 1,pp #sec2/ft4

v2 = 2570 ft2/sec2
F = 1.99/2 x 2570 = 2558 lb/ft2

The forces on the foil may readily be represented in non-

dimensional coefficient form, by referring them to the dynamic

&$,.. !, ' _
pressure and the foi area. Thus

Lift Coefficient, c, =

Drag Coefficient, CD = wp

Pitching Moment Coefficient,

where L = lift of the foil

D = drag of the foil

M - pitching moment on the foil about some point on

the foil chord - taken positive when tending to
increase the foil angle of attack

S = planform area

C - foil chord

C). = dynamic pressure

The pitching moment coefficient is essentially constant when

moments are taken about the aerodynamic center., (Theoretically, this

.^-._ _...-._. ---- .- ----

-. I__-___-._.-.---..~--r-.---.~“.I.” I... “I_.
 AIRFOIL PRINCIPLES

point idha~~.er-ch~.~~...distance behind the leading edge and Is thus

4
a good reference point for wing geometry, as shown in Figure 1.2.)

For such moment center, the pitching moment coefficient is identified

The basic aerodynamic notation and the units usually employed

are as follows:

A aspect ratio non-dimensional

b projected foil span ft.

6 foil chord ft.

D drag lbs.

f foil camber ft.

b lift lbs.

M pitching moment ft.lbs.

dynamic pressure, lbs/ft2

e
5 projected foil area fta
.

t foil thickness ft.

v speed of advance ft/seo

d angle of attack radians

P mass density of the fluid lbs.sec2/ft 4

II - 1,8
 AIRFOIL PFUNCIPLES

A angle of sweep degrees

r angle of dihedral degrees

Cp drag coefffcient, D/cqS non-dimensional

CL lift coefficient, L./s5 non-dimensional

&, pitching moment coefficient, M/%sc .non-dimensicqal

II - 1.9
 AIR.F'CIL PRINCIPLES

3. Lifting-L,ine Airfoil Theory

General

Classical airfoil theory originally developed by scientists

such as Frandtl, Lanchester, Glauert, Munk, et al, is; available for
study in many treatises on aerodynamics 3,495. These works are

based on the concept of a single lifting line,. which1 is sufficiently

accurate for wings of large aspect ratio. Although in recent years
advances have been made in the development of a lifting-surface

theory (most applicable for foils

of small aspect ratio), the lifting-

line theory is still of foremost

importance.

Two-Dimensional Theory

The basic theory of lift is

derived from the concept of a

rectilinear vortex advancing in
a fluid In a direction normal to
‘ORIGIN OF LIFT
the vortex axis. The super-
SUPERPOSITION OF CIRCULATION
AND SPEED OF ADVANCE v IN(h)
position of the circulatory LEADS TO THE LIFTING VORTEX
(b). A FOIL SECTION(C) PRODUCES
THE SAME KIND OF FLOW PATTERN.
motion of the vortex upon the
free stream velocity, as shown
FIGIJRE 1 . 4
in Figure 1.4, results in an

_-__” ----._
 AIRIQIL PRINCIPLES

frmwms in veloe,fty on one side of the vortex and a decrease on the

ot:tsr side. According to Bernoulli's theorem, therefore, the static

presyfire js decreased in the "suction"' side and fnereased on the

"pressure" side. The resultant lift force is perpendicular to the

free-stream direction, and has the magnitude:

L = bpV@ (lbs.) (1.3)

where is the mass density of the fluid (#sec2/ft4)

P
V 1s the fluid velocity ( ft/sec 1

0P is the circulation of the vortex (the integral

of the vortex velocfty over a closed p th
encircling the vortex center) (ftS/sec)
b is the foil span in ft. (assumed to be infinite
in the considered two-dimensional flow pattern)

The relationship between circulation and the lift coefficient of

a foil section can then be expressed as

(1.4)

The value of the circulation around an airfoil of known geometry

can be determined by fluid potential theory. Thus, for a Whin air-

foil'! section 9

0L = PcVQq’ (due to angle of attack, and

at the quarter-chord position)
(1.5)
= 2SVf (due to camber and applied at the
half-chord position)

.- -____-- ..------
 AIRFOIL PRINCIPLES

mencombining equations (1.h) and (1.5) we get

where CL is the lift coefficient

c~~~ is the moment coefficient taken
about the leading edge of the foil

c&, is the angle of attack in two-dimensional flow

f/c is th e camber ratio of the foil.

From equation (1.6) it is seen that for a cambered foil, the

lift is not zero at zero angle of attack (as measured in Figure 1.1).

Rather, the angle of zero lift is seen to ber

&o = - 2 fk (1.7)

However, zero angle of attack does have a particular significance

in cambered sections. In two-dimensional flow, it approximately

indicates the flow pattern where locally there is no flow around the

leading edge of the section from one side to the other. For this

condition of **smooth entrance", the ltsymmetrictl or *toptimum" lift -"I '

-,;
coefficient is approximately 1 I ,.
: ,
:’

(1.8)
 AIRFOIL PRINCJXJB

Example

WhaL is the lift coefficient of a "thin" foil section

(in two-dimensional flow) having a camber ratio of

f/c = 3% at an angle of attack of d; = 5" ?

The optimum lift coefficient of this section is

_ theoretically (equatS& ~8)s &Lopt =$f1o,ag=a38

In pp'actiee, air%oU~seetfons are not VhfngV:, but have a fX"S.n%t.e

thf@kness, as indicated iv Figure l.,L above. Properties 0% practl.ca3_

aIrfoIl. sections can also be determined by fluid potentfal m&hods.

The ealeulations l.nvolved are arduous, however, and the resulta are

not; really appPfcable due to viscous basses 3.n the boundary-layer

KLOW, Eacperimental data on the properties of afr.fof$ B~C~%SXS~ a.3

discussed below should be employed fn engineering appl.l.catfons.

However, a review bf the data for modern airfo%l secti.ons' l,%

l.ndicates that the values gfven in equation (1,6) for the lift
 AIRPDIL PRINCIPLES

and moment coefficients are reasonably accurate, and may be used for

preliminary purposes,

W5ng Theory

A wing, with chord c and span b, is the equivalent of a lifting

line having finite (limited) length. As illustrated in Figure 1.5, "

around the ends or tips of this line, a flow is caused from the

pressure side of the foil to the suction side. The circulation of

the bound vortex, therefore, finds continuation in the form of a

pafr of tip vortices, one originating from each wing tip, approxi-

mately as sketched in Figure 1.5. In other words, the single vortex

(as vfsualfzed in Figure 1.4) is replaced by a vortex system, which fs ,

remotely similar in shape to a horseshoe, (thus, sometimes called a

horseshoe vor%&],,

Withfn the space between the two tip vortices, their circulations

combine, thus '%nducing" a downwash velocity and causing as a con-

sequen1c.e a downward deflection of the basic stream from its undisturbed

direction. This deflection reaches a final angle at some distance

behind the ting, as a component part of the vortex pair in Figure 1,s.

At the position of the foil or lifting line, the deflection of

the fluid (one-half the final downwash angle) results in the "induced

angle of attack". Because of this angle, the lift force is now

 AIRJUL FRIKCIPLES

Al V(AOVANCINQ FLUIOI

LIIFTING LINE *

(b)

1 IP VORTICES

PLAN VIEW

,SUst@N - - -
(c)
-t-t++++++
PRESSURE

I- bvY4 ---I

FRONTAL VIEW

FLOW PATTERN PAST THREE-D IMENSIONAL WING

(a\ W I N G PLANFORM

(b) EQUIVALENT VORTEX SYSTEM

(C) O R I G I N O F T I P V O R T I C E S

(4 VORTEX PAIR FAR BEHIND WING

FIGURE 1.5
 AIRFOIL PRINCIPLES

__
displaced from its direction

normal to the free-stream

fl6r and is tilted backward

by the induced angle Oci,

as illustrated in Figure
1.6. The lift (defined as
ORI’GIN OF THE INDUCED
R E S I S T A N C E . CDL’ C+noCi
the vertical component of ,

the foil force) is accordingly

FIGURE 1.6

C‘ I c, cos d; z CL” (1.9)

and there is now a component of drag or resistance (defined as the

force component in direction of the motion) - the induced drag

corresponding to

Since within the scope of application of vortex-line theory, the

induced angle is comparatively small, it is usual to assume

cos eci z 1.0

scnoci z tanq Zdi

a s indicated in equations (1.9) and 1.10).

II - 1.16
 AIRFOIL PRINCIPLES

For an actual foil, the loading (or circulation) is not con-

stant al'ong the span as was assumed for the simple horseshoe vortex

but varies depending on the section geometry at each point along

the span and on the downwash at that point due to the trailing

vortices (for instance, the horseshoe vortex system leads to an

infinite downwash at the tips, an untenable assumption).

The classical treatment of

this consideration by Prandt13 __.

k .I SPAN‘, k
indicates that an elliptic, -7
7 , ,- ELLIPTIC
distribution of lift along the

span of the foil as shown in

Figure 1.7, results in a con-

stant value of downwash along ELLIPTIC LOADING

the span (constant value of
FIGURE 1.7
induced angle) and a minimum

value of induced drag.

For the case of elliptic lift distribution, the derived values

of induced angle and induced drag are, in coefficient form:

OCi 2 =L/VA
(1.11)
CD; = CL di = cL;;rA

where A is the foil aspect ratio.

-~~ --_l__- --- --- -- --

 AIRFOIL PRENCIPLES

The expression for the lift coefffcfent based on the angle of

attack fn two-dimensional flow, equation (lo6jp must be modified f'or

the ffnfte wing by taking the induced angle into account. Thus,

(1012)

where d 1s the geome&fc angle of attack

d e. is the angle of zero Ifft, from equation (S,7)0

The lift curve slope fs an important concept in airfoil (and

hydrofoil) analy&res, Rearranging the terms of equation (X.12) and

tS&fng the der%vative, the slope is

_-

1% -fa frequently more eonvenfent and useful, :to fnver% equa!z!.an

(b,U) and determine the angle of attack necessary to produce 1

desired %ift coefffofenb, Thus, the Yinverse slopePB or Bli9f't;

angle pr f.s

is the lift angle dye to the lifting-line uort~lr

ia the lff% angle due to the trafling vor%ida~

II - 1,18

--- -
 AIRFOIL PRINCIPLES

What fs the angle of attack of a wing with ellfpti.cal

loading, having A = 5, needed to produce a lift

coefficient of CL = Cl.!??

It should also be noted tha% from equa%ion ('b,ll)~

whQh is also a useful relationship in airfoil analysis,

.I
 AIRFOIL PRINCIPLES

4* Modifications to Lifting-Line Theory

Effect of Planform

The lifting line theory as outlined above is exact for wings of

large aspect ratio, having elliptical lift distribution. Considering

wings of simple geometry (i.e., straight, non-twisted, constant

section shape), elliptical distribution is found only for wings of

elliptical planform. It is therefore necessary to determine the

effect of other planforms on the induced oharacterfstics.

The method proposed by Glauert 4 can be used to determine the

corrections to be applied to the induced angle and induced drag terms

for wings of various planforms, The induced terms of equation (1,15)

now become

where TJ are the corrective terms for the induced

angle and induced drag, respectively.

Figure 1.8 gives the T and d corrections for wings of

rectangular planform in terms of the aspect ratio.

,_-.--
_.--_
~___.
- - .-..
--“~_-~--~-
 AIRFOIL PRINCIPLES

-f-ad FOR RECTANGULAR PLANFORMS

‘FIGURE I.8

Huebe# has calculated the correction terms for, wings having

straight taper from midspan to tip, for several aspect ratios as
a function of taper ratio, as shown in Figure 1.9. The value of

T can be read off directly from Figure 1.9(a), interpolating as

required for other aspect ratios. The value of J for any aspect
ratio can readily be determined from the function
& #A
where d' is given in Figure 1.9(b) for any aspecrt ratio (as
suggested by Hoerner').

II - 1.21

,__ -,-_ -_^l-. ~. ----.- ------

 AIRFOIL PRINCIPLES

r’

(a) .f FOR INDUCED ANGLE -

0.00
I
d
/ 7

(b) $ Fof? INDUCED DRAG

PLANFORM CORRECTIONS y&d

F I G U R E I.9

I I -1.22

_-_.__._ _. .-_-.----.------“- ---

 AIRFOIL PRINCIPLES

The above planform corrections are derived on the basis of the

lifting line theory, without consideration of the .finite chord. In

the practical case, there is some advantage to be gained from hating

square wing tips (both in planform and in lateral section) as they

prevent the flow from "getting around)) the wing tip, thus increasing

the effective span and aspect ratio, It will be shown, in the dis-

cussion of lifting surface theory below, that for wings of low aspect

ratio the rectangular planform is more effective (has less induced

angle and drag) than those of rounded or tapered form.

Effecta of Sweep

The effects of sweep on a foil have been investigated theoretically

in several references8j9. A simplified method of approach is presented

herein.,

For a swept wing (as illustrated in Figure 1,2), the reference

angle of attack is that of the foil section at midspan. Thus, for the

flow normal to the quarter-chord line, the nofinal angle of attack Is

increased by I/osn but the dynamic pressure is decreased by cas'A o

The net effect is to decrease the section lift curve by the factor c05A O

Equation (1,14) may then be written for the swept foil:

(1.17)

es II - 1.23
 AIFU?OIL PRINCIPIXS

and induced lift angles

foil
sweep

k is a factor to be determined.

For elliptical lift distribution, the factor 'k = 1 and the

induced term is the same as for a straight foil with elliptic

dfstribution. However, the low taper ratios required for swept-back

wings to make the lift distribution elliptical (as shown in Figure

124 below) are not practical, because of wing-tip stalling,

From an analysis of test data of swept wings, Hoerner7 shows

that k varies approximately as ‘/cosA e Thus

(1018)

where is the lift angle of the foil without sweep,

The induced drag term may also be expressed approximately as

Tests conducted by NACA1O*ll on various swept wings indicate that

the lift curve and induced drag do not obey these simple formulations

in all,cases, particularly at large sweep angles. Wings with

II - 1.24
 AIRFOIL PRINCIPLES

sweep f'orward have different characteristics than those with sweep-

back, and in some instances the induced drag increases with increase

in aspect ratio0 Therefore, the above relationships should only be

used fn the absence of specific test data,

Effects of Dihedral

According to a simple theory12, each panel of a dihedraled or

) _
V-shaped wing has an induced angle (in the direction normal to the
I *' 4
panel) equal to that of a complete wing with an aspect ratio which ',
is twice that of the panel,

Referring the lift and drag forces to the projected area and

span (see Figure 1,2 above), and noting that the reference angls of

attack is that of the foil section at midspan (in a vertical plane),

the lift angle may be simply expressed as

where k is the dihedral angle.

..

.._..... -- ---.---li -. -..-__--

.““Iyorr_rrrr AIRFCIG PRINCIPLES

Example

What are the induced characteristics of a wing,

i
having A = 25" and r = 291 According to

equation (1.18)
ck L
dCL A &A .
For A = 25" (COS 29 - 0,91), the induced

angle and induced drag are roughly increased by'

lo%, The angle of dihedral does not affect the

L- induced characteristics,

Lifting Surface Consfderations

The theory of lifting surfaces has been promoted in recent years

to determine the aerodynamic characteristics of w3ngs of small aspect

ratio, Weinig13 and Weissinger 14 have produced theories capable of

being applied readily for engineering purposes. L,awrence1S has more

recent:Ly produced generalized functions that agree with Weissingergs

results and further allow the determination of the chordwise load

distribution.

The complete formula for the lift of an airfoil may be expressed

as8

(1*21)

II - 1,26

__-____ __-“I_ -_I-

 AIRFOIL PRINCIPLES

where cX includes the angle of zero lift, for cambered foils

.
E is the correction factor applied to the basic lift-
ing line theory. Values of E are! given in Mgure
1.10.
2Sln'tX is a normal force coefficient, which can be derived
rigorously for sero aspect ratio bly any of several
concepts* The simplest concept isi that it represents
the viscous drag uoefficient normal to a two-
$mee;;;~l flat plate (CD= 2,0) with a fluid speed
0

Equation (1.21) is thus seen to represent the corrected lifting

line theory (with Srnd substituted for Oc for greater accuracy at

hi& angles) plus a non-linear term which is actually derived for

zero aspect ratio but which appears to be valid throughout the range

of low aspect ratios (A ( 2).

The drag,due to lift may be given approximately as2

where 6~~ is the first term of equation (1.21)

6~~ is the second term of equation (L21).

Values of E can be derived from Figure 1.10, where I/E is

plotted against aspect ratio for rectangular foils. It is seen that

Wefnfg and Wefssinger show different values in the range of A>! 0

An empirical factor E = 1 +'/A2 for A p2 falls between the two

theoretical values.

,-

I I - 1,27

.-- -
 AIRFOIL FRINCIFLES

0.6

0 I 2 3/4-G. v56 7 8 9 IO
f2T.A ‘ 1 ~,- ,
VALUES OF & FOR RECTANGULAR FOILS
AS A FUNCTION OF ASPECT RATIO

Lawrence (and Weissinger) also derived values for triangular

wings. Figure 1.11 gives a comparison between the rectangular and

1
triangular values of the complete factor 5 +L and the lift-
trt lPA
ing line theory (E - l), for small aspect ratios.

II - 1.28

_ -,-l_l_--_- .-.
 AIRFOIL PRINCTPLES

I . R E C T A N G U L A R P L A T E S - SOUARE E D G E S
+RECTANGULAR PLATES -VARIOUS ROUND
ma*-” cb c1Am.IcII

0 I 2 3

ASPECT RATIO

tiOMPARlSON OF S M A L L A S P E C T R A T I O T H E O R Y 8 D A T A

FWJRE Hi

Various test data for rectangular foils have been analyzed by

Hoernerje, and tke results are also shown in Figme 1.11. It is seen
that the characteristics are dependent on the edg:e shape of the

sect?Lons; those having square edges showing slightly greater lift

than theory indicates while those having rounded edges and rounded
corners are close to the theory for foils with triangular plan forms.

s in an unpublished memorandum

. .
.I

II - 1.29 '

--. _” - -----
___.~__-_-..l.-_l
.___-
 AIRFCIL PRTXTPLES

For foils of aspect ratio larger than 2.0, th'e angle of attack
is usually small so that the second or non-linear term of equation

(1,21) may be neglected, and the equation may then be rewritten:

(COP 02) (1023)

to which the planform, sweep and dihedral correci&ons can readily be

applied, Thus, for the complete wing:

&,I
dcL co5 fi
L
l- ZiGi
+ l+r
VA 1 (1.24)
dCo I+6
a?) = IfA
where E is the lifting surface correction, l+2/A2
c FM a )2J
A is the angle of sweep

f is the angle of dihedral

r,d are the respective planform corrections.

I T - 1.30
 AIRFOIL PRINCIPLES

5. The Influence of Fluid Boundaries

When the wing is operating close to a fluid boundary, its "free-

flfght" aerodynamic characteristics are modified, to an extent dspend-

ing on the type of boundary and the distance of the wing from ito
c -- I *
There are two main types of boundary4 Y"~~ICE- w 94~" ' +-

(a) the rigid boundary - the walls of a wind tunnel in whfeh a

wing is being tested, and the ground

when an airplane is in the process of

taking off or landing.

(b) the free boundary - the boundary of an "open-jett* tunnel,

where a wing is bel.ng tested in a jet of

air that is freely surrounded by the

atmosphere. As will be shown below, thfa

is also equivalent to the effect on one

wing of an equal biplane.

The influence of fluid boundaries -is treated in detail in many

classical works395 and will only be discussed briefly herein. FCC

convenience, the term "ground effectIv is used for the effect of a

rigid boundary, and "biplane effect It for that of a free boundary.

While the biplane effect need seldom be considered in modern
 AIRFOIL PRINCIPLES

aerodynamics, it is basically involved in the influence of the

water surface upon the characteristics of hydrofoils.

The presence of a rigid boundary at the distance k from the

wing, as shown in Figure 1.12(a), affects the motion of the fluid

about the tip vortices originating at the wing. By reasons of
symmetry, this infuence is
.identical to that of a '

"mirror image" of the foil

RIGID WALL
and vortex system at the

same distance k beyond the

boundary. (That is, the

boundary can be replaced by

l---b--l
(I 1
h .
the mirro:r image, in consider-

E
- - - - - - - - EOUIVALENT
FREE BOUNDARY
h
ing the e.ffect on the wing.)
c 1
It is easily seen that the (b) BIPLANE EFFECT

downwash and the induced

angle at the wing are reduced. FIGURE 1.12
As a consequence, the lift

curve slope d=hc is somewhat increased and the induced drag

decreased in the presence of a rigid boundary.

The biplane effect, on the other hand, produces the opposite

results. Consider the biplane represented in Figure 1.12(b) with

I I - 1.32
- AIRFOIL PRINCIPLES

the separation of 2h between the wings. Here, each wing provides the

same lift and has identical vortex systems in the same direction, It

is easily seen that the downwash and the induced angle at one wing

are increased due to the vortex system of the other wing, in direct

apposition to the ground effect discussed above. It can readily be

shown that a free boundary a distance h from a wing can be replaced

by a "biplane image" a distance 2k from the wing, and thus is

equfvalent to the "biplane1 effect on one wing,

The fnfluence of either the ground effect or biplane effect on

the fnduced characteristics of the wing may be expressed as a factor,

K, fn the equation

dor;
-= dcoi
'- I4
(1025)
dCL d(G.T) =

The factor K is given in Figure 1,13*, a,s a function of b/b9

Kb denoting biplane effect and Kg denoting ground effect, It is

seen that at a given b/b the ground effect would decrease the fnduced

terms by the same amount as the biplane effect would increase them,

Whenh = og the induced terms would vanish due to ground effect and

* See Figure 2.7, Chapter 2, for a

more detailed graph.

-
 AfRFOIL PRINCIPLES

would double due to biplane effect, When Kb= Kg = 1.0, the

"free-flight" condition exists, i.e. no boundary effect.

B I P L A N E A N D G R O U N D EFFEGT FAC:TOR,K

For small values of h/b, corrections to K, must be made for

such things as the actual tip clearance of dihedraled or swept

foils. Furthermore, there are the effects of the image of the

lifting vortex which is also present (but which has not been shown)

in the complete flow system near a boundary. Such image influences

the foS1 section characteristics when the distance k is small as

compared to the foil chord.

I-
 AIRFOIL PRINCIPLES

Usually, for aircraft the ratio h/b is sufficiently large so

that these secondary influences are negligible, For hydrofoils,

however, h/b is generally small and these effects must be considered,

Therefore, the treatment of these secondary influences, as well as

other boundary effects on the characteristics of hydrofoils, is given

in Chapter 2,

-_ _ _ _ _ .,.- .^. .--

 AIRFOIL PRINCIPLES

6, Pitching Moment Characteristics

FoSl Section Characteristics

As indicated in equations (1.5) and (1.6) above, the lift

originating in a symmetrical foil section (due to angle of attack)

can be represented as a single force applied approximately at the

quarter-chord point, The lift due to camber, hlowever, is centered

at the half-chord point.

Taking the moments about the quarter-chord point, it is seen that

the moment is only due to the lift developed by camber* Thus, from

equation (1,6)

Gny, = - If 0t-c (126)

and is a constant for any given foil section, f/c being the foil camber

rat;:Lo, and is thus approximately independent of the magnitude of the

lift,

Experimental data on airfoil section&g2 indicate that the value

of the coefficient and the chordwise position are slightly different

f~orn the theoretical values given above. It is therefore usual to

.
tiefine the constant moment coefficient by

=Tlc
which is taken about the aerodynamic center, or a.c,

II - 1.36
 AIRFOIL PRINCIPLES

Effect of Planform

The aerodynamic center and the corresponding moment coefficient

for a complete wing depends on the geometry of the line of aerodyneunfc

centers and the spanwise life distribution,

:For w5ngs employing the same section shape, 'c&c and the aocO
position are the same as for the foil section when the a.c, line is

straight and normal to the direction of advance. (This applies to

rectangular and tapered planforms without sweep, and is independent

of lift distribution,)

For other cases, such as when sweep, twist and variations in

section shape are employed, the effects must be determined as a

function of geometry of the foil and the resultant lift distribution.

Methods for determinlng the spanwise

loading distribution for foils of
*-r
arbitrary plan form are given by.DeYoung and Harper 9 .,

In the partklar case where the taper ratio bears a certain

relation to the angle of sweep as indicated in Figure 1*14, the fo91

loading is ellfptical and the aerodynamic center is given byx

ax. = $ + * 0.342 -0.567A - 0.408 Aa ,q +an A

lo(l+X+?e) (lS7)

where h = taper ratio (tip chord/root chord)

A = sweep angle
from which cm,, can be determined.

II - 1.37

____ .-.. .- - -
 AIRFOIL PRINCIPLES

30 45. 60 734
ANGLE 0F SWEEP, A, DEGREES

PLANFORM PARAMETERS FOR APPIROXIMATE

ELL SPTIC LOADING

FIGURE 1.94

Pitching momek characteristics of wings of various con-

figurations are, however, most readily determineld from wind tunnel

tests. Thus, available test data on wings of various form (such
as presented in references 10 and 11) can be utilized to determine

the pitching moment of similar wings under cons%deration.

I I - 1.38

--____--_ ---
 AIRFOIL PRINCIPLES

Small Aspect Ratio Wings

In his treatment of lifting surfaces, Lawrence15 also derives

values of the aerodynamic center for rectangular and triangular

wings of small aspect ratio. (Triangular wings were assumedto have

the 'apex leading and the base trailing.) The va:Lues given in Figure

1.15 are for the linear a

portion of lift (the
first term of equation -
1.21), measured from the
leading edge. The non-

linear lift (second term

.-
of equation 1.21) is
assumed to be uniformly
applied over the planform,
so that for the rectangular ASPECT RATIO

RERODYNAI4lIC CENTER
wing its center is at mid-
FOR SMALL ASPECT RATIO FOILS
chord, while for the
FIGURE I.15
triangular wing it is at

the 2!/3 chord position.

The effect of camber on the moment of small aspect ratio wings

has not been investigated specifically. It is aasumed that the'lift

due to camber is,applied at the mid-chord as for wings of larger

aspect ratio.

I I - 1.39
 AIRFOIL PRINCIPLES

7. Airfoil Drag

There are three basic types of drag (or resistance) that are
experienced by an airfoil, which are

(4 the induced drag resulting from lift, as discussed above.

w the friction drag, resulting from the shear forces in the

fluid at the body. As a function of the fluid velocity,

these forces are set up tangential to the contour of the

body, and the sum of the components in the direction of

the free fluid flow represent the friction drag.

(4 the pressure drag, resulting from unequal pressures on

the front and rear of the body. In a perfect fluid, the

pressure (which acts normal to the contour of the body)

would have components in the direction of free fluid flow

which would sum up to zero, However, due to viscous

effects, the pressures on the rear of the body are

reduced,.resulting in a net pressure drag. The magnitude

of this drag depends primarily on the shape of the body,

and is thus sometimes called "form drag".

-- --- .__I.I--.
 ATJRFDIL PRINCIPLES

The basic difference between

friction drag and pressure drag
444-+
can be best illustrated by con- 44-+*

sidering the drag on a thin flat FRICTION DRAG

plate, as shown in F'igure 1.16.

When the plate lies parallel to

the direction of flow, the 'drag

is almost entirely frictional;
when normal to the flow,
PRE9SURE D R A G
entirely due to pressure. It

is common procedure to couple

F R I C T I O N A,ND PRESSURE‘DRAb
friction and form drag together ON A FLAT PLATE
-
as "parasite drag", which is
FiGuRE 1.16
defined as that drag independent
of, or not due to, lift.

I?arasite drag is the most importantcomponent of the total drag

on a foil in the high speed ranges of aircraft and hydrofoil craft,

and is indeed an important consideration in all fields of fluid

mechanics. A comprehensive treatment of the subject would be ex-

haustiive and is not within the scope of this work,, Reference should

be made to standard texts, Prandt13 for a presentation applicable to

airfojils and to Hoerner7 for a detailed treatment of the parasite
drag of various aircraft components.

-s_ “-_ -... -. .----_.- _. . . _

 AIRFCXCL PRINCIPLES

Since the standard naval architectural texts explain the concept

of friction drag (frictional resistance of ships) and pressure drag

(eddy making resistance), it is not considered necessary to dwell on

the general aspects of this subject at this time. Particular appli-

cation to hydrofoils is given in detail in following chapters.

It should be pointed out, however, that there

are slight differences in the procedure of
analyzing parasite drag as employed in aircraft
and ship design. Airfoil drag coefficients are
referred to the plan area of the wing, and other
aircraft components (nacelles, fuselages, etc.)
are referred to on the frontal or cross-sectional
area; whereas the standard ship practice is to use
the total wetted area of all submerged components.
Furthermore, the marine industry has adopted the
"Schoenherr line" for turbulent friction, whereas
in aeronautics no standard has been adopted
although the "Prandtl-Schlichting line" is in
general favor.

