Professional Documents
Culture Documents
Review of Added Mass (1982) BRE052
Review of Added Mass (1982) BRE052
January 1982
An investigation Conducted by C. E. Brennen 360 Olive Tree Lane Sierra hladre, California
REPORTDOCUMENTATION PAGE
1 R E P O R T NUMi3FR 2 G O V T 4 C t E S S I O N NO.
R E A D INSTRUCTIONS-
CR 82.010
4
T I T L E (and Subc~rle)
AuTrORis;
C O N T R A C T OR G R A N T NUMBER(s)
C. E. Brennen
b
P E R F O R M I N G O R G A N I Z A T I O N N b M E AND ADDRESS
10
Ill
14
C O N T R O L L I N G O F F I C E N A M E AND ADDRESS
~k !To24
a
5 0
15
I
DISTRIBUTION S T A T E M E N T (of t h e a b s t r a r r e n t e r e d I..? BJock 2 0 . r f drllerenr from R e p o r t ) 19
K E Y WORDS I C o n r r n u e on r e v e r s e srde ~f n e r e s s a r ) and r d r n t : l y bv b l o c k n u m b e r )
Thi s r e ~ o rr te v i ews t h e e x i s t i nq s t a t e o f know1edge concern. i n g t h e e v a i u a t i o n o f t h e f o r c e s imposed on a body i n a f l u i d due t o a c c e l e r a t i o n o f e i t h e r t h e body o r t h e f l u i d . I t concentrates on those f l u i d i n e r t i a l f o r c e s due t o a c c e l e r a t i o i r a t h e r than on t h e d r a g / l i f t forces due t o steady motion. The f i r s t p a r t o f t h e r e p o r t presents a -survey o f t-he a n a l y t i c a l
DD
M :7 :3
1473
EDITION O F I N O V 6 5
1s O B S O L E T E
Unclassified
S E C U R I T Y C L A S S I F I C A T I O N O F THIS P A G E ( W h e n D a t a E n t e r e d )
. .
background i n c l u d i n g t h e d e f i n i t i o n o f added mass, t h e s t r u c t u r e o f t h e added mass m a t r i x and o t h e r e f f e c t s such as t h e in f 1uence o f v i scosi t y , f 1u i d compress ib i 1it y and t h e p r o x i mi t y o f s o l i d and f r e e surface boundaries. Then t h e e x i s t i n g data base from experiments and p o t e n t i a l f l o w c a l c u l a t i o n s i s r e viewed. Approximate empi r i c a l methods f o r bodies o f complex geometry are explored i n a p r e l i m i n a r y way.. The p o s s i b l e dramatic e f f e c t s o f t h e p r o x i m i t y o f t h e ocean bottom a r e f u r t h e r high1 ighted. The confused s t a t e o f a f f a i r s r e g a r d i n g t h e p o s s i b l y major e f f e c t s o f v i s c o s i t y i n c e r t a i n regimes o f frequency and Reynolds number i s discussed. F i n a l l y a number o f recommendations stemming from ocean engineering problems are p u t forward.
S E C U R I T Y C ~ A S S I F I C A T I O NO F T H I S P A G E (When D a t a E n t e r e d )
CONTENTS
1. INTRODUCTION ................................
... ................................................................................... 1
2. GENERAL EX.
.........................................................................
3. ANALYTICAL A
TENTIAL FLOW
.................... ....*....,........................
; VARIIATEQN
........................................................
W ETH
TION TO DISPLACED M A
S MATRIX ............................................................................................... 6
OF FLOW SOLU-
s ...................................................................................................................................... 8
.5
TRIX .............................................................. 10
N OF THE PLUI
ER THAN THE
HEITY ...................
.........9~e.....D.e9.e............
..........
11%
.................................... .
16 16 17
. ...................... . .
.........................................................
19
BUS EFFECTS ON
Y .................................................................................................................................... 25
MMENDATIONS .................................................................................................................27
7. REFERENCES ................................................................................................................................ 28
U dike
d to arbsorb ev
fRef.l?),
L EXPL#bNATION OF ADDED M A S
mas" i that it determines the mxesay work s
associated with the motion o the fluid. Any motion of a fluid such f
h e t i c energy assodated with the fluid motions. This kinetic energy, T, aan be
sented by
domajn or volume of fluid For simplicity we sllall assume t,hmughoutthat the fluid is i n r x , m b l e
with a density p.
