Korvin-Kroukovsky and Jacob
Korvin-Kroukovsky and Jacob
Korvin-Kroukovsky and Jacob
AD NUMBER
AD134053
FROM
Distribution: Further dissemination only as directed by Director, Office of Naval Research, Washington, DC, or higher DoD authority.
AUTHORITY
ONR ltr, 9 Nov 1977
CLEARED
KEWL~C RILW
~20e
~ D~D R200
AND
SUSE
AN4~Dt(OS
J .
f'>
,A
IL~e
NN~
OOCU
EN'
EN `E R SERV1'.1 C
rr
fwcrnCEAR M
~r1jj,IFDR11V EE A OVMX:MO
FOR AN ULMEME NM
ZSPCMAIMH
OHE ATA
s-ZCVXR: AND THE FACT THAT THEL XYr, NOR ANY OBLKGATI -4 *W %)NS NO RESPOMI IN ANFY WAY SUPPLIED THE TDeMW1nOR G-OVERtNMENT MAY HAVE M Y: ERGRE SNTT ~ , R0.K SAM) DRAWINGS, SEII~-M OTHER ANT OR HOLDERt THE :RWE A LKAN IMPLXCATION ORC ANArR, TO PRMUMMON OR *jf~Y-,Mj COKNVEYVI'' OR PERSON OR CORPOATION, ETDTERO MTrNINATWYB USE OR SELL ANY AV,,INXTM
lf this block is checked, the copy furnished to DTIC contained pages with color printing, that when reproduced in Black and White, may change detail of the original copy.
FE
-lo
~:tI
own,-
/i
i;i
177
PT
&
I.7
!
PITCMIJG AND HBAVING MOTIONS OF A SHIP IN REGULAR WAVES 1
by B. V. Korin-Kroukvsky
and
Member
mEZ
T 659
22 May 1957
Issue of this advance copy is on the express understanding that no publication, either of the whelvb or i abstract, will be made until after the paper is mads available for discussion i canmection with the Awa1 Meeting of the Society of Naval Architects and Marine Engixers in
SNvemier
1957
Is" a seq,--l to the 1-0,55 paper ': p~resents an additional. st~e= in tbe grd-a
h~su~
:o ,ok'sy e-
ship motions in reg-aar waves which v~as started %rKlof ihach 'as g'a~id noet
Dends, Kor-1-n-Kroukvsk.y =r
o)re'rious work are corrected and the thbeory- in i-ts :-resert forn is -e-i fie-d byr c~omarilson of the theoretica~l> cco--ted motioas o eigh.t
ie
different s-d~p models w-ith the notions &o;served in towinmg-tank test~s over a range of model speeds and at several wave I-engths. ALn
~~c
cor-.elation -is de~nstrated in the case -f a .Cestro-jer =odel, the sh,.e of unichi con~forms most closely to the assu~tions of -.be theory. azreement is foundi in the cases of several. t7PicalI co~zerciall '-ood
-"-iz fo.and
The agreemmt, is generalliy satifator.7 both in ::=-itnudes ol "'iz pitxhing motion and in the Dnhaze relationships "betveen the =tior-s zand
RTh material in, th'is p aper w4as prepa-red at 'he Experirnntal Twing -ar~k, Ste-ens :nstitute of Tecl~logy as a part of tha activit-ir -zner Office :)f Nava! Research Contract No. 'lonr 2633-10 (E7-*T ?roject _I 1-4, ). eduction in whole or i-n part i- z--rriixtted for any -r:mcese of the S.. . o ve r.-rer 2 erc-?:-fessor .---of -'--:d Dynamics, Ex;aeriment-al :ouing T7ank, Ste ve.ns
:-ititAe
f Technology,
iiolboken, N.
j.
3-.athenat~ician, Sg,-aerinental T7owing T-ank, Stevens :nstitute ,f Tec ?- ' o '--e rve dat tne Ar-ua. 'Ieetin:-, ',ew -,f the Society of Navral Architects am-d Mr rzne
o-' og-y.
NovemberU.
R-659
-2-
the waves.
The theory is
racterized by large slopes of sides at the LWL throughout its length and by a pronounced bov overhang. A discussion is given of the usefalness of the theoretically
computed responses to regular waves in predicting ship motions it, irregular sea and a 3atisfactory agreement is shown for the destroyer
model between the statistical averages of heaving Lnd pitcling ampli-' tudes of motion obtained in this way and those obtained experimentally,
INTRODUCTION
In a papor written for the November 1955 meeting of the Society of Naval Architects and Marine Engirners, the first author presented
a theory of the heaving and pitching motions of a ship in regular head or following seas. That paper added one more chapter to the historic and which
development which was started by Kriloff (1896, 1898)4, received a strong impetus in
and S-. Denis (1950), St. Denis (1951) and Korvin-Kroukovsky and
Lewis (1955).
4he Bibliography at the end of the paper is arranged alphabetically by authors' nasm., with the date of publication given in parnthesis.,
R-659 -3-
calculation of the exciting forces exerted on a ship by waves the results of which were in measurements. satisfactory agreement with experimental
The calculation took into account the interaction betneglected in the heretofore universally
The motions of two ship models computed on the basis of the thory were compared with the motions measured in good agreement in a towing tank. A reasonably
In
not only
because of its
the diotribution of hydrodynamic loads along the hull could be obtained. This distribution combined with the distribution of inertial loads re-
sulting from the ship motions can be directly applied to the analytical calculation of bundLg moments Since publication of that paper, ,,-ig on a ship structure in waves.
evaluating ship bending moments on this basis perimental results. It was found in
UJnder the auspices of Panel S-3 of the Hull Structure Coluittee Panel H-7 of the Hydrodynamics Committee of the Society of Naval Irchitects and Marine Engineers.
and
R-659
-4 -
calculations require a much greater precision than calculations of motions, and that the theoretical method which gave reasonabl good
amplitudes of motion was not yet adequate for the calculation of bending moments. In the course of re-examination of the original method occasioned by this difficulty, several errors and omissions were discovered and rectified. Most important has been the realization that the velocity-
dependent (or damping) terms of the equations of motion consist not only of energy dissipative parts but of dynamic non-dissipative parts as well. The existence of the latter had been demonstrated by Haskind but not sufficient attention had been paid
perhaps because of the difficulty in following the and because of the relatively recent appear-
Introduction of the necessary corrections has improved the rqorr1lation between the calculated and experimental anplitudes of motion and has resulted in ships. Furthermore, a generally good correlation of the phase relatiQnconfidence in the method has increased with more
extended application from the two models studied in 1955 to a total of eight models of widely diffprent types at present. heaving motions of these models in The pitching and
R-659 5
measured 6 and analytically calculated, and the results are described in this paper. In his paper of 1955 the first author emphasized that his objective was to place in the hands of naval architects a working tool to use in the process of developing better ship forms. It is the aim of the present
paper to present an improved tool and a broader basis for appraisal of its success in application.
ESEARCH
gular head seas, it may be advisable to demonstrate how this particx:!ar subject fits into the general pattern of research into the seakeeping qualities of ships. The complete picture would require an investigation
-
-z.l.,
pitch and
vestigation has not yet been made, but as a firzt step the simpler problem of ship motirns in head or following seas has been tackled by several writers, beginning with xriloff (1696). A further simplifica-
tion has been to assume regular seas, so that side sway, rolling7 and 6 The authors are indebted to Prof. Edward V. Lewis, Mr. Edward Numata and Mr. John Dalzell for all experimental data. Grim (1952) shows that at certain speeds in a following sea rollirn instability develops despite all symmetry.
R-659
-6-
yawing notions are eliminated and only surging, heaving ard pitching motions remain. Of these, surging appears to have little effect on is sufficient to consider t7. notions
nique has surging appeared to be important (Vedeler, 1955, and Reiss, 1956).