The procedure employed herein is to use the reference areas

employed in aerodynamics (except for hulls, where marine methods are

retained) but basing the friction drag coefficients on Schoenherr's

function,

.--
 AIRFOIL PRINCIPLE3

8. Airfoil Data

The aircraft industry has published a wealth of informationlon

airfoi.1 characteristics.* The evolution of the characteristics of

hydrofoils, as presented herein, has been accomplished by drawing

heavily on this information, utilizing airfoil theory with such

corrections as have been indicated by experimenta data. This

material has been applied directly or modified as required for

surface effects, as can be seen in the chapters that follow.

Some of the data is of basic nature and is published in readily

usable form so that it can be applied directly to the hydrofoil.

Thus, NACA has published correlated section data Ion a wide variety

of airfoil shapes1r2j16, from which the basic section characteristics

can be readily ascertained. Figure 1.17 shows a typical set of data

available for one of the NACA series.

9 See Yndex of NACA Technical Publications"

National Advisory Committee for Aeronautics,
Washington, D. C,

--
 AIRFDIL PRINCIPLES

.I

4 0 0
0

-.I

- 1.0 - .2

I I I I I I 1 I
- 29I, -10 0 IO 20
ihCtlON ANGLE OC ATTACU Cl
a(g,DLG
R6YNOCOS NO.* CKlO’

SECTtOY N A C A 63,- 2 1 2

TYPICAL SECTION CHARACTERISTICS

From this particular data, the following two-dimensional character-

istics can be found:

section lift coefficient

section lift curve slope

section drag coefficient (smooth and rough-various
Reynolds numbers)
. section moment coefficient (about the l/b chord and
about the a.~.)
angle of zero lift

maximum lift coefficient

.. .
 AIF@'UIL PRINCIPLES

Data is also available on the characteristic:s of miscellaneous

shapes and bodies, flaps, etc. as may be required for a detailed

analysis of hydrofoils but which are too varied to allow individual

treatment in the text that follows.

_-
 AIRFC3IL PRINCIPLES

References

:I, Abbott, von Doenhoff "Summary of Airfoil Data"

and Stivers NACA Report 824, 1945.

2. Abbott and "Theory of Wing Sections"

von Doenhoff McGraw-Hill, 1.949.

3. Prandtl and "Applied Hydrcl and Aero-

Tietjens mechanics"
McGraw-Hill, 1.934.

4. Glauert "Aerofoil and Airscrew Theory"

MacMillan, 1944.

1;. Durand "Aerodynamic Theory" (6 Volumes)

California Institute of Technology,
1943.
6. Hueber "Properties of Tapered Wings"
Zeitschr, Flugt, Motorluftsch,
1933.
7. Hoerner "Aerodynamic Drag"
Published by Author, 1951.

8. Toll and Queijo "Approximate Relations ,and Charts

for Low-Speed Stability Derivatives
of Swept Wings", NACA Technical
Note l!%l, 1948.

0
.e DeYoung and Harper "Theoretical Symmetrical Span Load-
ing at Subsonic Speeds for Wings
Having Arbitrary Plan Form"
NACA Report 921, 1948.

10. Hieser and Whitcomb, "Investigation of the Effects of a

Nacelle on the Aerodynamic Char-
acteristics of a Swept Wing and the
Effects of Sweep on a Wing Alone"
NACA Technical Note 1709, 1948.
 AIRFUIL PRINCIPLES

11.. Purser and Spearman 'Wind-Tunnel Tests at Low Speeds

of Swept and Yawed Wings Having
Various Plan Formsrl
NACA Technical Note 2445, 1951.

12. Purser and Campbell "Experimental Verification of a

Stiplified Vee-Tail Theory and
Analysis of Available Data on
Complete Models with Vee Tails"
NACA Report 823, 1945.

1%. Weinig "Lift and Drag of Wings with

Small Span"
NACA Technical Memo 1151, 1947.

IL. Weissinger "The Lift Distribution of Swept-

Back Wings"
NACA Technical Memo 1120, 1947.

15. Lawrence "The Lift Distribution on Low

Aspect Ratio Wings at Subsonic
Speeds"
-.
Journal of Aeronautical Sciences,
October 19510

16. Jacobs, Ward ?Zharacteristics of 78 Related Air-

and Pinkerton foil Sections from Tests in the
Variable-Density Wind Tunnel"
NACA Report 460, 1935.

_-
 CHAPTER
- 2. SUBMERGED FOIL CHARACTERISTICS

1. General Consfderationa

2, Hydro.fofl Theory

3. Bask Foil Characteristics

40 Effects of Struts9 End Plates and Nacelles

50 Effects of Planform, Sweep and Dfhedral

6. Foils of Small Aspect Ratio

70 Pftching Moment Characterfstics

8, Summary of Submerged Foil Characterfstics

Formulas are given for the lift produced by a fully submerged

hydrofoil and for the effects due to lfft, such as fnduced angle and

induced drag, wave drag, and pfkhing moment0 The effects of foil

submergence and configuration are shown and examples given illustrating

the uaev of the formulas. The prrneipal formulas are summarized for

convenient use and reference.

 The hydrodynamic characteristics of hydrofo-fls operating fn water
are closePy related to those of airfoil.3 operating fn air. When the

hydrofoil is SO deeply .submerged a3 to be considered operatfng fn an

infinit;e fluid, it is dlrecLly equivalent to the afrfoik, taking into

account the dffference fn density between water and air (and assuming

that spe,edT are below those at which cavfta$lfon in water and compress-

ikdlfty in air would be lnfl,uent,4a~‘>.

The basic hydrofoil tiheory, Lhex~forei, takes fnto account all,

surface effeet>s on a srmple, fui1y submerged foil. From the relation-

ships estabLished from the theory, t,,be I"airToil" effect3 and the "wave"

effects can he separated and independently considered for general

engineering appEcaFioa.
 Csrreetions to the basic hydrofoil. characteristics for varfatEons

in foil geometry, the effects of struts and other appendages, and the

fnfl.uence of various operational factors can be readfly and clearly

fndkated as fumctfons of the "*airfoflV" and "wave" effects considered

separately,

The hydrodynamic characteristf@a of the fully submerged foil are

treated in this chapter along these Pines; surface-piercing foils are

treated in Chapter 3. The 9-Jl.uence of cavitation has not been con-

sfdered in the derlevation of these foil chaPa@terfstfcs. The con-

dft5ons under Whfch cavitat,Eon will occur, and the consequent effect

on the foil properties are treated fn Chapter 12,

 SUBMEFGED FOILS

2. Hydrofoil
- - Theory

Two-Dimensional
-~ Theory

The concept of a rectilfneas vortex advancing in a fluid has been

used %n the development of airfoil theory, as fndfcated in Chapter 1,

It can be similarly used to develop hydrofoil theory, wfth the added

condition that the vortex %s located below a free water surface.

This case has been Snvestlgated by Kotchin.12 with the results that
the lift is modified by the free surface and a drag on the vortex is

predueed (in association with a surface wave), The relationships are

given as8

(2.1)

(2.2)

where L is the lift

Qw is the ""wave drag"

b f the span (considered to be infkite

in the two-dimensfonab case)

k is the depth of submergence

v 5s the fluid VelG@ity

@ fs the value of t;he c%rculatSon

B is the fluid density

is the acceleratfon‘o.f gratity

9

- ---....__---- __- .-.

 SUBMERGED
-. FOIL!3

and

Equation (2.1) indicates that the lift of the vortex in infinite

fluid (the first term of the'equation, as shown in Chapter 1) is

modified by an image system due to the proximity of the free surface.

This image system changes the magnitude and direction of the fluid

flow, resulting in a change in lift, indicated in equation (2.1) and

a drag of the l'inducedt' type, equation (2.2). It can be seen from the

equations that both the drag and the change in lift reduce to zero as
the submergence approaches infinity, thus reducing to the case for

the foil in infinite fluid.

Values of the factor CL are given in Figure 2.1 as a function

 SUBMERGED FOILS

of the Froude number, w4ii& O It is seen that in the limiting

cases

xl = - I.0 cd F,s,

J-l = 1.0 a& pI f m

and thus, from equation (Zdl), the image system is- seen to be that
of a"mirror" image at FbsO and a "biplane" image at F, o ~0 9 as

indicated in Figure 2.2. In these

cases p the image sfproduceslt a

change in fluid velocity past the

vortex, but no change in fluid

direction. Thus, the wave drag

must be zero and is so indicated

in evaluating equation (2.2) at

VORTEX SYSTEMS AT
these limiting Froude numbers. k8MlTlNG FROUQE N U M B E R S

(At t= 46 , only when the sub-

FBGURE 2.2
mergence is not zero.)

For practical applications of hydrofoil craft, the speed is

high and the foil submergence is relatively low so that the Froude

number9 'Fh e V/F 0 is high, Insofar as the effect on fluid

Yelocity is concerned, the value fl* Il.0 (from Figure 2,l) may be
generally used without serious error.

Then, equation (2.1) may be written

II - 206

_---- .---- ._“.

 1, = P @ ( \I-k)b (2.3)

where 4Q = 0/4rk p the decrease in fluid velocfty at the

vortex due to its bfplane fmage.

Thus, the equation for the lift of the vortex near the free

surface can be treated 8s that of a vortex in infinite flufd with the

velocity (\I-&) %nstead of v. From Chapter 1, then, the circulation

around the equivalent foil section can be written

where e fs the foil chord

4 fa the angle of attack of the foil in two

dimensfons (measured from the angle of zero
lift for cambered foils),

From equations (2,2), (2.3) and (2,L), the lift and drag can be
derived for the foil, in coefficient forms

W3
 SUBMERGED FOILS

where cD+, = o,/,, w

F,, = VA-g-
b’ is given in Ffgure 2,6 as a function of Fh
($7
hZ

Notfng that nc=0/4ah B the value of ( 'FL)' can be deterrnfned

from equatfon (2.%Jg

I
V-AZ 9 (2,7)
lv) l I + ='dv&l J

At normally small angles of attack, the effect on the flufd velocfty

is negligible for submergences of over 1 chord, and

when b/c >I,0

As the submergence decreases, however, the effect on lift becomes

fnportant. Furthermore, tith a fofl, of f%n%te chord, the Influence of

the vortex image must be considered over the entire chord, instead of

only at the liftfng line as heretofore assumed,

Fini%e Chord CorrectSons

When the submergence is small, the curvature of the flufd flow due

to the
. vortex image appreciably affects the lift charaeterfstfcs of the

foil. seetfon. Thfs can be treated by a simple approximate method

familiar in aerodynamics2, which %s to consfder the flow components at

 SUBMERGED FOILS

a OVcontrollf point l/2 chord

behind the location of the

lifting vorte$, as indicated

in Figure 2.Jo The downward

velocity component due to

both vortices a% this point

is then

and the reduction in horfaontal velocity

(2.9)

??ow to satisfy the condition that there is no fluid flow through the

foil, the angle of attack must be equal to the angle of fluid flow,

or

which by combining equatfons (2,3) ahd (2,9) fs

(2.10)

to which must be added the PPinduced10 wave angle0 From equatfon (2,6)
 SUBMERGED FOILS

Gc, = (2,ll)

Combining equatfons (2.3) and (2,9), and rioting that for Small

values of b/v, hfgher powers can be disregarded,

(2*12)

and

(2.13)

The total foil angle (measured from the angle of zero lift) and

the total wave drag coefficient can then be found by substitution of

terms:

(2.d-b)

The derivatives dGw&c, a) are simflar in form, except that

the respective factors for the effective fluid velocity are

.x
 SUBMERGED FOILS

(14 ZUCL)
( 1 + 312 UC,)
instead of
( I + UC-L)

Equations (2,14) and (2.15) are the basic equations for the hydro-

foil in two dimensions* and indicate the following effects of the free

surfaces

(a) The wave effects, as a function of submergence Froude number

VkjK and submergence/chord ratio vc l

At fnffnite Froude

number or infinite h/, , the wave effects disappear.

(b) The finfte chord effect, as a functio:n of h/,, At values of

Q>l.a 9 the value of u becomes negligible,

++ Complete- theories for the two-dimensional foil tith f i ite chord

have been advanced by Kotehin3 and Keldysch-Lavrentiev ii but the
effects are not as clearly fllustrated as in the treatment above,
Krieness evaluated the case for large Froude numbers (i.e.
negligible wave effects) and the results agree closely with those
given above.

-
II - 2.11
 SUEMERGED FOILS

The- Foil of Finite Span

-

For the foil of finite span, there is the additional influence of

the traflfng vortex system0 In Chapter 1, the induced effects on a

ffnite ting in the presence of a fluid boundary were shown to be

which were added directly to the airfoil properties in two dimensions,

to give the total ting characteristics,

However, this procedure cannot be used in the case of the hydro-

foil, Both the f%nfte chord and wave effects in two dimensions are

modfified by finite span considerations; and the afrfoil fnduced effects

are modified by the influence of gravity, Therefore, the hydrofoil of

finite span must be investigated as an entity. The concept of separate,

add:Ltfve terms will be useful however in evaluating the characteristics

of the foil, as till later be show%

In the development of the theory of the hydrofoil of finite

span, the vortex line concept is used0 Finite chord effects are not

considered herein, but will be re-introduced 1ate.r.

The hydrodynamks of the vortex line of fFnfte

span in the vicinity
'r
of the free surface (and in infinitely deep water) have been investi-

gated by several authors 6,V. Reference should be made to the original

works for the details and proeedure required to determfne the theoretkal

.,

.__._ - -.---- _ ---.... -- --

 SUBMERGED FCXILS

characteristics of the hydrofoil, sfnce the expressions derived are

rather complicated and do not lend themselves to simple formulation

or evaluation.

Qualitatively, however, the characteristics can be shown to be

composed of three major components, as followsr

(a) The aerodynamic induced effects of a foil in the presence

of a rigid wall, This is identical to the ground effect

discussed in Chapter 1, wherein the induced angle is

modified by the factor, %I *

(b) The lateral wave induced effect, due to the trailing vortex

system (tqhe trailing vortices from the foil tips and their

fmages above the water surface), The induced factor varies

tith Froude number, from the limiting value 0~F=~)to

(~(b-$)~~~~) j where Kb is the biplane factor,

The total effect of the above components added

together results in a function similar ton g
which is given in Ffgure 2,1 above.

(c) The transverse wave induced effect., due to the lifting

vortex system (the lifting line vortex at the fofl and its

fmages above the free surface), This is sfmilar to the two-

dimensional wave effect discussed above, but decreases in

magnitude as the submergence/span ra%fo fncreases.

If - 2.13
 SUBMERGED FOILS

Bres:LfnB evaluated the drag of a foil of aspect ratfo 10 at a submerg-

ence/span rat&o of 0,084, The results for an assumed elliptical. lfft

dfstribution are shown in Figure 2.4, with the effect of each component

befng indicated,

Practical Consideratfons

As indicated above and as shown Bn Figure 2.4, the combined aero-

dynamic and lateral wave effect $s simSlar to the function given for

the two-dimensional, case (compare Ffgure 2,1>; and as in that case, it

may be considered equivalent to the biplane effect at high Froude

numbers0 That is, the induced angle of the foi:L in infinite fluid is

modiffed by the biplane factor, Kb -

However, Figure 2.4 indicates that the difference in transverse

wave drag between the three-dfmensfonal theory and the two-dimensional

theory is significant. Evaluation of the three-dimensional formula is

so a.rduous that it appears more reasonable for 'engineering purposes to

use the two-dimensional formula, reduced by a suitable factor.

Arbl.trarily the factor ($,-I) i s used, which factor has a simple

correlation to that used for the effec't of the trailing vortices0

Thus, the equation for the _I_g_-_o_

induced lift angle is in the presence of

the free surface:

SUBM3-G YD FOILS
- -
 SUBMEXGED FOILS

and the transverse wave lift angle is

c&J =
dCL

where -ki fs the indueed lift angle in infinite fluid

fdG ) go
is the wave lift angle in two-dtiensional flow.

From this, the following concept can be stated:

The fnduced angle of a fofl of finite span in infinite

fluid is increased by two additfve terms as the foil

approaches the free water surface, One of these is due

to the effect of the surface on the trailing vortices,

the other on the lifting vortex, Each of these terms

is modified in the same degree by a common factor

that depends on the submergence-span ratio of the foil.

The us'e of thfs factor (cb-\) for the transverse wave effects may

have little theoretical justfffcatfon, but is considered to be of the

proper magnitude, Applying this factor to the ease shown in Figure

2.4, the total drag is seen to agree reasonably with that derived by

the complete theory,

 SUBMERGED FCXCLS

30 Basic Foil Characteristfes

As indicated in the preceding section, the h,ydrodynamic charactsr-

istics of the hydrofoil are approximately given as the sum of two

distinct components:

(a) The aerodynamic characteristics of a fo:il operating near

a free fluid boundary. These are identiffed as the basic

airfoil effects, denoted by the subscript U, in the equations.

(b) The additional transverse wave effects (occurring in the

,presence of the water surface. These are identified as the

wave effects, denoted by the subscript W in the equations.

The basic airfoil effects are given in Chapter 1. For the hydro-

foil, however, the finite chord and reduced fluid velocity corrections

must be added to account for small foil submergence. The correction

terms involving &!, (introduced fn the preceding sectfon for the two-

simensional case) must then be added. For finite foils, the values

of 11 have been calculated along the span, and thle average value

determined. Thus, the airfoil terms can be shown to be:

(2.16)

(2.17)

I I - 2.17
 SIIBNERGl?D FOILS

where u is the factor denoting change in fluid flow, and

is given in Figure 2.5
E is the lifting surface correction, I+ '/Aa

p, J are the respective planform correction terms,

given in Chapter 1
U is the biplane factor, generalized to account for
various foil configurations, as later indicated.
For the basic foil, the factor is tib given in
Figure 2.7.

The wave effects are readily determined from the hydrofoil

theory given in the preceding section:

(2018)

(2.19)

I I I -T---I

II - 2.18
SUBIWKXD FOILS

-.-- t . q--j.y-.-.I

.,.,....izLLA
0.2 0.4 0.6 0.6 9.0

II - 2.19
 SUBMERGED FOILS

Shallow Water Corrections

Operations in shallow water affect both the airfoil and wave in-

duced characteristics. For the airfoil components, the induced effects

are reduced by the usual ground effect, but this involves such extremely

shallow water (depth from foil to bottom of l/2 foil span or less) that

it is not a practical operating condition and can be ignored,

However, the wave induced components are influenced in moderately

deep water by the presence of the bottom, depending on the speed at

which the foil fs travellfng. Thus, when the depth of water is less

than v'/s ( or the craft speed is above the critical value,$=q )

the transverse wave effects vanish. In the subcritical range, the wave

effects are modiffed to some extent which has not been determined.*

Tentatfvely, the following factor is proposed for correcting the

wave drag in shallow water operations:

I- (8)' z I- v& Jar iJ< Jyz

Y) =
(2.20)
9 z 0 4ov vsjz

by which the wave induced effects, equations (2~8) and (2.19) should
be multiplied.

* A two-dimensional theory of wave drag in shallow water has been

advanced by MeyerP, but shows lfttle correlation with model tests
of finite span foils. The test resultslO, are also obscure in
that, the actual shallow water effect cannot be determined, because
there are no experimental results on the wave drag in deep water
to afford comparison0

II -
 SuBplERGED FOILS

Summary of Basic Foil Characterfsties

The total effects of the fofl are then determined by adding the

two components, thust

($ = (Sk +&L (2.21)

(2.22)

where for the stiple, horizontal foil the airfoil effects are given in

equations (2.16) and (2.17) and the wave effects in equations (2.18)

and (2.19). In the following sections of this chapter, the effects of

.-
struts and other appurtenanoes, changes fn foil geometry, etc, are

considered as they affect the afrfoil characterfstics of the simple

foil. The wave fnduced effects are not considered changed from that
given above (except where speciffcally noted), The total character-

istics would still be expressed as the sum of the airfoil and wave

effects, but wfth modified values as required.

 SUBMERGED FOILS

DESIGN EXAMPLE NO. 2.1

DETERMINE THE CHARACTERISTICS OF A
FULLY SUBMERGED FOIL
THE FOIL IS RECTANGULAR, WITH A SPAN OF 20FT.
AND A CHORD OF 2 FT., OPERATING’ AT A SUBMERGENCE
O F 3 F T .

DETERMINE THE EFFECTS AT SPEEDS OF 30,35 d 40KNOTS

AT AN AVERAGE WATER DEPTH OF ‘2DO FT.

PARTICULARS A = IO
VC = 1.5
h/b = 0.15
v = 50.7 59.2 67.6 (f't/sec)
Fh= v/gh : 5.16 6.02 6.89
V2/yd = 040 0.54 0.71

AlRFOlt EFFECTS
E = I.02
U = 0.04 (Flyure 2.5)
K~Kb437(Ftgure 2.7) .
*6' :: 0":0':8;} (Flgure I.8 of chapter I)

Frolm Egwoh(2.16):
I.02 0 04 l.228xl.3-l
&i, =( l+o.o8q-~:+ *+
IO 9 I
=0.219(1+0.08CL,I
From Equation (2~1)
dcoa (j+0.06 c‘,(qp
d) =
= 0.047(It0.06~.)
 SUBMERGED FOILS

DESIGN EXAMPLE NO. 2. t

W A V E ‘EFFECTS
From Figure 2 . 6
e - i/Fh’
--- 0.017, 0.013,0.0\05
2 Fh’
bib-1 = 0.37 I

Fvom Equatlons(2.18),(2.19), te14c2.20)

0dA
-
& w=ho*08CL) 0.77x
i.5 0.017( I-0.40) =0.0025( I tO.08 Cl)
O.O13CI-0.!54)=0.0015(l+0.08 CL)
0.0105(1-0.71)=0.00075(I+0.O$CL)
dCW
= 0 . 6 0 2 5 (HO.06 CL)
d k2)
-. 0.0015 (1+0.06 C L )
o.ooo75( I+ 0.06 C,)

T’TAL EFFECTS
Assure a destgn CL= 0 . 5 0 a t 3 0 K n o t s

TOTAL

V
KNOTS 5 CL94 Mw $3) $1 $& f&j

30 0.50 0.228 0.003 0.049 0.003 0.231 0.052

3'; 0.368 0.225 0.002 0.048 0002 0.227 0.050
40 0.288 0.224 0.001 0.048 0.001 0.225 0.049'

II - 2.23
 SUBMERGED FOILS

.-

4. Effects of Struts, End Plates and Nacelles

The changes in induced effects due to struts, end plates, nacelles

and other similar attachments to the foil can be #considered due to two

separate causes:

First, a favorable effect due to the retarding of spanwise flow.

This is the commonly known "end plate effect" and may be

considered as a redistribution of trailfng vortices along the

surface of the attached member rather than the foil, result-

ing in a reductfon of induced angle and drag. Theoretical

analyses have been made for airfoSls and the results are

readily adapted to hydrofoils,

Second, an adverse effect (generally) which is a function of the

thickness (spanwise extent) of the attachment. This effect

fs generally fdentffied as llinterferenc:ett and is due to the

interruptfon of spanwfse lift distribution and/or due to the

constriction in flow past the body,, Analysis of test data

is the most suitable means for evaluatfng this effect.

End Struts and'End Plates

.Analysis of the biplane, boxplane and end plate effects on airfoils

have been made by several authors, and alPe as given by Durand12 .

Hoerner13 has shown these effects referred to a hydrofoil, and the
..

II - 2024
 SUBMEXGED FOILS

functions are reproduced in Figure 2.7 in terms o:f the factors, K, bY

which the basic airfoil induced effect must be modified, Thuso

Kb is the hydrofoil "biplane" effect, a function of

Kbc is the hydrofoil with end struts, "boxplane*! effect,

a function of

Ue is the end plate effect, a function of

The following relationships are seen to exist, approximately%

tbo = Kb ,+;,b (2.23)

I (2.24)
Kc * 1. + 2 b/b

The end plate effect,&, is derived for a foil in infinite fluid

with a pair of end plates of actual height, he. :Kowever, for the hydro-

foil, the end plate effectiveness is modified by the presence of the

free surface and an effective end plate wefght must be substitued for

the actual height in equation (2,24)e

._______- ___^-._ . ..-- ---- -___I_--. _-_.

 F I G U R E 2.8

From Figure 2.8, considering the upper portion of the end plate

only, the end plate effect approaches the boxplane condition when

the ratio "p/s, approaches 1.0, and the effective end plate height

is then l/2 that of the actual height, from equations (2.23) and

(2.24). For the lower portion of the end plate, the effectfveness

is increased as “/i; approaches zero (due to the influence of its

a
image), so that at k/hA = 0, the effective end plate height is

twice the actual height.*

Thus, depending on submergence, the effective height varies

between l/2 and 1.0 for the upper end plate and between 2.0 and

1.0 for the lower end plate. An empirical formula to determine

s This can be visualized as follows! the biplane image of the trail-

ing vortex reduces the spanwise flow of fluid'on the upper foil
surface (inward) and increases the spanwise flow on the lower
surface (outward), Therefore, the upper end plate is less effect-
ive, the lower end plate more effective than in infinite fluid.

--.-
_---_1_-..
_-.-. -
.-. .-_.- I. -_.. “-____l__(_-_l__.” --,. -..*
 SUBMBRCXD FOILS

the total end plate effectiveness at a submergence, k , is as follows

(2.21;)

Effect of Struts and Plates Inboard of Foil Tips

The effectiveness of struts and end plates is reduced as the

member is moved inward from the foil tips. Manglerlk calculated the
effectiveness of end plates in various spanwise positfons on an air-
foil and the results are shown by Hoerner13 to be a function of the
spanwfse position, @& p and the end plate height, he,& o It can be

shown that a good approximation for any he/b is that the effective
height is reduced in the ratio
t S0
b 1
where QL/e is the distance of the plate from the midspan of the foil.

Assuming this correction to apply for both plates and struts for the
hydrofoil, the total &( factor can then be determined for the hydro-

foil. Thus, for a foil of the configuration shown in Fdgure 2.9:

.-- ---.-. -.-~-_-.___.--

 (2.26)

where is determined from equation (2.25).

II - 2.29
 SUBMEZGFD FOILS
-.--

DESIGN EXAMPLE NO. 2’ii-l-~~f oF1

DETERMINE THE K FACTOR FOR THE
F O L L O W I N G CONFlGURATlON . ’

JOn?L ty FUCTOR

From @id/oh (2.26)

K = /'A93 I i I.183
/+(U.3)30.3 + 2x0.1363 /. 2 8 0

= 0.924

II - 2.30
 SUBMERGED
- FOILS

End Plate Effect of Nacelles

It has been shown by Hoernerls that a nacelle (wing tank) of

cylindrical shape has an effective end plate heig:ht of l/2 the

diameter of the nacelle., When located at the wing tip, however, this

effect is counteracted by the rolling up of the trailing vortices more

readily due to the round tfp, with a consequent reduction in effective

span equal to l/2 the nacelle diameter, Thus, in this case, when con-

sidering the overall span to the outside of the nacelle, there is no

end plate effect,

However, in order to formulate the end plate! effect for all span-
- wise locations of the nacelle, the nominal span may be considered to

be between centers of the tfp nacelles* Then, adding the end plate

effect, the same net relationshfp is derived. As indicated in Figure

2.10, the total end plate effect of nacelles can be estimated by

consideringg

(a) the height fs equal to l/2 the rounded part of the nacelle

plus the full height of any flat sides

(b) the spanwise location is measured to the center of the

nacelle,

With these dimensions, the effect can then be calculated on the basis

of the end plate formulas derived above.

I I - 2.31
 SUBMEKGEXI FOILS

(al CONSIDERING FULL SPAN -NO E ND PLATE EFFECT

(b) CONSIDERING SPAti TO 4 NACELLE-

PLUS END PLATE EFFECT

! a g--l
1
(Cl NACELLE INBOARD OF TIP
[D) TYPICAL END PLATE VALUES

FIGURE 2.10

Effect of Lateral Area on End Plate Effect

-

The above formulations for struts, end plates and nacelles have

been derived on the basis of these members having a chord (longitu-

dinal extent) equal to that of the foil. However, depending on the

total lateral area of the member, and the shape of that area, the

effect may be greater or smaller than indicated above. There are no

definite formulas available to account for this effect, but

tendencies can be noted, as in Figure 2.11. Generally, where these

I I - 2.32
 SUBMERGED FOILS

members are full at the foil and tapering towards the upper and/or
lower ends they are more effective than those (such as struts) that

are full at the ends and taper towards the foil, For rectangular

or near rectangular members, the effective height may be considered

to vary directly as the ratio of the mean chord of the lateral area
to the chord of the fofl.

Interference

The "interferencel' effects arise due to ,the disruption of flow

at the foil caused by the superposition of other bodfes on the fo%l.

Such interferences cause additional viscous effects (as treated in

Chapter 4) and additional induced effects.

 The lift induced effects are due to disruption of spanwise dis-
tribution of lift4 i.e. the superimposed body removes a portion of lift-
producing foil area, causing a reduction in lift and a redistribution of
lift along the remaining span of the foil. Opposing this effect, the
constriction of fluid flow at the sides of the body results in super-
velocities of the fluid and thus a local increase of foil lift at the
intersection.

