Wave Equation in Fluids
Wave Equation in Fluids
Plane wave propagation means that the acoustic variables, such as sound
pressure p, have a constant instantaneous magnitude throughout any given plane perpendicular to the direction of wave propagation
Rarefied Pressure p / p0 p0
Wavelength
Direction of propagation
Figure 1 Symbolic depiction of plane longitudinal wave propagation for a sinusoidal disturbance. The fluid particles are sketched in the compressed and in the rarefied regions.
1/48
2/48
Pressure:
where
p0 the pressure in the undisturbed medium, and v p (r , t ) the pressure disturbance in the medium, the sound pressure.
v v p t (r , t ) = p 0 + p (r , t )
(1)
) and time (t),
r r r r r Particle Velocity u (r , t ) = u x e x + u y e y + u z e z
(2)
r r where u (r , t ) is the particle velocity vector ux, uy, ur are the corresponding velocity components, r r z e x , e y , e z is the unit vector r r t (r , t ) = 0 + (r , t ) Density:
r where t (r , t )is the total density,
(3)
0 is the density in the undisturbed medium, r (r , t ) is the density disturbance in the medium.
3/48
Equations of continuity
one-dimensional case
y
ux t
z y x x
u ux +( t x) t x
x
Figure 2 Mass flow in the x-direction through a volume element fixed in space.
4/48
( t xyz ) = ( t u x yz ) x ( t u x yz ) x + x t
(4)
The second term on the right-hand side can be expanded into a series, for small variations about the undisturbed equilibrium state, and if higher-order terms can be neglected, then ( t xyz ) = ( t u x yz ) x ( t u x yz ) x + ( t u x yz ) x x (5) t x This can be simplified to
t + ( t u x ) = 0 t x
(6)
5/48
(7)
r r r = e x + ey + ez y z x
permits a simplified expression of the continuity equation, as
(8)
t r + ( t u ) = 0 t
(9)
Putting the total density (3) into the equation and taking advantage of the fact that the undisturbed density 0 is independent of time and position, and ignoring second order terms, that consist of the product of two acoustic disturbances, gives the linearized wave equation
6/48
(10)
ux + 0 =0 t x
(11)
Equation of motion
Consider a specific fluid particle, with a fixed mass Dm and a fixed volume V = xyz, as in figure 6-3, that moves with the medium.
y
( p+p ) y z 0 z y x x
- [( p+p ) + 0
( p+p ) x y 0 z ] x
Figure 3 Force in the x-direction on a particular fluid particle moving with the medium.
7/48
Fx =
p xyz x
(13)
r r p r p r p F = e x + ey + ez xyz y z x
Putting in the del operator, as
(14)
r r r + ey + ez = e x x y z
(15)
8/48
r F = p V
(16)
r r
For a given fluid particle, the velocity u ( r , t ) is a function of position and time. At time t and position (x, y, z), the velocity is (x, y, z, t). At a later instant t+t, r the position is (x+ x, y+ y, z+ z) and the velocity is u ( x + x, y + y, z + z, t + .t The differential change in position (x, y, z) can be written x =ux t, y = uy t and z = uz t, so that the acceleration can be written
r a = lim
r r u ( x + u x t , y + u y t , z + u z t , t + t ) u ( x, y, z , t ) t
t 0
(17)
The first term is reformulated with the help of a Taylor series, so that equation (17) can be rewritten r r r r r r u u u u u ( x, y , z , t ) + u x t + u y t + u z t + t + K u ( x, y, z , t ) r x y z t a = lim (18) t t 0
9/48
r r r r r u u u u + uy + uz + a = ux x y z t
(19)
r r r u r a= + (u )u t
In one dimension,
(20)
ux ux + ux ax = x t
(21)
For acoustic fields with small disturbances, the second term in equation(20) can be neglected, so that
r r u a= t
(22) 10/48
ux ax = t
Making use of equations (16) and (22), as well as m = (0 + ) V , the equation of motion can be formulated as
(23)
If second order terms can be ignored, then the linear, inviscid equation of motion is
r u p V = ( 0 + ) V t
(24)
In one dimension,
r u 0 + p = 0 t
(25)
ux p + =0 t x
(26)
The equation of motion, gives a relation between pressure and particle velocity in a sound field.
11/48
( p 0 + p ) = ( 0 + ) RT M
where (p0 + p) [Pa] is the total pressure, (0 + ) [kg/m3] is the total density, R = 8.315 [J/(mol K)] is the ideal gas constant, M [kg] is the mass of a mole of gas, and T [K] is the absolute temperature.
