.

- a - - :
Mariiie Technology, VoL 13, No. 1, Jan. 1976, pp. 14-45
Small-Craft Power Prediction
Donatd L. Blountl and David L. Foxl
A valid performance prediction technique lor small craft is an invaluable tool not only for the naval archi-
tect, but also for the operators anC buiHers. This presentation describes the methodology for making
spae6power predictions ior hard-chine craft ol the tvpes tound in tha offshore, military, ar|d recreational
applications. Tha distinct acivantaga of this method is that existing technical data have been organized into
a logical approach, and aieas of limited data have been overcome by the presentation of engineering fac-
tors basd on model tests and fuli-scale trials of specific hull forms. This speedjower prediction method
accounts for hull proportions. ioading, app6ndage configuration, propeller characteristics (including cavita-
tion), and resistance augmentation du to rough water,
Introduction
THE spEED-powER RELATIoNSHIP of a crafr is of prime
interest to all parties from the design agent tc the owaer/oper-
ator. The initial cost of installed power is iollowed by corre-
sponding maintenance and operating elpenses, panicularl-v-
fuel, direcily related to horsepower. i\Iany tcchnical papers on
small-craft design
(with references
[r], [2],
anti
[3]2
being no-
table exceptions) have been related to
just
determining the ef-
fects of variation of hull form. Savitsky, Roper, and Benen
[4]
presented an outstanding paper on the design phiiosophy of
effective hydrodynamic tradeoff studies for smooth and
rough-water operations. In addition, useful data have been
published giving propeller characteristics under cavitating
conditions, appendage drag, and propulsive data. The object
of this effort is to present the development of a small-craft
power prediction method which allows the ciesigner to appiy
these existing data to select, with imprc;e confidence, hull
proportions, engine power, reduction gears, arrd propellers. A
less obvious, but important use of this prediction metiiod is
that it serves as a baseline for determining ihat a crafa has at-
tained its technically achievable performalce during i'ials
and in service.
Resistance prediction for the hull
There has been almost no correlation of model and ftrll-
scale trial data for hard-chine craft, bui consisuent experience
has indicated that model tests for specific designs are the best
source of resistance prediction data. This experience also indi-
cates that zero correlation allowance produces the best full-
scale extrapolation of these model data rvhen using the
Schoenhen friction formulation. Another source of resistance
prediction data can be obtained from published test data from
geometrically varied hull forms. Notable exampies of these
type data for hard-chine craft are Series 62 and Series 65. In
addition to these, mathematical techniques such as that re-
ported by Savitsky
[5]
are widely used.
The significant dimensions of the hull which affect the
-lGi-srtip
Engineering Center, Norfolk Disision, Combarant
Craft Engi neeri ng Department, Norfol k, Vi rgi ai a.
2
Nurnbers i n hrackets desi gnate References at end of Pl per.
Presented at the Februar-v 14, 19?5 meeti ng of the Gul f Secti on
\Yest of THE SOCIETY OF N.AVAL ARCHITECTS ,\ND II-ARINE ENCI.
NI : ERS.
The vi ervs expressed herei n are the personal opi ni ons of the authors
and do not necessari l y refl ect the offi ci al vi ews of the f)epartment of
Defense.
1 4
f
i Ff t O CHr Nl t
( 7 K o l v
L ) , o , J
porver requirements of a craft have not all been documented
as to their relarive importance. The predominant prediction
method used within the small-craft technical community has
been that developed by Savitsky
[5].
For the case where all
tbrces are assurned to pass through the center of gravity, dis-
piacement, chine beam, deadrise angle, and longitudinal cen-
ter of gravity are required geometric data. The Savitsky meth-
od is baseci o:l prismatic hull form, that is, on craft having
cons[ant beo- and deadrise. In as much as few craft have
these prismaric shapes, designers have irsed various geometric
features of their designs to represent an
"effective"
beam and
deadrise to use the Savitsky equations. Hadler and Hubble;
using the extensive model test data from Series 62 and Serieg
65
[6-S],
used a statistical approach in referencs
[9]
as orre
method of establishing
"effective"
proportions for use with
this analytical prediclion method.
In an eftbrt to ineprove the predictivd process without intro-
duci:ig a new analltical approach, it was decided that
"modi-
f1"ing" the Sarirsky method might produce improved accuracy
in the hunp-speed range while retaining the experiencc an<i
use of exisiing computer programs. This process consisteti of
first making Savitsl<y resistance predictions for a select num-
ber of hull tbrnis for which model test data existed. The pur-
pose was to isolate the effective chine beam which would pro-
duce the best analytical prediction. Figure 1 shows a typical
sample of the resuits for a Series 62 hull- The comparisons
made indicated that the ma-ximum chine beam produced the
best high-speed predictions for craft s'ith constant afterbody
deadrise. In the hump-speed range, which is normally outside
of the vaiid range of the Savitsky method, the resistance of the
hull was ahval-s underpredicted no matter what chine beam
was used.
Likewise, a similar approach $'as attempted to isolate an ef-
fective deadrise for craft having nonconstant afterbody dead-
rise. This effort was much less rew-a.rding. as indicated in Fig.
2, rvhich shorvs a typical comparison of model test data with
predicted results for a huil having longitudinally varying
deadrise. The center of pressure for dynamic lilt in the limit-
ing case is approximately % of the mean wetted length forward
of the transom. The longitudinal dynamic pressures which
were measured and reported in ref'erence
[lOf
show this distri-
bution. In practice, the mean weited length of a commercial or
military craft is seldom less than one half the chine length. In
effect, the latter statement, and the fact that the high-speed
dynamic center of pressure approaches 7r of the mean r*'etted
length forrvard of the transom, virtually eliminates the after
MARINE TECHNOLOGY
d
.i..iEnl*r*&ie.a.i-*&.4ri-r* ffi
I
,
r_
,
,.;
r
..*:1.
' ,*i
i;!i$ogs \J
i ' l z
\
a
Flgr 1 Effec{of chin boam
on rcsbraocc predictlon -.
. ii-i.;r;
G
c
\
t
ll
F
t
$
- M. eel
4{ t bL L. / $n. a. ob
-
Flg.2 Etfect ot deadd3e angla
on resistance prodiction
q
3l,'
il
UI
el to
!l
fl
t
I
d
d
--T
al
*rooet seeso
(*+ \
+ Ercr
Tast I
.catcl D-{+
hoJel 351
-l
PT.A
ctftn
. o. 83
'///\-
(-o.,..
- k ' 7 '
+
4'
3 r "
Q . o
I r "
!
r r o
i , o
\
o .
o
4
,o tt z Za
lrtooit sP.Ep
(Kl
)
JANUARY 1976
?-. ct al ol Ch' aa L2dt h-
i--
I
I
portion of the hull as the location of the effeetive deadrise
ansle. '"iii"
the beam, the effective deadrise angle varies as speed
i".t"*"t iro- ""to- The maximum
chine beam' howevet' was
iJentined as the representative
value for best
prediction
-at
;l;il-;;;"d;.
The
"effective"
deadrise angle. was arbitrarilv
;;il;; fu angle at mid-chine
length,-as this location ap-
;;;;;h"d
trt" pt.iti""t aftmost
high-speed longitudinal center
of
pressure.
"^
ti;;;;;;"nt
afterbodv deadrise
(warp or twist) has been
considered
by some to reiult in a higher resistance
when com-
paredwi thaconstant-deadri setrul l .Thi sconceptmaynotbe
fi;;;i; ttpp"ti"a
tv experimental,data,when
compari-
sons with constant aod wa"ped afterbody
h'rlls are made for
;;;;; where both have equal deadrise an-gles at the center
.] "i"rr*".
For this case tliere is little difference
in relative
il"fi-t-.-t-i"t"".e,-b.rt
there is some difference in dynamic trim'
il;;;;i;!'i"
cor,.id"ted
to be a designer's tool to control
iy""-i" tiin" in the same fashion that bottom
plate exten'
.io* "t" built into craft and bent down to change dynamic
trim if desired after builder's trials'
The establishment
of the
"effective"
bea- asthe mazimum
,h;;; Erii, "ia the
"effective"
deadrise as rhe deadrise
i"sl" ii i;i'chine length,-allows-
the development of an
"en-
ni"""".i"*"
factor that * i" used to modifl' the eristing Sav-
i"tv
ptEai"iion method. The modifying factor reported he-re
*.."".t"Uti.tted
in a rather simple manner' Resistance
predic-
iio"t *"t" made for a number of n"il desiSns for w-hich model
il d"; existed.