Formerly the response of a ship to reg-lar waves was thought of as an approximation to the actual behavior ;t sea. Since the publica-
tion of papers by St. Denis and Pierson (1953) and by Fuchs and McCamy
(1953), and the subsequent work of Korvin-troukovsky and Lewis (1955),
Lewis (1955), Lewis and Numata (1957), Cartwright and Rydill (1957) and
Cartwright (1957), the point of view has changed. ObtaLing the res-
It should be emphasized that the ship properties Msch as shape and mass distribution are considered only in the hydro-mechaniical phase; the
statistical phase is a strictly mathematical treatment of t= resuits obtained in the first. The results of the ffirst phase are required
only to be in the form of a plot of amplitudes or.' pitch and heave per
R-659
7arious methocs of
man-dimensional plotting proposed and used from time to tim ave Ittle usefulness in this connection. T ship respoonse to regular waves can be meas-=-e in a toaing
tak....
~g "
reference to theory.
wot bring out readily the significance of the various ph7ysical properties of a ship. It is necessary -o develoz a reasonaably cc~rehers-ive
t.heory in order to gain an uderstanding of thp effects of t-hese properties on the motions of a ship. Such a theory would also proride a There are. =ay pitfanls in
towing tank testing of ship models in warves which can be avoided -.hn know-ing what to expect on theoretical grotznds. The theoretical approach to the p.)oole= of ship responses to regular waves has taken two general directiozs. One is to u=certake e. to
m-d the lhyodynamics of a fltidi with a free su-f ace. most complete study of this kimr is that of Haszind
translation (under the auspices of the H-z Pa;ei of the Society of Naval Archdtects and Marine Engineers). These st-.'dies have been
R-659 -8-
but their
mathematic&l difficulty ad the need for extensive calculations to bring out the final results prevent their being used directly by naval architects. The other, simpler, approach has been taken by Kriloff
(1896, 1898), Weinblum and St. Denis (1950) and St. Denis (1951).
Here
the differential equations of motion are written on the basis of the dynamics of rigid bodies and various coefficients are independently and individually evaluated by hydrodynamic considerations. They are
usually derived by analogy with a fully submerged body and the effects of a free water surface are allowed for by separate considerations. Specialized investigations of virtual masses and damping such as those by Holstein (1936, 1937 a,b), and Grim (1953) are available, Havelock (1942), Ursell (1949, 1953, 1954.), mans
porting literatmue the simpler theoretical approach has become more and more realistic. The present paper follows this simpler approach. objective is
Its immediate
lowing oeas, with the expectation that this material can then be statistically treated to permit prediction of ship behavior in a realistic
R-659 -9-
irregular sea.
TE EQUATIONS OF MOTION
Neglecting the surging, the equations representing the equilibrium of forces and moments acting on a ship moving in regular head or following seas can be written in the form m z
+
c z
Hi
m where m is the mass o* the ship itself and z is the heaving displacement and c - C J 9 is its noment of the angular displace-
inertia;
ment in pitch;
are the unit restoring force and moment in the buoyant forces arising from the Hf is
respectively dile to c
deviation of a ship from its attitude of normal flotation; the force and
to water pressures generated in the flow by the action of the wave as well as by the ship's own oscillations. Thase forces and moments are
computed in the Appendix, and are3 found to be expressible by polynomials the terms of which are divided into those corresponding to the ship's oscillation in smooth water and those corresponding to direct wave action on a restrained 3hip. When the terns of the first kind are
R-659
iu
+g@0
Pe
(2)
+ E, +Gz Mei
The complete expressions for the coefficieut6 on the L.H. side are
given in complex form in order to facilitate the algebraic work of the solution and it :As understood that only the real part of the exponential on the R.H. side ef Equations (2) F F eic
M e i't
is to be taken.
and the real part of F ei)t and the real par+t of^
Ut
o_ ('.t +
Here
F
0
and
M
0
and (27)
in the Appendix; Or and I are the phase angles at which the maxima of the force and moment occur (positive angles are leads, negative angles are
lags). ( or Z is zero if the maximum occurs at the instant when the wave
crest is
amidships.
The harmonic oscillation of force and moment is in Equations (2), where LA), written for
The terms
the interaction between heaving and pitching motions. usually referred to as "cross-coupling terms".
It is assumed that the ship is moving in a uniform sea, and that a steady state of heaving and pitching is established. In such a case,
the transient responses have been damped out, and only a particular solution of Equations (2) is required. Since the exciting functions
are assumed to be harmonic, the solution is likewise harmonic, i. e. of the form z = Z e iut and e = 4 ei"t
(4)
Sand
and
I - 0 eiE
0
where
0
o
that of
pitching. and
to
(A
+t.
is readily obtained as
Z
FR-'(
S -P -QR S(5
(5)
(6)
QR-PS Q where the capital letters represent the groupings of the coefficients of Equations (2) as followss
R-6~9
-a
LZ2
+ i bu)+
S =-A LA2 + i B U+
zero, i.e. d - e - g - D - E - G - O, the two equations would re resent two simple harmonic oscillators in forced motion, and Equations ( and a)
The heaving and pitching motions in regular head seas have been
(6)
The properties of these models are given in Table 1 These hulls are currently subjects
of research at the Experimental Towing Tank of Stevens Institute of Technology for various proj,.zt:, sponsored severally by the David Taylor M-cdel Basin, the Office of Naval Research, and the Society of Aaval Architects and Marine Engineers. This collection of models does not it
R-659
-
13-
Model 12,45 is the Series 60, 0.60 block coefficient fdrm and Model
1616 is an experimental modification of it bry & V. Lewis (1955) which has the sw dimensions and the same arterbody but an extreme V-type nhese two models were investigated by Korrin-Kroukorsky
in his 1955 paper but have been submitted to re-analysis by the corrected and improved procedure given in the present paper. values of all the coefficients for wave length L, X/L ) The newly-computed equal to ship length
1, are listed in Tables 2 and 3, which replace Tables 4 and 6 B, b, e2 and E2 versus
of the earlier paper, and the new curves of frequency and of C and g
and exnerinpntally V
lengths,
1mRS1rIri
rtYPRwt-
of
Model ILU is the T-2 Tanker (Fig. 2) which has been the subject of many experiment-i eonducted by E. V. Lewis (19-54) in connection with
the measurement of bending moments (under the sponsorship of the S-3 Panel of the Society of Nival Architects and Martine Engineers). calculated and experimentally measured motions for in Fig. 9. - I 1/ ar The shown
R-659
-lii-
towing tank tests for four different wave lengths are shown in Fig. 10.
described by Captain W. Mdckel (1953) as possessing particularly good seagoing qualities. ratio 204. It is a short stubby hull of displacement-length
Model 1699C has the same section forms as the trawler but
spaced wider apart and dimensioned to give the low displacement-length ratio of 60 like the destroyer. Fig. Ii shows the calculated and ex-
perinentally measured motions of 1699A and 169% in waves of two lengths, X/L = I and 1.25. Attention should be called to -'=diffezct s-Le%
The lengthened counterpart of the yacht, section forms spaced wider apart and a (is-
placement-length ratio of 60 like the destroyer and the lengthened t-r- er. Fig. 12 compares the calctlated and experimentally measured
motions of 169-9B and 1699D in vwavs of -,w lengtts, IL/L X/L .25.
ad siwes
of thiis figure.)
in
op~erating efficiently in ann open ocean., it vas thotzhft -,b-zt t-bir fcr=~ say hare some partic-1aazy cdesizable feat--res vorth E. . Lc-is 'k:95c) andNm~ata and I~viz "-OC7) ~r ratio so e~aig
-
atg~
the lengtheme
zoaxparxison of Maodel
reducim. the
!-659
-
16-
i.
can be seen by comparison of the original and the lengthened versions of the trawler and the yacht on Figs. 11 and 12.
Excellent agreement between analytically calculated and experimentally measured results is found on Fig. 10 which depicts the responses of XA/L
-
1, 1.25,
1.5
It
should be
noted that of the eight this model comes closest to being "vall-sided" at the LWL, and in this sense conforms closely to tAV- ---
tlons made
in the theoretical derivations of the Appendix as well as in th ammorting literature on damping. Next in degree of correlation co Fig. 8, ment is and the T-2 tanker (Model 1444), Lae Series 60 hull (Model Fig. 9. Here,
1445),
excellent between the calculated and experimentally measured A.iase relationships. However, there is a signifi-
cant overestimation of the heaving motions at and near sy-chronism, for both models, and, while the calculated phase lags in heave are close to
1445,
some discre-
R-659
-
17-
only at the stern where the relative ship-water motions are generally smaller and the effect of section form is less important. The calculated and observed amplituades of motion for Model 1616 (lower half of Fig. 8) there is agree very well at X/L = 1, but at X/L = 1.5 The
phase relationships show good correlation in pitch but not in heave. This model, with its extreme V sections at the bow, has, of conse, a deviating
pronounced slope of the tangents to the sections at the IML, in this respect from the assumptions of the theory.