If the motion of the body is one of ste
st then dearly the amount o f
numbers.
the kinetic energy i the fluid n
es. If
energy must be
body in orrler to increase the kinetic energy o the fluid. And the rate of additional work required f
eo f
T with .respect to tirlne, dT/% This additional work is therefore emrperitional work done, -FU is simply
ccxlsknt it f ol-
md',
i no such ida3nmle s
the total kinetic
31 EXAMPLES: RECTIEIN
W FLOW
It should be dear that the methodology could be extended to more gmcanid out i the foll~wingsection n
I, and the~fore added rmss can be dcul&d for rectilinear motion the
For the purposes of this example let us examine the ideahzed potential flows past, a sphere and a
cyhnder, The geometry for both is as depicted in Fig.1. The s p h e ~ cylinder d radius R is assumed or
with time varying velocity U ( t ) ( t is time) in the positive z b t i o n . Polar cooldis (r, 19) a~ used where z =rcosT?
Fig. 1
?-
r who is unfamiliar
th these solutions may wish to satisfy himselP that kro solutiom satisfy
the bun-
n&tion that the relative velocity nonnd to the surface of the body is zero ((v,.),.~= U cos @.
se
How-
ever, it is c o n ~ p w y
for the case of the a r d w cylinder). Furthemre we shall see that in general, the value o Ule added f
mass depends on the direction o acceleration. For example, the idealized potential flow solution for f
the j.nfM&ly thin
th the
sdme
is &md, namely that the fom! on tlke body due to ~ l e ~ is o n wt necessarily in the same direction
on in one &&ion
leraban. If, for example, one w m liftmg a body from the ocean bottom by means d a cable then
e r e m a r mtioa
s a method o f
corn-
on of the inertial forces, If the center of mass o the body is chosen as the origin the f
only seven
and the six &ff e ~ nmmponents of the moment-of-inertia matsix [Yh, p.lO2]. However one cannot t
in general relate any o the 36 M e m t corrqponmts of f
the added m a s m
o symnma3tryI for f
need evaluation a m
The last two, wh& with MBE=Ma and MsS=MSJ represent #he only mmmn off-ihqpl terms,
m m w n d to the moment
the z ,
axis gene
the
xz c
axis g
about the
xg3
direction wiU
elements
Ma , i= I+ 6 M
( 1) 4
if potential flow is assumed all of te off-&agond terms are zero. Only i this bunple aase o h n f
e
34 A
MASS MATRIX
the mere definition of M y i Eq.(P) requires certain n o the flow and the f
six direcstions. The question of- the
preconditions n c s a y i eesr n
Equation (9) is one which wl not be addressed hem. It is however dear that these il
as mop1 as one
ed by each o the f
that
lin~ar.This i mt s
mperposabiuty is not, in
gened, applicable. However there are two mdels of fluid flow which do saiisfy this condition namely
(i) the poknhd flow m d e l for hgh Reynolds
asymptotically d Reynolds
fern Furthermore provided if one is dealing w t rigid or mdef ih
rsents, j =4,5,8 of
fluid velocity
and % a identical m
zrr
Furthermore it is dear+ fmm this evaluation that the added mass mataix must be gmmetric since
mersmg j and k in Equation ( 1 ) does not change the value of the intqml. Hence superposability 6
uy is
the fluid velocity due to unit velocity of motion of the body in the j dirrrction. Consequently the solut o o six steady flows for j = 1-r 6 a l w evduakion of all 2 distj.net values in te added maas m&rix. in f los 1 h
Hence one can rrnake use of exisbg methods for solving steady f o s mund bodies of quite complex lw geometry in order to evaluate the added mass mtrix. References 1,2,3,4 9 provide info-on and
se exi-
on
metho
O n e other Em o E s f
(17)
pj
o ~ p ~ s the h n velocity p t n i l of the steady flow due t unit motion of the body in the j-direction oeta
then it follows that
n o d t~ the surface, S which mpresents the body surface. I m y cases of n f o s around complex bodies it i dearly eader to evd* lw s
the volume
i (17). Indeed the form (19) is ideally suited for use with p t n i l f o n o e t a lw
a s N e m code.