Test data13 indicates that the overall effect of a foil-strut

intersection ia to reduce the lift (or increase the induced angle and
indumd drag for a given lift). A eimple analytioal expression can be
derived for this effect by amuming the lift to be lost over that
portion of the foil span occupied by the struts, 'ts , and neglecting
the effect of super-velocities, By further assuming the upper foil
surface develops 2/3 of the total lift, it can be shown that the inter-
ference effect on the lift angle slope and the induced drag may be
approximately expressed as

(2.26a)

are the values derived without inter-

ference effects

--
II - 2.31r.
 SUBMEXEJ3D FOILS

is a factor which depends on the strut thickness,

‘3 foil span and spanwise location of the strut, This
is given by

h/b is the maximum strut thickness to foil span ratio

"/b is the strut location ratfo, as given above.

The above relationships are given for each strut on the upper

surface of a foil, where the junction is well-fflleted, and as such

agrees well with experimental data. For other conditions, the value
of 7 given above must be multiplied bys

(a) the number of atruts

tb) l/2 - for lower surface struts

(4 2 - for unfilleted junction8

It should be noted that the above relationship represents

the interference effect of a foil-strut junction as a
function of lift (or angle of attack). It is not known
exactly whether this effect is fnduced or parasitic in
nature; for convenience it is representNed herein as
induced, The interference effect at zero lift is essentially
parasitic and is given in Chapter Lo

The induced interference effect8 of nacelles are more difficult to

evaluate, being sensitive to nacelle size and position. Theoretical

investigation8 12, 16 indfcate that the effect8 of superimposed

nacelles of small extent on the induced characteristics of wings are

small. Test data17 on ting-fuselage intersections indicates that,

,
I I - 2.35

.
an increase in induced drag is accompanied by an increase in lift,

both effects being minimized by appropriate fairings at the junctions.

Therefore, in the absence of specific test data for considered foil-

nacelle configurations, the interference effect of nacelles may be

neglected.
 SUBMERGED FOILS

DESIGN EXAMPLE NO. 2.3

DETEtiM’INE THE K FACTOR FOR THE FOLLOWING
FOIL- STRUT -NACELLE CONFIGURATION. *

BIbLANE FACTOR
From Fogure 2 . 7 ; K,, = 1 . 4 2 ( a t h/b =0.125)

END - PLATE EFFECTNENESS

struts - Jhe strua effect we helg”f 1s 2.5 y $L = I.875

hheelles
Assume an effectrve n a c e l l e icngth df 3 . 5 f’.t. Then
the effectwe heights W-C

h, 0 0.5 (J.5 - I.s)/~ * O.!;O (dcductrng strut e&J)

$ * I.0 y 3.5/2 l I.75

Then, hm~ equahn ( 2 . 2 5 )

R tO.SO. 4S,G$B.5 .* 0.133
ce
b >ePC 20 6.75

TOTAL K FACTOR

FF-0m Equatm (2.26)

pg = 1.42 I = 1.42 *
1 +(o,7)3 1,875 + (0.7Px
2 0
o.tss
I 1.234

= 0.15
 SUBMERGED FOILS

5. Effects of Planform. Sweeo and Dihedral

Effects of Planform

Corrections to the induced angle and induced drag for a wing not

having elliptical lift distribution have been indicated in Chapter 1.

The corrective factors are (I+'f) and (I+&) respectively, f and d

being derived from the methods indicated,

In the presence of a free surface or in an equivalent biplane,

the value of the planform correctfons and their relationship to the

biplane factor are not exactly known., Olauert" proposed that r and 1

d derived for the monoplane be added directly to the biplane factor

to give combined factors (7~ K) and (a+ tic) o It appears more con-

venient to apply each factor fndependently, so that the combined

factors are

u (1+‘r) and K(d)

with little loss in accuracy0 This latter method is used herein, as

indicated in equations (2.16) and (2,1?) above.

In regard to the wave induced effects, it has been shown by

Breslln8 that uniform lift dfstributfon results in less wave effects

than elliptical dfstribution. Dependfng therefore on the relative

magnttude of airfoil effects and wave effects,, the optimum planform

is probably somewhat between the elliptical and rectangular, However,

 SUBMERGED FOILS
.-

the difference is small at high Froude numbers, and in view of the

empirical correction already applied to the wave effects of a finite

foil, equations (2,18) and (2,19), the effect of planform may be dis-

regarded in numerical computations of wave effects,

Effects of Sweep

As shown in Chapter 1, the effect of sweep may be expressed as

(2.27)

where A is the angle of sweep

is the section lift angle, a function of

the liftfng vortex

is the induced lift angle, a function of

the traflfng vortices
k is a factor, determined from specific test data.

It was also shown that k Z '/&A generally, and so in the

absence of specific test data,

(2.28)

,where $ and ~ are the complete airfoil functions

t L1 d(G’) in the absence of sweep0

-,-~-- X._ ..-.. I ^_-_sl___l__l_ --_-

 SUBMEFGED FOILS

For deeply submerged hydrofoils, the above relationships can be

used, substituting the appropriate values of and dGa given
d-1
in equations (2,16) and (2.17) for the airfoil effects. The. wave ."

effects are unchanged, the average submergence of the swept foil being

used in equations (2.18) and (2.19).

However, at small submergence, the above re:Lationships are not ex-

pected to be accurate when large angle of sweep are employed. The
biplane effect derived from the relatively simple lifting line theory
is considered inadequate to show the influence of the swept geometry

and.the variation in submergence (between the midspan and t!.p of the

.-
foil at angles of attack).

The wave induced effects would also be modified 'by these factors, con-
sidering the interference in the waves generated along the span.

It is therefore necessar!t to obtain test data on hydrofoils of

large sweep and low submergences before accurate characteristics can

be determined.

Effects
- - of Dihedral

The effects of dihedral on a fully submerged foil can be estimated

from the two established conditions for which the foil properties are
known. Thus, as indica%ed in Chapter 1, dihedral does not affect the

induced characteristics foI* a foil in infinite fluid; while for a foil

-

. . -.-- -.-.” - “...^..--- _^... --_--“-~ -.-_“-.~-..9..-111- .--- -.-.. - -....--. --.--l___ll-.--
 StJBMEiEEiD FOILS

with the tips touchfng the surface

(surface-piercing foil), the in-

duced effects are equivalent to

those of a foil with end struts

(boxplane condition) at the' sub-

mergence k ' bbJ4 9 as is

shown in Chapter 3 following.

The effects of dihedral for any submergence can then be

estimated from the f&lowing

- (2.29)

where $ is the dihedral angle

'$$ is the tip submergence/foil span ratio

b/b is the average foil. submergence ratio, "yb 9 +aax

Kb is the biplane coefficfent (as may be modified by

other factors given previously) for the average
submergence ratio,,

It fs seen that at lasge values of “Q& 9 Kp,approaches LO.

At kt/$ = 0 (surface-piercing fofP), Kr Es equal to the boxplane

effect gfven in equation (2,23) for a submergence of b Qarwl$ 0

--T
As indicated in Chapter I., the effect of dihedral modifies the

foil section angle by the factor $&S$ ti The total effect of

l

--_
* I’

.
 SUBI4EXSD FOILS

dihedral on the airfoil lift angle can then be written

(2.30)

where ($1 and: ($ ), are the lift angles for the foil without
L h dihedral (at the mean submergence
t,s hk + k+*nP
T-1
d is the foil angle measured at the foil
midspan in the vertical plane.

The wave effects can be estimated by using the submergence

h:b+ by in equations (2.18) and (2.19),

II - 2,42

-- __^- _---____ --._.-.

SUBMERGED FOILS

II - 2.l.o
 SUBI4i%ED
- - - FOILS
- -

6. Foils of Small Aspect Ratio

---_I---_-

For hydrofoils of small aspect ratio operating at a greater

submergence than 1 chord or 1 span (whichever foil dimension is

larger), the airfoil formulas for lift and drag given in Chapter 1

should apply.

At smaller submergences, the biplane images will have an effect

as heretofore shown, with several

important distinctions. As in-

co-
\\
dicated in Figure 2.13, the bound \\ t-l
vortex fmage acts on the foil at \ I

a point c.z~ behind the lifting -

vortex (as suggested by DeYoung

and Harper19). Also, since angles

BOUNO VORBEX WAGE EFFECT
of attack tend to be large for SMALL ASPEC? RATIO f00b

small aspect ratio foils, thfs

must be considered in determfnPng me. and w 9 as shown. Thus, the

image effects are not linear and must be determined for each
indfvidual case,

Generally, the expression for the lift would have the following

form:

.
---_-_
II - 2J.llc
 SUBMETUXD FOILS

where il.) and i(W) must be determined,

Since foils of small aspect ratio are not considered of general

.,
or important application in hydrofoil craft, the above factor'a have

not been evalhated. Tests" have been conducted on foils of A - l/L

and A - 1.0 at various small submergencea and the data can be utiliz;ed

directly in estimating the lift, drag and pitching moment character-

id&8 Of Simihl? fOik3.
 SUBMERGED FOILS

7. Pitching Moment Characteristics

--

For deeply submerged foils, the pitching moment characteristics ..

given in Chapter 1 for airfoils may be used without serious error.

Thus, the pitching moment and the aerodynamic.center can be determined

*from airfoil section characteristics (either theoretical or
experimental) and applied over the foil when the "equivalent wingtl

lift distribution is ascertained,

For foils of shallow submergence (one chord or less), the pitoh-

ing moment characteristics are expected to be affected by the influence

of the vortex images, particularly when'the foil employs large angles

of sweep and dihedral. It is considered necessary to run tank tests

on the configuration when accurate pitching moment characteristics are

required (partic.ularly fn cases where controllable foils are employed,

pivoting around some given axis).

For preliminary purposes and where simple planforms are employed,

the pitching moment characteristics can be estimated from the following

considerations,,

The biplane image of the lifting vortex has been shown to cause

an increase in section lfft angle; or, inversely, it causes a decrease

in section lift at a given angle of attack. This is due to the

curvature of the fluid flow at ihe foil, which may be considered

 SUBMERGED FOILS

equivalent to a reduction in effective foil camber,, Thus, there is

an increment of lift acting downward at the mid-chord position. This

increment of lift can be shown to be

AC, s - d$.$AG
'(2.32)
= L 2"*g,c,
, E 8%
where d& ia the lift slope of the foil in infinite fluid
dd.
~6 is the increase in section lift angle due to the
vortex image

u is given in Figure 2,s.

For foils of large aspect ratio A2 10 , the two-dimeneional value

of Scan be used without serious error (see equation 2.13), and

equation (2.32) reduces to

(2.33)

II - 2.47

-_- --.-_-____ ----_.. --_ ---. . --_- -. I---.-. -.

 SIBMERC~D FOILS

The forces on the foil can

then be represented as shown in

Figure 2,lL emaL and the a.c.

pos%tfo:n in Pnfinete fluid, and

can be found as indicated in

Chapter 1. The total foil pitch-

ing moment characteristics can

then readily be determined.

It is seen that the value of bc, increases with decrease

of submergence, and the center of pressure correspondingly moves

forward towards the leading edge. Available test data2C on the

center of pressure of hydrofofls shows qualitative agreement with
the relationships given above,

-.
 SUBMERGED FOILS

8, Summary
- - - of Submerged Foil Characteristics

Anrrle of Attack

The basic equat-lon for the t%.ft angle" (inverse lift curve

slope) is

from which

(2.35)

is the foil sectfon lift angle, a function of the

lifting vortex and i-ts biplane image
is the airfoil induced lfft angle, a function of
the trafling vortices and their biplane images

is the wave induced lift angle, a functfon of the

gravity image system

*Id is the angle of zero lift, measured in the same

plane as A 0 The section angle of zero lift
must be corrected: dhO z kq $-

1’

II - 2e49
 SUBMFXGED FOILS

Drag Due to Lift

The Xnducedl~ drag is given by

(2.36)

(2.37)

where &ii is the airfoil induced drag (as a function of lift)

d(CLI)
d Cow is the wave induced drag (as a function of lift)
Z@)
Fautors

The various factors in equations (2.34) to (2*37) which have to be

evaluated on the basis of given foil parameters are as followat

u the term denoting a reduction in fluid velocity,

given in Figure 2.5 as a function of h/c

E the lifting surface correction, !* z/Aa

the biplane induced correction factor, given in

Kb Figure 2,7 as a function of h/b

U the generalized biplane induced factor, which ia derived

from the biplane factor corrected for various configuration
effects. Thus

,-

.--___ _... .^...._.-.. ~... --.-.-.--* II_ --.. -._ . . .”

_c SUEMEXGED FOILS
-4

where fi is the strut or end plate factor such

as .in equation (2,26)

A is the sweep factor '/,osA ; equatioq

(2.28) or as derived experimentally

f3 is the dihedral factor &ven in equation (2.29)

other factors as may be required.

submergence Froude number, v/f@

the wave factor given in Fzigure 2,6 as a function of

the planform correction factors, from Chapter '1.

Foils of Small Aspect Ratfo (A < 2.0)

The following must be used for foils of aspect ratio less

than 2.0:

(2.39)

where E is given in Figure 1.10 of Chapter 1.

Equations (2.37) and (2.38) are taken from Chapter 1, and are

applicable for h/b > l.O< h/c . For smaller submergences,

surface effects must be included as fndicated previously in

this chapter.
 .

l 4wgmmme
suBMER.oED FDILS
.-

Pitching Moment Charaoterietics

F o r foilrs of h/c > 1.0, the pitching moment, may be detemnined a8

in Chapter 1 for airfoils. At smaller aubmergencas, a aorrection must

be applied II indic!ated in this chapter,
 SUBMERGED FOILS
cl
c.

References

1. K&chin, Kible & "Theoretical Hydromechanics",

Rose Chapter VIII, Article 19
(Translated from Russian by Prof.
Kortin-Kroukovsky, Stevens
Institute of Technology,
Hoboken, N. J.).

2. Wieghardt Yhordwise Load Distribution of a

Simple Rectangular Wing", NACA
TM963, 1940,
3. Kotchin "On the Wave-Making Resistance and
Lift of Bodies Submerged in '4atern,
SNAME TechnJ_cal and Research
Bulletin Nos. l-89 19slo

4. Keldysch & Lavrentiev stOn the Motion of an Airfoil Under

the Surface of a Heavy Fluid, i*e.
A L1qu-S.d" (Translation No, 306, by
F, X, Prior,, Science Translation
Service, Cambridge, Mass.).

5. Krienes As reported in Critical Review of

%BR--SF Hydrodynamdsches Archiv" by
S, F. Hoerner, Jan. 1953* Enclosure
(B) to Gibbs & Cox, Inc. Letter
(Conffdential), 13531/A9(1-1180)
dated 23 Ja:nuary 1953 to ONR
(R1-23-53 - 4006).

6. Meyer VITh.ree-Dfmensfonal Vortex-Line

Theory of a Hydrofoil Operating in
Water of Large Depth", Hydrofoil
Corp. Confidential Technical Rpt,
No, HR-k9 February 1951,

7. wu pYA Theory for Hydrofoils of Finite

Span", Hydromechanfcs Lab,, Calif.
Institute of Tech., May 1953,
Rpt. 26-8,

.__ _.-_ ---

 SUBMERGED FOILS
-

8. Brealin '*A Linearized Theory for the Hydro-

foil of Finite Span in a 'lluid of
Infinite Depth", Bath Iron Works
Corp. by Gibbs & Cox, Inc. Confi-
dential Techniical Rpt. No. 16,
January 195!1~

9. Meyer "Two-Dimensional Vortex Line Theory

of a HydrofoKL Operating in Water
of Finite Depth", Hydrofoil Corp.
Confidential Rpt. HR-1, Nova 1950,

10. ‘Wadlin, Shuford & "A TheoreticaIL and Experimental

McGehee Investigation of the Lift and Drag
Characteristics of a Hydrofoil at
Subcritical and Supercritical Speeda",
NACA RML%?D23a, July 1952.

11. Bath Iron Works Corp. by Gibbs & Cox,

Inc. Confidential Techniaal Rpt. No.
2, "Detailed Analysis of Hydrofoil
Model Data", Contract No. Nonr-
.- 507(00), October 1951.

12. Durand "Aerodynamic Theory", Vol, II,

California Institute of Technology,
1943.

13. Hoerner "The Influence of End Flates, Struts

and Nacelles IJpon the Characteristics
of Hydrofoils"', Bath Iron Works Carp,
by Gibbs Rc Cox, Inc. Confidential
Technical Rpt. No, 14, Contract No.
Nonr-507(00), August 1953.

14. Mangler "Lift Distribution of Wings with

End Plates", 3Gufo 1939.

15. Hoerner "Aerodynamic Drag", (Published by

the Author), :L9511

16. Vladea "'Effect of Fuselage and Engine

Nacelles on Some Aerodynamic
Properties of an Airplane Wing",
NACA 'I'ec'Ti!-i:,Ll Memo No. 736.

II - 2S4
 SUBMHRCED FOILS

17. Jacobs &.Ward r'Interference of Wing and Fuselage

from Tests of 209 Combinations in
the N,A,C.A. Variable-Density
Tunnel" p NACA Rpt. No. 540, 19350

18. Olauert t'Aerofoi1 and Airecrew Theory",

MacMillan', 1944*

19s DsYmng Q Harper "Theoretical Symmetrical Span

Loading at Subsonic Speeds for
Wings Having Arbitrary Plan Form",
NACA Rpt, No; 921, 1948.
20, Wadlin, Ramaen C "The Hydrodynamic Characteristios
Vaughan of Modified Rectangular Flat Plates
Having Aspect Ratios of 1,CO and
0,25 and Operating Near a Free
Water Surfacen, NACA Technical Note
3079, 1954d

I.
 ,
CHAPTER 3. SURFACE-PIERCING FOIL CHARACTERISTICS

1, General Considerations

2, Char+acteristfcs Prior to Ventilation

3* Ventilated Characterfstics
ho Design Considerations

Formulas are given for the lfft and drag characterfstfcs of

surface-piercing foils in non-ventilated and ventilated condftfons.

These formulas are based on a limfted series of tests and are

con&dered primarfly applicable only to foils hating sMlar

geometry and operating under sfnflar conditions as those tested.

 SURFACE-PIERCING FOILS

1. General Considerations

Foil
- - Geometry and Notation

The characteristics of a surface-piercing foil are referred to

the projected dimensions of the submerged foil area as indicated in

Figure 3.1, and all lift and drag formulas can be referred to these

dimensions, after appropriate corrections.

.I PROJECTED SPAN - b

SURFACE

FIGURE 3.1

The specific notation for surface-piercing foils is as followst

b the projected span of the submerged foil (the
horizontal distance between foil tips)

C the foil chord

* r. the dihedral angle

L angle of attack, measured in the vertical plane

f;/c camber ratio, measured on the actual foil section

(in the plane normal to the quarter-chord line)

A the projected aspect ratio (b/c for a rectangular foil)

II - 3.2

.- .-- "vl_"..--- -. -.
 SURFACE-FTERCIRG FOILS
~.--- _-_-----.-_-___

s, the projected foil area (bc for a rectangular foil)

s/s the ratio of actual foil area to projected.foil area.

P

Status of Existing Information

.----

Available information on the hydrodynamic characteristics of

surface-piercing foils is limited to preliminary theoretical in-

-
vestigations' and several sets of tests on specific configura-
tions2J3,4. Neither the theory nor the test data is sufficient to '
allow accurate prediction of the hydrodynamic characteristics cover-

ing the wide range of configurations that employ surface-piercing

foil elements. There are many varieties of shape as shown in Figure

3.2 and variation in planform, camber, twist, etc. that are con-

sidered to have noticeable effect on the overall foil characteristics.

TRAPEZOIDAL L A D D E R TYPE

TYPICAL SURFACE - PIERCING FOILS

FIGURE 3.2
 SURFACE-PIERCINQ FUILS

Sottorfqs Fxperiments

Sottorf2 ran a series of tests on various surface-piercfng

V-foils and trapezoidal foils of 26.50 and 45" dihedral angle9 tith
a varfety of aecrtional shapes (constant along the span) and sub-
mergence over a range of angle of attack. Reference should be made
to the original report for the complete results of these testsj
however, certain representative data are reproduced herein in
Figures 3.3 and 3.4 to illustrate the adequacy of the formulas
proposed below.

Application of Data

The data and formulas proposed are conside:red stri.ctly.to apply

to those types of foils tested, although they may be applied genemlly
in the absence of more specific information for other types,

At this stage, it is recommended that tank tests

be run on particular surface-piercing foils con-
sidered for use, in order to determine more
accurate characteristics over the range of speeds
and angles of attack described.

.- -.“---..>. -l-Is_--- __-_.

 SURFACE-PIERCING FOILLS
__- - -

0.6
/-- --i

0.2

-5” 0 5"

0.9

8.4

0.3

0.2

0.1
SURFACE-?IERCING FOILS

..

-lll!m
 SURFACE-PIERCING FOILS
-

2, Characteristics Prior to Ventilation

Airfoil Effects

The airfoil induced characteristics of a surface-piercing foil

may be given approximately as .

those of an equivalent submerged

foil with end &ruts (boxplane

condition), as indicated in

Figure 3,Se The equivalent

"boxplane" has the same pro-

jected span and encloses the

same water area within its

perimeter as the foil. it re-

places, Thus, the boxplane

submergence is the average

submergence of the surface-piercing foil;for example, it is P/2 the

maximum submergence of a V-foil and 2/j that of ,a parabolic foil,

The airfoil effects of the surface-piercing foil may then be

given, from the equations in Chapter 2 for the s,ubmerged foflg

 SURFACE-PIERCING FOILS

where k Is the average submergence of the foil

cbO is the boxplane factor, based on the average
submergence, Values are given in Figure 2.7
s/sp is the foil area ratio (equivalent to g/ccsT
for a V-foil, where p $8 the dihedral anglej
other notation as given above and in Chapter 2,

Wave Effeats

The wave effects for a surface-pieraing foil may be estimated to

be the same aa that for the submerged foil at the squivalent average
aubmergencs.

(363)
3

where, h is the average submergence of the foil

other notation as given above and in Chapter Ze

Effeate of Sweep

The effects of sweep on the a%rfoil or wave effects of a surfaae-

pieraing foil are not definitely know% There are no known reports in
the literature dealing with the theory or tes;t results of surface-
piercing foils employing .sweepe Specific tea;ts would have to be made,

at the present time, to determine such effects.

II - 3.8
1 SURFACE-PIERCING FOILS

Comnarison with Test Data

Comparison is made between the test data on V-foils, given in

Figures 3e3 and 3*4, and the formulas es derived above. It is nckedthat

the test speeds are very high (v/5e z 9,0), and therefore the wave

effects, equation (3.2), may be neglected.

For the cambered foil tested, the angle of zero lift may be

simply derivedg
@%i =

The lift curve was calculated from equation (3.1) for the foil
tested, and the results are shown in Figure 3.3,, Agreement with the

test data is seen to be reasonably good.

However9 analysis of the drag data given in Figure job indicates

that equation (3,2) holds only at an "optimumtl ILift coefficient, Above

and below this value, the drag is noticeably higher.

The theoretical curves shown in the figure are derived

by considering the section drag coeff:lcient in the order
of 0.01 (at the test Reynolds number (of 6 x Id) whfeh
leads to the total expression
0101
c, = COIO
+

It is seen that this is valid only at an optimum lfft

coefficient,
‘- SURFACE-PIERCING FOILS

From the data shown in the ffgure, it is seen that above the optimum
point, the slope of the drag curve is approximately twice the theoretical

value, while below the optimum the slepe is roughly one-half the

theoretical value. What signffciance this relationship has, and over

what range of configurations it can be applied is not presently known,

Qualitatgvely, however, this increase in drag appears to be due to two

considerations:

(1) The interference effect at the surface which causes

spray and replaces circulatory flow at the tips by

less efficient wplaning" flow.

(2) The sectfon pressure drag (see Chapters 1 and b)b

For airfoil shapes, this drag increases according

to the factor (.I + AC? ), where Aq fis measured

from the optimum point. However, for the sharp

nosed, circular arc sections at the low Reynolds

number of 6 x 16 employed in the tests it is

considered that the pressure drag is much higher

than this value.

The relative magnitude of these factors could be determfned by

further test data.

 - S&FACE-PIERCING FOILS

DESIGN EXAMPLE NO. :r

DETERMINE THE LIFT CURVE FOR THE FOLLOWING
SURFACE - PIERCING FOIL (PRIOR TO VENTILATIONi
CHARACTERlStlCS ’ P = 30’
A85
RECTANGULAR PLANFORM
SECTION ’ IO % THICK, ARCFORM WITH
FLAT PRESSURE SIOE

-.

- ..-. _.--. --~-. ---..__---.” --

 SURFACE-PIERCING F"Ig

3. Ventilated Characteristics

Inception of Ventfbation

The phenomenon of ventflatfon occurs when cafr from the atmosphere

displaces the low pressure water on the upper surface of the foil.

Some aeeumulatfon of boundary layer fluid at the water surface is

necessary to P1trfggerBl such afr entrance, which condition fs most

readily met at large angles of attack,

From a study of the existing test data fop surface-piercing

foils (and for vertfcal struts as shown in Chapter 7) the fnoeptfon

of ventflatfon appears to be a local condition, depending on the foil

geometry at the water surface, An empfPfca1 expression has been found

that agrees well wfth the data of Ffgure 3,2, 3:n terms of the foil.

angle of attack,d B the fneeptfon pofnt ls

where is the angle between the chord line and the upper
B surface of the fofl sectfon at the leadfng edge

r is the dihedral angle at the water surface,

Equation (job) can only be applfed when the speed

of advance fs large enough for: the foil to normally
develop sub-atmospheric pressure on the suction
s%de over the entfre span,
 SURFACE-PIERCING FOILS

The angle, 9 is easily

s
determined for a foil with

sharp leading edge, but has no

exact definition for an airfoil

section, It appears reasonable
to use the angle formed by a

circular arc tangent to the

section at the point cl" rn<k-imum

thickness and passing through

the leading edge, as indicated

in. Figure 306,

When the foil angle exceeds that given in eqM4.sn 1(3~&)

above, the lift of the foil falls off more or less rapidly until
the fully ventilated condition is reached, as indicated in l?igure

3*3 (and as further indicated in references & and sJO

It is not apparent why, in some instances, the

lift falls off immediately after.& is exceeded,
whereas in others the lift falls off gradually
with angle of attack. In the fatter cases, there
is some evidence of instability with the lift
alternatively assuming non-ventilated and
ventilated values for different test runs under
seemingly identical conditions. Further tests
and analyses are required to resolve this
phenomenon,

. __I.. -1.-
 e S&FACE-PIERCING FOILS

Ll.ft and Drag Characteristics - Fully Ventilated

The lift and drag of a

fully ventilated foil (with

flat pressure side) can be

readily determined by the

use of Rayleigh's formula6

for the lift of a flat plate

in discontinuous flow. This

formula is expressed for the

normal force on the plate as

c, = 2 II- Swb4.
4 + ‘lr Slvtcc, (3.5)

which strictly applies only to a two-dimensional flat plate,,

However, since for a foil with both ends piercing the surface
the actual flow across the tips is limited (in a spanwise direction),

it is expected that this formula should approximately hold,

Therefore, referring this force to the case of the surface-

piercing V-foil, we get for small angles

CLIJ = zfloc cosr

4 +-II-oc cesr

which is in excellent agreement with the data shown in Figure 3.3*

The angle of zero lift is dependent only on the shape of the lower

--_-.,-~ --.“_.-- .._.

 SURFACE-PIERCING FOILS

surface of the foil. In particular, it is apparently that angle at

which the trailfng edge of the pressure side is in line wfth the flow.

The drag of a fully ventflated foil having a flat pressure side

is evidently a simple function of the angle of attack. Including

one-half of the conventional friction profile drag (since only the

lower surface of the foil is wetted) the total drag coefficfent for

a V-foil is then

(307)

where Ck is the basic section drag coefficient

(see Chapter 4)

cl,, is the lift coefficient of equation (3*6),

Note2 Wave drag considerations have not been included in the

fully ventflated condition. In practical applications,

such ventilation is expected to occur only at high

speeds where the wave drag is negligible. Where eon-

sfdered to be a factor, the wave effects may be

calculated in accordance with equatiion (3*3) gfven

for the non-ventilated condition.,

--. . .-.. --._“-~--- “.-,-I.. - . *...I- ---___ ---- . ..-.-.. -.---_l_- -.-_ ---.-- .-_-
 SURFACE-PIFsRCIK FOILS
-

DESIGN EXAMPLE NO. 3.2

) SH IOF 1
ESTIMATE THE VENTILATING CHARACTERISTICS
OF THE FOLLOWING SURFACE-PIERCING FOli

P = 3o”
A =5
AS IN EXAMPLE 3, I
10 % THICK, AR&ORM SECTION
WITH FLAT PRESSURE SIDE 1 ’

Inception o f Ventllafron

For the se&Ion employed, 6 l 0.385 radlrns (22Y’oppx)

F r o m Equcdlon (3.41

D(” : 0.385 X0.577/= : O.Ik7 radians

= 9.0"

Lift

F ram Equation (3,6)

2huG cos 3 0 ” : 3.624,

c,” =
4+n N cc3 3ob 4 +1.81~
 .