In the case of the lengthened trawler Model 1699C (right hand of Fig. 11), there is on the average a very good agreement between the
calculated and test results, but the calculations underestimate the pitching amplitudes ar.d eo not show the exaggerated narrc. heaving at synchronism. peak in
The calculated phase relationships agree very As can be seen from Fig. 4, the slopes of
the tangents to the bow s'rctions are more moderate for the trawler than for the U-bow Model 1616 but greater than for Y:.Iel 4L5.
The original trawler Model 1699A combines these section characteristics with the high displacement-length ratio of 2C'14. The experimental
R-659
1.25.
The agreement in heave amplitudes and phases at the latter waves length is excellent.
The only models for which the computation mmt~hod fails completely are the yachts, both the original Model 1699B and the lengthened Model
1"99D. These hulls have not only large slopes of the tangents to the slope at
of ship motions appear sufficient to mutilate computed results. Tt w&v expected that estimates of the motions of t,-c original yacht would be poor not only because of the slopes of the sides at the LWL and the large overhang model was included in but also because of its stubbizess. This
the analysis in
an attempt to establish the limits However, the failure in the case of illu-
have much greater veight in the calculations than the displacementlength ratio. Tb. relatively small importance of the latter could have calculations of three-dimensional
a-659
and
from the comparison by Korvin-Kroukovsky (1955a) of his inertial force calculations for a spheroid of length-diameter ratio 5 with Kavelock's for an infinitely long spheroid. To sumarize, there is excellent agreement between the calculated and experimentally measured ship motions in the case of a destroyer,
and a generally satisfactory agreement in the case of normal ship forms. Only in the case of a yacht is the cm~putation method completely in-
the LWL indicate a strong non-linearity in the restoring forces and th corresponding cross-coupling terms, and G it is of tle equations of motion. i. e. in the coefficients is c, 0, g
However it
damping coefficients which is responsible for the discrepancies between the calcul2'ted and the observed motions. Attention should be called to
the fact that although the forms for which Grim has derived the damping forces include what appei.r to be extreme V sections, these sections hawe a sharp turn of the bilge and are '.angent to a vertical at the LWL, whereas the bow sections of a Maier form or of the yacht discussed
1-659
wide V sections used in speed boats and in seaplane hulls has been derived for very high Froude numbers, glected in at which the Lorce of gravity can be neand is not valid for the
As a result of all investigations made so far, the effectz of certain properties of ship forms have become established. The natural
pitching or heaving period of a ship oscillating in smooth water becomes somewhat modified in coupled motion, but anverthe'ess remains a most important criterion. The shorter tle natural pe-iod, the higher the
frequency of wavo encounter ct synchronism, and therefore the higher the ship E ced for synchry.idsm in waves of any given length. causing lzrge chip motionw The waves
E. V. Lewis (1955) has shmin that the limiting snip speed in rough weather therefore cieases with ship length. This effect i3 clearly shown
Note that the speed s'cales are different for the original and lengthened hulls.
R-659
"-owever, a naval architect will arrive at a certain ship length and a certain displacement-length ratio by many considerations other than seakeeping anility. Whatever the dimensions chosen, it is desi:-
able to see what improvement in seakeeping can be obtained by selecting a suitable ship form. here is the d_mping. 1937b), Havelock (1942) Evidently the most im;>ortant characteristic
The theoretical work of Holstein (1936, 1937a, and Grim (1953) shows that da.ing increases Yet avoidBoth
ance of bow emergence and slaxing dictates a large actual draft. conditions lead to the requirement of a low section coefficient,
usually associated with a V form, or at least with a large deadri-se of the ship bottom9. T-hs is indeed a characteristic of the small ships
These principles are well illistrated by the romparison on Fig. 8 of the Series 60 Model 1U5 with its modified V-bow version, Model 1616. The larger beams of the bow sections of Model 1616 increase the moment T he advantage of a low section z-iflcient in reducing the maplitude of heaving motion in waves is vividly demonstrated Ln the recent theoretical work of Grim (1957). Of two sections having draft equal to halfbeam, the one with section coefficient of 0.555 is shown to oscillate with a magnification factor close to unity through a wide range of frequencies, while. the one with section coefficient of 1.015 has a magni.ication factor of 3 at a sharp synchronous peak.
R-659
-
22-
the damping forces and thereby markedly decrease the amplitudes of pitching and heaving. The advantages of the trawler and yacht section forms over those of a conventional ship like the Series 60 lie largely in the increased damping due tc smaller section coefficients. There is rather little
difference in the behavior of the destroyer (Model 1723) and the lengthened trawler and yacht (Models 1699C and 1699Df), which were made to match the displacement-length ratio of the destroyer, because the destroyer also has t:i desirable low section coefficients. Indeed the very low
mesa
desirable as a more unixorm distribution of damping between bow and stern. This conclusion is in agreement with the often-expressed opinion
that "Pouble-endedO boats are preferable for good seakeeping qualities, and in fact trawlers, and f~ihing boats in general, show a distinct trend
This DOCumnent
-5 R-659
P-ePrI~ced Jroar23
in this directiorn
The beneficial effect on the behaTior of a hull resulting from a low section cteffirient is with its corollary a large damping coefficient
quite evident qualitatively from the familiar definition of "ali, i. is e. the solution of the simple uncoupled equations however, to judge tba effects of the c.oss-
impossible,
coupling coefficients by inspection of the coupled equations of motion or their solution. Experimentally also these effects are difficult to
trace because of the nuaber of different m.odels required and the tediousness of towing-tank testing. expended in validity. So far all analytical efforts have been demonstrating its
series of calculations necessary to bring out the ini'luence of the crosscoupling coefficients; but mov it appears that at least for ship forms sufficiently
wor~Iwhilp,
A Popt
ok @o~sis
of systematic changes of form leading to changes of cross-coupling coefficients could be easily or7maized and carried out economically on many of the currently available computing marlcinms.