fluid. This =view would be incomplete w t o t some corwnent on the cases i whch t e fluid far ihu n h
witha
would be me
at a velocity, l$ (j = 1-+ 6) T .
conditions
identical for the two flows ezcqt that where thc3 pmsam, p
dR
second flow. Consequently the stresses and forces whch the fluid exerts on the body m identical
where VD is the!volume of fluid dlsplmd by the body. Consequently the inertial f orre is
But as stated previously the arxlelerabon of the body in the second flow i now Aj s
and hence in the ame of the second Aow
+ dWjM
red ~
whe
delta (6+
the "added masd' assodated with the fbdd acceleration, dW j/& i i t e s e o ~ n d is the n nh flow
of the true added
s,
However we
now examine more closely the general validity of Eq.(24). The Brst and second
even h
~ the h body velodties and far field fluid velocities are identical. In general the solutions ta
s egy&ons will also depend on al o the prwious time h&ry of the body and far l f
the two
fluid veld@%l&.
viscous effects a neglected then the fluid has no mmry and the Auid velocity fields will be identical; m
velocity,
( Uj - W j ) i iclenucal s
plane and treating the total flow due to the body and its image. Equivalence o the two pmblems f
allows the tramfe~noe f o
see the cases of two cyhders and a cylmder plus a plane boundaPy in Table f I.
38 THE EF'FECT OF A NEARLY
A s a cMlflSequc?nmthe
s flow problem. Though this subject is outside the scope of this report it. i necessary to makie a few
brief r e &
and, in piut;Lcul~t id&nMythe wanditions under w i h one must account for free suro hc
face effects.
ng b&es the reader is refened to excellent reviews of the analybcal techniques by W ehausen (Ref.12). Newman (Ref. 13) and Oghrie (Ref.1 ) W r n e ~ e d 4. bodies are only W t l y
easier t handle. Some o
e
paor
atisn or mtalhond m u o n s those bodes would have generated free surface waves which f
~ndscte
10
on the body. T
rized by a velocity, Ul say) then the waves mated will be neghgibly small and these prior
1
The results of Table III do a l w one to estimate what constitutes pxirnity to a free d a c e providlo
mg the above conditions b l d It can be seen that the free surface has liWe effect (less than 5%)provided the mbo o the depth o the body to the body dimension is g ~ a t e than about 4, For lesser f f r
of the
eerhg appbcatim. T
4 REXEW
jm th fluid d . ki o
methods d o g o u s t those described i Section 3. Bodies for which the steady flows can be genemimi o n
and i many mechanics texts ( e.g. Ref . 1, p.104) . A p r b c u l d y useful tabulation of many of the availn
able results is given i a paper by Patton (Ref.$) and h Tables 1 and 2 are mpmduced here as Tables I n s ixnd 111. Note that; the thud co1un-n of these 'tables contains the added mass denoted by q;the
f values given cornspond to the &agonal terns in the added mass m a . , Mu, the dmdion o dfied in the second column (No off-diagonal m ~ n e n t o the added mass sf
r e d & me also li&d for bodies o or mar t a solid or fm smface and comm o
ment an these is delayed md later. FWbn has hduded bath t h e o ~ t i d t potential flow added masses
detemined added masses in Tables I and 111. These are di-shed
by the letters
n h f f T and E i t e fourth mlumn o these tables. Another excellent s s m o tabulakd added masses is
on that Wse tables are more tlkan adequate for mst engineering
is not in close pmpdmily to a solid or free d a c e . The
3 and detailed in many mferences (e.g. Ref.9). Modem potentid flow computer progmms
(e.g.Do
~ euman code)Nfor ~
se as
in
added
TO THE GEOY ET
mtatmg the projected area of the body in that duection about m m i s defined by the smaller of the two
f pal dunendom, 2q, 2@, o the projected area For an eUptical projeckd area this would yield
4 M ~ --pn&(a$ = 3
&> q
(2 ) 6
where the^ is no implied summtion over the index i This would yield a onabl ably conservative
approximation for the poeckng case of the cylmiers. However it would substantially overestimate the added mas for a body like a sphere w h e h has a small aspect mkio. Then the above estimate would be twice the potential Aow value. Perhap a better emptricd qpmrdrnation would be
prolate elhpsoid acceledmg 'IxDmlslde on" (see Table IV) we And a value of the added mass usng
Eq.(ZI) whch is withm 7% of the exact value. Further irnpmve~zmts could dearly be made but are
o &or f value considering the other uncertainties dlscussed below,
The ~