SURFACE-PIERCING FOILS
-_

4. Design Considerations

The various correction factors for struts,: planform, etc. can be

applied to the airfoil effects of a surface-piercing foil in a manner

similar to that for a fully submerged foil, the principles of which

are given in Chapter 2.

However, it is necessary to re-emphasize that the formulas given

herein for the surface-piercing foil are tentatlive, beiLng based on a

minimum of test data and theoretical investigat:lon, and should be

checked by tank tests where the configuration is different from those

on which the formulas are based.

-
 SURFACE-PIERCING FOILS
.-

References

1.3 Weinig "On the Theory of Hydrofoils and

Planing Surfaceslt NACA - T.M,
#845, 1938,
.
2. Sottorf "Experimental Investigation of
Hydrofoils" German ZWB Document
FB 1319 (1940) by Inst, Seeflug-
wesen Hamburg

3* Hoerner Critical Review of %BR==SF Hydro-

dynamisches Archiv", Jan, 19530
Enclosure (B) to Gibbs & Cox, Inc.
Letter (Confidential), 1353l/A9
(l-1180) dated 23 January 1953
to ONR (m-23-54-4006)
4. Benson and Land "An Investigation of Hydrofoils
in the NACA Ilank" NACA Wartime
Report L-758,, 1942.
_-
50 Hoerner lfSome Characteristfes of Spray
and Ventilation" B.I,W, Carp, by
Gibbs & Cox, Inc. Technieal Report
M;3ls9 Contract No. Nonr-507(00),
0

6. Milne-Thomson nTheoretical Hydrodynamican

Macmillan, l!?sO

II - 3018

. -- -_-__l^“l_l_-l_l___- -.-. . ..--____“_- .--- --

 FOIL PARASITE DRAG

3. Drag of Small Projections

General Considerations

On the foil surface there is likely to be small irregularities or

projections in the form of plate joints, rivet or screw heads, control

surface gaps, e-tc. that have unique drag characteristics due to their

presence within the fluid boundary layer. (For large protuberances

that extend beyond the boundary layer, see Chapter 9.) The drag of

these projections can be determined as a function of their shape,

height and chordwise location on the foil.

The material which follows is t en almost entirely

from Hoerner~s('Aerodynamic Drag""1: . For a more detailed,
complete coverage, reference should be made to that work.

Spanwise
- - Plate Joints and Other Spanwise Projections

The drag coefficient of continuous spanwieie projections can be

represented in the form

cD. = \cd*33fi (4.14)

where CD. is the drag coefficient based on the frontal
area of the projection (or depression)

cCDl is an llindependent"coefficient, which is a

function of the type of projection.

bi is the ratio of projection he!-ght to distance of

the projection from the leading edge of the foil.

This relationship is valid for h/~ less than 0.02.

-

II - 4.16
 FOIL PARASITE DRAG

Values of b are shown in Fi~gure 4.6 for various plate jointa

El
and projections.

VALUES OF Icd FOR SBAWWOSE PROJE:CTIONS

Rivet Heads and Other Local Projections

The drag coefficient of rivet and bolt heads and similar ttspot-

Pike” projections takes the same form as equation (4.1,$) above for

plate joints B except that the coefficient is referred to the plan

area
- - of the projection. Thus

3%. (4.15)

where CD+ is based on plan area of the projection,.

Again, the relationship is valid for

- ‘18 less - than 0.02,
 FOIL PAP&SITE DRAG

Values of @ for various rivet and bolt heads are shown in

Figure 4.70

R ao.2d
CYLONQRICA 0.42 FLLi? .Q25 (AVG)

RQwNo FLUSH 0.002

VALUES OF Ic,l FOR LOCAL PROJECTOO~NS

Control Gaps

The drag due to the gap

between a foil and a control

flap can be estimated on the

\\
basis of the gap (measured

as indicated in Figure 4.8),

%y
CONTROL GAP MEASUREMENT

Thus, based on the plan area

of the gap,
 FDPL PARASITE DRAG

l-“Dt = 0.02 for the pressure side

b I 0.03 for the mx2t;fon side

or for the simple arrangement shown in the figure

(4.16)

where e/c is the gapg'ehard ratio

Qls is the spanwise extent; of the control flap

b is the foil span.

86;~~ ia then added dire:tly as a component of foil parasite drag.

 FOIL PARASITE DRAG

DESIGN EXAMPLE NO. 14.2 ISHI~~

D E T E R M I N E T H E E F F E C T O F A IBUTT S T R A P
ON THE PARASITE DRAG OF A FOIL ’
FOR THE 24” CHORD FOIL GIVEN IN EXAMPLE 4.1
THERE IS A 2” X 3/16” BUTTSTRAP ON THCi EXPOSED
UPPER a LOWER FOIL SURFACES,LOCATED AT MID-CHORD,
EACH SECURED BY A DOUBLE ROW OF l/4” DIA. COUNTERSUNK,
OVAL HEAD MACHINE SCREWS SPACED 5 DIAMETERS.

For the Strap

equation (4.14) at-d Figwe 4.6
Ft-om
h/&& o.o151Ies5 than O.OZ,SO IS appllcablel
a = 0.7

CDo = 0.7 ‘(3 (0.015)$ '(J.52

and b 0, = ‘2 x 0.52=0.0041

Fov the SCW.N~

lid dia. COUh tet-SUh k, ova\ heah screws have a d~atrdcr z 0.5”
and height * 0.025”
From eguatlon(4.ls) and Fqure 4.7
h~x=o.025/12 = 0.0021 (agaln, Iessthah o,oz)
jg = 0.012
ThevcFovc
Co+ = 0.0\2’3(o.ooz1)y3 = 0.0046
The swecu plah area ,per utwr OF span to5
2x$(+i2Xj: 0.314 rn’/;*r
ad &o= O* ~0.0046 = 0.0001

fatal
From Exam Ie 4.1, CD&~@= 0.0086
Thekefove, 1 R e tolal 1s Coo = 0.0Q36~0.0041 +O.OOOI = 0.0) 28

- . -- .- .-.- - ---- ---- -- ___-

- .

.x .

FOIL PARASITE DRAG

4. Interference Effects

Foil-Strut Junctions

At the junction of a foil and strut, there is an additiorial drag

due to reaction between the fluid boundary layers of the two bodies.

Analysis of test results7 indicates that this drag can be approxi-

mately expressed as a function of the thicknesses of the strut and

foil. Figure 4.9 shows the drag coefficient of various tested con-

figurations in terms of the Wlicknessl':

C4.17)

where c=Jt,
for junctions with and without fillets.

INTERFERENCE BRAG COEFFICIIENT

FIGURE 4.9
 FQIL PARASITE DRAG

-.

These test results are for fofls and struts of the same chord and

are probably not exaet when the respective chords are different. How-
ever, it is felt that the differences would be small and it is suggested

that the data shown can be used for all cases, basing the co-ordinate

t)c on the chord of the fofl,

It is also considered that the f%llet should have a radius in the

order of d- for the corresponding location along the chord.

There is an additional interference effect when the foil has an

angle of attack, as indicated by teat data7. However, ft is not known

how much can be attributed to viscous influences and how much to

induced drag (due to loss fn lift), For convenience, it has been con-

sidered an induced effect and an empil-ical relationship is gfven in

Chapter 2 (equation 2.26a),

Foil Nacelle Junctions

There fs a similar parasitic drag increment that arises at the

junctions of fofls and nacelles. However, since the nacelle replaces

a cestafn portfon of foil area9 the reducMon in foil drag (calculated

on the basfs of a plain wing, as is usually done) offsets the inter-

ference drag to a large extent.

 FOIL PARASITE DRJKI
.-

That is, adding the parasite drag of .a nacelle (see

Chapter 9) to the parasite drag of a .foil results in
a drag value that reasonably represents the total
parasite drag of the combined configuration, since
the mutual interference drag is offset by the
reduction in surface area0

The actual total effect of the nacelle-foil configuration is

dependent on many geometric factors., The shape of the nacelle, the

relative spanwise, fare and aft;, and vetiical locatinr; of the nacelle

with respect to the foil, and the type of fairing employed at the

junction are all important considerations and no general relationship

can be found to cover all conditions,

Rather, it is necessary to refer to test data for configurations

similar to that which is being considereda Reports such as that by

Jacobs and Ward8 on tests of wing-fuselage or wing-nacelle configur-

ations can be utilized for this purpose,

II - 4023
. .
l

FCIL PARASITE DRAG

bESIGN &XAlVP&& NO. 4.3.

DETERMINE THE STRUT INTERFERENCE .
EFFECT ON THE PARASITE DRAG OF A FOll,
FOR THE FOIL GIVEN IN EXAMPLE 4.1 y A S,PAN
OF 12’ IS TAKEN (WITH RECTANGULAR PLANFORM),
SUPPORTED ON THE SUCTION SIDE SY TWO STRUTS
OF 24” CHORD, 10% MAXIMUM THICKNESS.

From exumple 4,Lthe fotl I-IQS 24”choud, 12*/o thdness

Therefore,from equa.tron(4.17)
tl/c = mTz?E SS.II
Amd from flgwe 4.9
f& = 0.2 2 Cwithout fmmg)
o.oa(Nlth frlkd farwlg)

Convertmg CD+ fothe proper wu-ease wfo~l drag cocffuent,

II - 4.24
 FOIX PARASITE DRAG

References

1. Rossell and Chapman flPrinciples of Naval Architecture"

Vol. II The Society of Naval
Architects & Marine Engineers 1941.

2. Prandtl & Tietjens "Applied Hydro- and Aero-Mechanics"

McGraw-Hill 1934.

3. "Uniform Proscedure for the Cal-

culation of Frictional Resistance
and the Expansion of Model Test
Data to Full Size"
The Society of Naval Architects &
Marine Engineers Bulletin No, l-2,
1948 o
4. Hoerner l'Aerodynamic Drag"'
Published by the Author, 1951.

..-. 5. Abbott, von Doenhoff YSummary of Airfoil Data'

& Stivers N.A.C.A. Report No. 824, 194se

6. Hoerner ‘nThe Effect of Roughness Density

Upon the Frictional Drag" to be
published in the Journal of the
American Society of Naval Engineers,
19540
7. Hoerner "The Influence of End Plates, Struts
and Nacelles Upon the Character-
istics of Hydrofoilsfi, Bath Iron
Works Corp, by Gibbs & Cox, Inc,
Technfcal Report No. 14, Contract
No, Nonr-507(00), August 1953*

8. Jacobs & Ward "Interference of Wfng and Fuselage

from Tests of 209 Combinations in
the NACA Variable-Density Tunnel",
N.A.C,A. Report No. 540, 1935.

II - 4025'
 CHAPTER 5. FJAP CHARACTERISTICS

1, General Considerations

2, Lift Characteristics

3. Moment Characteristics

be Drag Characterist%s

Relationships for plain flaps and ailerons are given for

hydrofoils, derived from airfoil theory and data, Methods to

determine characteristics of flapped foils with full or partial-

span flaps are indicated. It is indicated tha-; airfoil relation-

ships may not be sufficient for flaps piercing or within one chord
of the surface, and specific test data is required for such cases0
1. General Considerations
I__-

Flaps and ailerons are applicable to hydrofoil configurations

for several purposes. Ailerons may be used on controllable foil

craft and for assisting turns, while flaps may be used on fixed

foil configurations for changing trim and 'submergence as desired

for changes in craft speed or load.

Some of the flaps developed for airfoils are illustrated in

Figure 5.1. On a practical basis, the plain flap or aileron

appears to be t:>e most attractive

for hydrofoil usi due to its c =e-

PLAIIN FLAP OR AiL&tOPd
mechanic and hydrodynamic

simplicity and its ability to

SPCDT FLAP
serve as an aileron or a trim

flap, as desired. The-:etical CD\

EXTERNAL AIRFOIL FLAP
relationships and test data are
cI>q
more readily correlated for SLOTTED FLAP

plain flaps, and formulations

more readily applied to design.

Therefore, the characteristics

shown below are primarily

applicable to plain flaps or

ailerons, FEURE 5.8

 For details of flap theory and addit.ional data on
flaps of all types, reference should be made to
pertinent airfoil treatises l-5.

There is no available data on the characteristics of flaps on

a foil in the vicinity of the water surface. Generally, it is
considered that the effects, of a flap on an airfoil will apply also

to a hydrofoil, taking into account the propertj.es of the unflapped

hydrofoil. This concept should apply where the submergence is

greater than 1 chord, but for smaller submergences (including

surface-piercing foils) the flap characteristics may well be

different due to the local flow conditions set up by the biplane

image system. In this latter case, model tests should be employed

to determine flap effects more accurately.

The notation employed for

plain flaps is indicated in

Figure 5.2. The flap chord,

C , is measured from the

+
pivot point and the flap

deflection is measured from

the chord line, as shown.
FLAP NOTATION
Flap hinge moments are taken

about the pivot point.

FIGURE 5.2
 FLAPS

2. Lift Characteristics

Section Characteristics

The deflection of a flap has an effect on the foil. similar to

that of a change in camber of the section, expressed as an

increment of effective angle of attack:

where k~ is a function of the

flap chord ratio. The theoretical

value of k# , shown in Figure 5.3,

gives results that are somewhat

higher than those derived from

experiment. For flap deflections
of not over 10" to 19, the

experimental data is well fitted

by the empirical formula

below the value %/a~ * 0,7.

\
The section lift equation.for a flapped foil in fnfinite fluid

is then

-.” ._-- __.-_.. ..^. ._ .l_l.l__----__....l.ll.l -- ._-- I ,.._ -.l.-^_. -. I_.- -..^.--.----.1 -
 FLAPS

(5.3)

where Or, is the foil section angle

40 is the unflapped angle of zero lfft

d is positive downward,

Foil With Full Span Flaps

The lift of a hydrofoil with flaps extending the full span can

readily be det'er&qed from

64)

where ( g ) is the lift curve slope of the unflapped foil,

as determined In Chapters 2 and 3e

Foils With Partial Span Flaps

When flaps extend only over a portion of the foil span, the

flap effectiveness must be suitably modified by an additional factor

/
so that the total factor is 4%' o This; additional factor
4 9
is shown3j5 to be a functfon of the basic spanwise lift distribution

of the unflapped fofl, and thus for any but an elliptically loaded

foil a function of aspect ratio, An approximate value for the

factor +' can be readily determined once the basic spanwise dis-

tribution is known, by the ratio of the basic lift over the span

._--l__ ___---
_l---m*_ --I..
--_-- - . . _ l _ l , _ _ _- ~ _ l - . . . “ . - - . - -
 FLAPS

of the flapped foil to the -BASIC LIFT DISTRIBUTION

lift of the entire foil, as

illustrated in Figure 5.4(a).

For an elliptically loaded

foil (of any aspect ratio)

the factor &f' can be

determined from Figure 5.4(b)

in terms of the ratio Q./b,

where Q/' is the span of the

flap that extends to midspan.

For partial flaps at the tips

or within the span, the

,factor can be determined

from:

The elliptically loaded case may be used for all.foils without

serious error.

A further correction is required for the .hydrofoil xhere

greater accuracy is required, due to the biplane 5mage a? ?&?

incremental lift over the flapped span. This may be determined

from the relationships derived for a biplane with unequal spans

 FLAPS

(888 Dursnd6 ). However, for preliminary purpoaIe,8, such corrsctiolir

nUry bo nrglSOti#d,

gift o f flag8

The nO!%al forC8 on the flaps i8 giv8n rp~roximat8ly b y tm

equation

(5.5)
whom CJ ir the esotion lift of th8 Ella sd foil
drtorminad for eaoh point -1-4!hK0*
the flap from the bario lift dirtributlon,
n,,a, are oo~ffioisntr given i n Fij~~ts 5.5, ’
 FJAPS

DESIGN E~XAIVIPLE NO. 5.1 I SH I OF I

ESTIMATE THE FLAP EFFECTIV’ENESS

FACTOR FOR A PARTIAL-SPAN FLAP

A PAIR OF FLAPS HAVING 4 CHORD EOUAL

T *-------3
O 250/ ; T H E FOILCHORD A N D A N
OVERAL; SPAN OF 60% OF ;HE FOIL
SPAN ARE INSTALLED OK A 2 ‘0 TAPERED
FOIL,A§ SHOWN ON THE SKETCH
.I
.+5

From equation (5.2)) the effectiveness of lhe

flap section i5

The foil has approximately an elliptic lift distribution,

(as indicated in Chapter I);
Therefore, from Figure 5.4,or-d pcnqes 5.5 and 5,6

Then, the f-da\ flap effectiveness factor is

= 0.325

‘-
II - 5.8

---_-.. ----.- --___-

--
5

3. Moment Characteristics

Pitching Moment

The section pitching moment about the quarter-chord line OI

the foil, due to flap deflection is given by the equation

(5.6)

where dCmv4 is given in Ipigure 5.6, and is seen to

db be a function only of flap chord ratio.

For full span flaps,

the total pitching moment

about the quarter-chord

can then be determined

readily by the methods

indicated in Chapter 2.

For partial span flaps,

the total pitching moment

due to flaps must be de-

termined from the basic

lift distribution (span-

wise) of the foil, as

indicated above for the

f
lift due' to flaps.

___.-___--. .^ .-
-.---II_
-.-
,-
FLAPS

Flap Hinge
- - Moments

The section hinge moments on a flap are given.by the equation

(5.7)

where % is the coefficien-t of the moment about the

flap hinge in terms of the flap chord

c' is the rflapped foil lift coefficient

are factors given in Figure 5,7 and are seen
to be functions of the flap chord ratio.

The total hinge moment for

a full span flap is readily

determIned from equation (5.7)

for the section, by the pro-

cedure given in Chapter 2 for

foil pitching moments, For

the simple case where the

flap chord ratio is constant

across the span, the total

hinge moment is found merely

by substituting & (for the

foil) in place of Q (for

the foil section).

 FLAPS

For partial span flaps, the hinge moment at each section must

be determined from the basic lift distribution of the unflapped

foil, as previously indicated,

 FWLAFS

DESIGN EXAMPLE NO. 5.21

ES%IMATE THE PITCHING MOMENT 8r HINGE MOlV!ENT
CHARACTERISTOCS FOR A PARTIAL-SPAN FLAP ,

FOR THE FOIL-FLAP ARRANGEtiENT

G1VEf-j IN EXAMPLE 5.1

and tb\g tot-1 m o m e n t ~factors f o r tRe fail are

R-b ‘/4 s 0,66 ~0.59 = 0.38q

-c; >

J CH = O,lO x s.sq = 0.-q

cx

G
= 0.60 x 0.59 = 0.354
( d8 )
 FLAFJS

4. Drag Characteristics

Parasite Drag

The parasite drag of a foil section increases with deflection of

the flaps, as indicated by experiment ,233. However, such increase is

small, being of the same order of mangftude as ,that experienced when

the unflapped foil is given an equivalent angle of attack,

In other words, the simple relationship of the foil parasite drag

varying as the factor (1 + b'c;) given in Chapter 4 appears valid

whether the lift is due to angle of attack of the foil or deflection

of the flap, and no additional parasite drag term need by considered.

Induced Drag

For full span flaps, the expression for the induced drag in terms

of the lift as given in Chapters 2 and 3 is valid, without further

correction.

For partial span flaps, however, the induced drag must be derived

in terms of the redistributed spanwise loading due to the flaps. The

airfoil case involves several additional parameters (see reference 3)

and the hydrofoil presents the additional consideration of biplane

images e Qualitatively, partial span flaps increase the induced drag

due to departure of the lift distribution from an optimum whereas

_.__.I .^_. -_-- . ^---~- -----

_ I,___ __- .- _~ __.-_- --_. - ..- .--. -.---- -I.-_ I-..--. ^---.-_l_~
 FLAPS-
-

the biplane factor is reduced (when considering the effect over the

entire foil. See Durand6 for biplanes of unequal spans).

It is presently considered that such additional factors need not

be considered for preliminary estimating purposes6 Test data on

partially-flapped foils should be obtained for actual configurations

under consideration.

.’ -amlIP
II - 5.14 1
 FLAPS

References

1, Abbott & von Doenhoff "Theory of Wing Sections"

McGraw-Hill, 1949.

2. Cahill "Summary of Section Data on

Trailing-Edge High-Lift Deticesl'
NACA Research Memorandum No.
~80~9, 1948,

3. Young "The Aerodynamic Characteristics

of FlapsIt, ARC Technical Report
No, 2622, 1953.

4. Ames & Sears l~Determination of Control-Surface

Characteristics from NACA Plan-Flap
and Tab Data", NACA Report No, 721,
1941.
5. Pearson & Anderson l'Calculation of the Aerodynamic
Characteristics of Tapered Wings
with Partial Span Flaps", NACA
Report No. 665, 1939.

6. Durand ~*Aerodynamic Theory", Vol. II

California Institute of Technology,
19430
 CHAPTER 6. HYDRCIQILWAKE

1. General Considerations

2. The Wave Pattern

3. Sub-Surface Flow

4. Sfgnificance of the Wake

5. Design Data and Procedure

The pattern of the wave produced by a hydrofoil is discussed

,
I qualitatively and illustrated by speeiffc test results. Methods

to determine the sub-surface flow are given, the significance of

the various flow factors for different types of hydrofoil craft

are discussed, and data is presented on the waves and flow fields

behind a foil.

II - 6.1
 HYDROFQILWAKE

1, General Considerations

The wake produced by a hydrofoil operating close to the water

surface consists of two distinct flows. First,, the foil produces a

gravity flow, expressed in a surface wave with its concomitant sub-

surface flow field. Secondly, the foil also produces a flow similar

to that of an airfoil operating in afr but which is restricted due to

the proximity of the water surface. This ltaerodynamic" flow is

independent of the gravity flow and does not p:roduce a surface

disturbance.

All components of a hydrofoil configuration that are located in

the wake are affected by the flow characteristics of the wake, Thus,

it is necessary to investigate the effects of the flow field on such

components located in the wake asz additional foils, struts and

rudders, flaps, and propellers,

.^” _-_.-._“._-.___^ _- . ---.---....--..---. ----_I__-.---“-.

-.-. I .-...-. l..~.--.-.-._
 HYDROFOIL WAKE

2. The Wave Pattern

The Wave Produced by a Point Disturbance

The pattern of the wave produced by a hydrofoil is essentially

that produced by a series of point disturbances distributed along the

lifting line of the foil, and the observed characteristics of the wave

pattern can most readily be explained on this basis, Therefore, the

characteristics of the wave produced by a single point disturbance

will be illustrated briefly,

The pattern of the wave produced by a point disturbance was

derived by Havelockl and later by Lunde2 and is the well-known ship

pattern consisting of lateral and transverse waves, as illustrated in

Figurn 6,1, The formulas for the wave amplitudes are only valid 'Ifar

behind" the origin and do not represent the conditions close to the

origin. Furthermore, the exact amplitudes are not readily ascertained

and the relative amplitudes throughout the wave region are extremely

difficult to evaluate. Generally, however, the amplitude of each crest

is greatest near the ltcuspll line, where the lateral and transverse waves

combine to produce a high, short crested wave, called the cusp.

The amplitudes of successfve crests and troughs decrease in

proportion to the distance from the origin, with the transverse waves
 HYDRGFOIL -WAKE
-

1.0

.9
X
I
va

AN .
3 .8

s+ .I
P
L

a
Ea .3
J .3 1334

COORDINATES OF CRESTS 8 TROUGHS

WAVE PAT TERN

CHARACTERISTICS OF .WAVES PRODUCED BY

A POINT DISTURBANCE

FIGURE 6.1

- -___^ ...--” ..--

II - 6.b

I---..-.- --- .__-_._ _~----....---I ----- .- ---I.. __l_ll_

 HYDROFOIL
-. WAKE

decreasing at a greater rate. Far aft of the origin, the lateral

waves at the cusps are greatly predominant.

In shallow water of depth,d, the characteristics of the wave

are markedly different from those in deep water given above. Above

a speed V~QB@ the cusp line angle increases from its deep water

value o.f 19”28f, with consequent spreading of the ILateral waves and

decrease in transverse wave amplitude. Theoretically, at the critical

speed vc=q, the cusp line angle is 90" with a solitary wave pro-

ceeding along in line with the origin and no following disturbance.

At' supercritical speeds, the pattern is as shown in Figure 6.2,

where the transverse waves have disappeared, the lateral waves are

concave to the centerline, and the "wave front" is defined by the

I WAVE AT SUPERCRITICAL

FIGURE 6.2
SPEED

II___ _,
.__--“------ .-.- --_.-_____-
 HYDROI;UILWAkE

The Two-Dimensional
- Wave

A continuous distribution of point disturbances along a line of

infinite span would result in transverse waves only, with the crests

parallel to the line of disturbance. The wave has been shown by

Meyer3 to be a simple sine wave extending aft with a superimposed

local disturbance in the vicinity of the foil. Figure 6.3 shows the

wave profile and indicate6 that the local disturbance disappears

approxjmately l/4 wave length downstream. The equation of the sine

wave is given as

(6.1)

and the slope of the wave

dt. = &0 (6.la)

d x

h
S’J
A (4 FREE STREAM VELOCITY

_-
L WAVE PROFILE IN TWO DIMENSIONS

FIGURE 6.3

-----___ ______--.-.-----_... ---- _______

 The Three-Dimensional
- - Wave

From the above considerations, the pattern of the wave produced

by a hydrofoil of finite span can be shown. In the region between

the converging cusp lines emanating from the foil tips, there'is a

cancellation of lateral waves and a strengthening and straightening

of transverse waves, simulating the two-dimensional sinusoidal

pattern. Outboard of these convery& cusp lines, the pattern
approaches that due to a single point disturbance, and is equivalent

to it at the cusp lines- diverging

- - from the tips. Figure 6.4 shows
the pattern, as described above.

WAVE PATTERN FROM A HYDROFOIL

FIGURE 6.4
Wave contours measured from hydrofoil model tests are shown in

Figure 6.5, showing the general pattern indicated above.

.-
 HYDROFOIL WAKE

.-

PARTICULARS CONTOURS
~--

S U R F A C E W A V E S P R O D U C E D B Y A H Y D R O F O I L O F L A R G E 4SPECT R A T I O

FIGURE 6.5

II - 6.8
 HYDROFGILWAKE

3. Sub-Surface Flow

The sub-surface flow in the wake of a foil is due to two effects,

namely:

1. that gravity flow associated with the surface wave

2, that due to the t'aerodynamicl' effects of the foil.

The relative importance of these effects is a function of Froude

number, as will be discussed below,

Gravity Flow

The gravity flow at any depth,h 9 below the surface is easily

determined from the surface wave by the classical ttdecayl' formula:

where &w is the flow angle at the submergence,4

&0 is the flow angle at the surface.

Aerodynamic Flow

The basic airfoil wake will first be considered, after which the

influence of the surface will be taken into account, The vortex

sheet emanating from the foil wraps up into two distinct vortices

approximately within one span behind the foil, as indicated in Figure

6,6, The separation between vortices, b' 9 and the radius of the

vortex: core, /7J , are functions of the foil load distribution, and can
be determined by methods indicated by Durand14 0

II - 6-9

.____-._-. .-....__._.._.I. -.--l---ll--- ---- -_- -_- _.. ._

 HYDX?2ILWAKE
.- -

’ .-< VORTEX .CORt

TRAILtNG VORTEX’ SYSTEM FAR BEHIND FOIL

POWNWASH FACTOR FOR ELLIPTICALLY LOADED FOIL

~ERODYWAMIC WAKE

FIGURE 6.6 ‘.

I I - 6.31
 HYDROFOIL WAKE

It is considered sufficiently accurate to asa\une elliptic load-

ing for the foil when de&mining the flow in the wake far behind the

foil. Then Durand4 gives the following values

(6.3)

\.

The downwash angles in the plane of the vortex lines can be readily

determined from the following expression

(6.4)

tihere E is considered positive for upwash

.
is a factor depending upon spanwise
3 location, and is given in Figure 6.6.

Low Froude Numbers

At very low F'roude numbers,

c
the tlaerodynamicll effect is that of (
h
RIGID WALL
an airfoil near a rigid wall at the I - -
h
water surface, as indicated in

Chapter 2, Then the aerodynamic

flow angle would be determined by WALL EFFECT

the trtaiklng vortex system and its FIGURE 6.7

mirror image as indicated in Figure 6.7. In the plane of the vortices,

II - 6.11
 HYDROYUIL WAKE

the downwash angle given in equation (6.k) would be modified to

the value

where G( is the generalized factor determined for

the foil, as Indicated in Chapter '2.

Then the total sub-surface flow angle at low Eroude Number is

E, = && 4 Ei (6.6)

High Froude Numbers *

At high Froude numbers, the wave effects become negligible and

C the total effect is that of the llbiplanetl, as indic,ated in Chapter 2 .

Then the total flow angle is

determined by the trailing

FREE SURFACE
vortex system and its biplane
If---
h
image as indicated in Figure

6.8. Thus, at high Froude

numbers, &may be neglected.