R-659
-24-
tical treatment of these responses in combination with the spectrum of sea waves can give results which are significant fr in the realistic sea. the ship motion
by St. Denis and Pierson (1953) and it was applied to model tests by E. V. Lewis (1955). (1956, 1957). A simple exposition was presented by Korvin-Kroukovsky
per unit wave height, for a given wave length at a particular speed (i.e. fLr a given frequency of encounter),is multiplied by the ordinate of the sea spectrum (corrected to that ship speed) at the same frequency of encounter. These products plotted against frequencies of encounter
(resulting from a range of wave lengths at a particular speed) form the spectrum of the ship motions at that speed. Since the ship motions in
an irregular sea are also irregular, varying continuously in amplitude and period, they can only be described statistically, as for instance by mean amplitude, mean of the 1/3 highest amplitudes, 1A0 highest amplitudes. near-maximum. or mean of the
These means were derived by Longtiet-Higgins (1952) as E under the response spectrum
SR-659
-
25-
curve:
equal to .88
1/10
highest is
equal
ocean waves as well as to the ship mot.ions in them. As an illustration of the above, pitch and heave spectra were predicted for the destroyer model No. 1723 from the calculated responses to regular waves of eight lengths at two ship speeds. The squares
of the computed responses were divided by the square of the wave height and the quotients were multiplied by the ordinates of the 2rregular wave spectrum derived from analysis of the waves in the towing-tank tests, thus yielding motion spectra. These spectra were compared with the
spectra obtained experAmentally by Prof. E. V. Lewis and his associatcs at the Experimental Towing Tank of Stevens Institute of Technology (Lewis and Dalzell, 1957) who used two different methods: first, the
spectra were obtained directly by statistical analysis of iM-lzi model motions, and second, thm spectra were computed from experimentally
A comparison of the
records of ship motions in irregular waves, the dash lines the spectra computed from the experimental responses to regular waves, and the
R-659
-26-
crosses the spectra predicted by the methods of the present paper. It should be emphasized that the shape of the spectrum curve has Components of all
frequencies exist simultaneously and only the area of the spectrum, Z, is significant. Means of the 10% highest amplitudes, cor-puted from The amplitudes of
heaving motion obtained by the three different methods are in excellent agreement. The pitch amplitudes computed from responses to
regular waves are slightly smaller than those resulting directly from a statistical analysis of the irregular model motions. On the average
there seems to be no important difference between the amplitudes computed from the experimental and those computed from the theoreetc'-l responses to regular waves (second and third lines of Table 4); zere speed the theoretically computed values are near, tically derived ones, at
Uie statis-
at 2.53 ft./sec. the values based on experiments The deviations are well withSince there is always a
rather large statistical uncertainty in evaluating sea spectra, further improving the accuracy of the calculation of responses to regular waves would have little meaning. It is only important to look for ways to
R-659
-27-
The discussion so far has been limited to amplitudes of motion. These are readily obtainable by both experimental and thaoretical methods in use at present but they are not i'portant just by themselves. Other
phenomena may be more decisive in qualifying the seakindliness of ships. The accelerations, amplitudes, there is which for a given frequency are proportional to the
a direct connection between accelerations and sea-sickness and Shaw, 1954). On fishing trawlers the accelerations
(Geller, 1940t,
1953)t
tions in pitch and heavR are of minor importance except for the structure supporting the cargo in th. lower decks in No. 1 hold. The critical
'onditions limiting the operation of these ships appear to be shipping of water and slamming, the latter mostly in the light condition. These
conditions probably also limit the operations s, such naval ships as destroyers. The statistical treatment of the azcelerations is of the amplitudes of the motions. bow, for example,
Wimilar to that
can bc darivad analytically from the computed motions 0 2S, where a is the amplitude of vertical acce-
by the forula
leration, s is
R-659
-28-
combining the heaving and pitching motions in the correct phase, W is the frequency of encounter.
and
Acceleration spectra can be obtained the mean of the I/10 highest ampli-
and from these the mean amplitudes, tudes, etc., in irregular waves.
E. V. Lewis (1955)
the ship's bow on the regular wave profile, the bow submergence or emergence can be evaluated for each wave length and ship speed. If
these values are treated statistically by the method outlined above for amplitudes of motion, the frequency of deck submergence or forefoot emergence in an irregular sea can be estimated. The frequencies
thus obtained in one case were found to -ree in a towing tank irregular wave test.
analytically here for hulls of normal form are close to the model test data, the calculated motions can be ueed instead of model tests to predict bow submergence, hence shipping of water, and frequency of forefoot
emergence, which would qualitatively indicate likelihood of slamming. It is well known, however,
the frequency of slamming a more elaborate statistical treatment of either experimental or calculated responses to regular waves is needed. The
R-659
a high descending velocity of the bow and approximate tangncy of the keel to the water surface at the instant of impact. A beginning has
been made in the work of L. J. Tick (1954) where the probability of the joint occurrence of two of the conditions, forefoot emergence and high descending velocity, was investigated. It is believed that the phase relationships are of greatest importance in this ccrnsction. Sea captains and other observers are
bahavior.
T-A--rtotion
in tii
ck,,
settle quietly into the hollow of a wave and at 5 R.P.M. more it may strike with force the flank of an oncoming wave, causing either slamming or shipping of water or both in succession. Since the phase relationships in regular wava-s are give~en satisfactorily by the method of calculation presented here, it can be expected that this behavior will be predicted correctly. Thus the
stago has been reached at which the significant aspects of ship performance at sea can be predicted by analytical methods.
fR-659
-30-
CONCLUDING HMARKS
two lines of
'A
research were suggested to improve the theory of s'iip motions lar waves. These were
regu-
rimental evaluation of the various coefficients of the equations of motion. The need for such research appeared to be acute since there the analytically cuputed phase lags.
;jhas been shown, most of this error was due to omission of Application of the -.tad li-ax
theory described in the present paper has yielded generally sat!sfact-ry estimates of both amplitudes and phase relationships for ships of n.rmal
commercial form. Apparently the deviations of most of thesa ship forna
from the simplifying assumptions of the theory are not large enough to affect the results seriously.
rat'-r
characterized by sloping sides at the LWL throughout its length and by a pronounced bow overhang and a cutaway forefoot. theory along the lines proposed is Development of the
R-659
estimating the behavior of such hulls and would make the theory more reliable for ship types like the Maier form and *ha fishing trawlers. Since publication of the theory of ship notions in irregular seas by St. Denis and Pierson (1953), a large amount of practical experience
in its application has been accumulated by E. V. Lewis (1955, 1957), Lewis and Numata (1957), and Lewis and D. Lzell (1957) in connection and by CarTwright and Rydill (1957) and The vali-
Cartwright (1957) in connection with an actual ship at sea. dity and usefulner established.
The theory of notions in regular waves can therefore no It is only a part of the picture,
e
f a s 1vis
I shes th
this part, the ship responses to regular waves or 'response factorsw, are then treated by methods of mathematical statistics in conjunction with a measured or zassumd spectrum of a realistic irregular sea to give the realistic ship notions. These statistical methods assti that the motions vary linearly
with wave height and require that the results of the hydro-mechauical phase be supplied in 'Inear form as ratios of response to wave height. The recoendation that solutions of non-linear equations of motion be
R-659
solution of a non-linear problem but rather a solution of a substitute linear problem which would approximate the true solution. Such a pro-
cess of "equivalent linearization" has been used by Golovato (1956) to take into account the effects of non-linear damping in heave. However,
non-linearities are usually pronounced in both the damping force coefficients and the restoring force coefficients, and the coupling of
heaving and pitohing notions further aggravates the difficulties of this problem. It is to be noted, also, that in the statistical method the ship
responses to a number of wave lengths are averaged, so that nonsystematiO errore tend to compensate each other. Because of tUhs, and
the statistical uncertainties in the evaluation of a sea spectrum as well, there is less need for a high degree of accuracy in estimating the ship response. The theory of ship motions can thus in its present The application of the
theory is not limited to amplitudes of motions and accelerations; the frequency of shipping water can be readily predicted and prediction of the frequency of slamming is not too far in the future.
B-659
-
33-
CAiG'LEDQDWEN
A large part of the stimulus for the work reported in this paper vas furnished by the S-3 Panel of the Society of Neva3 Architects and Marine Engineers, and particularly by its chairman, Prof. Edward V. Lewis. For the experimental material the authors are indebted to
Prof. Edward V. Lewis, Mr. Edward haata, and Mr. John Dalzell, and for assistance on certain theoretical questions to Dr. Paul Faplan of the staff of the Experimental Towing Tank of Stevens Institute of Techn-olc.
SBIR IOGRAPHY
Cartrright, D.E. and Rydill, L.J. (1957),"ri* Rolling and Pitching ofi a Ship at Sea - A direct comparison between calculated and recorted notions of a ship in sea waves", Transactions of the Institution of Naval Architects, Vol. 99, No. 1, Jan. 1957, pp. iCO-135. Cartvright, D. E. (1957), "On the Vertical Motions of a Ship in Sea Waves,' Netherlands Ship Model Basin Symposium on the Behavior of Ships in a Seaway, Wageningen, The Netherlands, Sept. 1957. Fay, J.A. (1957), "The Motions and Internal Reacrtions of a Vbssel in Regular Wavesa, Submitted to S-3 panel of the Hull Structure Comaittee, Th Society of Naval Architects and Marine Engineers; publication pending. Fox, U. (1938), 1938. Fuchs, R.A. and MacCamy, R.C. (1953), wA Linear Theory of Ship Motion fRacing, Cruising and Design". Chas. Scribner Sons,
in Irregular Waves", University of California, Institute of Engineering Research, Technical Report Series 61, Issue 2, July 1953.
R-659
-34-
Geller, W. (19LtO),
Klinische
"A Study of the Forces and Moments on a Surface David Taylor Model Basin
Ship Performing Heaving Oscillations", Report No. 1074, Grim, 0. (1952), Seegang", Aug. 1956.