mass for a body o cornplex geometry (an implane). Though it would require further detailed analysis f
it would not be masonable to suggest that a complex body be considered disassembled
component parts (wmgs, fuselage, W) and that t e added rnasses for each mmponent h f evdulakd for m e perpendicular dmctions o amelemtion usmg the technique outlined in the previous section Then we mud ask whether it is appm*te1y
corn& tn sunply add the added masses for
the mmponents in each of the tlme diredions. Fmrn an engtneenng p i n t of view it seems masonable
w n;dmely
'the
Fmm a
dose p m u t y
effects of t e pmximity o a solid fluface on the inertial forces recped to move a body may from that h f
surface. CoMder the two clunensiod problem of a Rat plate o width, 2a, lying on an ocean f h r . f ? f A vertically upward f o m , l per unit length of the plate is applied at the center o the plate to lift it
away from the floor. Due to thls force the plate has risen to a uniform he&& h(t), above the floor
a time t. The velocity and i3ccele&on of the pl& in the upward direction a therefore &@ and t m
d2h/& (see Fg.2).
F i g u r e 2.
if -i-:a77;;7-77-7 -~7;-7;i*"-/-7,;r;rr-r;;r7-7--7777777-7~/ / //
t-----------
~rn T
i h (t)
/
,/ -
/ / / A /
This
stic of m f&y y
grpidy only prtions of the mdeISUllQacewould be in contact with the w a n floor. However one
between t e undersuface and the object Wtmg up o the bbject could., o mum, make ho very h f f
small. In any case some ho would be pertinent to the moment t =O, when the hft force is applied.
tuti
I t is o intenst to examine brieAy the corn- o this khwior of the fluid inertial f oms. If, f f
for wnphcity one neglects the o the plate itself, then the upward form apphed to t e @ t by the f h ae
cable wl be equal and opposite to the fluid inertial force, For illustrative purpose suppose a constant il
upward cable fonz is applied. Then internon of the ecphon of motion represented by Eq.(31) if F
leads to a kind o ''catastrr,phic" mlease from the botbm in which. the upward acceleration increases f
with time. I t is unlikely the~fore that a constant upUt f o m could be maintained under these cinc
. Consequently the actual initial mtion would be dependent on other factors such as the
cable elasticity.
and analybcally.
ed by the viscous effects in the flow m n the body. These viscous effects on both t e fluid herud h
d drag foms have been the subject of a number of debled studies as represented for exam-
Refs.10, 11 and 15. The essence o the mmdplidon is that in certain regms o flow t e f f h
viscous processes o flow f
e
is embodied in what is known as M orisn's equafion (Ref.16) in wh& the total forte on the body,
p,
where C . is a lift and drag coefficient rnatrix, and A A a typical m a for the body. This equation is
Carpenter. The Reynolds numbers for the Skop, Ramberg and Femr experiments a in the range m between 230 and 40,000 and are t h e ~ f o r e a r to those o Keuiegan and Carpenter. It is quite dear M f
that further detailed m e m m e n t s using mare sophisticated measurement and data analysis techniques
are needed to resolve this question. Tho@ it has little value, I have formed the very tentative opin-
ion that the experiments and data reduction techmques used by Skop, Ramberg and Ferer axe superior to those o Keulegan and Carpenter and therefore I would place mare confidence in their d t s . On f
the other hand the data o Keulegan and Carpenter is much more widely known and used; this I[ believe f
For the present it is necessary for engineering purposes to be awm that pathologid behzrvior of the Auid i n d d f o m s m@W occur for body motiom whose typical amplitude is water than about half of
te body dirrnension h
mg this Subject it is of value to S m r d a few o the results of the experiments cartied out f
by 3
The relibnship between the foms when the fluid is amelemtmg past the body as opposed to the
reverse is discussed in Sedion 3.8. I t is shorn that a relation ononly be Annly established if either (i)
suprposabfity is possible (e.g. potential flow) or (ii) if the e& n
city is identical i the two cases. Then the qprupriake Auid mass i the aase o n n f
Auid arxe1erat;ion i s
dab on this is limited and a,nM&ry. For the present one can only point out that pathological
behavior rmght occur in certain ranges o frequency (or typical time of acceleration) and Reynolds f
numbero
me, A t the p ~ s e n t
engineer could use w t confidence. ih
7, REFERENCES
Dwer
Pub., NY.