BIPLANE EFFECT
FIGURE. 6.8

In the plane of the vortices, the total sub-surface flow angle

at high Froude numoers is

(6.7)
 HYDROFOILWAKE

40 Significance of the Wake

The significance of the flow conditfons in the wake of a foil as

they a.ffect other components naturally depends on the type of hydro-

foil configuration investigated. Thfs in turn depends on the type of

craft to be designed. Generally, it is necessary to evaluate the

influe:nce of the wake although some aspect which may be important in

one design is negligible in another, and vice-versa.

How the emphasis on different aspects of the wake changes with

differlent designs can be illustrated by examples Iof two different

types of craft.

(a) The large, slow-speed craft

For a hydrofoil craft of several hundred tons or more

and a design speed of less than 40 knots, the hydrofoil con-
figuration would be of the large span, tandem foil type,
The wake from the forward foil would be essentially two-
dimensional fn nature, as fn Figures s83 and 6,s. Assuming
the two-dimensional case for the purpose of illustration,
the upwash on the aft fofl would be, from equations (6.1)
and (6,2)s

where the subscrfpts 1 and 2 denote the forward and aft

foils respectively,
Since the upwash changes the direction of the lift pro-
duced on the second foil by the angle C$ resulting in a
drag component, the total drag due to lift would be

- --
II - 6613

-___1_- -.-I-.-.-.- __.____-.-... .I.._..^ -..-.-- -_.- ..-.---- -- ..-__- -I -..,, __ --_.. “.- -.... -_._ I__
 HYDROFOILWAKE
-

which for foils of equal geometry, submergence and lift can

be shown to be

-qh+
(since CQ,=G,' gz e for the foil in two dimensions,
as shown in Chapter 2).

Thus, depending on the separation between the foils, the

total wave drag could have a value varying from zero to twice
that of the configuration without upwash,

(b) The small, high speed craft

For a hydrofoil craft of under 100 tons and a speed of
over 40 knots, the foils would be small and short in span.
The wave produced would be of the type shown in Figure 6.5
(for high Froude numbers), and the wake would have the
following features:

a. neglfgible transverse wave pattern

b, strong ltaerodynamiclt downwash
c. fnportant lateral wave crests and roaches

For this type of wake, the transverse location of the com-

ponents of the configuration is important as compared to
example (a) where the fore-and-aft location is important.

Foils and control surfaces should be positioned so as

not to be in the strong flow of the ae:rodynamfc wash; and
furthermore, vetiical struts and rudders should be positioned
to avoid roaches (particularly when located at or near the
centerlfne) and strong lateral cusps, On the other hand, it
may prove beneficial to locate propellers directly in the
trailing vortex field to take advantag,e of the rotary flow.

The above examples illustrate the relative importance of different

aspects of the wake for dffferent types of craft, In any case, all

of the effects should be investigated for a given design.

---------_I,.. _ -.,. _-” __-.- --.~ _.-II .._.. - _-.. -- -... -~.--_- .-.. -_---.I..
 HYDROFOIL WAKE

5. Design Data and Procedure

Existing design data on the wake of hydrofoils is limited to that

produced in two exploratory tests. Breslin 5 conducted wave measurements

on a foil of aspect ratio 20, the wave contours o.f which are presented

in Figure 6.5. E.T.T. W,8 conducted a series of tandem foil tests from
which the average upwash angle along the span of the after foil have

been calculated, as shown in Figures 6.9 and 6,10,

All the above tests were conducted at a submergence of 1 chord,

with foils of large aspect ratio at relatively low Froude numbers.

There is a need for additional data, particularly for small foils at

large Froude numbers, in order to make reasonable estfmates of the

effect of the wake for a greater variety of designs.

The existing data given in the figures can be used for such con-

figurations where interpolations or extrapolations are reasonable,

Methods for calculating the wake effect are given, as followsx

(a) Use of Wave Contour Data

From the contour map of the wave produced by a foil of
given aspect ratfo at a given Froude number, the flow can be
determined for any point in the wake. Determining the sur-
face flow angle, t5* 9 from the contour map, the flow angle at
a given depth can be established by use of the methods
described in Section 3 above0

The determined flow angles along the span of a foil in

the wake can be simply average over the span for a reasonable
approximation,

II - 6.15
 -.05
20 2.7 3.0 Fc 3.5 4.0 4.9

UPWASH A N G L E O N A F T F O I L F O R A = 2 0 E Q U A L . FOILS;h/,=l.O

,-

.I0

NOTE: CONFIGS DENOTED

Y .05
cl,

I
2.0 2.5 3.0 6 3.5 4.0

UPWASH ANGLE ON AFT FOIL FOR VARIOUS COMBINATIONS OF

LARGE ASPECT RATIO; h/c= 1.0
FIGURE 6 . 9
 HYDR@I"OIL WAKE

.-,
9h
.oz5

-.025
[A,- “/b, - b;i,,]
1
2.0 2.5
3.0 Fc 3.5 4.0 4.5
UPWASH ANGLE ON FOILS AFT OF A= 12 FOIL. h/c = 1.0

O -
I
/
NOTE : CONFtG5 DENOTED
‘-I

[A,- %, - b2/i-j
]
b.0 Fc 3.5 4.0 4.5
UPWASH A N G L E O N F(IILS \ AFT OF A = 8 FOIL h/C . I.0

FIGURE 6.10

.-.---~~--.._. _-
 HYDROFOIL WAKE

(b) Use of Tandem Foil Data

The average upwash data can be int'erpolated for similar
configurations to those tested, and is :more direct than the
contour data when determining the average upwash on the aft
foil, as long as the submergence is the same as that given
in the data, When the submergence (of the aft foil) differs
from that tested, the data must be corrected by use of the
methods of Section 3 above.

(c) Location of Lateral Cusps and Roaches

Where wave contour data is applicable, the location of
cusps and "roaches" can be readfly observed, and the surface
and sub-surface effects readily calculaked. (The interfer-
ence between converging lateral crests, behind the foil mid-
span, causes a large upsurge of water, commonly called a
roach,) Where data is not available for the conditions to be
investigated, the approximate locations of crest and troughs
can be determ%ned from the theoretical values given in Figure
6.1 for the wave produced by a point disturbance, The waves
are assumed to be generated at the foil. tips, The theoretical
location of roaches directly behind the midspan of a foil (or
midway between the tips of port and starboard foils) can be
determined from FBgure 6.11, Generally, only the first few
waves are of consequence and these only near the cusps and
roaches where the amplitudes are large,, Where more than one
foil is considered (such as with small foils, P/S), the
effects are combined, Since the exact amplftudes of such
waves cannot read'ily be determined, care must be taken in
evaluating the importance of different cusps and roaches,
particularly when a combination of foi:Ls are under study,

--.--------
 HYDROFOIL
-- 'dAKE

I I I
4 2 0
FOlL S P A N - FT. LOCATION OF ROACH - FT,
AFT OF FOIL

50 40 30 20 IO 0 IO0 200 300 400 500

FOIL SPAN - FT. LOCATION OF ROACH - FT
AFT OF FOIL

THEORETICAL LOCATION OF ROA,CHES

BEHIND A FOIL

FIGURE 6.11
 HYDROFOILWAKE
“ -
-

References

1. Havelock "The Propagation of Groups of Waves

fn Despersive Media, With Application
to Waves on Water Produced by a
Travellfng DisturbanceIt, Proc. of
Royal Society, 1908.

2, Lunde "On the Linearized Theory of Wave

Resistance for Displacement Ships in
Steady and Accelerated Motfon",
Society of Naval Archftects and
Marfne Engineers, 1951,

3. Meyer "'Surface Waves Produced by Hydrofoils"

Hydrofofl Corp, Confidential Technical
Memo HM-00, 1950.

4. Durand "Aerodynamic Theory", Volume II,

California Institute of Technology,
1943.
.-
rse Breslfn "An InvestigatTon of the Character-
istics of the Waves Produced by a
Hydrofoflfl, BIW Carp, by G&J Inc.
Confidential Techn%cal Report No, 13,
19.53.
6, Sutherland "Extensfon of Single and Tandem Foil
Tests of 2-l/2@' Chord Aspect Ratio 20
Hydrofoil", ETT Ltr, Rpt. 410, April
19.51.
7. Kaplan "ExtensSon of Single and Tandem Foil
Tests of 2-l/2" Chord Aspect Ratio 20
Hydrofoil and 5" Chord Aspect Ratio 6
Hydrofoil", E!TT Ltr, Rpt, No, 428,
1951.
8, Kaplan and Dingee "Tests of a 5" Chord, Aspect Ratio 6
Hydrofoil Used as a Strut and Tests
of 2-l/2'" Chord Hydrofoil for Various
Aspect Ratios in Single and Tandem
Conffgurationsll, ETT Ltr. Rpt. No.
429, 1951.
 CHAPTER 7. STRUT CHARACTERISTICS

1. Drag Characteristics at Zero Yaw

2. Characteristics in Yaw Prior to

Ventilation

3. Characteristics in Yaw When

Ventilated

4. Height of Spray

The drag and side force characteristics are given for surface-

piercing struts. Spray drag and ventilated characteristics are seen

to be functions of section shape, while side force characteristics

are shown to be similar to hydrofoil lift characteristics. The

hydrodynamic results given include experimental data on spray height,

 1. Drag
- Characteristics
- at -Zero- Yaw
-

Drag Components

The drag of a surface-piercing strut at zero yaw consists of

three components: the section drag of the wetted strut and the

localized drag at its two ends. The drag at the upper end where the
strut pierces the water surface is called "spray drag", being assoc-

iated with the production of spray at that poiht. At the-lower end,

the drag is either "tip drag" for a free-ended strut, or Vnterfer-

ence drag" when the strut is connected to a foil or other body.

A wave drag also exists at low Froude numbersl, but

becomes negligible above v/,& = 3. Therefore, at
the relatively high speeds associated with hydrofoil
craft, the wave drag may be ignored.

DRAG

SPRAT
a
T I P DRAG
-
%
(OR INTERFERENCE 1

COMPONENTS OF STRUT DRAG

, FIGURE 7.1

-
m
...
II - 7.2

i _._ ._ ^_. ^------ -^

 STRUTS

Section Drag
-

The section drag of a strut can be treated in the same manner as

for a foil, including all considerations of turbulence, roughness, etc.

This material is given in Chapter b, "Foil Parasite Drag". From

equation (k.9), the section drag of the strut can be given:

b (7.1)
where CD, is the basic section drag coefficient based
b
on the side area of the strut

(t/,), is the strut thickness ratio

5L.L is the flat plate friction coefficient based

on total wetted surface.

The value of c+ can be determined as outlined in Chapter 4, for

smooth turbulent, transitional, or standard rough conditions as may be

required.*

Spray Drag

The drag arising at the point where the strut pierces the surface

is manifested in the development of spray along the forebody of the

Y Tank teeit results on laminar-profile strut sections indicate that the

section drag coefficient is in the low drag "bucket" region at test
Reynolds numbers as high as 6 x 106. However, foils of comparable pro-
file at lower Reynolds numbers showed fully turbulent section drag
coefficients, probably due to a higher level of turbulence in the tank
generated by the foil. It is considered that in open waters, the
turbulence level is high and the section drag is essentially that in
turbulent flow,

__--._ ._..-.
 STRUTS

strut. This spray drag is a function of the magnitude and distri-

bution of pressure along the forebody, and thus a function of

fineness or thickness ratio, t/c, and the sharpness of the leading

edge.

At the relatively high speeds associated with hydrofoil craft

operation, the spray drag coefficient apparently does not vary with

Froude number, according ta test results.

Correlated, systematic test results of surface-piercing struts

are few in number, and the lack of sufficient data prevents the

establishment of a universal function to take into account the fine-

ness and leading edge sharpness. Figure 7.2 shows the spray drag
coefficient (based on the area,tC) for several strut sections

tested2y3 .

SPRAY- REFERENCE
Co ct- D**c
SECTION

NACA 66-012 0.031

2
N A C A 66-021 0 . 0 2 3

BICONVEX
“/# ’ 0.15 0 . 0 1 5- 0 . 0 2 5
I- I
Mb24 TYUNESS 3
eg%$ OIAMON 0
Vc8 0 . 1 5 0.01s

SPRAY DRAG COEFFICIENTS FOR SEVERAL STRUT SECTIONS

FIGURE 7.2
 STRUTS
.t * -

Generally, it is indicated from these tests that the spray drag

coefficient decreases with increasing thickness ratio (but the absolute

drag fncreases, as should be expected), and that the coefficient is

less for sharp leading edges than for rounded leading edges such as

employed in airfoil sections.

Tip Drag or Interference

The -tip drag of a free-ended strut (i.e. not attached to a foil

or other body at its lower end) can be evaluated from airfoil data.

HoerneA indicates that for well-rounded tips, there is no tip drag,

while for square tips the drag of one tfp is expressed as

co, = p
Dtlp ,=: o, 085 (for square tip) (702)

When attached to a foil, the strut experiences an interference

drag at the junction as it similarly imposes an interference drag on

the foil, However, the total interference drag at such a junction

experienced fn tests has been ascribed to the foil for convenience

(see Chapter h), and need not be further considered here.

Effect of Foil Lift

For struts attached to the upper surface of a foil, there is an

increase fn fluid velocity past the strut due to the circulation

around the foil, and therefore an increase in section drag. (Th ere
 STRUTS

is no increase in fluid velocity at the water surface and thus no

increase in spray drag.) This effect is small (being less than 10%

for most applications) and can be approximated by multiplying the

strut section drag, equation (7.1), by the following factor

where J depends on the spanwise

location of the strut,
Cr/b, and the submer- I
gence-span ratio of the
foil, h/b . Values are
given in Figure 7.3. J
c, - foil lift coefficient
A = foil aspect ratio

Effect of Rake

Raking a strut either forward or aft will reduce the section

drag from that value produced when the strut is vertical. This is
obviously due to the rcductionin strut thickness ratio in the

direction of fluid flow, and thus the thickness ratio to be employed

in equation (7.1) is

where (%\,is the nominal section thickness ratio

y is the angle of rake.
 STRUTS

It is not fully evident from existing test data whether there is

any real benefit in rake insofar as spray drag is concerned~ Generally,

it is felt that there is some advantage, but significant gains have not

been experienced2.

Ventilation of Struts

Ventilation is characterized by an air-pocket that is formed at

the after-body of a strut, which air-pocket is open ‘to the atmosphere

and extends downward along the strut a d&stance depending on speed and

configuration characteristics. (Ventilation should not be confused with

cavitation, which is a condition not directly related to the water

surface,)

The air-pocket is formed only in the presence of some sharp

discontinuity in the flow. The discontinuity produces a vortex which

allows the air to penetrate the water surface of constant pressure,

after which the full cavity can be formed. Discontinuities arise due

to abrupt changes in strut section, local protuberances, momentary yaw

angles, etc.

The effect of ventilation is to reduce the section drag of the

strut, since air of atmospheric pressure replaces water of sub-

atmsopheric pressure at the afterbody, resulting in a forward force

component. Quantitative results of this drag reduction are meager

and at present fairly inapplicable for the following reasons:

II - 7.7

.- _^-_ __.__” “^ _ -, .l-.-~ ___- .“_,. ..----.-- 11--- -.- ” -.

 STRUTS

(a) Streamlined free-ended strut sections tested at zero yaw

show no tendency to ventilate at moderate and high speeds 23

3 .

Whether this condition would prevail in practice where

transient yaw angles would be experienced or where local

roughness elements might develop sufficiently to "triggertt

ventilation is not known., If such were the case, there is

no data available on such a condition,

(b) Wedge-shaped strut sections which ventilate throughout the

speed range have such a high section drag :Initi.ally that

the reduction in drag due to ventilation d'oesnot appear to

result in a net gain, as far as existing data indicates 3 .

(c) The effect of foil lift on strut ventilation is not known.

Depending on the foil submergence, the low-pressure region

above the foil may %rigger" ventilation (which in turn will

lower the lift produced by the foil).

Further experience with actual craft or testing of various con-

figurations under simulated operating conditions is necessary to

produce accurate design information regarding ventilation. At present,

it is recommended that for streamlined strut sections, the effects of

ventilat-ion can be ignored in normal operations at zero yaw.

II - 7.8

---. -- ---llllll~-.-------- ---- .-..- -.--1__ ..--

 STRUTS

2, Characteristics in Yaw Prior to Ventilation

The vertical strut having an angle of yaw, Y , produces a side

force and a resultant induced drag. For a fully submerged strut, this

is equivalent to the characteristics of an airfoil with a horizontal

lift force. For the surface-piercing strut, necessary corrections must

be made to account for the boundary effect and for spray-producing drag.

Surface-Piercing Strut

The effect of the surface on the side force characteristics of a

strut is a function of Froude number, similar in nature to the effect

on the foil discussed in Chapter 2. At low Froude numbers, there is

a wave effect and a "rigid wall" effect, the first decreasing the strut

efficiency and the latter increasing it.

At high Froude numbers, the surface effect is similar to the

"biplane" effect on a foil; hydrodynamically, the strut is then

equivalent to one-half of an anti-symmetrically twisted wing.

There are no theoretical analyses presently known that give the

force characteristics as a function of Froude number* However, as is

indicated in Chapter 2 for foils and in reference 1 for strut wave

drag, the region in which wave and rigid wall effects are important

is at Froude numbers lower than considered practical for most appli-

cations, Therefore, the "biplane" effect is considered to prevail.

. .-.. -.- --l----*_l~- -I~~---I---- .----~.--I__

 STRUTS

On this basis, the aide force characteristics of a aurface-

piercing strut are approximately as followst

(7.3)

(7.4)

where C; is the side force coefficient based on

the submerged strut area (one side)

Ch is the strut drag coefficient, based on

strut area

As is the strut aspect ratio, V&

Y is the angle of yaw

E is the lifting surface correction

.- (see Chapter 2)

Effect of Foil

For the typical case of a strut attached to a foil, the foil exerts

an end-plate effect on the yawed strut. This single end plate has a

"height" equal to the foil span. Equations (7.3) and (7.4) are thereby

modified to the expressions:

dry E 2 (I+ b/h,

(7.5)
dc, = 25 +AAS (1+2%)

(7:6)

.
 STRUTS

where (I+ b/h) is the approximate induced factor for

(ix) a single end plate

b is the foil span

h is the strut length (foil submergence)

Equations (7.5) and (7.6) can also be used for a free strut that has

an end plate at its lower tip, with the actual end-plate ttheight"

being substituted for the foil span, b .

Low Aspect Ratio Correction

Strut aspect ratios are generally low for most hydrofoil appli-

cations, and the E factor becomes important, as indicated in Chapter 2.

.I
The non-linear effects, which are based on flow across the tips (as

shown in Chapter 2), are not expected to be present on a surface-

piercing strut supporting a foil, since there is no flow across the

tips in this case, and equations (7.5) and (7.6) still apply. For a

free-ended surface-piercing strut, there is flow across the lower tip,

and in the absence of specific test data, it is assumed that the non-

linear effect is l/2 that for a submerged foil with both tips free.

Thus,, rewriting equations (7.2) and (7.3) in different form and

adding the non-linear terms, we get for free-ended struts:

(7.7)

_-

_ ___...-. -_ ^-~..-.---.,“. ll__-___--_

_l -
 STRUTS

(7.8)

where tps, is the parasite drag coefficient of the strut

(including spray drag) which is also a function
of yaw angle, v .

Correlation with Test Data

Tests have been conducted on yawed struts at E.T.T.5 , and results

for free-ended struts are shown in Figure 7.4. In the non-ventilated
region, equation (7.7) is seen to agree closely with the experimental
results for side force.

For the drag, %I was estimated to be 0.012 including spray,

for the section at the tested speed. Equation (7.8) was then evaluated
(for A = l/2 and 1, which are identical. A - 2 gives slightly lower

*ad * The agreement is good.

II - 7.12

-_l_l_- __I_-
-_- --- -1-11.” -~__I
 STWITS
.

A.
3-
EQ (7.7
I I 0
Cs .2 / I

.I r / /
-r
I I .oI I I 1 ’ 4oo

A h/c = 2.0

3
-

.2
C
D5
.I

0
IO0 20” 30” 400
A N G L E O F Y A W - Iu
2-l/2”? 12% OOUBLE ARC STRUT .
TEST SPEED 25 FPS
‘IDE F O R C E A N D D R A G C H A R A C T E R I S T I C S
OF A SURFACE-PIERCING STRUT

FIGURE 7.4
 STIllITS

- DESIGN EXAMPLE NO. 7.1

DETERMINE T-HE DRAG 811 SIDE FORCE CHARACTERISTICS
O F A S U R F A C E - P I E R C I N G STRUT,PRIOR T O VENTILAT,lON

A SINGLE STRUT OF CONSTANT SECTION SUPPORTS

A RECTANGULAR FOIL AT MIDSPAN. THE STRUT IS

I VERTICAL WITH A I2”CHORD AND IO% THICKNESS,

THE FOIL IS 6’SPANX 12”CHORD,WlTH A
SECTION
SUBMERGENCE OF 2’,AND SUPPORTS A LIFT OF
6500 Lb. AT 35 KNOTS
I 0 J
PLAN

..’

-.
 DESIGN EXAMPLE NO. 7.1 I SH 2 O F 2

*S-l-RUT
- PARASITE
.----_ DRAG (COr\rTb)
_s4ray Dy
F r0l-n fye 7 . 2 , take the avera e value
9
- a . 0 2 0 f o r a btconvex set3 ton
St

Then, assunin 3 frnnttd -th,ekbess va’arres directly a s t.k yw awje

t =O.,OC + vc = c~o,10+ Ir)

and refetrloq the spray drag to t Le st t-d a r e a

.
c Dc~ = &pray = J&may x 3 = A c,. * A,!!
qct qch
Then
a CD* = & (O.lO + w) x 0 . 0 2 0 = 0 . 0 0 1 * 0 . 0 1 VI
h

_-5dOE FORCE_-& INGUCED DRAG

---

Fram ES uatlon (7.5)

dw + J& (1+3) 0.421
dc,‘$% 2.n(TXjc

dcs = 2.375
-2-F
F ram Equation (7.6)

cos
d = 2 l-1 +d = 0 I82
cpCs’ .?n (1+6) ’

TJJTAL CHARACTERISTICC,

Sde Force
cs 3 2 . 3 7 5 (+’

otag
c, = 0.093 + 0.001 + 0.093 (2.375 ul);L +o.oJ v to.182 (2.375YY

= 0.094 + 0.01 I# + 1.55 y2

 STRUTS

3. Characteristics in Yaw When Ventilated

Intention of Ventilation

As indicated in Section 1 above, ventilation is due to some

discontinuity in the fluid flow at the water surface. In the case
of the yawed strut, it apparently occurs when the angle of yaw,y/,
exceeds the entrance angle,@ , of the strut section at the surface,
as indicated by the test data of Figure 7.4. As has been stated in
Chapter 3, the angle, p , is
readily determined for a
section with sharp leading
edge, but has no exact ARCFORM SECTION

definition for an airfoil

section. It appears reason-
able to use the angle formed
AIRFOIL SECTION
by a circular arc tangent to
MEASUREMENIT OF ANGLE /3
the section at the point of
FIGURE 7.5
maximum thickness and pass- - .
ing through the leading edge, as shown in Figure 7.5.

Lift and Drap;

The characteristics of a fully ventilated strut are similar to

those of ventilated foils, as treated in Chapter 3, Thus for a flat

II - 7.16

_ - ..- _ .-.._ ___-_-__ ~ -..._ -.- - ---- - ---- ----.---

 STRUTS

plate strut with no flow across the tips, the normal force is

CN = 2-K s\n ly (7.9)

4+ mnq
For a surface-piercing strut with foil attached at the lower tip,
there is no cross flow at the tips and equation (7.9) should apply.
For a free-ended strut, there is cross flow at the free tip and the
resultant normal force is somewhat reduced.

In practical applications, strut sections are not flat on the

pressure side. The angle v in equation (7.9) must therefore be
measured from the angle of zero side force, which is indicated in
Chapter 3 for ventilated foils to be the angle at which the trailing
edge on the pressure side is in line with the fluid flow. For
symmetrioal struts, this is the angle of run (the angle for the
P
trailing edge). Thus, the equation for side force in ventilated con-
dition is

G V
= h sm vcos q’ (7.10)
4 + n-srn p’

where kV is a factor depending on the conditions at

the strut tips (1.0 for no cross flow)
v' is the yaw angle measured from the angle of
zero side force (in ventilated condiition).

The drag is not similarly reduced, however, bel.ng mainly dependent

on the frontal area of the strut exposed to the stream flow. Thus, it
is primarily a function of the actual yaw angle, modified to some pre-
sently unknown degree by section shape. Tentatively, it is considered

_,--.--_.. - .~
 STRUTS

that the drag component of the basic flat plate normal force be

applied (modified as required for free tips). Thus:

(7.11)

where Ly is the actual yaw angle

G* is the parasite drag coefficient of the strut

(including spray). The friction drag in this
case is l/2 that in regular flow since only
one side of the strut is wetted.

Comparison with Test Data

For the 12% double arc struts used in the E.T;T.S tests, the

entrance angle is - 13.9" (for leading and trailing edges). The

P
data of Figure 7.4 indicates the inception of ventilation to occur

between 1L" and l!?'.

The lift data in ventilated condition shows a reasonable fit with

equation (7.10), using a factor hv = 3/4an d measuring from the angle

of 13.9".

For the drag in ventilated condition, $ 0 was taken as 0.006

(l/2 of that for the non-ventilated case). Using the factor h,, - 3/k

derived above, equation (7.11) is shown in the figure, and appears to

be a reasonable estimate of the drag.

As previously stated, more test data on a variety of struts under

various conditions is required before more reliable design formulas
can be advanced and reliable factors established.

II - 7.18

_____--
-----.l”----.-
 STRUTS

4. Height of Spray

The spray produced by a strut is primarily a fhnction of the

frontal area of the strut (projected area normal to the fluid flow)

in association with the speed of advance. Hoernerl has analyzed avail-

able data on spray height, for struts at various angles of yaw. Hi8

results can be summarised.? in the following formula.

(7.12)

where h, is the m&mum height of spray

=S is the strut chord

Fx is the Froude number based on forebody length

f5 is the Froude number based on strut chord

Pa, is the strut thickness ratio

y/ is the angle of yaw.

Equation (7.U) must be considered preliminary, being based on a

minimum of data. In particular, the latter terms which include. yaw is

based on tests run at low Reynolds number (Rr 16) and Froude number

( Gs = 2.9), and may not represent conditions at usual operating

speeds.

c
II - 7.19

-------. --_____
 .; .Ilr STRUTS
.-
References

1. Hoerner Wome Characteristics of Spray and

V Ventilation", BIW Corp. by Gibbs--_&
Cox, Inc. TechnigallReport No. I&'\
Contract No. Nonr-~5O'l(O~~%~~.

2. Coffee and McCann ti ~*Hydrodynamic Drag of 12- and 210

Percent Thick Surface-Piercing
Struts', NACA Technical Note 3092,
1953.
3. Kaplan \ t!Tests of Surface-Piercing Struts"
E.T.T. Letter,Rpt,, No. b88, 1953.

4. Hoerner "Aerodynamic Dragfi

Published by Author, 1951.

5. Ding;ee Wharacteristics of Four Skids and

Y' a Surface-Piercing Strut", ETT Letter
Rpt. No. 472, 1952.
_-

II - 7.20
.
 CHAPTER 8. RUDDER CHARACTERISTICS

1. Description of Types
2. Characteristics of All-Movable
Rudders
3. Characteristics of Flap Rudders
4. Rudders on Hulls

The types of rudders likely to be employed on hydrofoil craft

are discussed and categorized. It is shown that rudder character-
istics can be derived from pertinent foil and strut properties
presented in previous chapters.

II - 8.1
 RUDDERS

,l. Description of Types

Rudders on hydrofoil craft may be considered in two broad

categories: first, as applied to the foil configuration when in flying

attitude; second, as applied to the hull, prior to take-off.

For some simpler configurations, one rudder or system of rudders

is used for both purposes, whereas in others independent rudders are

used for each type of operation. The choice lies in the selection of

the configuration to be employed and is a matter of overall design.

However, the pertinent hydrodynamic characteristics of the rudder

depend on its relative location and intended function.

Thus, rudders that are essentially part of the foil configuration

and are used in foilborne operations can be treated in accordance with

the foil principles set forth in previous chapters. For rudders that

are essentially part of the hull configuration (being used primarily

for hull .borne operations), the various influences of the adjacent

hull must be taken into account.

There are two main types of rudders to be considered: first,

the all-movable rudder which is also referred to as a balanued rudder

(since the pivot point can be located at the center of pressure), and

secondly, the flap rudder which is located behind a streamlined post

and is essentially a flap pivoting behind a fixed strut (similar to

II - 8.2 -
?qlmB RlJDDERS
- -

a flap on a foil). The character:-

istics of each type also depend on

whether the rudder if fully sub-

I ALL MOVABL E
merged or surface&piercing.

There are variations which ) ----czT&

combine both types to some degree, FLAP

the characteristics of which can

B A S I C AUODEH T Y P E S
be reasonably determined from

those of the basic types. FIGURE 8.1

II - a.3

.ll~--.^ .._,.” .-.-~ --- I.