"Rollschwingungen,
bautechnischen Gesellschaft, 47, Band, 1953, pp. 277-299. Griu, 0. (1957),"Die durch eine Oberfl'dchenwelle erregte Tauchbewegung', Schiffstechnik, Forschungshefte fdr Schiffbau und Schief-maschincnbau, 20 Heft, Feb. 1957, 4 Brd: n. 9p 1O..
Haskindi H. D. (1946 (a) "The Hydrodynamic Theory of the Oscillation of a Ship in Waves" (in Russian), Prikladnaya Matematika i Mekhanica, Vol. 10. No. 1, 1946, pp. 33-66.
(b) "Oscillations of a ship on a Calm Sea" (in Russian). Bulle-
tin de l'Academie des Sciences de I'USSR classe des Sciences Techniques, 1906, No. 1, pp. 23-2i. ENGLISH mANSLATIONs "Two Papers on the Hydrodynamic Theory of
Heaving and Pitching of a Ship", Society of Naval Architects and Marine Engineers Technical and ResearcY Bulletin No. 1-12,
April 1953.
Haskind, H.D. and Riman, I.S. (1946), "A Method of Determining the Pitching and Heaving Characteristics of Ships", David Taylor Model Basin Translation No. 253, 1955. "The Damping of the Heaving and Pitching Motion Vol. 33, 1942, pp. 666-673.
of a Ship", Philosophical Magazine, Havelock, T.H. (1955), of Advancew, 1955, pp. 246-251.
"The Coupling of Heave and Pitch due to Speed Transactions of In~,titutimof Naval Architects, April
P--659
Havelock, T.H. (1956), "The Damping of Heave and Pitch: a comparison of two-dimensional and three-dimensional calculations", Transactions of the Institution of Naval Arzhitects, Vol. 98, No. h, October 1956, pp. h64-468. Holstein, H. (1936), *Untersuchungen an einem Tauchschwingunen Aus-uhrenden Quader', derft-Reederei-Hafen, Dec. 1, 1936, pp. 385-389.
Holstein, H. (1937a), "Die periodische Quelie in dassei mit freier Oberfl'ache", Zeitschrift frir angewandte Mathematik und Mechamik, Vol. 17, 1937, pp. 38-h7. "Uber die Verwendung des Energiesatzes zur Ingenieur-Archiv, 8 Band,
von Oberflachenwellenproblemen",
Korvin-Kroukovsky,. B.V. and Lewis, E.V. (1955), Vol. 2, No. 6, 1955, pp. 81-95. Korrin-Kroukovsky, B.V. Moving under waves", on Fluid Mechanics, (1955a),
Korvin-Kroukovsky, B.V. (1955b), "Investigation of Ship Motions in Regular Waves," Transactions Society of Naval Architects and Marine Engineers, Vol. 63, 1955, pp. 386-435. Korvin-Krorkovsky, B.V. (1956), "Irregular Seas - A New Towing Tank
Problem," Eleventh American Towing Tank Conference, 12, 13 and 14 Sept. 1956, David Taylor Model Basin, dashington 7, D. C. Kcrvin-Kroukovsky, B.V. (1957), "A Ship in Regular and Irregular Seas", Netherlands Ship Model Basin Syzqx ium on the Behavior of Ships in a Seaway, Wageningen, The Netherlands, Sept. 1957. Kriloff, A. (1896), "A New Theory of the Pitching Motion of Ships on daves and the Strmsses Produced by this Motion", Transactions of institution of Naval Architects, Vol. 37, 1896, pp. 326-359. Kriboff, A. (1398), "A General Theory of the Osci-lations of a Ship on Waves", and "On Stresses Experienced by a Ship in a Zeaway", Transactions of Institution of Naval Architects, Vol. hO, 1898, pp.135-196,
R-659 -36 -
"Ship Model Tests to Determine Bending Moments in "Waves", Transactions of the Society of Naval Architects and Marine Engineers, Vol. 62, 1954, pp. 426-490. "Ship Speeds in Irregular Seas", Transactions of I the Society of Naval Architects and Marine Engineers, Vol. 63, 1955.
"Developments at Stevens on Behavior in Irregular Seas", Netherlands Ship Model Basin Symposium on the Behavior of Ships in a Seaway, Wageningen, The Netherlands, Sept. 1957.
Lewis, E.V. and Numata, E. (1957), "Ship Model Tests in Regular and Irregular Seas", Stevens Institute of Technology, Exper.mental Towing Tank Report No.567, 1957. Lewis, E.V. and Dalzell, J. (1957), "Motion, Bending Moment and Shear
Measurements on a Destroyer Model in Waves", Stevens Institute of Technology, Experimental Towinrg Tank Report No. 656. Lewis, F. M. (1929), "The Inertia of the Water Surrounding a Vibrating
Ship",
Transactions of the Society of Naval Architects and Marine Ziiginera, Vol. 37, 1929. Longuet-Higgins, M.C. (1952), "On the Statistical Distribution of the
Height of Sea Waves",Journal of Marine Research, Vol. XI, 1952, No. 3, pP. 245-266. Michell, J.H. (1898), "The Wave Resistance of a Ship", Philosophical Magazine, London, Vol. 45, 1898, pp. 106-123. MW'ckel, W. (1953), "Seaworthiness and Safety of Trawlers in a Seaway", papers No. 29 A and B, International Fishing Boat Congress, F.ALO. of U.N., 1953, also FISENG BOATS OF Thl WO'4D, Edited by Jan-Olaf Traung, Arthur J. Heighlay, Publ. Ltd., London, pp. 326-336. Numata, E. and Lewis, E.V. (1957), "An Experimental Study of the Effect of Extreme Variations in Proportions and Form on Ship Model Behavior in Waves", Stevens Institute of Technology, Experimental Towing Tank Repo.t.. 643, March 1957.
R-659
-
37
Reiss, H. (1956), "Dynamics of Towing Systems", Eleventh American Towing Tank Conference, 12, 13, l1 Sept. 1956, David Taylor Model Basin, Washington, D. C. Shaw, C.C. (1954), "On the Dynamics of Motion Sickness in a Seaway", Th Scientific Monthly, February 1954, Vol. LXXVI=I, No. 2.
SU.., 0.J. (k95i6, "_7he Effect of Method of Towing on Ship Model Motions," Eleventh American Towing Tank Conference, 12, 13, 14 Sept. 1956, David Taylor Model Basin, dashington, D.C. St. Denis, M. (1951), "On Sustained Sea Speed", Transactions of the
pp. 758-763.
St. Denis, M. and Pierson, W.J., in Confused Seas", Jr., (1953) "On the Motion of Ships pp. 280-357. Transactions of the Society of Naval Architects
-nd Converging Stream of Fluid", Aeronautical Research Council (Gr.--------+-n----------------. --:_ Lyv
I
"Certain Probabilities Associated with Bow Submergence and Ship Slamming in Irregular Seas", New York University, College of Engineering, Scientific Paper No. 1, Engineering Statistics Group, December 1954.
I I
I
T
Ursell, F.
(1?99),
R-659
Vedeler, G. (1955),
and Motion in Waves", Schiffstechnik, Forschungshefte fMr Schiffbau umd Schiffsmaschinenbau, 12/13 Heft, November 1955, 3 Band, pp. 55-58. Vossers, G. (1956), Discussion of Havelock's 1956 .aper. Transations of the Institution of Naval Architects. Vol. 98, No. 4, Oct. 1956, pp. 467, 468.
Weinblum, G. and St. Denis, M. (1950), "On the Motions of Ships at Sea", Transactions of the Society of Naval Architects and Marine Engineers, Vol. 58, 1950; pp. 184-231. NCHENCLATURE The following nomenclatuare is used throughout the papers A, B, C,) D, E, G ) - Coefficients of miscellaneous tex- of the equations of motions, Equations (2). Af-ttial
a, b, c,)
, ,
b
a0 B b, s
Instantaneous distances of ship bow and stern from the nodal point of wave as defined in Fig. 14.
Wave celerity. Area under spectrum.
c E F
-
R-659
"-39-
F g Hf
=
eT).
0
Force due to water pressures generated by waves and the ship's oscillations.