5. Paton, K .t 1965. Tables of hydmdynmic mass factors for tmdational motion A SM E P p r , 65W A NNT-2.
N o.260. Kmsler, L .E. and Frey, A .R. 1962. F m h n t a l s of A caustics, J.W iley and Sons,NY.
7. Pabn, K.T. 1965. An experinrental investigation of h y d m d m c mass and mechanical
K e&
Report 2299,
9. Keulegan, G .H. and Carpenter, L .H. 1958. Foms on cylinders and p b s in an 0sdtlzh.q fluid
J.Kes.N&Bum:Standawls, V~1.60, Q . ~pp.4553-440. N ,
! Sarpkya, 0.
W ehausen, J.V. 1971. M otion of float;mg bodies. A n n Rev.Fluid M e&. , Vol.3, pp.237-268.
13. Ogdvie, T.F. 1964.Recent pmgress tow& the understanSyfn;p.Naval W ydmdynamis, Bergen, pp.3-128.
14. 53op, R . A , , Rarnberg, S.E, and Ferer, K.M. 1976. Added mass and darriping forces on d r d w
T. d
n, M. 1981.Kil
TABLE
(From Reference 5 )
TABLE
11
(From Reference 9)
T~ =+ pn a '
u2,
as in Equation @$i],
Elliptic eyrinder in translation parallel t o an axis, called the a-axis, either a > 25 as s
TABLE
Plane lamina rotating about one edge:
II
(continued)
a4 a2,
with Apparent increase in moment of inertia Moment of inertia of fluid displaced bya cylinder of radius a rotating a s if rigid about a generator
3 p,
a4/2
Fluid inside semicircular cylindrical shell "gtating about axis of the semicircle
/ II Qu I \ I
A x i s of
Tipd(l-T) " i n r -1
a4 a2,
as in Equation [102e],
Fluid inside equilateral triangular prism rotating about its central axis: Axis of
1.
T l = m P S a'
a s in Equation [103k],
Fluid
8
TABLE
11
(continued)
Cylinder with contour consisting of t w o similar circular arcs; see Section 89.
grees
=
180 f
c2
2 ( 1 - f ) n+sin 2 0
degrees
1 'f', - pk 2
SO^,
k= -
SU~,
- -4~~
iT3
0.525.
----- - 1 = 2,049.
n3
fi/z
fOr translation i n any direction
For K , see the preceding ease, See Section ouble circular cvlirnd~r,each c ~ l i n d e r radius a; see Section 90: of
n2
\6
-1
k=--1-2.290, 3
TABLE
II
(continued)
E
Wall
Mi=
p s2 sin 0
Ilere 8 i s in radians and lr s t a n d s for t h e gamma function. Rectangular cylinder in translation parallel t o a side; see Section 91(b) for references.
M;'= k M;
apparent i n c r e a s e in mass,
s,
TI
1 =y u2, kMI
as in Equation [91g],
Cylinder of radius a in translation and instantaneously coaxial with enclosing fixed cylinder of radius b:
TABLE
II
(continued)
Cylinder of radius a in translation i n any direction across axis of enclosing fixed square cylinder of s i d e s , a / s small; see Section 9 ( ) 11.
(Only the force required t a c c e l e r a t e the cylinder i s consi o Cylinder of radius a moving symmetrically etween fixed infirrite w a l l s K a small:
Wall
Wall
Wall
Y-.