 RUDDERS

2. Characteristics of All-Movable Rudders

Fully-Submerged Rudders

Rudders that are fully submerged are essentially considered to have

the same hydrodynamic characteristics as a foil and thus the lift or

side force, drag and moment of the rudder can be readily determined by

the principles set forth in Chapters 1 and 2.

Since fully submerged rudders are usually of small aspect ratio,

the lifting surface theory should be applied.

Reference should therefore be made to those chapters for the

methods and formulas to be used, with the following changes in notation

to be employed:
C, the side force coefficient, instead of 6

6, the rudder deflection angle, instead of d

hR, the I'span" of the rudder (height).

Surface-Piercing All-Movable Rudders

All-movable rudders that extend through the water surface are

essentially surface-piercing struts, insofar as the hlydrodynamic

characteristics are concerned. Thus, the material given in Chapter 7

for struts can apply directly to the rudders of the same type, the

only distinction being that the rudder is referred tot

_-~
II - 8.4

--____ ._-.-----
 RUDDERS

d&, the rudder deflection angle instead of

the strut yaw angle.
Y#

This applies to ventilated as well as non-ventilated condition,

including the end-plate effect of any foil, nacelle, end plate, etc.
on the lower tip.

II - 8.5
 RUDDERS

3. Characteristics of Flap Rudders

As illustrated in Figure 8.1, a flap rudder is essentially the

after portion of a strut so arranged to pivot (as a flap) behind the

fixed forward portion, As such, it is similar to a foil with plain

flap and its hydrodynamic characteristics can be determined from the

principles of flaps given in Chapter 5.

Side Force of Flap Rudders

Thus, from equation (5.4), with appropriate changes in notation

(and noting the angle of zero lift is zero for symmetrical sections,
as usually employed on rudders), the side force can be given for sub-

merged or surface-piercing rudders2

VW

where Cs is the side force coefficient, based on

total side area of the strut-flap

3 is the side force curve slope of the unflapped

(W 1 strut, as determined above or in Chapter 7, for
the submerged or surface-piercing case, as
required.
v is the angle of yaw of the fixed strut section

is the flap deflection angle

4
rh is an effectiveness factor.
j

-
II - 8.6
 lUJDDEFC3

The effectiveness is given in Chapter 5 as

where g/c is the flap chord/total seotion chord ratio.

Equation (8.2) ia considered valid for flap deflections of less than

I.!? and flap chords less than 708 of the total ohord.

Other Characteristios

&meralPy, then, all of the flap rudder oharacteristioa oan

readily be determined from the principles set forth on flap8 in

Chapter 5 in conjunotion with the basic strut charaoterfstios set

forth in Chapter 7. Hfnge moments, partial-span flaps, drag, rtq.

oan all be obtained with the same degree of accurao,y as for foil flap&

Ventilation

The ventilating charaoteristics of surface-piercing flap rudders

are expected to be somewhat different from those of struts and ati-

movable rudders e It is considered that since the flap rudder gets

its Sift due to change in oamber instead of change in angle of attackp

the flow aoroas the leading edge of the section ia not as great as in

the other oases and thus the flap rudder should not ventPBate aa

readily (Le. it should be capable of higher side foroe prior to

rentflation). There is no known data available to suptort (or disprove)

this codention, and therefore tests are required b'efore design

formulas can be presented.

I I - 8e7
 RUDDERS

4. Rudders on Hulls

Auxiliary rudders are sometimes employed to provide steering when

the craft is hull borne, prior to take-off. Such rudders are attached

to or in close proximity to the hull and usually rise clear of the

water after take-off.

For the type of hull usually employed in hydrofoil craft, these

rudders would be located beneath the hull. Thus, no consideration need

be given to surface-piercing rudders for the low speed range (low

Froude and Reynolds numbers) prior to take-off.

Hull rudders may, therefore, be treated as lifting surfaces in

an infinite fluid, and the theory of small aspect ratio foils can be

employed to determine their hydrodynamic characteristics, taking into

account the influence of the hull.

Theoretically, the influence of the hull is to d'ouble the

effective aspect ratio of the rudder, when the rudder is attached

directly to the hull (zero clearance). As the clearance increases

the effectiveness is decreased, so that at about a clearance equal to

l/2 the rudder height the rudder is free of any hull effects. (This

can be shown quantitatively, from the treatment of air gap in a wing,

given by Durandl). To what extent viscous effects influence the

rudder eff'ectiveness (boundary layer effects on the clearance and on

._ ---. .-~--
 KUDDERS

the rudder itself) is not known specifically and cannot be generalized

due to the variety of possible arrangements on the hull-rudder geometry.

Kuch information on the characteristics of hull rudders, the effect

of various influences and practical approaches to design can be gotten

from various marine sources 2,3".

II - 8.9
 RUDDERS

References

1. Durand llAerodynamic Theoryl', Vol. II

California Institute of Technology,
1943.
2. Rossell & Chapman "Principles of Naval Architecture",
Vol. II, SNAME, 1941.

3. Mandel Y3ome Hydrodynamic Aspect of

Appendage Design?,, SNAME
Transactions, 1953.

II - 8.10

-... ---____
--_1--1__
____-- -,------
-__.-- -. .-. -.
 CRAPTER 9. CHARACTERISTICS OF NACELLES AND SMALL APPENDAGES

1. Nacelle Size and Shape

2, Characteristics of Isolated Nacelles

3. Characteristics of Nacelles in a Foil

Configuration

ho Drag of Small Appendages

The typical geometry of nacelles used in hydrofoil application

is indicated and data on the lift, drag, and pitching moment of

isolated nacelles is given. The characteristics of various nacelles

in configurations are discussed qualitatively and procedures for

determining and using various data are given. The drags of misc-

ellaneous small bodies are tabulated,

 .
. . .‘-L NACEUES
-

1, Nacelle Size and Shape

Sn hydrofoil applications, nacelles are used primarily for pro-

pulsion and transmission machinery. Characteristkally, nacelles are

streamlined in shape, with cross-sections varyfng from cfrcular to

rectangular (at their largest sections) which generally are faired

into :rounded noses and conical tails,

The hydrodynamic characteristics of nacelles are commonly ex-

pressed in coefficient form, usually based on the wetted surface of

the nacelle (in marfne applications).

The drag of a long slender nacelle (submarine, airship,

etc.) with its axis in the direction o.f flight is pri-
marily due to friction, and thus the wetted surface is
the most important reference area in this regard.

However, it is conventient in most cases to refer the nacelle

characteristics to the frontal area (maximum cross-section), which

is more readily determfned in prelfminary investigations and which

is important in evaluating the optimum nacelle g,eometry for a given

applfcation,

The ratio of the wetted area to the frontal area is approximately

given' for most nacelle shapes as:

(9.1)
 .” - NACELLES

where e is the length of the nacelle

d is the diameter (or equivalent diameter for non-

circular shapes) at the maximum cross-section

&& is wetted surface of nacelle

SD is frontal area (mzbnum cross-sectional area).

-_---_ -_---- _.I ..- . . . ..-. --.- --___- ~_I_ --.-._ - .-.. ---.. __ .-_
..-- g-l-_ -_.,_--..--.--.~.
 NACELLES

2, Characteristics of Isolated Nacelles

Parasite Dran with Flow Parallel to Nacelle Axis

The parasite drag of a nacelle in an inffnite fluid is composed

of friction drag and pressure drag. For a long slender body (hfgh l/d )

the frfktion drag is most important; for a blunt body (low e/d ) the

pressure drag is predominant, Thus, the total parasite drag is a

function of the various shape parameters of the nacelle.

Experimental data on a variety of aircraft bodies (airships,

fuselages, etc.) is available in N,A.C.A. reports4%. Gertler* gives

results for streamlined bodies of revolution of high p/d ( d/d=

4 to 10). From such sources, the parasite drag can be accurately

determIned,

From an analysis of such data, Hoernerl gives an empirical formula

which Ican be used for preliminary purposes with good accuracy0 In

terms of the wetted area, this formula is

%et = c+p+z 1- dh + 6(d4)4] (9.2)

and by substituting equation (9*1), the formula in terms of the frontal

area isr

* See N.A.C.A, Yndex of NACA

Technical Fublicatfons", 1949

_.-
 where is the frictional dr~~ir_ coef‘ficierit based on
cs
wetted area (see Chapter ,!I) at. the aparopriate
R.e.ynolds number based on bhe 1enp;th of the
nacelle.

42 is the length/diameter ratio, as indicated

above.

Figure 9.1 (taken from,rcPerence 1) shows the frontal drag

coefficient as given in
0. IO
equation(9.3), and as
0.08
derived from experimental
0.06
data. The optimum p/d is
0.04
seen to be close to 2 for
0.02
such isolated nacelles. (As
o t
shown in the following 0 2 4 8 IO

&d6
section, the optimum 4
/d of NACELLE FRdNTAL ORAG COEFFICIENT
the nacelle is larger when FIGURE 9. I
in conjunction with a foil.)
 NACFLLEZ
-~-

Wave
- -Dr,sg - of -Nacelles

In his tests on streamlined bodies of revolution, Gertler"

also determined the wave drag of bodies with -e/d= 7 at sub-

mergencles slightly greater than 1 diameter. Prom this data, it is

indicakd that the wave drag component is appreciable at low Frroude

numbers (based on the body length) but is negligible above Froude

numbers of 1.0 to 1.5. Figure 9.2 shows the results for one of the
bodies tested, from which estimates may be made for other cases in
the absence of specific test data.

6.0 h
DEPTH
'h/d
DIAMETER ,
=j.24 1 1 1
5.0 .

4.0

“%M

3.0

2.0

1.0

FlGURE 9 . 2
 . NACELLES
.

Characteristics at Angles of Attack

Nacelles may be considered to be foils of small aspect ratio in

analyz.ing the lift, drag and moment character&tics at angle of attack.

Howeve:r, due to the shape of the body, the large %hfckness" and
different "tip conditfonsl', such analysis would be qualitative only;

and rellable values ean only be determzined from experimental data, as

is available in reports on airship and fuselage characteristics (see

NACA Index of Technical Publfcations). Where the submergence and

Froude number are small, tank tests should be run fn order to determine

the characteristics more exactly as influenced by the water surface*

Figure Ye3 gives some representatfve data, as shown by Hoerner',

and as given by Jacobs and Ward3 for several fuselage shapes. The

lift and drag coefficient s are based on frontal area, The moment

coefficient about the quarter length is based on frontal area and

fuselage length,

-. ..--. -. .-- -- ---___.*.___ __-__... _.--__ -_-

 FROM REFERENCE I.

CIRCULAR NACELLE RECTANGULAR NACELLE

&&6,7 R=3X106 -e/b 6 . 9 R=106

EFERENCE 3.
R = 3.1 x IO6
NOTE : MOMENT COEFFICIENT
BASED ON NACELLE
.50’ LENGTH

-40 -40

.30 .30

20 .20

.I0 .I0

.O 0

.02 .05

.04 .I0

.06 .I5

.20
~0800 4“ 8” 12.’ id” 0” 4“ 8” 12” 16”

RECTANGULAR NACELLE CIRCULAR NACELLE

&d = 5.06 L/d = 5.06

CHARACTERISTICS OF NACELLES
FIGURE 9.3
- - --_.,;...
 NACELLES

3. Characteristics of Nacelles in a Foil Configuraticx -

General Considerations

When superimposed on a foil configuration, the nacelle exhibits

somewhat different characteristics than when isolated, There fs a

mutual interference effect on both foil and nacelle, which is re-

flected in different values of lift, drag and moment for the entire
configuration.

In the analysis of such configurations, the standard

procedure is to maintafn the total foil area, as if
the nacelle were not present, The superposition of a
nacelle which replaces some of the fofl will actually
reduce the foil area, resulting in a re!duction in foil
- parasite drag from that basically considered, There-
fore, it must be kept fn m%nd that the nacelle char-
acteristics determfned from tests, wherein the basic
foil drag is deducted from the total configuration
drag, consists of several additive components (the
basic nacelle drag, the interference drags on both
wing and body, and an induced drag Increment due to
change in foil lift distribution at the nacelle)
and a deductive component (the parasite drag of the
foil enclosed by the nacelle),

The total effect of superimposing the nacelle on the foil cannot

readQf be ascertained for the general case, due to the large number of

factors involved, Several factors ares

(a) the characteristics and geometry of the isolated nacelle

(b) the characteristics and geometry of the foil

(c) the angle of attack
 NACELLES

(d) the relative size of nacelle and foil

(e) the relative locatfon of nacelle in vertical,

horizontal and spanwise directions

(f) the extent of fairing at the fofl-nacelle Junctions.

Qualitative Results of Tests

Examination of data on wind-tunnel tests of .various nacelle-foil

combinations does indicate certain trends, which are given for a

general guide in evaluating configurations,

(a) Typical airplane nacelles mounted on a wfng are of low

with an overall length of the same magnitude as t e
f%jchord Q The drag of such nacelles tend to be larget .

Nacelles and fuselages of higher e/d 9 and which also have

a length appreciably greater than the foil chord, show small
increases in nacelle drag depending on the vertical location
of the nacelles tith respect to the foil. Hoernerl shows
the nacelle drag for various "high-wingff arrangements (the
results of one analysis are shown in Figure P,l), which
indicates the optiimum .&d for the nacelle to be nearly
7.0, and the drag for nacelles of larger .&'& to be about
20% higher than when isolated from the fo3.1.

(c) For similar nacelles and fuselages as jin (b) above, but with
%id-wing" arrangement, the net nacelle drag fs slightly
decreased according to tests conducted by Jacobs and Ward3.
This condition generally holds true for the range of vertical
positions, where the drag would be generally as indfcated fn
(b) above, For the complete characteristfcs of these "mid-
wing" configurations, Jacobs and Ward (observe that the simple
adding of the independent foil and nacelle characteristics
(lift, drag and moment) shows very good agreement with the
data (without further consideration of interference, etc.)

II - 9.10

..-
 NACELIJZS

(d) The addition of suitable fairings at the foil-nacelle

junction tends to reduce the drag appreciably, particularly
for high or low wing arrangements.

Recommended Methods

For accurate determination of the characteristics of nacelle-foil

configurations, tank tests should be run on the contemplated configur-
ation through the range of spsed, attitude, and submergence anticipated
in the design. Under certain conditions, data taken from pertinent
wind-tunnel tests on similar configurations may be, suitable.

However, for preliminary purposes, a simple approximation may be

used that ie considered reasonably accurate for hydrofoil applicationst

(a) The parasite drag of the configuration is simply that of the

isolated nacelle (at zero lift) added to the rest of the
configuration.

(b) The lift and moment characteristics are considered those of

the foil alone, when the nacelle quarter-length is in line
with the foil quarter chord*

This procedure is recommended only when reasonable fairings are

employed at the foil-nacelle junction,

- .-- ________-
 NACELLES
,c

4. Drag o.f Small Appendages

In a practical hydrofoil configuration, there is apt to be small

protuberances or isolated bodies installed for purposes of control,

instrumentation, access, etco In preliminary consideratfons these may

generally be disregarded, but where careful estimates of drag must be

made for performance or structural purposes their characteristics

should be noted.

Hoernerl gives a comprehensive coverage of the drag of many

bodies conceivably to be found in hydrofoil configurations, and

reference should be made to that work for the details.

As a general guide, Table ye1 gfves the drag characteristics of

some of the more common shapes.

Where such bodies are isolated from major components of the con-

figuratfon, the given values can be used. However, when attached to

a foil or strut, consideration must be given to mutual interference

effects, boundary layer effects, etc. in accordance with ptincfples

previously set forth.

ll__-
--.___. __I__y_(---- ---- -_,--_. . . . . .- _- . ..-- -.-
 NACELLES

TABLE 9.1
APPROXIMATE VALUES OF DRAG
COEFFICIENT FOR VARIOUS BODY FORMS

Form of Body

Tandem diisks
diameter) oe93
LO4
1.54
Rectangular plate (I to flow)
( L = length) 1.20
( 0 - breadth) 1.50
1.95

Circular cylinder (axis 1'1 to flow)

0091
0.85
0087
0099

Circu:Lar cylinder (axis .l to flow)

( L = length)
( 0 - diameter)
1,20

Hemisphere: Hollow upstream

Hollow downstream
( 0 - diameter)

Sphere ( 0 = diameter)

Ellipsoid (183, major axis I! to flow)

( D = diameter)
 NACELLES

References

1. Hoerner 'rAerodynamic Drag"

Published by Author, 1951.
-8L* Gertler "Resistance Experiments on a
Systematic Series of Streamlined
Bodies of Revolution - For
Applfcation to Hfgh Speed Sub-
marinew, DTMB Confidential
Report C-297, 19500

30 Jacobs & Ward "Interference of Wing and Fuselage

from Tests of 209 Combinations in
the NACA Variable - Density Tunnel",
NACA Report No. 540, 193so

4. Smelt, Smith, Smith "The Installation of an Engine

& Davison Nacelle on a Wingtl, ARC R&M No.
2406, 1950.
 CHAPTER 10. SKID CHARACTERISTICS~

1. General Considerations

2, Lfft and Wetted Length

3. Drag and Pitching Moment

The application of skids is discussed, and data on the lift

of skfds of various geometry is gfven. Drag and pitching moment

are shown to be obtained from the lfft values.

 1.. General
- - - Considerations
-

Small lifting components that plane on the water surface are

*
used on certain hydrofoil confi,gurations for the purposes of control

(sensing the water surface) and for the combined purpose of control

and lift (as on the Grunberg

type configuration). Such com-

ponents are essentially flat

plates of small aspect ratio

and are commonly known as skids.

The main feature of skids as com-

pared to foils is their high

- PLANIING O F
sensitivity to change in FLAT PLATE SKID
submergence. FIGURE 10.1

As applied to hydrofoil craft, skids are comparatively small in

size with consequent high Froude numbers, and operate at large trim

angles. Since typical planing hull data (referred to in Chapter 11)

is in the range of lower Froude numbers and low trim angles, such
hull data is generally not applicable to planing s.kids. Therefore,

specif:Lc flat plate data for the appropriate range of operating

conditions must be employed.

The notation used in presenting and analyzing planing data is

.
somewhat different from airfoil notation, although conversion is
 SKIDS

readily made. The important planing notation (using NACA Vtseaplanelt

coefficients) is as follows:

C'V= Y@ "speed coefffcfentl'

CA = w/,b3 "load coefficfentll

c Lb = wppb= lfft coefficient based on the square of

the beam (Cib L 2 CA/c," )

A = lqb "'wetted length ratio"

(the reciprocal of the aspect ratio)

where Z is the angle of trim in degrees

'sd is speed, ft/see

a, is mean wetted length of the plate, ft

(see Figure 10,l)

6 is breadth of plate (beam, span) - ft,

W Ss load on plate (lift), lbs.

p 5s the density of the fluid

9 is acceleration of gravity

w is the specfffc weight of the flufd lb/ftg

b =ps,

._-__s_-- - __-___-- ---_ ---_--.. ..--^-I__

 SKIDS

.-

2. L";.ft
-__f- and Wetted Length
.
At the present time there is no complete three-dimensional planing

theory avaflable which correctly predicts the lift9 wetted length, etc.

over the range met in practice. An empirical equ,ation has been derived

by ETTl which fs good wfthin certain limits of trim, loading, etc.

NACA* has extended the range of variables for the high speed case

(where buoyancy effects are neglfgible). Both sources also give the

Influence of certain variations from the flat plate (deadrfse, chine

flare;, etc.)

Data based on the NACA tests is presented in Ffgure 10.2, wherein

the Ylift coefffcient" 6 can be determined as a function of trim
Lb
angle and x o

Generally, since the wetted length Ls difftcult to pre-

determine, the normal design procedure is to determfne
6~ from the trim and balance of the entire craft. Then
for various trim angles, the wetted length can be
determined,
.
In addition to the flat plate data, Figure 10.2 shows the lift of

skids with various deadrise and local chine conditions as a percentage

of the CorrespondBng flat plate lift. Thus, & must be corrected by

the factor, bz. 9 for the pertfnent case as given in the figure,

These values of k are average, computed from the NACA

data for operating condftfons expected to be met by
skids, and are considered sufficiently accurate for
engineering purposes* For greater accuracy, reference
should be made to the origfnal reports2,
-.
 - .:

CHINE DETAIL
A . - - - -
’ PLAIN HORl Z. FL ARE VERT. STRIPS ’
I.00 1.00 1.00
0*

I-

V A L U E S O F !t F O R
VARIOUS SKID FORMS

LIFT OF PLANING SKIDS

FIGURE 10.2

._^_” .,.-.. ..-_.. .._ . II_ _-_

-- - ---.-~__~_
 SKIDS
i. --

30 Drag and Pitching Moment

The drag of the skid may be found by calculating the fnduced drag

and the frictfonal drag separately, although the latter is seldom

important in a practical ease due to the high trim angles (i.e. the

induced drag is predominant.)

The induced drag fs simply given by

since the dynamic pressures act normal to the plate. The frictional

drag is easily determfned (see Chapter b) once the wetted length is

found, as above.

From test data, the center of pressure is found to be about 70%

of the mean wetted length forward of the trailing edge, This value

may be used for all high speed skids without ser?ous error*

-l.-l..--.---.ll- I_.-
 ,
- , ? ‘$lmis
SKIDS

References

1. Savitsky 'Wetted Len&h and Center of

Pmssure of Vee-Step Planing
Surfaces", E.T.,T. - Available as
I,A.S,-SMF Paper No. FF-6 - 1951.

2. Series of NACA reports by Kapryan, Weinstein, Chambliss,

Boyd and Blanchard descrfbfng resul.ts of an integrated
series of tests in planing surfaces (19!;2).
NACA Technical
Note No. Deadrise Tested Type of Chine Detail

2804 20" Horz, Flare

2842 4o” HOPZ. Flare

2876 20" and 40' Plain

2981 0" Plain

3052 20' and 400 Vertical Strips

 CHARTER 11. HULL DRAG

1. aeneral Considerations

2. Basic Hull Drag

3. Hull Unloading
Characteristics

4. Air Drag

Reference is made to standard marine and sea:olane data for

determining the basic hull drag. Drag through the unloading range

can be approximated by a simplified method, as a .function of the

basic drag, as shown. A formula for the air drag is given, as a

function of the hull shape, based on existing data,

 HULLDRAG

1, GeneraIL Considerations

T’ne drag of the hull is an important component of the total drag

at speeds below take-off. At comparatively slow speeds when the foil

system provides virtually no lift (or for controllable foils when set
at zer,o lift) the hull supports nearly all of the required weight due

to its displacement and/or planing action (if any). Thus, the drag

of the hull for this condition must be determined,

Through the take-off range, the hull is gradually %riLoaded" by

the lift of the foils with consequent raising of the craft and re-

duction in hull drag. The hull drag in this range is most readily

expressed as a percentage of the drag of the fully waterborne hull,

as is indicated below.

The air drag of the hull becomes important at the high speeds

beyond take-off when the hull is completely clear of the water.

In an important hydrofoil craft design, model test data should

be obtained directly for the hull to be employed. However, for pre-

liminary purposes, the methods and formulas proposed herein should

be adequate.
-

2, Basic Hull Drag

Theoretical methods of calculating the resistance of waterborne

craft are availablel92, but the calculations are arduous and the

results are not as reliable as those obtained from model tests,

There are several systematic series of model tests available 3 3 4,s

on various types of hull form along with some collections of results67
J
on large numbers of specific designs, Table 11.1 lfsts some of these
sources along with a descrfption of the variations' tested and the data

reported.

Since there is a wide variety of hull forms possible

for application to hydrofoil craft, and since all of
the basic data is readily available in the referenced
works, it shall not be produced herein.

The methods of apply%ng the data in each case is given in the

sources, and are also described in standard references by,, The main

difficulty in utilizing the results usually occurs in the selection

of coefficients in the series to give a hull form as similar as

possible to the one in question. In this respect!, a general guide

can be given, based on the Froude concept of ship model testing, as

follows*

The resistance of the model is assumed to be separable into two

components; the "frictional9 due to viscous effects and depending on

 TABLE 11.1
SOURCES OF E?PERIMENTAL HULL RESISTANCE DATA

Displacement Ships Fullness, Length- P'Residualg' and

"Frictiond're-

Center of Gravity,
Beam Draft Ratio,
Length-Beam FM&o

Model Resistance Data

Sheets-160 Individual_
Ship and boat model v&z-up to 1,7 None Resistance

Flying Boat Models

5 100 individual sea- C, 0 - 15 None Resistance, Trim,
plane models Stability
 ,-
I-IULLDRAG
*’ -
.- .A

the wetted surface and Reynolds number, and the Vesfdualn which

influences the wave resistance and 'teddy-making" (separation effects)

depending on the Froude number, The model is tested at the same Froude

number as experienced by the full sized ship, and the model results are

corrected by assuming the frictional resistance to be that of a plank

of the same area and Reynolds number fn each case (ship and model).

Test data on planks is available for makfng this correctionl'.

The Froude and Reynolds numbers are defined on the basis of speed
and length as follows;

Froude number =

Reynolds number = "e/y

where v = speed of advance in ft/sec

a = length in ft,

3 = acceleration of gravity in ft/sec2

9 - kinematfc viscosity in ft2/sec

The length IIiIf is usually defined as the wetted length of the ship in

the direction of motion with the exception of some data on seaplane

models where (for use in the Froude number only) the beam "bBt of the

model is used, In ship parlance the Froude number is usually

abbreviated excluding the constant 1P g )t and using the speed in knots

stv II. Definitions of these quantities are as followsr
k
 HULLDRAG

Categoryo Ships Seaplanes

Quantitya "Speed-Length Ratio" YSpeed Coefficientvl

.
Symbol: vk/JI; * CV

Definitfon: v&z = 33 v/ g M-3

T:he general guide, therefore, fs to attempt to match the hull

form c:haracteristfcs influencing the wave or residual resistance such

as the fullness and Froude number, and to make corrections for differ-

ences in frictional resistance (especially wetted surface). A

knowledge of the basic mechanism of resistance is indispensable in

this respect.

.-

JC "I," and llfi tt have the same meaning

 HULL
--- DRAG
-

-
- DESIGN EXAMPLE N0.F
ESTIMATE THE HULL RESISTANCE OF THE
FOLLOWING PLANING HULL FORM

L E N G T H (B.P.1 62.5’
6EAM 16.0’ HULL IS S’AME AS THAT
DISPLACEMENT 50 TONS GIVEN IN LIP’ENDIX A
L.C.G. 4.05’ AFT % >
SPEED 30 K N O T S - NO UNLOADING CONSIDERED

SERIES 5 0 PLANING D A T A ( R E F E R E N C E 2) i s usED.

DATA IS PRESENTED ON RESISTANCE , TRIM, L.C.G.
AND WETTED SURFACE FOR A SYSTEMATIC SERIES
OF 40” LONG MODELS.

A YYPICAL DATA CHART IS SHOWN BELOW. KNlD’rVING THE

HULL C H A R A C T E R I S T I C S ( SPEED ,DISPLACEMENT, LENGTH, ETC.)
VALUES ARE READ DIRECTLY FOR THE MODEL AND
C O N V E R T E D A S R E Q U I R E D F O R T H E F U L L SIZE CRAFT.

-.----
...-~--- .I--
---1-.11---.-I_ .^.
4
.

HIILL
---.-IDRAG

iDESIGN EXAMPLE NO, I I ,I1 I SH 2 OF 3

Fl?OM J’HE HULi PARTICULARS~ T H E LCG IS 5 6 . 5 % O F

LIENGTH AFT O F T H E FOR’0 PERPENDSCULAR. F R O M SERICS 5 0 ,
THIS CORRESPONDS TO ZERO SYATIC ?‘RfbtJ (r*O ON THE CHAWS)

THE DOSPLACEMENT LENGTH RATIO Of THE HULL 6

= 200
(&j3
C~HlCH I S O U T O F T H E M O D E L T E S T RAN’GE. H O W E V E R ,
fsy coNStOERlNG THIS To BE 20 % OVERLBAD CONDlTlON,
A N D T H E R A T I O T O B E WSED 95

.j&-y =‘67

O N -lHE C H A R T S F O R D =N+20%
-r-HE CORRESPONDING BEAM/DRAFT RATIO 1:; FoiJND ( IN ‘THE
:SERlES 50 T E X T ) T O B E
S/M = 4.77

‘WE S P E E D L E N G T H WA410 1s
= 3.80
vdii-

tHEF4, FROM THE CHARTS THE FOLLOWI NG INFORMATION

I S TAkEN :
[;r=O” j A =N+20% ; p/(L/;60)” = 1667

VjyL Alp rrp s>

3.0 2.47
3.5 0.175 7.2
4.0 0.080 4.3 2.03
INTERPOLATING

3.8 cl.378 1. 3 2.12

__) -.-._ .- --
---DESIGN EXAMPLE NO. I I.17
THE MODEL RES\STANCE DATA hlU5TNOW BE CORRECTED
FOR REYNOLOS F90.EFFECT~AhlDP,SERVtCE ROUGHNE5S '
FACTOR ADOED TO GET WE CORRECT RESf,TArJCE VALUE
FOR THEFUIS SIZE CRAFT

MODEL LENGTH 3.33'

WEIGHT t6.6# [FROMAk~/too)3; zoo7
SPEED Il.7fp.S. [FI%M‘IK/'fl= 3.87
REYNOLDS NO. 3.28~10~ [FM (6, 69'7
9 0.00354 [TURBULENTJ
FLU SIZE REYNO1DShlO. 2.23x108 [SW. tii159"J
5 O.OO\ 86
ROUGHNESS AUOWANCE 0.00040
C$ CROW-I) 0.0022’6
THU5,TIiE MOOEL REWTANCE MW BE REDUCED AN
AMOUMT EQWAKNT TO
(AC& OoOO128
THIS CORRESPONDS TO
(AR) 3 AQ X& V2 S (model)
= 0.3-P
(my& = oq& = o.oz:z

THEFS,THE FULL SIZE REWLTS ARE

WA = 0.\78-0.022
=O.t56

HULL RESISTANCE =0.156xI12,000 -17,500*

HULL TRIM 17.30

 HULLDFfAC

3. Hull Unloadine Characteristfcs

The drag of the hull during the %nloadingn process prior to

take-off is a function of many variables and cannot be readily

determined except when the hydrofoil configuration and operating

characteristics are specified, Model tests11912 have been conducted

on several hull models at various specifted unloading speeds and angles

of trfm, but the results are not necessarily appl:'Leable to general

cases since the trim fs difficult to predetermine (except for fully

controlled configurations),

Therefore, model tests of the proposed configuratfon should be

conducted for accurate determfnation of the hull unloading eharacter-

istics, For prelfnfnary purposes, hull unloadfng characteristics can

be approximated by the method proposed by War$3 for both displacement

and plating type hulls. Since planing type hulls are generally used

in hydrofoil craft application (for moderate size waft, at least),

the approximate unloadfng eharacterfstfcs of the planing hull are

proposed for use.