For each of these conditions' the rario of
model te.t data to predicted resistance was computed and an-
iiy".d for sensitiviiy
to hull form and loading paraneters'
-'tik"
many designers who have confidently
Ssed
the. Sav-
it.f.u
p."al"iion mlthod for hull resistance
predictions' it was
;;';"iil;;o
the authors to find that the model and predict-
"J .".ltt. were essentially equivalent
at pianing- speeds' As
-r"ii*.a
previously, however, the hump-speed resistance
;;;';;;;;iicted,-resulting
in a correction
ratio generallv
;;;;;
"ne' The collectiie results of obtaining these data
F.. t ri"tt n.rll fot-. have been reduced to an analltical form
;;;;iifutve
fitting
process' The resulting expression
is
/LCG\
t.nu
"_r{F._o.ra
M
=
0 . 9 8 * ,
\ n r * )
r LCGr
-
t
(f f i )
n-3t F: -085)
(1)
Thi sexpr essi onwasnot devel opedwi t ha. t heor et i cal model as
" -"ri.,'""J"ie
stould not expect to rationalize equation
(1)
wiih hydrodynamic
logic.
i;; n factor, from-equation
(1), was established so that it
*.;il;; " iitliptvi"g'factor
to the resistance
predicted bv
;i';t;-pltfi;J
S""ii.tv
method'
(Norr": The Savitskv equa-
;i";. ;t; given in Appendix I for convenience')
Figure
? Iupt::
;;;;G6;ti""
(1)'and Fig' 4 for
{:.-soe{
relationship will
ieduce-the effort required ti apply this modiff ing factor dur-
ine manual comPutations' "
"P;;.;--;t
tlke exception to thenixing
of several dimen-
sionless coefficient
systeins, that is, 'F; and C''' as speed coef-
ficients:
the mi*i.g of dimensionlese
coeffircients to minimize tbe effort
ilJrJio-J"udop
an improvement
in prediction in the
hump-sDeed
range.
"-fi";;;*y
g"oal which led to the development of equatio-n
(1i;;";i-p;o"ve hump-speed resistance
prediction for hard-
;iil;"tft. Any unf"-iliar method is normdly:eceived
wit'h
;;;; ;-k;iGm
(ana rightfullv so) until individual coufi-
;;;;";;fii""d
bv trving or testing the met'hod on known
ilu"t i; it iti"t" i"t"-reut in the proPosed met'hod' Fig'-5 is
"ff"i"a to show a comparison of model test da-!4
[1ot
":t'
-fot
ii" a"""f.p-ent
of equation
(1)l
qnd
$e
modified
prediction
;;t*;;.rted
herein- The speed-resistalcg
prediction is re-
.;;;;;-;"d
slightly consenrative'
Note that the
gelection of
;;J;; "."a'to-develop
t'he M factor favored relatively
h";t ";;
(AplYzts in thJrange of-6'0-to 6'5) and the com-
;is,o;
in Fig. 5 improves ." th" displacernent
ePPto-a9hes
ffit;; "o---Zi"i.r ""a -iut"w loading' The sirnplifie-d Slv'
iGLv
prediction is also shosn ott th"t figure
(computed. using
lilJ'"rF*-u""
t""- ""a dea&ise as defrnJd in t'his report)'
---tli"
irairvittg factor is not a prnacea' for the
qnknowns
can haunt anyone attempting
ptifot---""
predictions' The
hr*" fi-it iions for "jpli"itiott should- be
judicigusly fol'
lowed. These limitations interact with thooe
prevrously re-
;;;; bv iao'itslv and actually extend the usefulness to lower
il;"d".-"T;;
limi-tations for ipplication
of
.
this
prediction
;th"d .l"i At"t those reported in reference [5]
are:
Fv 21. 0
LCG
.
o.a,
Lp
Resistance
prediction for appendages
For hard-chine craft the detail design of apperrdages
may
well result in significant
performance differelces
between two
;;p-;;;;i; ;qd"alent cr'aft. Thus, this subject needs the at-
iJiti"" iftlt has been reported bv Hadler
[ll'which
describes
il"-".i""r"tion
of drag iot sk"gs, propeller shaft' strut boss'
rudders, struts, strum
palms,- and appendage
interference
a;;.-Th";" "quations,
in part, are repeated in Appendix 2 for
convenience. --ii
.Jaiuo" to these appendages,
other items such as seaws'
t"t .t""-i""i. and depth iouoaei tt"*ducers
which extend be-
r""a1l" l"ff.L"ta t" taken into account-when
determining
;;;;";;*;;"e.
Hoerne.
[11]
notes that drag coefficients in
;frJ;;;airltiz
t" 1.20 shouid be used based on frontal area
;-d";;;
of fairing. A value of Cpo
=
0'65-is suggested
for
"oo""a?*". of this tfpe extending
mJre than 20 percent of the
fiIb,il;i;;;J;;i,-l;v"'
thickn-ess
from hull- This is based
o., i"fit..f" trials coniucted on craft with and without
pro-
;;;il;;-**in"r..
N,t-"ticallv this would account for 90 shp
i"tl"Eft-tq"are
foot of frontal area of strainers on a 2O-knot
"i"tt op.t"iing
with 0.5 propulsive coefficient'
--
D;d;";"rid
."centiv
[iz,
rsl offer the best experimental
irf.;;;i,t" for rudder it.e in free stream and in the propel-
i.i-.iiJ stream for a rang-e of craft spee.+ approaching-
40
lnoi..in.."
references
giie results foriu{9ers having airloil'
p.i"l"fi., n"t plate, andwedge sections' With the availability
of th"." data, proper allowance of strainers' and the data re-
ported by Hadle., a very detailed appendage
drag allowance
i""-U" *'"a" for ihe final design of any hard-chine craft'
--Whlf"
tn"* appendage dra-g calculations are laborious' they
"r" ,,oi aiifi.ult'and ".J""""tt"ti"l
when considering the corrdi-
U"*1o. which the final propeller is selected' These detail ap-
"."J"*" drag calculations,
howeuer, are not completely
justi-
i't;-il
pr"ti*in"rv design studies where various craft sizes
and arrangements
are being considered'
For such preliminary
re=ffi
(2)
g. = -! :
(3)
! gBpx
This mixing resulted from practical rather than technical rea-
r;;-.-i{";;"ing
the exisiing Savitskv
prediction method
l*li.r, uses C'iwas a guiding philosophy, and the majority of
)Joj;i't"eii"d_ai^
""oit.lt" to"the authors
(used F.s) sussested
1 6
MARTNE TECHNOLOGY
. - - . - - . . - - - - - - - - . - - - - - - . - - - - - r i ! ! - -
AE
=
expanded area ofpropellerblades
(ftr)
=
EAR(Ao)
Ao
=
disk ara of propeller (ft2)
=
rD214
4r
= projected area of propeller blades
(ft2)
' =Aa(1. 067 -O-z?9Pl D' ,
Bpx
=
maximum chine beam ercluding
external spray rail (ft)
Elr
=
transom chine beam (ft)
Cao
=
drag coefficient for seawater
strainetc
Crrp
=
drag coefficient for strut palm
C oa
=
drag coefficient for rudder
Cr
=
friction drag coefficient
Cea
=
deadrise surface lift coeflicient
Ceo =
zero'deadrise lift coefficient
=
distance of center of pressure (hy-
&odynamic force) measured
along keel forward oftransom
C"
=
spged coefficient
I
-
t gBpx
Cr
=
load coefficient
=
W/{w Bpxr)
|
=
propeller diameter
(ft)
Dr
=
skeg drag
(lb)-
Do
=
dragofseawaterstrainers
(lb)
Dp
=
drag of strut palm (lb)
Da
=
drag of nonvented rudder
(lb)
D"
=
drag of nonvented struts
(lb)
Dsn
=
drag of inclined shaft or strut bar-
rel (lb)
d
=
dianeter of shaft or strut barrel
(ft)
e
=
2.71828
EAR
=
expanded area ratio
=
Ae/Ao
ehp
= go1"1.tt"ctive
horsepower
(hp)
ehpux
Rr V
=
3219
=
effective horsepower, bare hull
=
BHrr V
325.9
=
volume Froude number
t'
r g i i s
a
=
acceleration due to gravity (ft/
sei)
=
32.15 ftlsecl
Hr g =
significant wave height
(ft)
' ft
=
depthof propeller Q belowwater
surface at rest (ft)
hp
=
strut palm thickness
(ft)
J;
=
apparent advance coefficient
=
u
lnD
Jr
=
thrust advance coefficient
u Q- Wr l
= ---;6-
-./q =
torque advance coefficient
u 0 - W q l
'
Kr
=
thrust coefficient
T
=; FF
Ke
=
toroue coefficient
- Q
-
PnzDs
LCG
=
longitudinal center of gravity
measured from transom (ft)
LOA
=
length overall (ft)
Lp
= projected chine length (ft)
I
=
wet length ofshaft or strut barrel
(ft)
M
=
multiplying factor
[equation
(1)]
.