- Moment about c.g. due to water pressures generated by waves and ship's oscillations.
a Wave amplitude.
"1' 2
'C 2
Added iass coefficient in two-dimensional vertical flow about a ship secticn. Correction coefficient for effect of free water surface.
j L
M
= Ship length.
Hydrodynamic monent.
- Pitching moment imposed on a ship by waves (- M ei). Mass of a ship or of a ship section. ( ) Vertical damping force per unit of body length per foot per second. P, Q, H,) S - Groupings of coefficients of the differential equations
of motions defined by Equations (7).
U-
R-659
p= R r S= s0 t u V v v x y
Y
-
Pressure.
in fluid.
Vertical velocity.
- Vertical component of wave orbital velocity. - Longitudinal co-ordinate with respect to wave nodal point. - Vertical co-ordinate or local half-breadth of UEL plane.
- Complex amplitude of heaving motion (= Z
e0 ).
- Vertical co-ordinate or heaving displacement. - Polar co-ordinate. - Angle between longitudinal tangent to body surface ,nd x-axis.
6 E
R-659
- 41 -
g 9
x Wave length.
-
9
0 -C7
or u)
APPE ND IX EVALUATION OF HYDRODYNAMC FORCES This Appendix is arranged to follow as closely as possible Appendix I of t'.I earlier paper (Korvin-Krouloovsky, cen be easily seen. ence, the detai
Formulation of the Problem Consider a ship moving with a constant forward velocity V (i.e. neglecting surging motion) with a train of regular waves of celerity c.
Assume the set of coordinate axes fixed in the undisturbed water surface,
R-659
-
42 -
with the origin instanianeously located at the wave nodal point preceding the wave rise, axes remain fixed in in relation to them. x is as shown in Fig. 14. space, With increase in time t the
any distance
y.
dx
apart at a distance
and water with orbital velocities of wave motion penetrate these control
surfaces.
Assume that the perturbation velocities due to the presence between control
planes, i.e. neglect the fore-and-aft components of the pe4-burzation velocities due to the body, as in the "slender body theory" of aero-dynamics. This form of analysis. also known -; th put
ri-,
mt
."
a certain degree of interaction between adjacent sections is neglected. The cross section of the ship at circular; x will now be taken as semi-
the correction necessary to represent other ship sections Following F.M.Lewis (1929) and Weinblum and
basic derivation will be ass'ied to be identical with that about ,he lower half of a fully submerged body. Corrections to account for the
R-659
-
43 -
In considering the pitching and heaving motios of the body it is necessary to introduce a second coordinate system noTing with the ship with its origin at the center cf gravity of the ship. -iulongitmdinal
distamce of any section of the ship from the origin is deiignated (positive forward). is designated by z Vertical displaoemit of the C.G. (i.e. the beave)
(positive ugards) and angular displacemrnt or 9 (positive for bow displced x pWards).
& G , with
(It if
1.
9-V9.
(5)
2.
Vertical component of wave orbital velocity S2wy T oee -2xhc 2u/ 0os3 2 T(x -ct) y - -R oos c, (9)
where
is wave amplitde,
X is wave length,
the depth below the still-water levwol to any point in the Iuid,
R-659
All motions are assumed to be sufficiently small so that the derivatives of the potential can be taken on the surface of tha circle as if its center were at its initial position y
-
0,
sion for the potential in a uniform fluid stream can be applied, despite the slight non-mniformity induced by the waves which are asssc-d to be small.
The vertical hydrodynanic force acting on the length submerged semi-cylinder is given by dx of the
n./2
dF = 2 r p cos o dk (10)
where
4 -n V4 a.
F
I
the polar
-&-F ----
mgjle as defined
VjcjU1% ------0
where
v V)
The first
(12)
R-659
- 45 r2
2z h
r-
e2RYAcr80 d
coo-LA(x-ct). 2X
Ow0
2uyA Co
2(x-c)
(15)
(141) and
(15),
(16)
0bt*
Eaiting Forces
Attention will now be conceatrated on the second and third terms of Equation (16), the two component parts of the velocity potential which give the exciting forces due to waves.
4R2c2hr2(2
)A
2(
2 r4
h c
(-2?R cos)A
-2ct).
r (and since
c2
g X/2n),
R-659
sin
(e'(-2. (.ct)
coso
coso)A
(7
(X
ct)
c.
e (-2n r
coscs)A
(In the 1955 paper an error was made in substituting ferentiating with respect to time, since r
R - r before dif-
is a function of time.)
The corresponding component of the force due to body-wave interference is obtain3d by substitution of (17) in Equation (10): (dF)
-h 2
g n r
sin-
(x - ct)
R g hr___
os --
(x -ct
zf
cos 2 0
e(-2 r
cosc)A d 0
r coso)A
2ur C
2Srzcos,2
'
( cos2k
0
(-2n r coscy)/L
4 n r
3 n3r2
Neglecting cubes and higher powers of the small quantity first term and also
(d F)I M-2
rA in the term,
r2A
gh r
I
t
x2
32X it
S(
- 8n r) Co. L (X
ct)
(18)
is,
R-659
h7
2 it
h c2
(19)
2 g h r sin-L(xc='w
ct)
os t
(-2a r cos 0
d)A
equal to
n2r
4 i22r
3 X2
r/)k s
(dx F
2 ~
r(1-~-2
S2
3
2 2
2
r2 )sin-z(X-ct).
(20)4
The first
force resulting from the wave rise or fall and the accompanying increase
or decrease of volue. tion of this force, The second and third terms represent a Mditficaexponential variation of
pressure with depth, and this modification is known as the 'Smith Effect'. The entire equation represents the force acting under what is usually referred to as the "Froude-Kriloff Hypothesis". This force which is
R-659
h8
Designating by
d r j
d&
. -V tan
(21)
dt
With
(In the 1955 paper the sign was erroneously taken as positive.)
the substitution of (21), the sum of Equations (18) and (20),
which is
dxd xaiw
2p 2 g. h r
d x br CI --- 2
+ V tan c
-22r) - 8 R r
sin .
(x
ct) (22)
cos 2a (x _ ct) I It
3 X
he sign of the velocity-dependent terms, which are small, (r tan ?)A term, which is
must be general-
ized for ship sections other than semi-circular and second, corrections must be introduced for free surface effects.
R-659
It
2V g r h sin 21 (x
-ct)'
= 2
og r
r
represents the change in displacement force with wave rise and fall; in this case then is to be taken as the half-bed, B/2,
line.
(1928) for the force acting on a body placed in city greeient. Here k2 is
equal to
With a free water surface and the formation of a standing wave systef, the value of
a
k2 - 1
.faAi-i vi
ihs de;ignatea as
u,) 2/2,zk
.262
.8i8 .632
.592
.673
.738
.762 .818
.859
.883 will be assumed that
SR.-659 -50 -
From experiments with an oscillator, Golovato (1956). derived the coefficients of added (virtual) mass in heaving oscillation for a ship form s~mmetrical fore-and-aft with U-sections almost wall-sided at the load waterline. Fig. 8 of that reference shows a curve very sfilar Am
trend to the coefficients of the above table but with values about 20% higher and with the minium shifted to a somewhat higher frequency. The
Th
factor (I l.
k)/2
term=of
In the earlier was applied z-y
ting forces.
k4
ad
highly advisable to adopt this more accurate procedure for the motion calculations. sent paper. Since the modified *Smith Effect' terms are connected with virtual defined by the coeffiThis omission has therefore been corrected in the pre-
masses and since the effect of section shq>e is cient ks, the factor r in this case is
sectional area, i. e.