It is assumed that the hull Ss an essentially flat hull planing

at an angle of trim, and that the foil system unloads the hull at the

center of gravity, Tt is a characteristic of planfng that the center

of pressure location relative to the wetted length remains fixed

regardless of the trim, and that the lift itself ,is roughly

.--llll_-- -- “--. --.-.--- ,___ _I_-..-_..--.

 .
HULL DRAG
-

proportional to the trim. This causes the craft to pivot, about the
forward edge of the wetted area during the process of unloading, as

indicated in Figure 11.1.

.- NO TRIM OR LOAD
7
100% UNLORDING

ILLUSTRATING SIMPLIFIED UNLOADING

FtGURE I I.1

It is also evident that the ltresidualll resistance is equal to the

component of dynamic lift in the horizontal plane and therefore

proportional to the load on the hull and the angle of trim. The
drag (at a fixed speed) may, therefore, be related to the drag at

zero unloading as follows;

.-.
 HULLDRM

"Resfdualfl "Frfctional"

R’/R.,. wdo .ITo

=

combining2

where R - hull resistance

id - loading on the hull

f = angle of trim

5 - wetted surface

3ubscrfptss
V - residual

f = frictional

o refers to values of the parameters at zero unloading.

The assumed unloading Curve is therefore as given in Figure 11.2.

It should be noted that the resistance at zero unloading must be broken

down into frictional and residual components which is usually standard

procedure in resistance tests on series results, as indicated above for

the basic hull drag. The assumption 5/s, = 1 naturally does not hold

at low values of w/d6 and in fact S/s, must be equal to zero when the
 .
-
: f=iw!lz HIJLL DRAG

hull is completely unloaded. The dotted line on the graph (Figure

11.2) represents this transition. The graph is good only for one
speed & since the ratios ani R&/~ 0 will in general
vary with speed.

o ET.T. E X P E R I M E N T A L D A T A
- ASSUMED CURVE

I.00 r I I I I R

TYPICAL UNLOADING CURVES

F I G U R E 11.2

Tests points are shown on the graph which were derived from

unloading experiments on a series 50 hu1112. It j.s seen that the

assumed characteristic variation of resistance with unloading is

at least approximately correct and sufficient for engineering

applications.

.c

I I - 11.13

-,---, .^ -____
 :, r“lY, HULLDRAG

At the high speeds associated with foil-borne operations of

hydrofoil craft, the air drag of the exposed parts of the craft (the

hull and superstructure, primarily) must be taken into account.

However, while the air drag is important, it is generally a small

percentage of the total and may be approximated b,y a simple ex-

pression with reasonable accuracy.

Thus, the air drag may be expressed as:

(11e2)

where Dalr is the air drag (in pounds)

is the density of air (0,00238 # sec2/ft4,

P at sea level)

S is the cross-sectional or frontal area of

the hull and major superstructure, ft2

v is the speed in ft/sec

CD is the drag coefficient based on the area S,

and is to be determined,

This may be more conveniently expressed in terms of the speed in

knots (v~)~ or4

Duw = K 5 VKZ (where '& = 0,,0034 ct, ) (11.3)

-
 HULLDRAG
. -
--

Eiggert3 used a drag coefficient of 1~18 for ships based on an

assumed cross-sectfonal area of l/2 B'(where B is the beam of the

ship). However, thfs area fs probably less than the actual maximum,

incluclfng the superstructure, for such cases,, Furthermore, there is

no attempt at streamlining in the case of large ships as there is in

the typical fast craft such as a hydrofoil boat, The effect of stream-

lining is by far the most important consideration in determining the

drag coefficient, It is proposed that a coefficient of 0.60 (which

would check with Eggertgs formula using sz for the area instead of

l/2 8') be used for blunt ended hulls and superstructures 14 p and 0.30

for well streamlined configuratfons (seaplanes, with no houses, have

coefficients as low as 0.20). The corresponding \( values are there-

fore ,,002 and .OOl respectfvely,

The recommended formula for the air drag is then

DCLIY = (o.ool to oloo2) s v; (11.4)

where the choice of the factor depends on the amount of streamlining

fncoroorated in the hull and superstructure,

 HULLDRAG
_-

References

1, Lunde "On the Linearized Theory of Wave

Resistance for Displacement Ships
in Steady and Accelerated Motion",
SNAME Transactions, 1951.

2. Guilloton ItPotential Theory of Wave Resist-

ance of Ships With Tables for Its
Calculationt', SNAME Transactions,
19.510
3. Taylor stThe Speed and Power of Ships",
U,S. Goti, Prfnting Office,
Washington, D.C., 19h30

4. "Tests of Twenty Related Models of

V-Bottom Motor Boats", EME3 Series
50, D.T.M.B, Rpt, R-47, Mar. 1949
Revised Edition.

5. Davidson & Locke "General Tank "ests on the Hydro-

dynamic Characteristics of Four
Flying-Boat HuILl Models of Differing
Length-Beam Ratio", NACA Wartime Rpt.
W-105 (orfgina:Lly ARR 4.F 15 June 1944).

6. "Model and Expanded Resistance Data

Sheets wfth Explanatory Notes", SNAME
Transactions, NOV. 19L1, NOS, 14.60..

7. Locke "A Collection of the Collapsed

Results of General Tank Tests of
Miscellaneous l?lyfng-Boat-Hull
Models", NACA Tech. Note No, 1182,
March 1947.

8, Russell & Chapman "Princfples of Naval Architecture",

Vol, II, SNAME Transactions, 1941.

9. Murray ItThe Hydrodynamics of Planing Hulls",

SNAME Transactions, 1950.

.-

II - 11616
. m HULLDRAG

10 0 Wnffor-m Frocedure for the Cal-

culation of Frictional Resistance
and the Expansfon of Model Test
Data to Fu91 SBae'~, SNAME Bulletin
No. l-2, August 19LS.

11. Sutherland *%xploratory Model Tests for

Engineering Desfgn of a Hydrofoil
Vessell', E,T,T, Conffdentfal Rpt,
No, 40?$ X&lo
12, Kaplan QlHull Charaeterfstfes for Four
Basic Hull Types", Confidential ETT
Ltr. Rpt. No, 449, 19%'.

113 * Hoerner & Ward 81Take=Off Performance of Hydrofofl

Crafto'j Tech, Rpt. No. 6, Contract
Nonr-507(OQ), Bath Iron Works Corp.
(Gibbs & Cox, Inc.), 195'2,

tin... Foster & Pfziali llWind-Tunnel Investigatfon of a

lJi.6 Scale Model of a Hydrofoil
Landing CraftB7 9 DTMB ConfSdentfal
Aeso Rpt. 852, 1953.
.

CHAPTER 12, ASPECTS AND INFLUENCE: OF CAVITATION IN

THE H'YDRODYNAMIC CHARACTERISTICS OF HYDROFOIL BOATS

Notation

1. Physical Meohanism sf Cavitation

7
2, Pnoeption of Cavitation

3. Forces in Cavitating Flow

A preferred field of applfeatfon for hydrofoil boats is in

higher speeds e In fact, realization of speeds above .some b0 knots is

one reason for the development sf these boats. A% such speeds,

cavitation is ino longer avoidable, both fm the propeller and in the

hydrofoil system* Consfderation of cavitational effeots upon per-

formance and behavior is, therefore, necessary4

A basic review of oavitation fs presented by Aekeretl, a more

recent and more detailed analysfs is gfven by Eisenberg2, and an

extensive bibliography is available fn reference j. Thfs present-

atj.on here, reiterates the physical mechanism of eavltation, mentions

delay in the onset of eavftatfon in oertaiin conditions, makes

predictions of the critical speed (inoeptfion of oavitation) in hydro-

fo%ls and presents some force data ',n caviltatfng flowe

 CAVITATION
-

Notation
cl bubble or body diameter
chordwise distance
c foil chord
t thickness ,
f height of camber
b wing span
h submergence
A aspect ratio,
half apex angle of wedges or cones
air content of water
angle of attack
weight density
mass density
speed in ft/sec or knots
dynamic pressure, c0.q V2)
static pressure
static pressure coefficient, Ap/'$,
cavitation number, (pom~-p vop*r?/+
critical cavitation number
drag or resistance
drag coefficient, D/Q
lift.3 also subscript for lift
lift coefficient, Lf4S
Reynolds number on A w/y,

-
 GAVITATION

1. Physical Mechanism of Cavftation

The compressibility of water is negljigibly small when compared

w%th conditions in afro However, as a fun&Son of speed, water flow

reaches a critical condition characterized by vaporization and

cavitation.

Cavitation Bubbles

Vaporization is the growth of a gas- or vapor bubble. There is

-----....+/.'
an equflfbrium in such a bubble between a force (proportional to the
projected bubble area &/4 1 corresponding to the pressure differ-

ential between inside and outside, and a force (proportional to the

circumference d q) due to the surface tension of the bubble (see

reference 1). For a given pressure differential, thereforea

Suction -Diameter (124

Ten&on

This function means that a critical or minimum diameter is necessary to

assure growth of the bubble.

In %leanM water, there are no bubbles or other cavities existing,

! , 4 fl, c' " "I~';-, ; .% t L ,
As a consequence, water which has been distilled does not'readily boil
3
at the temperature which is commonly called boiling temperature; and
St does not Wupturel' or cavitate upon reaching the "vapor pressure"

corresponding to temperature.

II - 12.3
 CAVITATION

If certain nuclei (gas bubbles of microscopic or macroscopic

size) are available, however, it is to be concluded from equation (12.1)

that there is a critical (negative) pressure at which the diameter of

the bubbles is just sufficiently large to permit growth. Upon reaching

critical conditfons, the bubble may suddenly increase its diameter.

In concluding, a certain air bubble content or existence of gas-

carrfing particles is a necessary prerequisite of cavitation.

Air Content

Water with a "free" surface has the natural characteristic of

absorbing afr, There is a maximum amount of air which nsaturatedtt

water can retain* Under standard atmospheric conditions, for example,

corresponding to 59°F and 760 mm Hg, the saturated air content is in

the order of 2% by volume, which is only a ratio of 25/106 by weight.

It is the content of bubbles, rather than the dissolved amount of air,

which makes cavitation possiblea. One way of prclviding such bubbles is

locally reducing the static pressure by placing a body in a flow of

air-saturated water, In certain preferred places;, the water becomes

over-saturated, thus shedding bubbles,

Accordfng to basfc experiments4 ) cavitation starts at pressures

close to vapor pressure provided that the air content is close to the

satur,ated condition, In waves and turbulent water, air is ent;ained

II - 12.4
 CAVITATION

and suspended %n the form of bubbles, thus raising the total content
above the saturated value. Also, dfrt and
. organic substances seem to

carry gas bubbles, Therefore, in practical applications, the nuclei

content is always above any critical limits; and ships and hydrofoils

are faced with the problem of cavitation.

The air content also has another more direct effect. Existing

afr bubbles grow upon entering a low-pressure field, and may combine
r) I, i. :;
with each other, The result is a 19cavftatingl? flow pattern without

any vapor, which is simzilar to unreal"', vapor-type cavitation. Two

types of cavitation and two dffferent crftical cavitation numbers can,

therefore, be consfdereds one fndicating the onset of air-bubble

formation ("bubble" phase) and the other one definfng incipient

(l'steadyl' or "sheetl' or 9Qaminar*t) vapor cavitation. Naturally, air-

bubble eavitatfon usually starts above the vapor pressure0

Erosfon

There have been considerable discussions and arguments about the

mechanism of erosion, caused by cavitation, The predominant and

accepted effect is the mechanfcal hammering OP impact at spots where

vapor-filled bubbles or cavities suddenly collapse upon the surface

(see reference 2), In contradfstinctfon, air-bubble cavitation does

not lead to erosion because there fs no sudden cclllapse and the in:

troduction of air (ventilatfon) into a cavitating water turbine has

---
 ' CAVITATION

been found to reduce the noise (of collapsing bubbles) associated

with vapor cavitation.

As an example, Figure 12.1 shows lift-coefficient areas in-

dicating inception of cavitation and erosion past a certain hydrofoil

section. Impact and damage only take place within the dotted areas,
where cavity or bubbles end ahead of the trailing edge. There is

. . .
.I . .
. . . . DANGER OF EROSION

o\o
FULL CAVll~Al ION BEYOND
.a0
20 ‘-
TRAILING EDGE
. ..-*--- . . .
0.6

0.6

NON-CAVITATING

01 I I
0 0.5 I.5 2.0
C A V I T A T I O N ;&3ER G

PHASES OF CAVITATION ON CIRCULAR-ARC SECTION (REE 5)

FIGURE I2 .I

-
 CAVITATION
.

only a restricted'interval of lift coefficient (in the order of ACL

41 f, 0.1, at both sides of c‘,,t) where cavitation can be av'oided at

F in the order of 0.5. After traversing the phase with partial,

eroding cavitation, erosion is no longer to be expected within the

fully cavitating region.

C
 CAVITATION

2, Inception of Cavitation

As a Function of Pressure

Upon increasing the temperature or decreasing the pressure9 water

is likely to vaporize, boil or cavitate at the vapor-pressure level.

At a standard temperature of 5u"F (15°C) this pressure, pvapcr9 is

comparatively low (33 lb/ft' or 0,55 ft of water, in fresh water); it

is less than 2s of the sea-level atmospheric pressure. Therefore, fn

proximity of the free surface, water is expected to vaporize and to

cavitate shortly before reaching ~3,;~ = zero,

On the basis of the avajlable pressure differential (,pambfent -

Pvapor)9 th e cavitation, number is

%mbr'tn& - Pvapsr
G== 02.2)
q -

If disregarding the value of bapor which is e~mall in many practical

applications, the cavitation number simply appears to be the ratio of
the undisturbed static to the dynamic pressure of a considered flow

of water. The cavitation number is a feature of the flow as such it

indicates the t'preparednessl' of a water flow in respect to cavitation.

The smaller 6, thP stronger is the tendency of cavitating,

The static pressure in a flow around a body exhibits variations.

Tn certain places, the velocity is local!.y increased, and the static

II - 12,8

__.
..__I_ --
.- -._... . -- ,.__
____.__-__
- --
 CAVITATION

pressure is consequently decreased. This decrease is indicated by

, (12.3)

where q=
0.5 p v2 denotes the dynamic pressure of the undisturbed

flow, and Cpm;r\ indicates a negative static pressure coefficient

depending upon shape and attitude of the body involved. The minimum

static pressure on the surface of the body

(12.4)

steadily decreases from pa,b , approaching zero as the dynamic

pressure q increases from zero to the critical value. In other

words, at a certain preferred point at the surface of the body, the

pressure reduces to the level of the vapor pressure (Pmiv\ = Pvagor).

The critical cavitation number is, therefore, expected to be

with the subscript lt,~~~ (incipient) indicating the onset of cavitation.

Critical- CavitatJ.on
- Number

The onset of cavitation in experiments can be determined by

visual observation, or by a sharp increase in sound level which is

associated with the collapse of vapor bubbles, or by the divergence

of lift-, drag- or momen-t coefficients from the?r undisturbed valusa.

Proper definition of\ quoted critical cavitation numbers is desirable.

----
 CAVITATION

Figure 12.2 presents some experimental results on three-

dimensional head shapes, demonstrating the correlation between in-
cipient cavitbtion number (G+ taken from reference 7) and the
Avapor
pressure coefficient (taken from reference 6),. The graph also

indicates the influence of the shape upon the inception of cavitation.

OGIVAL SHAPES
‘0 Cpmln IOWA(S)
1) aiair DITTO
(3 d~~por CALPECt71

AND

0 D 2 3

THICKNESS RATIO ;

CRITICAL CAVITATION NUMBER OF VARIOUS HEAD SHAPES

F I G U R E 12.2

Figure 12.3 prescn:s the pressure characteri3t523 of two-

dimensional streamline shn;-es e For statistical analysis, the fore-

body of these sections 5s c~::~i.z!~rcd onl:/p on the basis of the *

 CAVITATION

0 NACA 0015 SECTION, REF. SO

+ JDUKOWSKY SECTIONS,REF. 9
A DVL,X/C =(40 TO 50)%,REF.l0
I 6~ OF 66-021 SECTION, REF. I I

0 IO 20 30

EFFECTIVE THICKNESS RATIO ii

2X

MlNlMUM PRESSURE COEFFICIENT AT THE SIIDES OF

SYRRMETRICAL SECTIONS AT ZERO ANGLE OF ATTACK

FIGURE 12.3

length X. It appears that the influence of the afterbody upon the

flow pattern past the forebody is of secondary importance. Hence,

the minimum pressure coefficient increases in proportion to the

thickness ratio of the equivalent, approximately elliptical section.

For symmetrical sections at zero lift, the pressure coefficient is

approximately

(12.6)

___. - ~.- ---.

with t and y asdefined in Figure 12.3. The constant is nearly

equal to the theoretical value of 2.0 for elliptical sections.

Unfortunately, there is only one suitable value for Gi available for

Figure 12.3. General experience confirms, however, that in slender

shapes G; 5=: -Cpm;n.

The pressure distribution around a lifting foil section is

basically composed of a component due to thickness (equation 12.6) and

a component corresponding to lift. The average and minimum additional

pressure differential at the suction side is s:imply

(12.7)

in the first approximation. The lift, however,, is not uniformly dis-

tributed along the chord and the value of the pressure minimum is higher

than indicated by the last equation. Distribution and minimum coeffl

icient depend upon the section shape.

Cambered sections have an troptimum" lift coefficient, defined by

a flow pattern in which the streamlines meet the section nose withbut

flowing around from one side to the other. This condition may be

identified by ltsmooth" or ltsymmetri.cal entranc'e". As a function of

camber ratio f/c , smooth entrance flow is existing at

Lpt = (IO to *
c
12)
,
(12.8)
 w CAVITATION

With respect to minimum.pressure and cavitation, operation in the

& ',
vicinity or somewhat above C LO+ is expected to provide the highest
;I,
possible critical speeds for a given foil section at the respective
lift coefficient.

Assuming now that the minimum-pressure differential due to lift

be simply superimposed on that caused by thickness, it is possible to

reduce experimental results to zero thickness, and to isolate approxi-

mate values due to lift. Available experimental data presenting the

minimum-pressure coefficient at or near the ltoptimumrl lift coefficient,,

have been evaluated accordingly, by subtracting a value attributable

to thickness as indicated by equation (12.6). The remaining component

due to lift is plotted in Figure 12.4 for a number of foil sections.

The experimental points are evidently grouped according to thickness

location. In each group, cavitation-tunnel results (with G' defined

by the deviation of drag- and/or lift coefficients from the non-

cavitating values) are seen in close agreement with the minimum-pressure

measurements. The component due to lift is aFp$oximately

(12.9)

where k is a function of thickness location (and probably of other

shape parameters as well). Combining the pressure minimum due to

lift (which is in the vicinity of 25% of the chcrd, at CL*+) with a

thickness location at 502, provides a factor k =C,,7. An appreciably

_---. -_--. -_._-----____ --_ --

 CAVITATION
- - - -

higher value results for a location at 3o%, where the factor ks O*85

0 P RINGBUCH (12)
0 C MARTYRER (14) + =30%
9 P GUTSCHE (131
0.0-
A P HOLL (15)
0 P GUTSCHE
0 P RINGBUCH (121
o,6- I C WALCHNER 06)

s PRESSURE OISTRIBUT ION

CAVITATION TEST

INCREMENT OF THE CRITICAL CAVITATION NUMBER, DUE

TO LIFT
FIGURE 12.4

Example

What is the critical cavitat-ion number of a foil

section having t/c = lO$ (at 0.5 chord) andf/c= 3% ?-

According to equation (12.6), the critical cavitation

number due to thickness is c~ = 2.1*0,,1 I= 0.21. For

CL = 0.33, equation (12.9),indicates a As a 0.7*0.33

= 0.23. )Ience, the number indicating incipient cavit-

ation is expected to be Gi = 0,,21 + 0.23 = O.bb.

-
 CAVITATION

Delay of Vapor Cavitation

Even on the basis of a sufficient number of nuclei, cavitation

may not exactly start at reaching vapor pressure. Evidently, the

growth of air- or vapor-filled bubbles and the transition from liquid
to vapor phase require some time. This time element may not be

important in many cases where the flow velocity is small and the

geometrical size of the low-pressure field is large. In fact, the

results in Figures 12.2, l2,3 and 12.4 demonstrate that up to -$t&

sGiaO.8, delay in the onset of cavitation is, in general, small.

Considering, however, as an example, conditions at the leading edge

of a foil, where Cpn;n may reach values which are ten times as high

as those just quoted, and where such values are orily existing in

narrow peaks (that is, during very short periods of time)- the

influence of small size and high speed may be important.

As an example, characteristics are presented in Figure 12.5 of

the NACA 6.4.AOO6 foil section, tested in a wind tunnel 17 and also in

a towing tank 1 8 . The incipient cavitation number I$ is decisively

lower than the value of the corresponding pressure coefficient C,,,;,

for values between CL= 0.3 and 0.5. For example, at CL = 0.5,

where -qm;i = 5, the incipient cavitation number is only 5E2.

---
 .
CAVITATION
- -

6 -

6 4 - 0 0 6 ; N A C A f.RPT 824

EQUATIONS 12.6 AN0 12.11

0 I I I I 1
0 0.2 0.4 0.6 0.0 1.0

CL

PRESSURE AND CAVITATION COEFFICIIENTS OF

64A006 FOIL SECTION
FIGURE 12.5

In the considered foil tests at CL+ 0.5, the time during which
a particle of water travelling past the point of minimum pressure, is

really exposed to pressures below vapor pressure, is only in the

order of

exposure time =length 0.002

-- D - = 5/100,000 second (12.10)
Vlocal 2 x 21

This short interval may be responsible for the discrepancy between

-cp = sar,d $=2, as observed at CL = 0.5.
I
 ”
.

. CAVITATION

Reference 19 presents experfmental results of another foil section,

the 44.3.2 section, as tested fn a cavitation tunnel. Good agreement be-

tween 6~ and -CP,,'ita is found at pressure coefficients or cavitation

numbers below one200 However, at negative as well'as at higher

positive lift coefficfents, where pressure peaks are to be expeated at

the respectfve sides of the sectfon, consfderable discrepancies of the

same magnitude as those in Figure 12,s are evident.

Cutting off the peak of a really narrow pressure minimum may not

mean losing much of the total lfft produced in a foil section, The

pressure distributfons during the cavftatlon-tunnel tests may not have

been the same as in the wind-tunnel investigations,, It is suggested

-
as a possfbilitythat in the water tests very smal:L air bubbles (nuclei)

may have levelled off the pressure peaks to some extent before visible

bubbles could be observed and reported, This levellfng-off effect may

be combined with the time effect, described above.

Whatever the explanations may be, Figure 12,s and reference 19

demonstrate that in the case of suction-pressure psaks as may occur on

lifting hydrofoil sections, inception of cavitation cannot reliably be

predicted from non-cavitating or theoretical pressure distributions.

This fact also discourages applfcation of the theoretically correct

method (reference 21) of correlating the critical cavitation number

with the critical Mach number which is available for many sections and

lift coeff$cients,
C

II - 12,17
 CAVITATION

The available results on the delay of cavitation (or whatever

the effect may be) are assembled in E?i.gure 12.6. It appears that

results of slender bodies, or streamline sections in the vicinity

of zero lift, having by comparison "no" delay, are compatible with

such extreme cases as the lifting section in Figure 12.5. A

statistical conclusions is that

(12.11)

6 X N A C A 64AOO6 SECTsON (1%)

0 @ALTEC 4 4 1 % SECT1ONP19)
e OITTO, A T POSITWE LIFT COEFF’S
d BALHAN ARG SECBIIONS (221 atTi w -(O.O% TO OnIt)C~*ipl
4 kB CALBEe HEMISPHERE (23)

0
f
- C p min
STATISTICAL SURVEY ON THE “DELAY” OF
CAWTATION IN PEAKED PRESSURE DISTRIBUTIONS
FIGURE 12.6
 CAVITATION

Vortex Catitation

The minimum of the statfc pressure does not always occur on the

surface of an obstacle, Tip vortices, for example, originating from

lifting surfaces, have a core with a static .pressure on a negative

level o According to Prandtf', the pressure differential in the core

of a wing-tfp vortex is

(12J2)

where :A = aspect ratio of the wing. Cavitation may start accordingly

within such cores, as has been observed behind the blade tips of water

propellers.

Vortices are also sheddfng from bodies exhibiting a separated flow

pattern, especially in the form of a "vortex street". As an example,

cavitation was found starting within the cores of such separation

vortices 9 orSginating from a blunt, cylindrdcal head shape (see

reference 6) at 6;;, = 10'i'6, while the minimum pressure coefficient at

the surface of thfs body was only -C~,.& = 0~64. Here again, the

pressure distribution would not be a reliable indication for the onset

of cavitation; and the deviation would be in the direction opposite to

that as found in pressure peaks,

In concluding, equatfon (12,5') only seems to hold for slender

bodies (foil and strut sections in Figure 12.3 or three-dimensional

II - 12*19
 CAVITATION

shapes in Figure.12.2) and in "smoothl' flow patterns (such as near

c,o,t in ' Figure i 2 . 4 ) .

Critical Speeds in Hydrofoil Systems

Considering boats with hydrofoils operating beneath the free

surface of water, the ambient static pressure is

(12.13)

where t = 62.43 lb/ft3 for fresh water and r = 64 lb/ft3 for

"Atlantict sea water . On the basis of a critical cavitation number
GA (assumed to be known for the system considered), the critical

dynamic pressure is then

The standard atmospheric pressure is patnt = 2120 lb/ft2 , correspond-

ing to a head of 34 ft.of fresh water. At a standard temperature of

59'F, the vapor pressure is comparatively low, in the order of 1.6% of

the atmospheric pressure; the corresponding head is roughly 0.5 ft of

water. Disregarding this small quantity the critical pressure reduces

and in close proximity to the surface (more correctly
I20 -4 =Pamb/~ 3
at hr 0.5 ft), the critical dynamic pressure is

(12.15)
 CAVITATION

Using the standard sea-water density f = 1.99 (lb sec2/ft 4 ), the

corresponding l'standsrdtl critical speed of hydrofoil systems is
found to be
2
2120 46.2
Vcd = 1.99 F = - (12.16)

In knots, the critical speed is

Vctit m 27/
Fi
\I (12.17)

This function is plotted in Figure 12.7.

47 EQUATION If: % R = # (KNOTS)

CRlTlCAC CAVITATION NUMBER C,‘i

STANDARDIZED RELATION OF SPEED 8 @‘RlTlCAL

CAVITATION NUMBER
FIGURE 12.7

II - 12.21
 . .

CAVITATION

Example

*at is the critical speed of a hydrofoil system

operating near the water surface, on the basis of

a minimum-pressure coefficient C&4<, = -0.44 (as

found in the preceding example)? - Using equation

Introducing the critical cavitation numbers indicated by equations

(12.6) and (12.9) into equation (12.17), the standard critical speed of

favorably designed hydrofoils can be predicted as a function of section

shape and lift coefficient:

(12.18)

This function is plotted in Figure 12.8.

Example

What is the maximum permissible lift coefficient

of a hydrofoil section having t/c = 10% (at 0.5
chord) for a speed of 40 knots? - Figure 12.8 in-

dicates CL = 0.36. For comparison, the average

lift coefficient in the blades of a destroyer

propeller is well below 0.1, to avotd cavitation.