71
=
propeller rotational speed, rps
=
N/60
1Y
= propeller rotationd speed, rpm
_
y( t __
wr )
( 101. g)
Jr D
NpR
=
numberof propellers
OPC
=
overall propulsive coefficient
=
ehpsnl' shp
P
=
propeller pitch (ft)
P/D
=
pitch ratio
P..r
=
atmosphericpressure
(lb/ftz) =
2116 (lb/ft' ) for 14.7 psi
Pr
=
static water pressure (lb
lft2t
:
pgh
P"
=
vapor pressure ofwater (lb/ft' )
Q
=
propeller torque
(lb-ft)
=
shp (5252)
r
Rr
=
added resistance in waves (0
for
calm water)
(lb)
Repp
=
appendage resistance
(lb)
Rsx
=
bare hull resistance (lb)
Re
=
Re:rnolds number
Rr
=
total resistance
(lb) =
RsH *
R..rpp * 8e
rpm
=
propeller rotational speed
=
N
shp
=
.6"L horsepower ihp)
-
2rQN -
ehP
33,000
qt
S
=
transverse projected area ofrud-
der or strut
(ft2)
So
=
frontal projected area of seawater
inlets
(ft2)
Sr
=
transverse projected area of skeg
(ft2)
T
=
thrust of each propeller (lb)
- R :
-
( l
-
t ) Npr r
Tr,,r.nr
=
total craft thrust
(lb)
Rr
=( Fl
f
=
strut thickness
(ft)
f/c
=
thickness to chord for rudders or
struts
V
=
velocity of boat
(knots)
um
=
rnan velocity over planing sur-
face (fps)
u
=
velocity of boat (fps)
uo.t*
=
square of resultant velocity of
watcrat 0.? radiusof propeller
(fps)'
=(
#
!=t' aq\ ,,
\ J r t l
u
=
weight density of water
(lb/ft8)
W= displac?ment
(lb)
Xp
=
distance fiom stagnation line to
'
appendage
(ft)
Y
=
width of strut palm (ft)
(1
-
c)
=
thnrst deduction factor
=
R7/
?torel
(1
-
We)= torque wake factor
=
Je
lJt
(l- Wrl
=
thrust wake 1r"1o,
=
JTlJt
F
=
deadrise angle (deg)
c
=
angle of sbaft relative to hlttock
(deg)
p =
massdensityof watcr (lb-.dl
ft+)
=
1.9905 for 59oF salt watcr
r
=
kinematic riscoeityof tttrr(ftt/
'
sec)
=
1.28L7 X 10-5 for 59
tF
ralt water
tr
=
mean wetted length-beam ratio
?.4
=
appendage drag factor
r
: - - -
0. 005Fc2 + 1. 05
qq
=
propeller opgn-water eflicieocy
=
(
[:\(L\
\
f f ol \ ! r /
?o
=
propulsive coefficient
eho
=
--
=
ao4xrlr
4s
=
hull efficiency
?n
=
reiative rotative efficiencY
F
=
volume of displacement
(fti)
_
L(2240)
p g
.f
=
displacement
(long tons)
=
F/
. 3 5
6
=
boundary-layer thickness
(ft)
ACr
=
conelation allowance
AD
=
interference drag (lb)
o
=
cavitation number based on boat
velocity
P. q + Pr r - P,
lhl
p u2
6o.7R
=
cavitation number based on re-
'
sultant water velocity at 0.7
radius of
ProPellers
f J''" 1
=dLZt +4r { J
rc
=
thrust load coefficient
T
=
6pApuont '
r
=
trim angle relative to mean but'
tock
(deg)
$'P=
I('lP?',0.0b4e3)
(l
-
we,)r
Vr ,Juu
I knot
=
1.68?8 fps
Fg
JANUARY 1976
studies the following approximation for appendage drag factor
has proven to be useful:
flA
=
0.005 Fvz + 1.05
(4)
This equation is based on a collection of data from twin-screw
hard-chine model tests made with and without appendages.
This expression is slightly more conservativethan the append-
age factor data reported in reference
[21.
Nunerical valuee for
equation (4)
are given in the following table:
The
by
magnitude of appendage resistance is then represented
R,tpp
= (RBH)
(A
-
l) (5)
Added resistance in waves
Craft performance in a sea is best predicted by model tests
conducted in a representative random sea in which tbe craft is
expected to operate. These tests give added resistance in
waves as well as motions and accelerations needed to design
hull structure and to estimate credequipment
lirnitations,
These types of tests are of great technical value aad return the
dollars invested when only a few craft of given design are pro-
cured. For design studies or a
"one
of'construction project
Fridsma
U4.
151 offers an excellent source of rough-water per-
formance technology- for hard-chine craft presented in a for-
mat for use by designers.
The calcu-lation procedure for added resistance in wavee
from reference
[U]
is reproduced in Appendix 3 of this reporl
Propulsive data
After identifying a rneans of predicting the speed-resistance
reiationship for a craft, it follows that the interrelation of the
hull-propeller must be described in order to properly include
propeller characteristics. The propulsive data are the transfer
I
Fig. 3 Variation of modifying tactor with volume Froude number
-
andLCG/9zy
F19.4 Variation of vobme
Froude number with speed
and displacement
Ft 1.0 1. 5 2.4 2-5 3.0 3.5 4. 0 | 4. 5 5.0
tl,n
0.948J.9420:9340.925).913
n
qnr
0.88510.86! c.851
- 24Mo
- j o,ooo
-
+qooo
-
5o,ooo
-toqooo
-2O4OOo
-3OO,OOO
-4o4ooo
w- La
SPE ED
. AT
MAFINE TECHT.IOLOGY
' a
ot o- d
8 l l & 1 "
sffir
,Jg
Z-
HarA.
tarx
)
.i
v
s
b
l*
q - !
I
\
' \
s
a
l\
C \ -
\
s
L
c . )
\o
5
+
I
L
-l
r T
I
L
q t
t
-l
s | {
s l
t
(.l
s
v'
I
a
L
!
s
c,
{
6q
{
{
\
' +
N
(J
6
6
!t
6
o
o
E
E
E
3
o
()
g
(t
6
6
o
!,
o
o
E
o
CL
o
C
o
c!
qt
o
E
o
C)
|.'
g
IL
e l : l
r l i l
$l l
'i
, N
\r
, a a r t l
\ n + r l n f \ S
A1-
t JNYJst s2v
' 77oaU
\j
\'
a
(
\
t
s
\t
L
l \
N
!
,ll
q
\
\
s
, [ ^
{ \ }
s :
s $
F A
{ r *
\ b
l v i
i c
i
s
v
\
i
* -
s l
U $
\ \ R
*\ i:
\ \
f , !
F S
:
V)
JANUARY 1976
l.l
s-"
l.a
0.t
l.t
a\"
ra
0. ?
l.t
1. 0
0. 1
08
/.5
e.0
e.5
tr
Fig. 6 Twin-screw propulsive daia
I t r
t ' l r ' r -
functions that describe this interrelation and, unforturiarely,
this area of hard-chine technology has the leasr published in-
formation.
Hadler and Hubble
[2]
presented a very complete synthesis
of the planing craft
fropulsion
problem for single, twin, and
quadruple-screw configurations. This work reports computed
values of
(1
-
Wl and
(1
-
t) for vdrious shaft angles and
speeds. Reference
[16]
reports experimental propulsive data
obtained frdm lull-scale trials of h twin-screw craft. In addi-
tion to these.sdurces, other model and full-sbale experimental
data for twiii-screw crdft have been collected and found to
consistentfy fall within reasonable bounds with some variation
with speed. The range ofshaft angles for these craft was from
10 deg to 16 deg measured from the buttocks and may be par-
tially responsible for the bandwidth of data. These data col-
lected for twin-screw eraft are reported in Fig. 6 with the
mean values and observed variations. Limited propulsive data
for a single-screw small craft with a skeg have been reported in
reference
[17].
One significant difference in definition used in this paper
must be clearly
lrnderstood
so that the thrust deduction factor
(1
-
f
) reported in Fig. 6 is properly appiied. Using (1
-
t) in
the classical sense, to describe a resistance augmentation
where the propeller pressure field changes hull flow patterns
for hard-chine craft performance prediction, requires iterative
computations to resolve equilibrium cooditions of the various
forces and moments of the hull, appendage, and propulsion
systems. The thrust deduction factor (1
-
t) reported here
was experimentally obtained and computed as the ratio of ap-
pendaged resistance
(horizontal
component of resistance force
20
25 ho
Oa/DF rr?5
r l
? L < t ' t h -
i . L
J 0 . ^ I t
s
!
J l ' t
(
B c a
' ) t
t
'
rtPal aDrD
Lo*a sfi3
$FaBrHai{fll-
OATA
$ r a
I
-
o.?
{J
I
J.O
'
- ' t ' t
i \/
'
fu"u yla
when towed in the shaft line) to total shaft line thrust
(when
propelled at full-scale self-propulsion point)- Thus, this modi-
fied definition of
(1 -
t) includes the effects of the classical
definition as well as that for the angle difference between the
resistance and thrust vectors, the trimming effects, and result-
ing hull resistance change, due to the propeller pressure field
acting on the hull, and sirnilar trimming effects for propeller
lift resulting from operation in inclined flow.