2/X
(23)
wbere
is
df
-S
dg
distzl-outIon of vertiar is a particzlar expressead W.
eaving foroes due to the action of waves on a shi. of the ship on the wave (t - 0)
insta4tazmous positio
d "
where and
g h B (KI
a r2 I1 " k2P.
si 2T 2 x * K2 c s
no--dimsez-nal om 12 ( moefficients: 4(
(25)
are.,o
the war.e
ko
tho
di
ec
in th
For am- analysis of ship motions the force distribut~m integrated to provide the total heaving force :ng _---=at R: F
must be
Fd
d
(26)
R-659
-
52 -
and
J
3dx
nd I !US)
dx
(27)
and
at
results from the consideration that the water pressure acts normal to the body surface; the moment arm is ( in the case of a body of varying circular section + r tan P) and, by the use of the relationships
(23) %id (24), for a normal ship form the moment arm may be assumed to be (g + d S/n dg ). Equations (25), (26) and (27) r;place Equations
(33) and (34) of the earlier paper. The integrals of (261 and (27) can be evaluated readily by Simpson's rule. By changing the ship's position relative to the wave
the maximum values or amplitudes of the exciting force and moment can be found as well as thg phase lags 0 and "T . amplitudes, F
0 0
Calculations of these
611, 0,60 block coefficient hull (ETT Model No. 1445) in waves of ship
length (X/L - 1) and wave height of 1.5 in., for comparicon with the experimental data described in Appendix 2 of the 1955 papor. A very
good agreement between calculated and experimental values was obtained except at zero and very low forward speeds (see Fig. 15). It has been
R-659
-53-
reflected by the walls of the towing tank, while at no longer encountered and the
The pressure in
from
--
-
2 r -cosc -2
v r
i0 C 8%s
r cos
R = r. in
v i cosa
(28)
+
is
-v
,
the vertical the
velocity due to the angilar velo-ity of pitching and the third is vertical velocity due to the instatancous angle of trim contral planes. Since is a function of tio and -V, 9
at the
G-,-2 V After substituting (8), (21), and (30) in Equation (29), the following
(0 ..
R-659
-54-
.................................
+ 2k
+
rt 2 Sr t
........................... .. . .
t.................()
) (2 ...
(3(.)
itE r tan P V} +
2it
(
9..... (6) Termas () and term (3) is e
r tan
n[ir V2 t-an
Equation (31) replaces Equation (39) of the 1955 paper. and (2) are identical with (3) and (6) of Equations (39)
the s= of the earlier (1) and (5) with the sign of the latter cn erms (4), (5) and (6) are twice term (h), (7) and (2)
respectively of
the earlier Equation (39), which were incorrect because substitution of R - r had been made before differentiation with respect to time. The factor
r 2 /2)
kc on the basis of the reasoning outlined in connection with exciting 1 forces, it can be expressed as
S k2 k).
The factor
( X r tan )
in the last three terms is the derivative with respect to utr 2 /2) and so it can be expressed as d(? S k2 l4 \)/
of The total
dg .
...
55
then
(32)
vi +
aad the noment is
dg
+ V
(V +B
dSS
- )
dt
Since
are functions of
z and Q
rad vwir a
w'--
tives, the terms of Equations (32) and (33) may be trasposed to the left-hand side of the coupled equations of notion. Their contribu-
tions to the coefficients of these equations are designated by the subscript 1. Thus
a,
fSk
kJ
d~
k. d(dS
-j
f dV2
A ( k2X,~ 1 ~~((
22
-~ B- -- 2 VO ((S k-kAd
,fd(S k21%)
D1
d
d 5 k.Z k
vr :-e
-,i
-t
I~:t I
~ =
_______
-99
Ere
rz
I'M
r
-J be
R-659 -58-
E2P,
and
S k21c,) d
pjkd (S 4,1c,) I V9
(35) [ Jd(S k 2 k
]V i
For the case of a half-immersed spheroid under the condition of a free water surface but neglecting wave-making, mic coupling terms as - pM V 6 and where is M + q MV i is for the heaving force for the pitching moment, In general p
,
q, and each
coordinates and associated Legendre functions of the second kind. a long spheroid Havelock finds that length-diameter ratio of 8 which is form. (Harkind (1946) - =
-
(1 +
.)/2,
or .515 for a
is
Sd
L
-L
dS
M/e.
where the bar indicates the value for the entire body. I
a-659
Thus
p - q -k 2
For a circularsection
=1
d, at the
oscillating frequency in the vicinity of synchronism for most ship forms, k4 is of the order oZ O,,. It appears from the application to the
is increased by the addition of the dynamic coupling terms. Dissipati-e Damping It was mentioned previously that the free water surface was not taken into account in the basic derivation of the presett paper and that a correction for it must be introduced independently. The effect of the by
free sv-rf,ce on the virtual mass has been allowed for approxizzel the use of the coefficient k. derived by Ursell for a semi-cylinder (195sh, and Discussion, Korvin-Kroukovsky, 1955b). (Grim, 1953, has
also calculated this effect for some ship-like forms but his material is not extensive enough for general application.) The other well known
effect of the free surface is the di.sipation of energy in the formation of waves which propagate away from the ship in all directions. In the
'strip" method of analysis it is assumed that waves from each length segment dg propagate laterally. If the ratio of the amplitude of
these waves to the amplitude of the heaving motion of a ship section is designated by A , the dampii-4 force per unit vertical velocity v of
Havelock 194~2),!6)
ee
wJave
HLstein U?936, "I937a and !l;37D) and~eoc. the heaving body b7 a diztributo of :6zomc~a~Ly
in
pi~
ai
riven ;
2e-ko f
vnrp-
LI-I
/g,
-s Ithe ha'-b~-eaa
BE/2,
fA
=
tn--me
~r
z ar
test tank.
and
-:-t-a
r~
ppocz
33
af
s~
e-..
a-ee -rt2* a
~-C4.
WMU l~hf= 49=t 'e-* .di-er t vs-t
~t -2
at& -
as.
3f
1-J-
!v
"i_
?apn~~ -,-
rtrlfvbe
-e
swhc~
a.e2
f2
-vm
elw
ti
aw71
mmi
R-659
-
62-
The computed coefficients are based on the "strip" method of analysis which assumes two-dimensional fluid flow, while Golovato's and Haskind's and Riman's tests were made with ship models, i.e. in threedimensional flow. The relation of damping in three-dimensional flow to
damping in two-dimensional flow can be estimated on the basis of the work of Havelock (1956) and of Vossers (1956). Havelock calculated the
damping coefficient by two methods, three-dimensional and "strip", for a submerged spheroid with length-beam ratio of 8, parison with ship models. a fair value for com-
Haskind's theoretical results (1946) for a "thin ship" in the sense of the approximations introduced by Michell (1396) in hs th-ory cf
0 C4+ n--.
,4
T6
__1
4.
o*-
__
-'-
"-~-
'-A
1-,--
4-
Figs. 16, 17 as ratios of the coefficients of damping in heave or pitch by the two methods. 1 0 It should be remembered that accurate
evaluation of damping is most important in the vicinity of synchronism The valnes of the frequency parameter u32 L/g (02 L/g in the
figures) at synchronism are given in the last two columns of Table 5 for the eight models to which the comp'tationpl method outlined here has been applied. (Also included in Table 5 are the parameters L/B,
in the figure,
R-659 -63-
Omitting from consideration the original trawler and yacht, the lowest values of the parameter u 2 L/g are 10.4 in heive for the T-2 tanker
and 11.7 in pitch for the Series 60 hull, while for the models with displacement-length ratio of 60 W12 L/g is above 17,.6. At these valuzo heave
and small corrections for dauping in pitch (0 to 15% based on Havelock' s computations and 10 to 20% from Vossers'). Since the corrections for
three-dimensional effect are negligible or small and since they have not yet been developed for ships of normal form, they will be ignored in present work; flow by Grim. the
tC1c damping will be taken as computed for two-dimensional It is gratif7ing nevertheless to h=ve measuxre of possible
v NC
where N
V ej1 N
( ,(8
),
the damping fc-rce perlunit vertical velocity of the ship The to damping force and moment of the
[+
-V
a(3)
Te contributions o
ts
B2
N2
d5
C 2-7
=t rical2.,- as
all2 cases iz
paperi, and :iver the '-.e=th of the1=1 *ihen allJ the forces and momts p"xtoa1~ derivatives vithi resemct -x ti~1 'velopmt of t* .pot~ent'ial 4t.eoreffect-v, z azzd and~e abc-re ufc -
are coti-rd with the imrtlal and rest..ri of th~eato. :f =,:iz z"=
the c-,effficiemt~s
a z - .(S
kz
&%
/2
l--
ca~
A-~~
'-"~2
=tgD-=a.r
:jIl a
66-
eveluated on the basis of the beam of a section* cients of the angular displacement
and also on the kinematics of the fluid flow resulting from the ship being at an instantaneous angle of trim. It is seen that the damping force coefficient b in heave is a
speed
V per se.