C
. CAVITATION

EQUATlON I2 .I81

LIFT COEFFICIENT CL+

STANDARDIZED CRITICAL CAVITATION SPEED OF HYDROFOILS.
FIGURE 12.8

II - 12.23
 CAVITATION
.

3. Forces in Cavitating Flow

Transitional Phase

At speeds (or lift coefficients) above the inception of cavitation,

lift- and drag coefficients do not immediately change their magnitude,

Within the "bubble" phase, they remain essentially at their non- \

I/
\
cavitating level; the lift coefficient sometimes increases slightly.

Subsequently, as soon as any real, coherent cavity develops, the lift

coefficient decreases, the drag coefficient increases. This transition

(see Figure 12.11 for illustration) can be rather extended, reaching up

to twice the dynamic pressure (1.4 the speed) at which cavitation first

sets on. No systematic information can be given on this phase as of

this time. Once in fully-cavitating condition, stable flow patterns are

prevailing, some of which are presented as follows.

Resistance of Blunt Bodies

IIn fully catitating condition, the resistance of an obstacle

evidently corresponds to the uniform negative pressure at the rear side

(within the cavity) and to the average positive pressure component on

the face of the body. For G = 0, the latter one has been calculated
for wedges24 and cones2S as a function of the hal.f apex angle t.

Upon increasing the cavitation number from zero, the flow pattern

past the forebody changes, as explained in reference 26, Accordingly,

the drag coefficient is expected to increase as

II - '12.24

--.- e----P
 CAVITATION
-

ca = CD* (r +q (12.19)
where Cd0 * drag coefficient at G = 0. Figure 12.9 presents ex-
perimental results of disks, confirming this function very well.

1.4

NON-CAVITATING
1.2 0
/’ 1’ ----.-‘-
------I

I.0

CIRCULAR CYLINDEIR ( 141:

0 AT SUBCRITICAL REYNOLDS NO.
0.6
m AT SUPERCRITICAL R’ NO.
EQUATION
+ CIRCULAR CYLINDER THEORY (2
\
./ A REICHARDT CONE (Ee 450) (26)
co + TM6 (28) HEMISPHERE
0.6 . 0 ACKERET( I) SPHERE (SUBCRIT’L)
-I \
Y. r- \ 0 ENTRANCE J. APPL.PHYS. 1946 RI109
. ‘4-.------ CI

0.4 P X - X
/+ Re< 10’
/ X’x
AA Y “:Niz--
, FULLY’ 1
1
0.2 v : (@@$!j Qm-;Ex S’TREET
/’
X
cn,
XxxRe > 105 t
I ’
C
0 Oh I.5 2.0
CAVITATION NUMBER 5
DRAG COEFFICIENTS OF BLUNT BODIES IN CAVITATING FLOW
FIGURE 12.9

-----_-- -.----.- --
 CAVITATION
-.

In the case of round bodfes, the minimum pressure point is

located somewhere on the forebody (and not at the trailing edge as in

wedges and cones). Cavitatfon, therefore, starts from that point.? and
the starting point may move further forward as the cavitation number

is decreased, As a consequence, the c&(G) function is no longer

strafght as according to equation (12,19), Such characteristics are

demonstrated in Figure 12,9 by results of a hemispherical body28 and a

Sphere'. Results of cireuldcylinders (in cross flow) also show a di-

vergence from the mechanism according to equation (12,19)

Drag in Ventilating Flow

Insight in the drag mechanism of blunt bodies may be amplified by

discussing results of a "streamlined" half section tested in ventilating

(rather than cavitating) flow in the Gibbs & Cox towing tank at speeds

between 1 and 10 ft/sec. A piece of strut having ,a chord of 3.1 inches

and an aspect ratio of 4 was towed between hollow end plates (connect-

fng the cavity tith the atmosphere) at an average submergence h=

1.5 c. The cavitatfon number is then simply

At higher cavitatfng numbers (above 0.6), ventilation (or cavitation)

does not take place; the drag coefficient is approximately constant and

practically equal to the base drag coefficient of such body shape.

xi - 12.26
_- . ’ Qam@
CAVITATION

Wfthin the range of small cavitation numbers (below 0.3), the

coefficient is approximately
-z- /
CiA
(12,21)

Flow pattern and drag coefficient would not be the same, however, in
cavitatfng flow* There is etidently a line of minimum pressure some-
where between leading and trailing edge. Cavitation would start from
there, rather than from the trailing edge as in ventilating flow3 and

the drag coeffiefent would be higher than according to equation (12,21),

Lift in Fully-Cavitating Flow

The force on a flat plate in cavitating flow (at 7 *= 0) has been

-
calculated by Kfrchkoff and Rayleigh. Using such a blate as a lifting
surface in two-dimensional flow, Betz predicts in reference 30 that the

coefficfent of the force normal to the plate's surface is

For G = 0 and for angles of attack G%,,be,low some 100, this function

reduces to
TT
G
dci =-3x- (12,23)

which is only one quarter of the basic lift-curve slope of an airfoil

section (which is 2 I?)* Experimental results (see reference 5) of

foils tested in a water tunnel (tith a geometrical aspect ratio of

II - 12.27
 CAVITATION
- - - -

one, between the test-section walls), are plotted in Figurea 12.10

and 12,,11. They evidently confirm the theoretical function - for
eufficiently thin sections (up to and including t/C - 7.5%).
Thicker sections and sections with rounded noses show lift
coefficients, however, which are sometimes higher (because of nose
suction), and at small angle of attack lower (cavitation shifting
from the upper to the lower section side), than according to the
fully cavitating theory.

SECTIONS WITH FLAT PRESSURE SIMS,

HAVING t/C’ 2.5 TO 7,$9&tESTED BEtWEEW

Ol IN 2-DIMENSIONAL FLOW

LlFf COEFFICIENT AT GO

F I G U R E 12JO

I I - 12.28
: .!iammk CAVITATION

l.O-

0.6-
0 REC. 5 t / C ’ 2 . 5 TO 8.9%
CL X AEf. 5 t&m 7.6 AIJO 7.SS$c
/’ a NUMACHI ml) t/c l 0%

0.4- ,J a : 6’ + CONSTANT

0.2- /d’
,!

LIFT COEFFICIENT AS A FUNCTION OF CAVITATION NUMBER a,

TESTEO BETWEEN TUNNEL WALLS
FIGURE 92 .I I

For a wing of finite span, the induced angle of attack

where k - biplane factor

4 = b”/s = aspect ratio
 CAVITATION

must be considered; the aids' in equation (12.22) is to be replaced

by the effective angle (u!-~$1. Consequently

This equation is compared (in Figure 12.13) with points tested in the
Gibbs Kc Cox Tank on (the flat bottom of) a plate.towed in ventilating

flow between-hollow end plates. On the basis of a biplane coefficient

indicating the effect of the free water surface, 14 - 1, for the tested

Itbox plane" condition at h/b = 0.25, magnitude and character of the

function according to equation (12.25) seem.to be verified.

Resistance of Hydrofoil Sections

W,alchnerS~16 is the only source of informatio~n available, Indicating

resistance of foil sections in truly cavitating ccindition at or near

zero lift. The pressure component of this resistance at G - 0,

plotted in Figure 12.12 shows some relation to thalt of wedges, having

the same forebody-thickness ratio. Between the values at c = 0 and

the non-cavitating phase, the drag coefficient decreases considerably

. (hardly without any first-increasing trend as per Figure 12.9).

II - 12.30

_
 SAVITATION

THEORY -REF. 24 /

0 0.1
$ AN&
-
DRAG COEFFICIENT OF SOME FOIL SECTIONS
A?’ c=O 9 FROM REFS. 5 AND 16
FIGURE 12.12

In lifting foil sections (with fully cavitating upper side,

the pressure drag is a geometrical component of the total force.
For flat pressure side is

(12.26)

where ~4 is indicating the angle of the lower side,in two-dimensional

flow. Utilizing equation (12.23), this angle is

I I - 12.31
 c

l
CAVITATION

(12.27)

Therefore, 'the drag due to lift of sections having flat lower *sides in

two-dimensional, fully cavitating flow is expected to be defined by

(12.28)

For G - 0, some experimental evidence can be found in references 5

and 1.5, confirming the trend of d% /de: =2/r , for circular-ard

sections with t(C 4 7.5% and o( up to 6".

For higher angles of attack, as in the case of Ngure 12.13,

the complete equation (12.25) has to be employed to find CL cd) first,

and then cDpressure as a function of C’ and cx .

I I - 12.32
 CAVITATION

a*
- 2 0 -10 0

wml CDp~3 titan o(C,

FULLY VENTILATING

EOUATION 12.26

STRUT RESIS?ANCE

L I F T A N D D R A G O F A F L A T PLATE HAVli’JG ,A:4

BETWEEN END STRUTS, AT NEGATIVE ANGLES
OF ATTACK

Fl GURE 12.13

Cambered Sections

Fclr a long time p only circular arc sections (with flat pressure

sides) have been considered, possibly to be .used in cavitating flow

conditions. It has been suggested, however, that (cambered pressure

---
I I - 12.33
 CAVITAT:O'\

sides shoulh provide higher sectional L/D ratios or lower resistance

ratios, respectively - on account of certain forward components of

the forces normal to the surface elements of such cambered sections.

Figure 12elb presents the characteristics of such a section;

0 0.2 0.4 0.6 0.8 -50 -40 -30 -20 -10 0

+I A ’ -1’

A* 4, BETWEEN END STRUTS

FLAT PLATE ( FIG. 13 I

LIFT AND DRAG OF A VENTILATING CAIMBERED

FOIL SECTION, AS TESTED IN THE GIBBS 8 COX TANK
FlGURE 12.14

II - 12.34
 CAVITATION

tested in the same manner.as the plate in Figure 12.13; that is, in

ventilating flow. The transitional phases in these graphs would not

be the same in cavitating flow, however: and they should be disregarded

if considering vapor cavitation. For the cambered section, the fully

ventilating-cavitating phase exhibits considerably higher lift

coefficients and.appreciably lower resistance ratios than for the flat

plate.

The performance of fully-cavitating foil sections, in tGo=

dimensional flow at F = 0, has recently been investigated by

linearized theoretical methods (see reference 32). The optimum shape

is found to be cambered (at the pressure side) according to

Disregarding frictional drag, theory predicts for this camber ratio

that the pressure-resistance ratio will only be :=11/6 of that of the

flat plate. The results in Figure 12.14 do not confirm this much of

improvement.

II - 12.35
 CAVITATION
-

References

1, Ackeret *SExperimental and Theoretical In-

vestigatfons of Cavitation", in
Tech, Mech. & Thekodynamik - Vol. 1
of Forschung Ingenieurwesen 1930
pa 1; Trans. TMB No. 20.

2. Eisenberg "Mechanism of Cavitation", TMB Rpt.

792 (19SO)and Rpt. 8L2 (1953).
3. Raven, Feiler & "Annotated Bibliography of
Jespersen Cavitatlonf', TMB Rpt. R-81, 1947.

4. CrumP llCritical Pressure for Inception of

Cavitatfon in Fresh and Sea Water
as Influenced by Air Contentfl,
TMB Rpt. NO, !%'s, 1949.

50 Walchner YInvestigatidn of Foil Sections in

Cavitating Flo+, in Hydromechan-
ische Probleme des Schiffsantriebs,
HSVA Hamburg, l932. See NACA Tech.
Memo, 1066,

60 Rouse & McNown YZavitation and Pressure Distrib-

utfon, Head Forms", Publication No,
420 Iowa State University (Bulletin
No. 32, 1948).

70 Knapp & Doolittle "Nose Cavitatfon, Ogives and Sphero-

gfves", Rpt. from Caltec High-Speed
water Tunnel, It8 Jan. 1945.

8. Hoerner "Inception of Cavitation on Hydro-

foil Systems", Gibbs & Cox Tech.
Rpt. 13431 No, 8, 1952.

9. Fage, Falkner & "Experiments on a Series of

Walker Symmetrical Joukowsky Sectionstt,
Brft. ARC RM 1241, 1929.
10. "Tested Pressure Distributions",
Jahrbuch D, Luftfahrtforschung 19&l
p,I 101 and German Rpt. ZWB F'B 1621
(1943)o
.-

II - 12,36

. ..--- _--- -_-...- _----

 CAVITATION

11, Coffee & McKann "HydrodyneunPc IDrag of 12- and 21-

Percent Thick Struts", NACA Tech.
Note 3092, 1953.

12. Schrenk wPressure Dfstribution Along the

Chord of Wing Sectionslt, Contrib-
ution IA11 in Ringbuch Luftfahrt-
tech&, German Ministry for
Aeronautfcs, 1938,

13. Gutsche ~~Characteristics of Propeller-Blade

Sectforqy a) Mftteilungen Pr, Ver-
suchsanstalt Wasserbau Schiffbau,
Heft 10 (Berlin 1933), b) Yearbook
Schfffbautech Gesellschaft 1936
p- 277; 1938 p0 1253 and Vol, 41,
1940,
14, Martyrer "Force Measurements on Cylfnders and
Foils at Cavitation", in Hydrodyn-
amische Probleme des Schiffsantriebs,
HSVA Hamburg 1.932-a

150 Ho11 "Investigation of Propeller-Blade

Sections Having Reduced Cavitation
Tendency", Forschung 1932 p0 109.

160 Walchner "Contrfbutfon to the Design of Ship

Propellerz+Wfthout Cavitationw,
AVA Monograph,, Reports & Trans-
lations No. 330, 1947, British
Ministry of Aircraft Production.

17. McCullough & Gault "Boundary Layer and Stalling of NACA

6!&006 Airfoi:L", NACB.Tsuh. Note
1923 (1949), :now Tech, Rpt. 903.

18, King &Land "Force and Cavitation Character-

istics of'As~~ec+Ratio-.!! Hydrofoils",
NAfW?@ J ' ntial Rpt, LS2J10, 1952,
190 Daily "Cavitation Characteristics of a
Hydrofoil Sectionw, Trans, ASME 1949,
p0 269 (Caltec).

20, /Pinkerton '"Pressure Distrfbutfons Over 4k12

Airfoil Sectfonlt, NACA Tech. Rpt.
5633 1936,

I I - 12.37

-.____--. -- .- ~-II___
 * - CAVITATION
-

21, Joehua Hendy Cofp@ Sfydrofoil Studies and Design Daban

Rpt. to ONR, 1950,
22. BiLlhan @%veetigation of Some Prbpeller-
Blade Sections in Cavitating Flow",
Mageningen Publication No. 99, 19%
23. lterlneen HObaervations of Cavitation on
Hemiapherioal Head Modeleti, Caltti
Hytf.ro~amice Laboratory Teoh, Rptr
E-354 1952.
240 Bobyieff Quoted in paragraph 78 of Lamb@8
Hydrodynamics

250 Please% & Iohafer @Drag in Cavitating FlouoP,J. Appl,

Physica, 994.8, pa 934 and Rev. Mod.
F'hwfcs,, 194.8, pm 228.
26, Refahardt "Law of Cavitating Bubble@, I)srman
Dct, ZWB UM 6628, Trans. 766 Bri.&
Min, Afrcraf't Production.
27. Brmm "Optimum Slenderness Ratio of Low
Drag Body", Caltec Hydrodynamic8
Laboratory Memo Rpt. N!%# 1949.
28, Eieenberg & Pond Water Tunnel Inv6stigation of
Steady-State: @avitiea", TM@ &6,
No, 64Bp 19b0,
29, Brodetsky Proo. Roy, 8ocs London Qer. A NOO
718, 1923.
30. Beta Qfluenct3 of Cstitition Upon
EffirSiency of Ship PropellerG,
Proceedings Third Internat, Congr,
Tech. Mechanics, Btookholm 1930,
V o l . H p* 4:Ll.s

314 Nwwchi %avitation Tests on Hydrofoils in

C,aacadeflo T:rans ABME 1953 p. 1257,

32. Tulin Q Burkat "Theory About Eifting Foils at Zero

Cavitation Numbertip TMB m
Rpt.t 1954.

II - 12e38

-- .- _--- -.- -__---

 APPENDIX .A, CALCULATION OF THE LIFT AND
DRAG CHARACTERSSTICS OF A
PROPOSED 50 TON HYDROFOIL CRAFT

1, General Consfderatfons

20 Charaeteristfea of the
configuPat8on

3. Design Examples

Design examples are given for the drag and required foil settings

II =-vA,aP
.
LIFT AND DRAG CALCULATIONS
Pj

1-0 General Considerations

The hydrodynamic characteristics of the components of a hydPofof1

craft, as presented in this volume, are used for the purpose of cal-

culating the various performance eharacterfstfcs of a proposed design.

Thus, wfth this information and havfng a knowledge of the methods to

be employed, one can calculate the total drag throughout the speed

range (fncluding take-off), the turning c%rele, required fofl and

rudder angles, et@,

This volume presents only the baste eharacterfstfcs of components,

and the methods of utilisfng th-fs fnformatfon fs the sub,ject of other

volumes 0 However, for the purpose of %Llustration, design calcula-

tia~na are included herein, These examples are for the drag and

requtred fa$l settings of a specific conffguratfon at two foil-borne

speeds.

c
 LIFT AND DRAG CALCULATIONS

‘2. Characteristics of the Configuration

The configuratfon chosen for these examples is that of a 50 ton

hydrofoil craft, and is based on the desfgn study reported in

reference 1.. Thfs craft ia a Vanax"dsl type wfth automatically con-

trolled, fully submerged fofls, and wfth the power transmitted through

a single nacel3.e on the mafn (rear) foflg as sh'own in Figure A.l. The

dfmensions of the foil system and other per%inent fnfo??matfon are given
in Figure A.2.

For the purpose of these calculatiions, the craft is assumed to be

at zero trim and normal submergence under full load condftion. Two

speeds (40 knot cPuisBng speed and h.$ knot maximum speed) were selected.

Reference 1 - Confidentfal Letter from Bath Iron Works

Carp, (by Gibbs & Cox, Inc,) to ONR, file
1353a/s1/3(a-X283) of April 3g 1953 -
enclos!ing
"Des%gn Study for $2 Ton Hydrofofl Craft"

---Ii - A,3

---
-. .--.. _. --.l____-~“l_-..- -- --.-. -_._.--
.
LIFT AND DRAG CALCULATION!3

DESIGN STUDY MODEL

OF 50 TON HYDROFOIL CRAFT
FIGURE A.1

--_-. __I. _-.--

 LIFT AND DRAG CALCULATIONS

3. Design
- - - Elxamples

The overall ealeulatfon is broken down into fndfvfdual design

examples, included on the sheets followfng:

Design Example - AP Induced Charaeterfstfes

A2 Wake Effeeff

A3 Fo3.1 and Flap Settzings

Sk ParasEte Drag

A5 ffihmmmtlry of Drag and Fofb Settings

 LIFT AYD DRAG
VP CALCULATIONS
-

,-
LIFT AND DRAG CALClJLATIONS
- -
46 LIFT AND DRAG CALCULATIONS
 DESIGN EXAMPLE NO. i-7
WAVE EFFECTS- M AI N F O I L CANAR? F OIL

Qua frw (2.34-1,deep hlaf-er 3 0 KNOTS 4 5 KNOT: 3oK~1tmi 4 5 KH&

h 2.44’ 2. 4 4I 2.Od 2.06’

v ( W5ec) 50.G 76.0 50.6 76.0
Fh = ~/~jx- 5.2 I 7.8 2: 6.22 q.34
h/b 0.123 O.iZI) 0.147 0.147
=/h o.qc2 n.QG:Z 0.801 0.801

c----?c- e25 -V) (F, .y,'T? 2.(7) 0.0~75 0.00 80 0.0125 0.0058
i--t,
I<?,--I) ’ (FI(JtJI’@ :.(; ) 0.43 0.43 0.38 0.38
c

p&~ ($p$$:~,,-l) 0.0012 a. 0033 0.0030 O.OOl8

h

d G, frtm, sh. 3 0 * 0543 0.054-3 o.os4r 0.054 I

d kb a hove 0.0072 0.00~~~ 0.0038 0. oorg

cl c C‘“)

d (C,l) o.obr5 0.05’76 0.0579

‘. 0.0559

-
I.026 I. 012:. I.028 I.012

:; T - A . ‘j

-____ _-
.---- -.--“-~ .”
 LIFT
-_ AK) DRAG-._-_
CALCIIT,ATIONS
-I-

DESIGN EXAMPLE NO. A.II

TNDUCED DRAG 8 t=-OIL /hUCLES

MAW FOIL. CANARD. FOIL

3 0 KNOTS 4 5 k:bJ8d 30 I(NOTS 4 5 I(NOtS

0.2459 0.242a 0 . 2 550 0 . 2 5 3 0

0.4Cl 0.2 as 0. Soa 0.222
0 . 1 1 3 4 o . 0 4 4 6 O.lZ75 o. 0562

I. 026 I.012 I.028 i.arz

0.1163 532 0.4

0.1311 0.0569

0.0615 0 . 0 5 7 6 0 . 0 5 7 9 o.a559

O.Ll2 0.042 0.250 0.049

0. 0130 0.00242 0.0145 0.00273

/.a26 1.012 1.070 1.4r2

0.0133 0.0~245 o.ol4q 0.00276

-/-OfAL I N D U C E D -G,

30 ~&JOTS D,= 0.0\33r67.82x2460 = 2304

0 . 0 1 4 9 L 23.04: rcZ560 = 8 7 9
3,1a3*

45 S(NOTS & = 0 . 0 0 2 4 5 g 67,82% 5 7 5 0 = 455

0.00276 F 2 3 . 0 4 ~5750 = 366
1,321*
.-

DESIGN EXAMPLE NO. jr;--

MAIN FOIL. CANARD. FOIL

‘b-JOT5 3 0 &JOTS d-5 KNafS

0 . 2 593 6.2530
0 . sea 0 . 2 2 2
0.1275 0.0562

f-028 r.ar2

0.1311 0.0569
0 . 0 5 7 9 o-a559

0 . 1 5 0 0.047

0.0162#2 0 . 0 1 4 5 a.00273

I.020 I.arZ

,245 0.0149 0.00276

30 ~&JOTS D,= O.O\33*67.82rZS6a = 2 3 0 4

0 . 014-9 x 23.0+x2560 = 8 7 9

3,1881f

45 KblOTS q = o.ooz45 167.82% 5 7 5 0 = 9 5 5

a.00276 z 23.04 g5750 = 366
r,321*
 LI?T A!li) D'tiG CALCI~LATIONS
-.-.---.--__- --_-__ ----

DESIGN EXAMPLE NO. u12m

CALCULATE THE WAKE EFFECT ON THE
MAIN FOIL OF THE 50TON CANARD
CONFIGURATION AT 30 AND 45 KINOTS

TI - A.11
 LIFT AHU CRAG CALCIJWTIONS

DESIGN EXAMPLE NO. A.3 I SH I OF 2

CALCULATE THE FOIL ANGLE SETTINGS .
AND FLAP DEFLECTIONS FOR THE 50TOlil
CANARD CONFIGURATION AT 30 AND 45KNdTS

DESIGN CONSIDERP;TIONS
The main foil is fixed in configuration with 25%
full-span flaps on a cambered foil section. The
foil is set so that the flap deflection is zero
at 45 knots,
The canard foil has no section camber, and
pivots about an axis thru its quarter-chord.
The main foil setting is referred to fhe section at
mid-span,and is thu5 the angle derived below, without
cwrection. The msin foil flap and the cam-d foil angles
are referred to planes normal to their axes, and the
angles derived must be corrected for sweep and
dihedral.
MAIN FOIL
45 KNOTS
The main foil 'IS set at zero angle to the effective
fluid flow at 45 knots (the fail section camber is
designed to account for the lift at 45 knots). .
The induced and downwosh anqtes are:

= [ 0.0566 + 0,0072] ~02~5~~.012(fromExample A.\)

= 0.0132
& = 0.0031 (from Example A.2)
Then, foil angle at 45 knots:
Jitdw + C= 0.0163 radians = 0.93"
 LIFT AND DFi.4ti CALC!JXTIONS
---

DESIGN EXAMPLE NO, At?-j-=-

MAlN FQ\ LCCONT’D)
Requwed fosi anqle = 0.05~02 (Example Al)
D e w e d Colt settmq =~-0.0163
Ecpwa\enl angle due to
= 0.0339
,fod c a m b e r
NOTE: The foil se&on must be chaser, to have
30 KNOTS ALO - - 0 . 0 3 3 9 ;go; 0 . 0 3 5 5 =Z.O”
Requwed Fool A n g l e = O.lldCS (Example A.1)
Dawnwash Angle + 0.0031 (Example A.2)
Equivalent Sdt~ny at 45 knots -- 0.0502
- -
Requtved fob flaps = 0.0692
F l a p efFed\ueness (EqualloM 5 . 2 )
kf = I.I&z- - 0.55
Reqowd f l a p angle equrvaletd
o.06gL/o.55 = 0.125-j’
Actual flap deFlectIon (a bout own axis)
6= 0.1’257 ;$$$ = 0.1317 s-adlahs

= 7.55”
CAbjAR! FOIL
Reqw-ed ,fod anqk= 0.\311 @ 30 Kvltofs
CE xampk &.I)
0.0569 @ 4s Knots3
Required anqks about own MS
CA= 0.1311 x cz:= 0. I434 i-dchans= 8. 2Z” @ 30 KNOTS

A = 0.0569 cs;o = 0.0622 r-adlans=3.56"@45 KNOTS

I I - A.13

_,“~__ ---
.. LTV AIJD !ll?‘?A"I
--___ --- CALCl'LATIONS
---

DESIGN EXAMPLE NO. A4m

CALCULATE THE PARASITE DRAG OF THE 50T.ON
CONFIGURATION AT 30 AND 45 KNOTS

FOIL PARASITE DRAG

(From &pter 41
MAIN FOIL CANARD FOIL
30 KNOTS 4 5 KNOT5 30 KNOTS 45 KNOTS

Meah C h o r d 2.83 ‘ 2.83 ’ 1.65’ 1.65’

Recpolds No 1.12 x IO’ 1 . 6 8 46’ 6 . 5 1 * IO6 9 . 7 6 x IO”
(SW. at 59°F)

c, (std) - Equatmn~4.12) 0.00369 0.00350 0.00393 0.00374

Cl+ HO&>‘] (t/c = 10%) 1.10 1.10 I. IO I. IO

2qfCstd) [I+ lowc)2 J 0.008lQ 0.007 70 O.oOe65 0.00822

20 = 0 0.500 0.22 2
A CL

C Eq (4.13) o.ooeIo 0.00770 o.oro8r 0.00863

Do (std)

I%! v2s = 4.5 174,000= 3 9 0 , 0 0 8 5 9 , 0 0 0 ” 137,500*

II 1,410’ 3,000 * 638* I, 143 -

TOTAL FOIL FARASITE DRAG

A T 3 0 K .N O T S 2 0 4 8 *

AT 45 KNOTS 4143 *

i
II - A.lh

_^",-~---.. ---.. -._-.-

 LIFT ArK! DRAG CAT,CI%t4TIONS
.

DESIGN EXAMPLE NO. Aii-j==

STRUT DRAG

APEAS

Lehqth c hod .Are;l N O . T&al Avea

2.58‘ 3.08’ 7,951 $I I 7.9 5
24.154
2.63’ 3.08’ 8.W 6 2 16.203
3.00’ 2.12’ 6.3169 I 6.36
9 . 7 4y-----r
’ 1.75’ 3.oNj 2 6.00
*

DRAG
3 0 KNOTS 4 5 KNOTS
CANARD CANhRC

1.83~10’ 1.26~10~ 1.04”10’

0 . 0 0 3 4 5 ?.0036\ 00037C

0.924 0.212 0 . 3 5 0
0.003 0.004
0.054 0 . 0 5 4

rota1 std b-q Awa

2
/2
PV
JOTAL STRUT I>RAG
 LIFT AND DRAG CALCULATTGNS
- - -

DESIGN EXAMPLE NO. 1-v

.NACELLE DRAG
(Chapter 3)

Nacelle 2'-5" dia. x T'-0" long

L/cl =' 3.72

Frontal Area = 71/4(2.42)"= 4.60 4

Drclq 3OKNOT5 45 KNOT5

Reynolds No. 3.56 Y IO' 5.35 % IO'
Cf bfd) 0.00318 0,00302
@Do (Equation 9.3) 0.0413 0.0392
Drag Area [Co, s 4,601 0.190 O*l90
P/2 v2 2560 5750
Dray 406ff lO35#

AIR DRAG(HULL)
(Chapter II)

Cross-Secttonal Argo of the

Hull t Main Superstructure:
S= 1Zs8+i6x9 x 0.8=21\$
From equation 11.4, use a coefficient of 0.0015 ,(for a fair
bow and square stern).

Then,

DAirr a y Dair= 0.0015 x 211 x 30'!= 204* (30 KNOTS)

= 0.00!5 x 218 ,c 45% =640* (45KNOl5)

II - A.16
. _ -.-. _. __“__.~ __---._ --
DESIGN EXAMPLE NO, 1-T
SUMMARY OF DRAG AND FOIL SETTINGS
FOR 50 TON CANARD CONFIGURATION AT .
30 AND 45 KNOTS

/ Rm70
5 L5 2