Propellef characteristics
Since most working craft operate at fairly high speed and
propeller loading, their propellers more than likely operate
with some degree of cavitation. Cavitation adds a new dimen-
sion to propeller characteristics. Operationally, this variable is
most often reflected in a nonlinear speed-rpm relationship
near top speed. It is generally detected as a
"gravel-passing-
through-the-propeller" sound which may be heard in the la-
zarette above the propellers, and as erosion of propeller blade
material.
Relative to the predictive process, cavitation must be ac-
counted for as a change of propeller characteristics as reported
in reference
It8]
and seen in Fig. 7. Systematic variations for
several types of propellers have been reported giving the ef-
fects of cavitation on characteristics. Of these sources, Gawn-
Burrill
[18]
represents flat-face propeller sections similar to
most commercial propellers made for small crafl A compari-
son of cavitation data for a four-bladed commercial propeller
with the equivalent blade area for a three-bladed Gawn-Bur-
rill propeller was made in reference
[te].
To quote this refer-
\ - - -
\ - l
. . . - - _ .
- - -*
- _/
MARINE TECHM)LOGY
It
(
t 2
,Ir
? / D, 1 . 2
EAR
'
o. t ob,
Flg.7 Effect of cavilation on propellrr charactsrbtics
t.8 20 t.?
cr
J
9
z2 zl 2.o l.?
,z .. .4 .t .(t .7 .e I t' o t' l 1.7
t' 3 t' 4 l ' 5
?
t Je
ence,
"A
comparison of these two sets of data indicates that
the Gawn-Buirill data can be used to make engineering esti-
mations of this type of commercial
propeller performance
when operating in i cavitating condition-" This' in fact, has
been consistent with other experiences where speed, power,
and rpm were measured on new craft, with the importdnt-ex'
ceptiin that Gawn-Burrill datd are uery optimistic where
piopeller sections were thick and the leading edge was blunt
with a poor-quality
finish.
The Gawn-Burrill
propeller characteristics are in the form
of K7, Kq,
?0,
versult Jr fot various pitch ratios, blade area ra-
tios, and'cavitation numbers
(o)- This familiar format, how-
ever, can be replaced with another which reduces the effort re-
quired to optimize the propeller; select the reduction ratio,
and make the speed-power-rpm
predictions. This format is
that of
?o
and Jr versus KrlJr2 for various PlD, EAR, and a.
The entire Gawn-Burrill
propeller series has been redone in
this format and is presented in Appendix 4. The effori re-
quired to recompute and rbdraw these curves is compensated
JANUARY 1976
for by the long-term savings in preparing performalce predic-
tions and optimizing proPulsion systems'
Power
Prediction
method
The material presented gp to this point have noted data
sources and logic leading to assumptions necessary to estab-
lish the data 6ase for making a power prediction for craft'
Now is the time to put it all tqgethdr.
It is important to keep one thought in mind when working
with propellers. Propellers produce thrust- While-engine
poor.ii. converted to thrust horsepot'"t by the propeller, se-
iecting a propeller to absorb power at a particular rpm and
speed-does nbt n"r"r"ority yield the rnarimu.m speed polen:
tial of a craft. The following
procedure describes the method
to effect a speed-power prediction, beginning with speed resis-
tance and then establishing the thrust requirements
The development of Xrl,lrz as the common variable be-
tween hull thrust requirements and the propeller characteris-
ii-lr"i'. i I
:-!:,9' i-i
tics has the.distinct aduantage of eliminating propeller rpm
from
the early prediction calculations.
For each propeller
T
,
u( l
- Wr l
Kr=;fu and Jr=:i-
Therefore
Kr/Jr2
=
pD2uz(L
-
Wr\2
,
RsHV
ehpgx
=
ffii
(14)
is used to compute bare hull or overall propulsivccfficient
as follows:
oPC
= *P*
snp
(15)
Kr/Jr2
=
pD2uz(t
-
Wrl2(t
-
t)(N pn)
Once the number of propellers has been established for a craft
design, the only significant variable that can influence Kr/Jr2
is the propeller diameter D. This relationship
[equation
(8)]
is
the basis for the format change in propeiler characteristics as
presented in Appendix 4. This format can be used for any tlpe
of propeller such as the Troost series, Newton-Rader series, or
supercavitating CRP series.
Once the thrust loadinc (Kr/Jr2) has been established, the
equilibrium condition between hull requirements and propel-
ler bapability leads, in general, to a unique value of open-water
propeller efficiency
(no)
and advance coefficient
(Jr)
for that
craft speed
(and
corresponding cavitation number aJ:
Pa * Pu - Pu
(s)
(10)
6 =
(\ )pu2
The appendaged propulsive coefficient is computed as
where
( 1 - r )
nH
=
(l=w;
The total shaft horsepower (shp) is computed from total ehp:
The difference between equations (10)
and (15)
for smooth-
water conditions
(zero
sea state) is mostly a rcsult.of the ap-
pendage drag factor (l,l) with minor effect due to tlrs tenden-
cy of propeller efficiency to reduce with incrsagiag tirust
loading. This latter factor becomes very important as the pro-
peller begins to cavitate. Thus, for moderate speeds and thrust
loading in smooth water
oPc
4D=
-
,lA
(16)
The speed-power calculation procedure, applying ttre ap-
proach briefly discussed, is best demonstrated by following
through a data calculation form. The sa.mple foru with col-
umn-by-column ."lsrrlation procedures as givsr in Table 1
wiil show, in practice, the interrelationships of hull resistance,
propulsive data, and propeller characteristics- The numerical
example depicts a 50-ft craft operating in rough water, and is
provided with the results presented in Fig.8.
It is important to note that engine characteristics play no
part in the speed-power requirements
(other
than impact of
machinery and fuel weight on total displacement) once hull
loading, size, appendages, and propeller geometry are fixed.
An engine and reduction ratio must be selected with charac-
teristics compaiible with predicted speed-ponrcr-propeller
rpm needs for fixed-pitch propellers since the propeller con-
trols the engine power oufput ai a given rpn up to the maxi-
mum power capability of the engine.
Appl i cati ons
Any rational power prediction method has many uses be-
yond that of
just determining the speed-power-rpm relation-
ship for specific hull and propeller combinations. Ingenuous
deSigners find analysis of full-scale craft performance relative
to predictive techniques often leads to improved performance.
Additional uses of this speed-power synthesis are discussed in
the following paragraphs, and it is hoped that theso will stim-
ulate other applications.
Hull proportions for smooth-water minirnum ehp
Whenever new requirements arise for craft operations it
may not be unusual for a new-size supply craft, crew boat, or
patrol craft to be developed around existing engines to satisfy
these requirements. The best economic resolution of craft size
relative to requirements (such
as payload, speed, range, sea-
keeping, maneuverability, and crew size) should lead to a de-
sign study to establish the technical and financial impact of
each requirement.
In the context of design studies, the modified Savitsky pre-
diction method, discussed here under the topic of hull resis-
tance, offers a reasonable means to establish craft proportions
for minimum bare-hull ehp in smooth water. As stated pre-
viously, displacement, chine beam, deadrise, and longitudinal
center of gravity are the significant factors affecting speed-
power when all forces are assumed to pass through the center
of gravity. Considering minimum smooth-water ehpss to be
the desired goal, an iterative series of calculations was made
for a wide range of these significant hull factors for speeds
from 15 to 45 knots and for displacements from 10,000 to
400,000 lb. Also, these calculations were made for 59oF seawa-
T
(6)
For the hull
r TorAL
=
O:t
Assuming that each propeller produces equd thrust for a mul-
tiscrew craft, the thrust required by each propeller would be
R7
f
=ffi
i
N'
rl,prr't, (7)
Equating the thrust requirements that each propeller must
produce to that required to satisfy the hull resistance leads to
Rr
(8)
tlD
=
rl,qHnR
by the following equation
,
RrV
enP
=
E 5s-
,
ehp
S n P = -
qD
*=W( 101. 3)
The corresponding value of Jr defines the propeller rpm
(N)
as
( 1 1)
( 12)
( 13)
Most designers of small craft are familiar with bare-hull
propulsive coefficient or the term overall propulsive coeffi-
cient (OPC). The magnitude of 0.5 for OPC has been used for
years for preliminary power estimates. For current design
practice a value of OPC
=
0.55 is readily artainable for twin-
screw craft. This is mentioned to emphasize that the propul-
sive coefficient
(lp), in equation
(10),
and OPC ore not the
same. Bare-hull ehp
22 MARINE TECHilOLOGY
ter and zero correlation allowance. The iteration was effected
by making incremental increases in chine beam until mini-
mum ehpsH was obtained while LCG/Bp;, deadrise, displace-
ment, and speed were held constant. Thus, both LCG and Bpy
increased at a constant rate during the search for minimum
ehpgH.