TABLE
I11
TABLE
I11
(continued)
Vcrticnl
Vertical
m,
t -
K -2- r p 3
ab'
a/b
1.00 1.50 2.00 2.51
E; for Latcrnl
i\!ctlor!
..
.NO 30 5 .209
.500
-156
2.77 3.97
4.99 6.01 6.97 8.01
. 13-2 .CS2
.059
.Od5
3.02 9.97
,033 .94 5
.954
.a ! !
1.003
TABLE
III
(conti nued)
I1:;rirody?arnic
ci a n c!!i;.scic!
>!ass
TABLE
I11
(continued)
Vertical
TABLE
I11
(continued)
,
4
:!rc,i
.,I;
iivr:;rl.nt;,]
oi I ? O ~ ! ~ >I:ri:num
r
1
i
~ ! o r i z o n t dSection.
Section.
I
,?re3 of iiorizon:nl
II
I
"Tail"
'l'ATZI1E
ITT
(continued)
TABLE
IV
( ~ r o mReference 9)
T = = - - ~ ~ ~ L I ~ ,a s i n Equation I127f1,
3
Only the force required to accelerete the sphere is considered here; s e e Section
T=-- p a 3
3
bJ+ -2 r z 3 u2, -
as in Equation [129e],
b"aJ
TABLE
IV
(continued)
Prolate spheroid (or ovary ellipsoid), a>b; see Section 137: Let e = eccentricity of sections through axis of symmetry,
2otation
d, Y
See Table
TABLE
TV
(continued)
See Table
TABLE
k2
k1
k '
Rotation about Equatorial Axis
Translation
- -
"edge on"
- 0.500 0.803
"
0
0.115
1.118
1.428
1.742
0.140 0.121
4.915
5.549 6.183
00
0.075
0.000
TABLE
---- --- .
IV
(continued)
T=-
4
3
p a3v2,
as in Equation
(apparent increase in mass) (spherical mass of fluid of radius a) Circular disk rotating about a diameter; see Section 138:
Axis of Rotation
(apparent increase in moment of inertia) (moment of inertia of sphere of fluid of radius a or 8 pn a5/15)
--
Elliptic disk of elliptici~ye in translation perpendicular to its faces, a > b; References (240) and (235):
=k"=
1 -
pn
E '
E=
f i 2
\/I- e 2 sin2 6
Jo
TABLE I V
(continued)
T = - pn abc
3
0 0 - 2, u
2-a0
4 M '= - pn abc, 3
2 T = -pn 15
abc o 2
( b 2- c 2 ) 2 ( y o - P o ) 2 ( b 2 - c 2 )+ ( b 2 + c 2 ) ( P o - y o )
F l u i d i n s i d e ellipsoidal shell rotating about its a-axis, any relative magnitudes of a, 6, c ( s w last figure):
t=
2 ( b 2- c ~ ) - pn ubc co 15 b2+c2
4
as in Equation [t40f],
I'= - pn a b c ( b 2+ c 2 ) , 15
49
TABLE
IV
(continued)
Solid of revolution formed by revolving about its a x i s of symmetry the lirnason defined by
b ( S + cos e)/(s2 - 1) where b and s are constants, The curve for s = 1 is a cardioid. A f e w values o f k are given by Bateman in Reference (240):
=
TABLE
(From Reference 9)
Radius of a circle or semiaxis of,an ellipse or ellipsoid, or half-width or width of a lamina Ellipticity Coefficient of inertia, a dimensionless constant
k
In translation,
k=
.
'
In rotation,
. '
Moment of inertia of displaced fluid rotating as a rigid body about the assumed axis of rotation See under T I Mass of fluid displaced by body See under T 1 Kinetic energy of fluid Values of T, 1', M' for fluid between two planes parallel to the motion and unit distance apart, in cases of twodimentional motion
U 8
An angle in radians
Density of the fluid, in dynamical units Angular velocity of rotation of a body, i n radians per second,
The fluid is assumed to surround the body and to be of infinite extent and at rest at
infinity, except where other conditions are indicated. i n regard to units, see Sections 18, 14%