(27)
Two idealizations can be considered: an isothermal process, implying such good heat conduction in the medium that the temperature is constant throughout; or, an adiabatic process, in which no heat conduction occurs.
12/48
( p0 + p ) ( 0 + ) =
p
0
(28)
where = cp /cv is the ratio of the specific heat of the gas at constant pressure, to that at constant volume. For the sake of simplicity, pt = (p0 + p) denotes total pressure, and t = (0 + ) denotes total density, in the series expansion. The total pressure can be expanded as
2 p t 2 1 pt pt = p0 + p = p 0 + + + ... t = 2 t2 t t = 0 0
(29)
where the partial derivatives are constants that remain to be determined for adiabatic disturbances about 0, i.e., the density in the undisturbed medium.
13/48
p=
p t t = t 0
(30)
or
p = 0
(31)
where
= 0
p t t = t 0
(32)
14/48
Time derivative of continuity eq. (11), Time derivative of continuity equation (10),
Using
,
2 t
2
+ 0
ux =0 x t
2
(33)
r u 0 + (p ) = 0 t
2
Spatial derivative of eq. of motion (26)
r u + 0 = 0 2 t t
r u 0 + (p ) = 0 t
(34)
ux p =0 + x t x 2
2 2
(36)
(35)
In abbreviated notation,
r u 0 + 2 p = 0 t
Indian Institute of Technology Roorkee
(37)
15/48
p x
2
t
2
=0
(38)
2p
0 2 p =0 2 t2 x
(40)
0 2 p 2 p =0 2 t
(41)
2p x
2
1 2p c
2
=0
(42)
2 p
1 2p c
2
=0
(43)
16/48
c = / 0
(44)
and is the propagation speed of a disturbance in the medium, the speed of sound. According to equation (32)
c = / 0 =
p t t
(45)
t = 0
pt t = ln pt ln p 0 = (ln t ln 0 ) p0 0
(46)
17/48
pt p t = t t
t = 0
(47)
Put into the expression for the speed of sound (45), this yields
c = pt t
i.e.,
(48) (49)
c = p0 0
c = RT M
The temperature dependence of the speed of sound is obtained by putting the ideal gas law (27) into (48)
(50)
If the speed of sound at 0 C (273 K) is denoted c0, then for other temperatures
c = c 0 T 273
Indian Institute of Technology Roorkee
(51)
18/48
T , after
c = T 0
Solutions to the wave equation
General solution for free plane one-dimensional wave propagation
(52)
For plane wave propagation in the x-direction, the wave equation (42) applies. Assume a general solution form of
p ( x, t ) = f (t x c ) + g (t + x c )
Indian Institute of Technology Roorkee
(53)
19/48
p 1 1 = f (t x c ) + g (t + x c ) c c x
2 p x2 = 1 c2 f (t x c ) + 1 c2 g (t + x c )
(54) (55)
p = f (t x c ) + g (t + x c ) t
2p t
2
(56) (57)
= f (t x c ) + g (t + x c )
Putting (55) and (57) into (42) shows that the solution fulfills the wave equation.
20/48
To interpret (53), consider a special point (x1,t1) in the wave, in figure 4. It represents a certain sound pressure p1(x1,t1). To have the same sound pressure at another point (x1 + x), at a later instant in time (t1 + t), then the arguments have to be the same, i.e.,
(t1 x1 c ) = (t1 + t ( x1 + x) c )
which gives the condition
(58)
x = ct
Thus, the solution f (t x / c) implies wave propagation in the positive direction along the x-axis, with speed c.
(59)
21/48
Figure 4 Instantaneous picture of the wave propagation in the positive x-direction at time instants t1 and (t1 + t). The propagation speed of the disturbance is c.
x = ct
(60)
The propagation speed c of a disturbance is called the speed of sound (or sound speed), wave speed or phase velocity.
22/48
p ( x, t ) = p + cos (t x c)
(61)
where p+ is the amplitude, i.e., the highest value of the sound pressure. The argument (t - x/c) = t kx is called the phase and
23/48
(62) (63)
T = 2p , = 2 /T
The second condition gives
(64)
(65)
i.e.,
k = 2 , k = 2
From (62), (64) and (66), one obtains the relation
(66) (67)
c = f
Indian Institute of Technology Roorkee
24/48
f=1/T
p(x1,t)
(68)
$ p+
Period T
Figure 5a The variation of sound pressure with time at a fixed position x1.