This optimization process ig illustrated graphically in Fig. 9
for one condition of displacement, deadrise, and speed. The
results of these calculations are presented in Appendix 5 as
contours of LCGlBpx and displacement
(W) relating mini-
mum ehpsH and marimurn chine beam
(Bpx). Each figure of
Appendix 5 is for a constant speed
(assumed
design speed)
and gives results for deadrise angles of l0 deg, 16 deg, and 22
deg.
These conditions for minimum ehpeH in smooth water per-
mit interesting speculation if one does not introduce extrane-
ous thoughts.
(The
authorc are well aware that other factors,
such as constructed weight, seakeeping, paylbad, cost, longitu-
dinal and traverse stability, affect craft proportions. This ap-
plication (Appendix
5), however, is limited to smooth-water
speed-po*er.) Most designers know, and these data show, that
"real"
hard-chine craft are too heavy relative to their size-
Since the ratio of LCG forward of transon to length overall
(LCGILOA) is usually in the range of 0.37 to 0.40, it might
take a zero payload condition for a craft to operate at a hydro-
dynamic condition for minifium power. (Exqmple:
At 30
knots,
6
=
16 deg, and a displacement of 100,0O0 Ib, an overall
craft length of 90 to 95 ft would result in ehpex of 1000 on a
craft with l6-ft chine beam.) Thus, a 90 to 95 ft craft could at-
tain 30 knots with approximately 1820 shp
(OPC =
0.55) at a
displacement of 100,000 lb. But, could a craft of these propor-
tions and power be constructed with adequate allowance for
fuel and useful payload?
It is interesting to note that for speeds of 30 kaots and
above, designers can relegate chine beam
(Bpx)
to a position
of minor consideration relative to powering requirements (see
Appendix 5). Thus Bpy cara be selected for other important
reasons such as seakeeping, internal volume, deck area, or
transverse stability as discussed in reference
[4].
Many tradeoff relationships can be extracted from Appen-
dix 5. Figure l0 shows the effect of design speed on the selec-
tion of chine bearn for minimum ehpsn in smooth water' Lile-
wise, other tradeoff relationships, such as deadrise effects o'n
ehpBH as shown in Fig. 10, may be extracted as user needs
arise.
Selecting best
propeller and reduetion ratio
The
"best
propeller" for a craft is that which satisfies the
craft thrust requirements within geometric, financial. and
power limitations. If there were no design constraints, an
"op-
timum propeller" could be designed for maximum efficiency.
With this slight distinction of terminology, optimum propeller
and best propeller are not considered to be equivalent and the
term
"best
propeller" will be used here.
The geometric constraints to be considered may rvell pre-
clude the'selection of an operationally suitable propeller.
Thus, it is important to establish the maximum propeiler di-
mensions allowable for the shaft angle, tip clearance, and
draft limitations. Most craft have shaft angles in the range of
10 to 16 deg measured relative to buttocks, and propeller tip
clearance of 15 to 25 percent of diameter. Smallest shaft an-
gles are generally employed on craft with highest design
speeds, and tip clearances are controlled to a large extent by
propeller-induced vibration, which is often traced to extensive
cavitation.
The key to selecting a best propeller to satisfy the craft pro-
pulsion needs rests with equating required crafi thrust lvith
JANUARY 1976
propeller thrusL Equation (8)
defines the thrust-speed-pio-
peller diameter relation as required for equilibriun oadi-
tions, and the speed-power calc-ulation form should be fol-
lowed. First, assume three or more values for propeller disme-
ter, not exceeding geometric constraints, and
qsqrrme
three
values of cavitation numbers corrwponding to speedr above
and below the design speed- Perform tJre calculations fron
Column I to Column 11 according to the procedure described
in Table I for each combination of speed and diametcr. Rec-
ord in Column 20 the propeller diameter used in each line of
the calculations.
Based on previous experience a designer will usually have
an estimate of the expanded area ratio of propellers on similar
craft. If so. the closest value of EAB available in the Gawn-
Burrill series
(Appendix
4) should be used with the propeller
characteristics for the remaining calculations-
(Some gurdes
for approximate values of EAR are as follows: Three-bladed
conventional stock propellers, use EAn
=
0.51; three-
bladed wide-blade stock propellers, use EAR
=
O.666; four-
bladed conventional stock propellers, use EAR
=
0.665)-
For each line, use the values of Kr/Jrz and o from Colunns
10 a-nd 11 to enter the appropriate propeller characteristics
curves in Appendir 4, and locate the 6s*ims1a value of efli-
ciency
(m) for that thrust loading. Record the marimum
49
and corresponding
(Jr)
and
(P/D)
in Columns L2,13, and 2l
respectively. The calculation form is then completed throtrgh
Column 18. Compute pitch and record in Column 23.
These data are plotted as shp, rpm, and pitch versus speed
for curves of constant propeller dia-sgg. Construct a hori-
zontal line at the installed power level that intenects the pre-
dicted speed-power curyes. Construct vertical lines passing
through each speed-power intersection point up to the speed-
rpm and speed-pitch currtes for the corresponding propeller
diameter- These ini.ersecting points are then plotted on a base
of propeller diamerer, that is, (i)
speed versus diameter at de-
sign power; (ii)
rpm versus diameter at predicted speed for de-
sign power; and
(iii) pitch versus diameter at predicted speed
for design power. This process is illustrated in Fig. 11 with in-
tersection points identified in both graphs- The ratio of pro-
peller rpm to engine rpm yields the desired reduction ratio.
Slight adjustments in propeller pitch are usually required to
match stock gear ratios. A numerical example of this propeller
selection procedure is presented in Table 2 and Fig. 11.
While this procedure establishes the best propeller diame-
ter and pitch for the assumed EAR, it does not establish that
the blade area is adequate relative to cavitation effects other
than from a performance point of view. Important factors af-
fecting both the hull structural design in the vicinity of the
propeller and the blade cavitation damage must be consid-
ered. Blade rate-induced hull pressures on the order of 4 to 5
psi can be generated by a badly cavitating propeller, and can
fatigue (crack) hull plating after short periods of operation.
Likewise, these blade cavities can be destructive to the propel-
ler, eroding blade material to the point of requiring frequent
propeller replacement.
These effects can be minimized by carefully selecting EAR
such that rc
(a
thrust loading coefficiency related to pressure)
does not exceed the 10 percent back cavitation relationship
defined by Gawn-Burrill
[1.8].
Thus, if
re
<
O.494(oo.za)o' 8ti
(17)
(an
approximation of the Gawn-Burrill 10 percent back cavi-
taiion criterion), one can be confident that the propeller has
adequate blade area. 16 and oe.76 take into account the resul-
tant of both rotational and axial velocities and are cornputed
as follows:
t
(tert
continued on page 261
Tabtc I P.opellct selsctlon and speorl'powcr
calcutatlon
procodurc
cppFT
4a,trr?c
oet-!%
cptcutsteo 8r 4'
z @'Wo
?.
#T
)
fu2eo',
r t d *
-.e*e
u ec
(r
)
-EL
27. e6t)t.zs'
rrOgl4n) JJ-'/'
g EP"
4,e'
1'""? * f;',t'ffi
I i . zt t I t . r o I I br . I t 7l
| ??t
l . t Ea l t
o o l . ? 6 | | 981 | 7o7
I t sat i t zza
I l . o a 1 i o 5 o |
b D
cta<UEzo4s
rcE PEt Df'1pN
sladu
lN F/4 UP E
' I
Note: Lines 1 to 10 may be used for propeller selection or
sPeed-Power
calcutallorui'
DescriPtion
Craft identification
Date of comPutation
F&!"" makiig comPuiations
biipiacemet
tbf craft
(l-b)
LCG of craft measured
from
-
aftmost Point
of Planing
bottom
(ft)
Maiimum' chine
beam exclu'
-' -di ne
ext ernal sPraY rai l (f i )
Oeaai i se at mi d-i hi ne
l engt h
( deg)
Mass densitY ol water
( lb sec' /ft' )
Ki nemat i c
vi scosi t Y of wat er
(f t 2l sec)
Cdrrelation
allowance
Nominal sea state
Vapor Pressure
of water
(l b/ f r'
)
O.ioitr tci t of ProPeller
hub
nieasured from
r*' ater sur-
f ace wi t h craf t ar rest (f t )
Propel l er di amet er
(i n.
)
Proi rel l er di amet er
(f t )
Probel l er Pi t ch
(i n. )
Probel l er Pi t ch
(f t )
Propeller Pitch
ratlo
Probel l er
6xPanded area rat i o
Nut i t b"t of
f roPel l er
bl aci es
Number of ProPel l er
shai rs
Data
Soeed
Rows 1 t o 10
Rows 11 t o 15
Rows 16 t o 20
Bare-hull resistance
Appendage
resistance
Added resistance in. waves
(0 f.or calm water)
Total resistance
Vol umn Froude No.