T.is appears to be confirmed by the experimental However, the damping moment coefficient -.-fficients V e and E B
are composed of
R-659
T&IE 1
MODEL PROPERTIES
1616 5.00
1444 4.79
-
1723 5.71
-.-
1699B 3.94
4.28
1699C
1699D
-
5.00
5.55
5.71
UdL
ft.
5.71
Beam., ft. Draft, ft. Displacement(FW), Block Coefficient A/(.OIL)3 Radius of Gyration in air, ft.
Natural pi .-hine 'lp__ver1afloat in c aim water) sec.
0.667
0.267 lb. 33.3 0.60 122 1.27
0.?
0.667
0.267 33.3
0.65
0.290 41.0
0.608
0.208 24.5
0.74
0.330 33-3 0.51 204 1.07 0.74
1.22 0.690
50.8 0.23 370 1.07 0.81
0.555 0.255
24.5
0.709 0.417
24-.5 O23 60 1.37 o.63
0.55
60 1.37 0.59
0. 51
60 1.37 0.59
0.70
0.70
0.75
0.61
0.73
0.78
0.61
0.60
where L
is the load WL
I
1
I I
TAEMX
3
|
VALUES OF COKI
EQUATIONS OF FOTIK IN HRAVING AND PITCHING FOR ETT MOML NO. Ii45 in WAVE LMGTHi - S= LENGM
fps) of
SymbOls a)
A
0 6.36 1.87
2.40
2.25
9.19 1.74
2.23
1-25
10.15
1 7 2.22
4.00
42
1
7
2.;2
b
B
5.6
7.7
4.7
6.8
3.9
6.2
3.0
5.2
2.3
L
C
0
lh7 186
C 0 2 i-5
1
185 -. 05
-2.0
C-+
d - D
e2 K2
186 -. 05
-2.3
138 -. 05
-2.2
137 -. 05
-2.1
183i -. 06
-1.3
el -- El
E
0
-2.3 0 02
-. 9
-1.3 -6
-23
-1.6
-10.5 -9
-26
-2-4
O.4 -10
-27
-301
1.2 -9
-26
i .
-17
-17
]
i
S3.6i
1.0 3--7i
0.8 3-6i
0.8 31Si
Based on :jt 13 is
at station 11.
AB.1I.2
.'
.,:I
fp-s) o.
5. 3 6
-. o
.1
. 2
Ia
A
5192
2.52
1.79
2.. .3 .3
b Si.!0.6
C
7.1
5.9
h.o 9. 7
227
-2
".6
.7
227-
226
223
22$
225
d
e2
_-
D.05
E "2 -. 1
.
.015
0 -. 9
.015
0 -1.6
.c
0
-2.L -3.0
C.
go
.2
-E.1-1.3
-!3
g2
0 .
- U.
-5 95. 2 0..7
.6
iI9
-=
5.. 22.
0.8
0.0
0.
I
!
I
II
IS
'TO
G ?3K
AS OB'TAI
BY' T
Heave, inches
?-tch, degrees
op 0
By dixeect analysis of axe! ntions in irregular aves By calculation from wave spectra and experimntally measured responses to regular waves By c.0iculation from wave spectra and ;wialytic ally compted responses to regular waves
2.53
t/sec.
0 speed
2.53
It/scc
5.2
6. I
.,
-2
!.l
L.3
5..0
.01
mlitudes
SFi
saze? 7.aJ
Tqc3kw95
sr-
'.
L.ai -B
-,c
na-deOyL5)
4--72
.2
1/
k-
k#
C7ode1 729-
3. 7-
-6 *7mde.
5-90
---
__
ee s-' D
amoc .66 ,
~1cor~Z
Figs. 6 and 7:
Sulbztu-r
-e2 =
for
A I (corrected for
Transactlns of SNAK).
Fig.
i. =.O,
i2
be showtn in
j$
;,,/
N.
t i
/
\ \
j
\
/
I
.1
K,j>\.K \\
I
-
j&Y
\\
-.
'1)
/1
.g44
'-
.L5 eSw
t4
/1!
Fig. 4
Model
1699A (Trawler)
10
j6
4
to
88
_ _ _ _ _
6 4
Il1
A 45
In
It
00 190 410
Circular Frequmncy,
23 220
.......... T-'
210J
ILT
20
00
____
3 4I
]___
2 3 4 Forward Speed, V, F_/Ssc
FIG.
FIG.
I. o e hose 4 4 5,LoX/7 1 -1 .0 -
is
300
-30AJ
SIbb.O
5c and--
Modelu .Leths
by 125
n.
Hgh0
0c
a
Doue A mpltudes
J f
P-Fase.TT Los
ns[!:
-6!
PnhaseLos
. . --
.o* ./LI X . !b 4
1444.6 X'-.
--
500
"_4" ) I2
Model
E jI ; o
FtiecPh 1Seed,
Pic
Fig. S triangles are D.T.M.B. data, squares M.I.T. and exes U.of Cal.)
LDouble
Arrnitud es
Phase L
Dole Amplitudes
Phase Log
,j
C 0Ch
00050
i~~~~
Pich'
r-' ....
0
-0-
SfiII.
Iu
oi 5. -F T. E. T. I. S~o c xizJ-2 3 In -Waveso l i.0o Mot -ions 1.5, and 2.0 Model Lengths by l.h3 in. High.*
1 25
*Circles indicat+ E.T.T. experimental data (open for heave, solid for pitch) and curves show calculated motions. Zero phase lag corresponds to maxim= pitch or heave with wave nodal point at midsection.
A~~itudeS Do~e
00
'IJ hVcK2
-. I45
0-
4? 60
24
68 Moe Spft
246 Ft'Sec.
Miodel 1699A ia~d 5*7-Fto E.T.T. FIG 11motions of 4.3-Ft. B.T*T,. !odel 1699C, in Waves of 1.0 and 1.25 Model Lengths., FIG.
J&8
Mo
TlLriwth
hi
Raiaht
[ThoAdeS1
2300g~
50-
F-ag
--
rrue
Phs7
30
Phase Logs 5001 I obeAmplihedes H -Mode~ 169 0,11 Vt 42 !r' Model 1699 B.x/LaMo; 5
Dcmkie Arrpietudes
'9D12'5eIj99
PO
Lags
l /LI 5
~i
o-j
Heav
FIG* 12 Motions of 4.3-?t. E.T.'r. Model 1699B and 5.7-F-1- B.7,?,. Xodel 1-69910, in it-mwas of 1.0 am-1 1.25 ',Kod1 Lengths, 1148 Model Length in 1Feight.*
- ki+
I
o044 __+-+
0Sped'
1723
__
__
II
__-
2.53
11it
*,
-
do
90
I ,\
so
il
.I
Freqvemey
of
Co
Itowve
SDestroyer
Model 1723
i) 2) 3) By dLzct analysis of notions in irregular ves (solid lines). By ca:lcuY .+ton from wave spectra andl e.Rpriment~aly measred
Sresponses
Sresponses Ii
to regular waves
L
-x
'-
dx
-i-
+ x7A Vu'c
WAR
G. il
0L
R~tI__
0 0
10
3xnt
FIG. 15
Coaparison of Coup ted and ExPerimentally Measvrv4d E~xctirg Force and oment, Amplitudet for E&T.T. Model iIa5 in Waves 5 Ft. by 1.5 in.
!*.c
1~
2L
I
"0 2
4 6 5
ILI
0 1
I
i
FIG. 16
Ratios of Three-Dimensional to TwoDixension.1 Calculations 8. of Damping Coefficients for Submerged Spheroid of L/B (Fig. 1 of Havelock, 1956)
I
I0+
04-
IX4~
9
FPG. 2
-_K-,
I
FIG. 17
Ratios of Three-Dimensional to Two-Dimensional Calculatiors of Damping Coefficients for Thin "Michell" Ship of B/H 2.
(Fig. 2 of Vossers, 1956)
i i