25/48
$ p+
Wavelength
Figure 5b The variation of sound pressure with position at a fixed time t1.
complex notation
p( x, t ) = p + e i ( t kx) + p e i ( t+ kx)
(69)
in which bold print means that the variable concerned is complex. The first term on the right-hand side refers to propagation in the positive xdirection, and the second term to propagation in the negative x-direction. For the sake of physical interpretation, the real part of (69) is needed, i.e.,
26/48
+ p e
i ( t + kx )
ux p + =0 t x
ux =
p dt x
(72)
1 ik ik + e i ( t kx) + u x ( x, t ) = p p e i ( t + kx ) 0 i i
Since k/ = 1/c, the particle can be written
(73)
27/48
The two terms refer to wave propagation in the positive and negative xdirections, respectively. The ratio of pressure to particle velocity is called the specific impedance Z,
Z = p ux
and, for the free plane wave case, is therefore
(75)
+ Z 0 = 0c
and
(76)
Z 0 = 0c
(77)
for propagation in the positive and negative directions, respectively. The quantity 0c is called the wave impedance.
28/48
r r W(t ) = F (t ) u (t )
A general expression for sound intensity is therefore that
(78)
r r r r r I (r , t ) = p(r , t ) u (r , t )
For propagation in the x-direction,
(79)
I x ( x, t ) = p( x, t )u x ( x, t )
The time-averaged sound intensity is
(80)
29/48
p( x, t )u x ( x, t )dt
(81)
0 Making use of the expression for pressure, i.e., the real part of (69), and the particle velocity, i.e., the real part of (74), in (81), gives the intensity in the form
2 2 I x = ( p+ p ) 2 0c
(82)
The first term refers to a wave moving in the positive x-direction, and the second term to a wave moving in the negative x-direction. For harmonic waves, the relation between the rms amplitude and the peak value is ~ = p 2 , so that (82) can be written p
~ 2 ~ 2 p+ p Ix = 0c 0c
(83)
30/48
0V0 = t V
E k ( x, t ) =
(85) (86)
For wave propagation in the x-direction, its kinetic energy is, from elementary mechanics, 2
0V0 u x ( x, t )
2
The potential energy of the fluid mass comes from the work that expended to compress it. Consider, now, the fluid volume shown in figure 7. The force applied to the piston is equal to the product of the pistons area S and the pressure p against its inside surface, i.e., Fx = p(x,t)S. When the piston moves a distance dx, the differential amount of work dEp = Fx dx = p(x,t) S dx is performed.
31/48
Figure 7
dE p = p( x, t )dV
(87)
where the minus sign implies that a positive sound pressure p(x,t), which gives a negative volume change (i.e., a volume reduction or compression), corresponds to a positive potential energy. We choose to express dV as a function of the sound pressure. Differentiating (85),
32/48
0V0 dV =
0V0 t
2
d t
d t
(88)
pt t = p0 0
d t
1 dp t d t = t dp t pt pt
(89)
dV =
V dp t pt
(90)
33/48
dV =
V0 dp p0
V0 p dp p0
(91)
dE p =
(92)
The potential energy is, finally, obtained by integrating from 0 to the sound pressure p; thus,
Ep =
According to (49), c = p 0 0 ; thus, for plane wave propagation in the xdirection, a general expression for the energy is
2
V0 p2 2 p 0
(93)
E p ( x, t ) =
V0 20c 2
p 2 ( x, t )
(94)
34/48
r r r (r , t ) = k (r , t ) + p (r , t )
2
(95)
Using expressions for the kinetic energy (86) and the potential energy (94), the energy density for propagation in the x-direction is then
( x, t ) =
0 u x ( x, t )
2
p 2 ( x, t ) 2 0c 2
(96)
According to (76) and (77), u x ( x, t ) = p ( x, t ) ( 0 c ) applies to plane waves. Using this in (96) yields the instantaneous value of the energy density
( x, t ) =
p 2 ( x, t ) 20c
p 2 ( x, t ) 20c
p 2 ( x, t )
(97)
T ~ 2 ( x) 1 p ( x) = ( x, t ) dt = T 0c 2 0
(98) 35/48
~2 p
0c
(99)
A comparison to (83) gives the relation between the sound intensity and the energy density of a plane wave, as
I x = c
General solution for free spherical wave propagation
(100)
The spherical wave is a basic cornerstone in the study of acoustic fields. More complex sound fields can be built up of combinations of spherical waves
36/48