Thrust deduction
factor
Thrust lvake t' actor
Relative rotative efficiencY
Thrust Ioading
Cavitation No.
hopel l er ef t ' i ci encY
Advance coef f i ci ent
based
on thrust
Appendaged ProPulsive
co-
ef f i ci ent
Source
ffi*:$Y?llli.' vr20
b.-iitt"a
for o of-Co.l' 11
tt3*?;'"'i'.3lo1fl"',$"?1"'"s,",
rudders,
etc. or estimated
rrom
"*:il"-*11f]. lrsr
Sum of Col s. 2, 3, and 4
6."iptt"a
frorirtsq'
(2) or Fig' 4
IVlodel tests or
llg- P
Irlodel tests or
f:rg' 9
U"S"J",'""'i"r:i'i-'f,
al1g1
i ombuted
from Eq. (9)
Obtained
from ProPeller
cnarac-
';;;;;i;
f"' civitllgn
No' and
k;ii;'at
proPer P/D and'
"J,f
'S'E:l*;3'l'i"ll"'li;r
Comput ed
f rom Eq.
(10)
Comout ed
f rom Eq-
(11)
Combut ed
t ' om po.
{}?)
Combut ed
f rom Eq- (13)
Repei t of Col ' 1
-. . .
Model t est s or
Predrcl l on
Assumed
values
^fl"*
ePP""dix
4 for
4o
oPti'
mum
fi""1fr,11i1':?
gok..2o
and 21
g:nixl:l
ilsil El:Ill]
CoI.
1 .
Dct a
O:aft
Date
Calculated
bY
Displacement
LCG
Bpx
0
p
v
LC.c
Sea scate
D
Deot h r of
proPel l er
D ( i n. )
D ( f t )
P ( i n.
)
P ( f t )
P/D
EAR
No. of blades
No. of shaf t s
2-
J .
^
i
o.
t .
8.
9.
10.
1 1 .
t 2.
l J .
L4.
20.
2r .
15. Tot al ehP
16. Shaf t horsePower
17. ProPel l er
rPm
18. Speed
19. Tt i m rel at i ve t o mean
but t ock
Propel l er di amet er*
or ext ra
Oofimum P/D* or exfta
EAR4 or ext ra
Propel l er Pi t ch*
or ext ra
Bare-hul l ehP
Overal l
proPut si ve coef f i ci ent
Ext ra
().,
23.
24.
25.
26.
I
I
I
t
i
N
I
t
a-
i
F
r
*
Data used for propeller selection'
MARINE TECHNOLOGY
U. tar&ro L6.
t C4. t 8. ?S Fr
92,
2
l4.o Fx
? .
t b '
o . 2 7 "
?/ o, 1, 82
6. r r l ?. o. ( 165
No, gLPDCS. 3
Uo.Sttt Bft .
3
FlS. 8 Predicted shp, rpm, and trlm
versus speed for calcdatad exampb
V . 3 o R
9 '
t t '
t / ' l OO, qr O l A5,
t5 /O t7 tt t? Zo Zt
BFr<
Flg. 9 Graphical reprasentation of process tor establishing minimum ehpgH
rl
b 3 '
seeeo (rrt)
rl
$
;t
sl
il
- , / 1 . . / i
| / l . z - @e e s o t
t l
I I
Mr Nt MuM
E
I t _ -
t5 /o 17
JANUARY
1978
$|
il
T
? ^ = -
rhpApuo.zn2
f J r z l
oo. ?R= r LFf f i , l
The data for these criteria are presented in Fig. 12. The pro-
peller selected in the example shown in Fig. 11 should have an
EAR
=
0.82 to satisfy the 10 percent cavitation criterion at
maximum speed. Since this value of EAR is greater than that
used for the propeller selection calculations, it should be re-
peated for EAR
=
0.82 to be certain the best propelier has
been obtained.
Should the value of EAR exceed 0.72 to 0.?5, then it is un-
likely that a stock propeller can be purchased. If this occurs,
the designer has the choice of preparing a custom propeller
design or obtaining relief from geometric
constraints to permit
us9
9f
a larger-diameter propeller to reduce 16, to an accept-
able level.
Full-scale performance
analysis
Builder's acceptance trials are the true test of the craft de-
sign effort as interpreted by the designer. The detail with
26
which the builder reproduces the design detail is reflected in
overall craft performance.
In order to ritionallyinterpret trial
results it is necessary to document the size and location of clt
underqrater appendages, measure the propeller pitch and di-
,heter,
note the leading edge detail of the propeller, measure
craft displace rnent and, LCG.
In order of experience with problems relatcd to low trial
speeds, the authors have found the No. 1 cause to be stock
propellers with blunt or thick leading edges, or both, or nomi-
nal pitch no better than *1 in- The second most frequent of-
fender is overweight construction relative to preliminary ac-
cepted weight estimates- (Either
better weight estimates or
bette-r rx,eight control during construction lre required to
avoid this problem.) In third place is the incorrect allowance
for drag during performance predictions. Also, craft that have
been in service for some time are often inflicted with a heaw
coat of marine growth which results in speed loss. Docu--
menting and solving craft performance problims is in itself an
intere:ting career not unrelated to that ofa detective. (propel-
ler vibration and loeal blade erosion problems do not come
within the context of this report.)
Data acquired from most limited trials will consist of visual
inspection of underwater portions
of craft, some estirnate of
displacement and LCG, and speed versus rpm up to dead rack
Ffg. 10 Effectof deslnspoedon Bp4ard
minimum ehpsl variatinn with deedrisc
' t
( 18)
(1e)
oEstoAt seeto
(za)
MARINE TECHNOLOGY
I
I
I
l l
h\ l
\ l
\ l
! l
t r l
il
" l
Flg. 1'l Plot of calculation example for propeilar selection
Table 2 Propollet sgleclion and speed'powet calculalion
Proceduto
RFPT
sl rP
'
l Soct
E^,7 . O-bC.,
CczFt
5a' Hrc
.
Drrs% Cl t<yt1zto.B' 3
?' 3
; r : ;
* - " " @'
; 7: . ' s.
f f i - - - . ' dv6l ) - 1e-
o, , r , c
o r
pt
o n t t e r ( n )
- zZL
Pa^rt rt . 2 Ot 72' D(u)
o(rt)-
cw)-{rr)-
"/c-
tR:!!! r'Jo.
gteoes;2- No saarts--Z,
z a 5
z 4
Vn
gu Ro Ro E" t - t ,-Yr
| 7. l*i?:
F Jr l ' / o
I Et l F s HP
i
Rn
Qa
P/o
ene | 4a
2pc
td 5rt 7t1 at 2t t.rt o.rz a f 7 . &1 a4 l l 8 4 l n l 9 za*
2
t 7 t I ?1
r t e ! 1 5 7
z @ t . 4 2. aa
:
!
t
. r1t
q 3 4 L b t . t t
l ?rt bt 12 b:x I5t ,)2' t -r7 ,t lJ2 o. 75
q ? 1a7 l Lot I Zt Oa tLtT t . 4 2. \ 2
I ? ! l f b l I t 5 a t 1 z l 2bo 1. 4 z. t . ,
at I 1 5 { a t I r a t o t b I z. t t l . t t
, 4t t ztl Uoa' t.a, ^t f o.9 3a $t6 2ot b 2444 t f b, z.e,
t
I 4 t t 517 I s70 t t 2
t l
I
2.tt 2. o
v - l t i P t s / P
l " P n
t7
, t
t 5
I
l r . a t i I i
t 7
l . t o I
,]
a. t
t o. 30
PaH.FX S
?. a?aLaaa t aLt cTeo Foa EPp t 0. 6t
o - a ? . ? . L , 2 .
? l o . I s z
7oa t ! af { a| Ar aet ct aYl cg
oF 9a t i P l T 29@ ' d
: . csro$ RArro t 3 zt oo / l l bg
.
1. 77 b
, t . r7 : ,
ctacuapTpNa FoE
ppaotcrlaN
sHoHN lN FtcuR.'ll
t./t ttrt
JANUARY 1976
27
(and maybe fuel rate) measured in deep water. Obtaining
power measurements during trials, however, is invaluable for
propulsion system andysis, with thrust measurements being
the most sought after but least often obtained data.
A common problem experienced during trials is failure of
the engine to reach rated rpm. The obvious solution is to re-
duce the propeller pitch until the proper engine speed is at-
tained and then accept the resultant speed. This is an example
of selecting a propeller to absorb power rather than attempt-
ing to obtain best craft performance.
It is very possible for the propeller to be too big in terms of
pitch or diameter or both. Trying to isolate the cause, how-
ever, could lead to a best resolution as to excessive hull resis-
tance or reduced propulsion efficiency. Using the resistance
prediction methods or experimental data sources available,
the hull contribution can be established reasonably well. Trial
speed, rpm, and propeller geometry can lead to a representa-
tive full-scale shp estimate as shown in the following table:
TRInr. onte
D-P/D- EAR- TY- LCG-
V N JA o Fv L ' we Ja Ko shp
Trial
data
Computed
from trial
data
As-
sumed,
Fi g. 6
Compute
from
J ^
arrd
(L-tre)
Prop.
charact-
eristics
Com-
pui ed
reference material that exists for the small-craft designer in
the area of performance prediction. Others, however, have and
will perceive different uses of these data. By exposing this ef-
fort for inspection, the true value will be established by the di-
alogue that follows.
2A
L Hadler. J. B.,
"The
Prediction of Power Performance on Plan-
ing Craft." Tr:ns. Sr\:{ME, Vol. ?4. 1956.
h.,
-1Z Hadler, J. B. ard Hubble, E. N.,
"Prediction
of Power Pertor- .. 1
l"
mance of the Series 62 Planing Hull Forms," Trans. SNAME, Vol.79,
)[,
1971.
3 Du Cane. P.. High-Speed SmaII Craft, John de Graff, Inc.,
Tuckahoe, New York. )
4 Savirskl', D.. Roper, J., and Benen, L.,
"Hydrodynamic
Devel-
opment of a High-Speed Planing Huil for Rough Water," 9th Sympo-
sium on Naval H1'drodyrramics, Aug. 1972.
5 Savitsky. D.,
"Hydrodynamic
Design of Planing Hulls," Ml-
RI i -E TEcH\ ot ocY. \ ' ol . l . No. l . Oct . 1964. .
'
Acknowledgments
This work has been carried out as part of the Combatant
Craft Improvement Program sponsored by the Naval Sea Sys-
tems Command. The authors would like to express their
thanks to the various members of the Navy communrty for
supporting this project.
Without the contributions of other authors, this materid
could not have been compiled. We are indebted to each of the
authors whose material was referenced and who influenced
this approach. Miss E. N. Hubble and Mr. G. O. Takahashi
deserve special attention for their respective efforts of compu-
tation and drafting to produce the propeller curves in Appen-
dix 4. Also, we wish to thank Mrs. Alice Waller for preparing
the manuscript.
The preparation of this paper took a considerable amount
of time from the normal family routine. We found the pa-
tience, understanding, and support of our families to be an es-
sential element to bring this effort to fruition.
References
Discus6er
Eugene R. Miller
- uJL
f u
/ ' I
An alternative method to determine shp for diesel engines is
6 Clement. E- P. anC Blount, D..I1,
':Resi{-alr99]99t-s
St
"
9yf-
to measure fuel rate and obtain shp from engine charicteris-
tematic Series of Planine Hull Forms." Trans- SNAME, vol' 71' 1963.
tics. By computing the estimared opc rhe d;6;;;;il;
c,"it *11[bs'j.3"";"li;i*ijlijXlT"',l,F.i;1,;flH,";ti,Tiiff.l,lt*:gf
comparisons with that value used for prelimina4r speed-power Gravity, andLoading," NSRDC Report 4307, April 1974.
estimates. A low value of OPC implies low propeller efficiency
^ I !_ol-l!ng, H.
D. and
liuQb-le,
E
-N.,.
"Model
Resistance D:a"."!
-t
or high appendage resistance. Detailed caclullo"s of ap--
i"d[iSl!rtlril';f.iyi';#e
to Hvdroroils and Planing crart"'
|'
pendage resistance can be made from the information summa-
{6
ff"ai"r, J. B.,'Hu6bte, S. N., and Hoiling, H. D.,
..Resistance
rized in Appendix 2 to verify or eliminate the appendages
Qtr!gg19{iqa_9f
aSyste'laiic.Serie_sof Planinf-HultForms-Series
d"
from consideration as the source of the propeiler overloading.
65," SNAIVIE, Ch-e-sapeake. Section lvlqv 1-97'!:-
-
Assuming re_= rr, the foregoing carcuiations.:l
!,"f:",:e
DJ|,ibftl3t*X#ts1,"f1*3;11;Ji;;j:il'-:?##fi.:,1;A':,:l:
on to rg and oe.7p. If these computed values'ex-ceed the Ka
ia'"iri.rn"ti" Surfaces," NACA fe;i;i;;f Nil"j4??,'5"pi. rgse.
breakdown curve in Fig. L2, the full-scale propeller is operat- 11 Hoerner, S. F., F/uid Dynamic Drcg, published by the author,
ing at lower efficiency than technicdly attainable. Should low
i\tidland Park, N. J- 1965.
pr:opetler erficiencv 6e the cause or engine ":"I1"?lt.lt
sr::d $i"dti"'JJ,3"%i,l*"tfl?n;,f,,"I;"ffsJrt'fffEfi#fi::l.
best solution is changing to a refined propeller design rather
iri;-Apr1 rgzs.
than
just
reducing pitch to absorb rated engine power at prop-
-i' 3 -i\Iarhis,
P. B. and Gregory, D. L.,
"PropellerSlipstream
Perfor-
er rpm.
mance
^of_Four
H-iq!-$pee{
-ludders
Under Cavitating Conditions,"
p-ower
measuremenrs during trials of r"*-lv
9:.ifl_d nlll
n:?o?fj#::3l1,,lt3I,13l'.tic
study or rhe Rough-ivater
per-
would be of
_great
interest to builders and operators. This
ior-""""-ofp-i"rii,g eoiL,;;--n""iJ"."'L";r;;i;t-ffip"rr R-r225,
would provide a rational basis for establishing technical Nov. 1g69.
achievement and refinement of stock designs- With the.e
15 Fli*d9mq,G*"ASystematic.gtudyof^theEgu.gh-lVaterPerfor-
data, performance analysis
(propulsion
nroblJm,s-gf1ns)
is re-
iliLTfi:iiiltl-"r"r:t lt'JftbtT.waves-Part
II)"'Davidson Labo-
duced to a technical exercise rather than a speculative art'
""x-almli?-it'';:rFJ'91*;r"ms::lt?;l
r,"i*:n?'ll'i;
26
Concluding remarks -J6t"'g'-'"
,3"-l;
1fil*.tance
and
propulsion
characteristics of a
we are in a period where dara obtained in the past or data
ff'lfh"r:*m"h"?63J."*""
Form of rMB series 63)'" DTMB R"-
X
spun off from other technology programs are being applied to 18 Gawn, R. W. L. and Burrill, L. C.,
"Effect
of Cavitation On the
craft design processes.
Performance of a Series of l6-Inch Model Propellers," Trans. /NA,
This effort was directed toward organizing and applyino
Vol' 99' 1957'
F-
l
MARINE TECHNOLOGY
- - - " + o - * '
i , ' : j i ! j ' _* ; .
' + i ' d , "
: . -
.ro
.40
.to
Appendix I
Savitsky equations
Equations for resistance and ehp computations ly
lflE$
method when all forces pass through CG
[5]
(Given: W,LCG,
Bpx,
9,
p, u, ACa, ul:
Computed from given data:
l 1
tf
li
n
t-
.ot
.o7
, ob
.2o
. t ,
. t o
, a2
.ot
.o7
. ob
, o,
.t, .zo .ra ' 4a ' to ' @ ' 7a ' & ' lb Lo
LocPl ce|tTeTtoN NUMBEE
aT o' ?o,pPDtUS
' Ft?(o' 7)
Flg. 12 Creneral trond ot Gawnurrill propelbr serio3 cavitation phcnornena
Solve for r:
cto
=
,Lt
[o.orr.a
*
o'qgj]'"J
Compute for u-.'
(24'
u - = u [ 1
'
tr cosr
^
u^|tBpy
Ke
= -
v
0.012rlf, rt'r
-
0.00658(0.O12{f,, t
U2
(25)
(26)
(n)
(28)
(29)
c"='J&E
Computed from given data:
""0=#"u"
Solved for )t:
^ - - L c G= 0 . ? b - [
t
I
tr=B;;=u.ro
Lu.,' i u' ,r*,,;J
Solve for Clo:
Cb
=
Cro
-
0'0065 F
Cuo
6
JANUARY 1978
0'242
'
TU=
l ogr o( Ke' uP1
(Schoenherr friction formulation)
. Compute for RsH:
I
BBx=wt ant -f f i
Compute for ehPsH:
,
RnxV
DPBH
=
g2*g
(20)
Compute forBe.'
Solve for Cr.'
(21)
(221
o"
4i
I
9
#,/'
(231
29
Appendix 2
Appendage resistance
Inclined cylinder, that is, shaft and strut barrel:
Dss
=
f,urrtt.tsin3c
* rCr)
(30)
Skeg:
Strut palms:
where
(31)
Dp
=
o.75 cn, {Fiii6 yn,
(t)
,^,
(82)
Appendix 3
Added resistance in waves
This work was extracted in part directly from reference
[151
with permission of the Davidson Laboratory. The following is
reproduced here so that the user of this prediction method
might have one complete reference source csntaining material
to account for all items to be considered. It is essntial that
reference
[tsJ
be consulted for complete understanding and
application of this method of accounting for added resistance
in waves.
Different notation was used for equivalent data descriptions
between this report and reference
[15].
These differences are:
Reference
[14]
(Fridsma)
b
L
R.t
w
A
Sruer,ucnerr
power prediction
(Blount-Fox)
Bpx
Lp
RA
w
Design Charts
The ultimate goal for this study is to enable designers and
those interesred in pianing craft to use the information gath-
ered herein in a practical and meaningful way. Working
charts, with appropriate correction factors, were constructed
so that the results could be immediately applicable to the pre-
diction of full-scale performance of planing hulls. Some details
of the effecls of individual parameters can be gleirned from
the charts and equations; but this is discussed in the next sec-
tion in a lrore generrli"sd way. In this section the reader will
be shown how to use these charts, and what corrections are
applicable, as well as a number of worked sro-ples-
To enter the charts and determine a prediction for a given
boat, seven quantities must be known; namely, displacement,
overall length, average beam, average deadrise, speed, smooth-
water running trim, and the significant wave height of the ir-
regular sea- Since realistic boats do not normally have a con-
stant beFm or deadrise, it is suggested that these quantities be
averaged over the aft 80 percent of the boat. It is understood
that the designer has recourse to smooth-water prediction
methods
[5]
which will enable an estimate to be made for the
resistance, trim, and rise of the center of gravity as a function
of forward speed.
The nondimensional parameters are calculated next, such
as Cs, L/b, V/\/L, andH6/b.
In using the charts, the designer should be careful not to
make gross extrapolations. The charts are accurate within the
ranges of test data. A reasonable amount of extrapolation has
been built into the charts beyond the limits of the test data,
and the results continue to be reliable. It is when parameters
go far beyond the test ranges that one must be careful. The
following guide should be helpful in establishing the limits of
the use of the charts.
Para-
ineter Ca Ll b c"lLlb
p H, l ' l b VIJl
lange 0. 3-0. 9 3-6 0.06-
0. 18
3-7
deg
10-30
deg
to 0.8 t o 6
Added resistance in waves (Figs. 13 and 14). The chart in
Fig. 13 is entered with a given trim and deadrise.
(Rew/
o*
=' r(2sdu^2cr
Copo 0' 65
and ,
6
ar
0.016Xp
Nonvented rudders and struts:
p
f r z r r l l
Dn/ s=i suz2cr l 1+2; * * ( t
j
( 33)
Interference drag:
^D
=;u^' t' lo.zs(l) -o-ooos
lGYl
Nonfl ush seawater strainers:
Oo
=
PiSeu-2Cpu
Coo
=
0'65
Experimental rudder drag coefficients in a propeller slipstream
Geometric aspect ratio
:
1.5
K7lJ7z
= g.2g
Propeller 0.55 D ahead of rudder stock
(3.r)
(35)
where
Dn
=
Plsu2Co^
Rudder
section t/c
Con
o = 4 . 0 o = 2 . 0 | o = 1 . 5
^ - 1 n
NACA
0015
0. 15 0. 0015
I
0. 0015 i 0. 0015
i
0. 0008
Parabolic
( bl unt
base)
0. 11 0. 0417 0.0427 0. 0433 0. 0425
Flat plate 0. 04 0.0278 0. 0325 I 0. 0371 0. 0433
6-deg
wedge
0. 11 0.0495 o.o4e5 | o.o+gs 0. 0487
3o MARITIE TECHNOLOGY
i : : . I
. . ' t . : I
:1' :0.010
r ' - ,
I
;' :
' .o.ot
. r I
; -, , -i o. oz
. .
: o. ol
" i
o
20
:
. . i -
Flg. 13 Maximum added resistanca and speed for Cj
-
0.60 and L/b =
5
i
Good for borh
Sea 5t at cs
,
4
JANUARY 1976
Generalized added resistance plot for Cr
=
0.60 and L/b
=
5
3l
.
urbs)-* and
(V/{L)t-" are read off for the three sea states.
An interpolation fbr the correct sea state can be made imme-
diately, or the added resistanc:e can be obtained as a function
of wave height. For a given V/{L or a series of speeds the
tatio V/V'*is calculated, and,R,qw/RAw-., is obtained from
Fig. 14. The added resistance is found by multiplying the re-
sistance ratio of Fig. 14 by the Rew/wb3l^
'
obtained from
Fig. 13. The result, however, is true for a C6
=
0.6 and
L/b
=
5, and must be corrected by means of the following
formulas:
(R
tw / wbs) nn,i-
= (R.c,w
/ wb3)cnst"
x E
(Cr
L/b, V/,/L, Hrnlb)
VNL E
Equa
tion
2 ,
-l{#''
-t]/rt
.
.[sil(H,t,/b- o.o6)t (1)
4 1 + t0H,
t,/b(c
"/L/b
-
0.12) (2)
6 L +
2II,
t,/bl0.g(Ca
-
0.6)
-
0.7 (C^- 0.6)'J (3)
Added resistance corttctions
For the patticular values of Cr and L/b, calcrrlate E and
--T-@..yLl^.a
or VfV^o is associated with the speed at which
plotasafunction otV/t/L- Readoff E attheV/JLof,inter-
(R,rw)-", ociiis. est to correct the added resistance value.
Appendix 4
Gawn-Burrill propeller characteristics,
?o
and J2" versus Kr/Jf
tL&re
ft.uLu.A6\
,f-ar".r.4
(,*
trl)
0"";44
J 2 0 ' O
hr/4,
Flg. ls
(Appendix 4 charts, Figs. 15-3.1, continue through page 40. Appendix 5 charts, Figs. 32-40, begin on page 41.)
MABI}IE TEC:III|(}LOGY
l a i
(
o.7
o. a
4
o. t
.o
a
a
a
r,
a '
. a
a
. l
. 2
ot o. 2
xr/ 4'
Flg. 16
. a
. a
. a
a
JANUARY 1976
, o
. e
. a
. a
2
34
MARINE TECHM'LOGY
a
7L
o.t
a a
a t
o-5
ot
L O
t.2
l . a
t . 3
2 0
'
.r,
,.o
a
JANUARY 1976
a ,
o-a
n
a 3
t.a
T
t 2
c
a
a
l(,/,r,'
F g.2g
IIARINE TEC}IMILOGY
36
a
a
,T,
o. a
t>1
o 5
JAfiUARY 1978
a
a
I
tr"
. a
a
]
2
d,
d a
a.t
o l
o. t
f,IARINE TECI{NOLOGY
o a
n
o.t
o.1
o 1
o
o.
a
a
q
a
39
JANUARY
1978
o.a
n-
a
o
o
a f
o.t
.a
.a
a
a
I'ARINE TECHT{OLOGY
Appendix 5
Conditions fsy minirnurn bare-hull ehp in snooth water computd from
sinplified Savitgky method as modified by equation
(l)
Cdcr'letionr for zero coneletion sllowanec, 59oF
Seaweter
I
$
I
I
I
I
(Appendi' 5 charts, Figs. 3240, continue through page 45.)
JANUARY
1978
rr
n
3
I
I
$'
t
I
I
I
t
rt
t
I
I
!
3
I
r
,a
ta
t 4 4 b a b a a a a a p
&,
n 4 r a r t & a t d 4 t a
4..
Fl g.3tl
u F ab tr D at 4 b L
l x
f
:l
12
MARINE TECH}TOLOGY
t
('
b
r
!
\
I
t
Flg.35
tl
t
I
t
I
-:
!
l'.**#1-L
i*./;,#f
-5"'l-
i - h. .
' . ! l /
l4r./,i
**l /
',/;
JANUARY 1976
ir
:
I
I
I
I
I
t'
t
t
I
I
t
!.
t
ll
3
I
t
I
t
\
t
-
.;
I
, . t a a b n b a v b t
b
l b a a D a a a a a a ,
Ea
, a t a t b r a & z a a a z a ? a
Dr'
F19. 38
a P b . a b 4 a a b a
,a
7ffi7x-
ilARNE TECIINOLOGY
'
- t
: a ?
' ?
,4/t
//l
J- f f i ?-
T f f i u s t t u
tn
P1939
!
t
I
I
,i
t
{
a
4
a
4
A
F
{
a:
{;
9:
g
*
*
'(,
*
s
*
si
F,
&='
ft
g
6
I.r
^l
I
r
:
I
T f f i - n
z z t t b t ' f f i - n
- a 4 L 4
Fl g. 40
777-12-p
u u
"
tfs
" r , { . r ! F + i F t . ' r ! r @' ! ! r . . t l ! | q