...... . .... . ........................... ... . . . . . . . .... .. ...... ....... .... .............. .. .. . ... .. .. .. .... . . . . . . ... ....... ...... . . . ....... ........ ... . .... ... ....... . .... ... ......... . . . .. .. . . . . . ... .. . . .. . . . . .......... ........... .................................. .............................. ... .. . . . .. .. .. . ........... . . . .. . ... . .. . . . ..... .. .. ... .. ...... ......... .. .. .... .. . . . . ............. .. .. . . .............. ... . .................. .................................................... ............ . ... . .. . . .. . ... .. . .. . . . ............... . ... .............................................. ........................ ..... .............. ... . . ... .... . ..... .. .................. .. . ... ..... ... . . . . . .. . .. . ... . ..... .... ... ............. .................................. . ......... . .. .... ..... .. . . . .. . . .......... .............. ................. . . . .................. ... .... .. . ... . ....... . . .. . . ..... .... . . ... .. . . .. .... .. .. . . ............... ............... ... ........... ... .. ................ .. .... ...... . ........... . ... . . . ... .. . .. ........ ........... . . ............ ..... ... ... ................. ... ................ . .... . .. . ... . ... . . .. .. . . . . ......... .. ... ...... . .. . ......... ... . . . .... ... ...... .. . . . . .. . . .. ....... . ...... . ... ............ ................................................. ............ .... . ...... . .. . . . . . ...... . .. . ..... .. . . . ................ .. ... ...... . .. .. ... . . . . .. ... ............ ... ................. ... .. ... . ....... .... . ... ............... ... .. ..... ..... ... . . . . . . .. .... . ... ......... . ... ... .............. .... . ..... ... ............. . ... .. ........ .. . . .. . . . . .. . . ............. .. .... . . ................................................ ....... . ... ..... . . .... . .. .... .............. . .. . . .. .. . ............ .. ..... ... . .. . ... . ... . . ............... .... ... ..... . . . ... .. . .. . . . . ... . . . . ...... . .. .. . . . . .. .. .......... . ...... .... ......................... . ... .... .... . . . . . .. .. ..... . . ...... ....... ....... .. . ... . .. ......
37/48
z (r, , )
r sin sin
38/48
(101)
For spherical symmetry, the sound pressure has no angular dependence, and equation (101) reduces to
1 2 1 2p r p = 2 2 r r c t2 r
or
(102)
2p
2 p 1 2p + = 2 r r c2 t 2 r
(103)
Unlike plane wave propagation, the sound pressure amplitude decays with increasing radius, since the sound power in the wave is divided over an ever-expanding spherical surface of area 4 r2. In a plane wave, the sound intensity is, according to (82),
39/48
p(r , t ) =
1 1 f (t r c ) + g (t + r c ) r r
(105)
Here, the first term represents an outgoing diverging wave and the second term an incident converging wave. The incoming wave seldom exists in connection with acoustic radiation from machines. The solution is verified by inserting it into the wave equation (103). On a term-by-term basis, we have
2p r
2
1 rc
2
f (t r c ) +
2
2
f (t r c ) +
2
3
f (t r c ) +
r c r 1 2 2 + g (t + r c ) g (t + r c ) + g (t + r c ) rc 2 r 2c r3
(106)
2 p 2 2 2 2 = f (t r c ) f (t r c ) + g (t + r c ) g (t + r c ) 2 3 2 3 r r r c r r c r
(107)
40/48
1 rc
2
f (t r c ) +
1 rc
2
g (t + r c )
(108)
Putting all of these into (103) shows that the assumption does indeed fulfill the wave equation. The equation of motion (25) relates the particle velocity to the sound pressure. In spherical coordinates, it takes the form
r 1 u r r 1 r 0 + e r + e + e p = 0 sin t r r
(109)
ur p + 0 =0 t r
(110)
ur =
p dt r
(111)
41/48
p(r , t ) =
A+ i ( t kr ) A i ( t + kr ) + e e r r
(112)
The first term on the right-hand side refers to an outgoing, diverging, wave, and the second to an incident, converging, wave. The sound pressure amplitude in the outgoing wave is A+/ r, where A+ is a constant. The amplitude is therefore a function of r. From (111), the particle velocity becomes
u r (r , t ) =
A p(r,t ) dt = + 0 r
1 ik i ( t kr ) dt 2 + r e r
(113)
Integrating, and applying the relation (62), k = /c, gives the particle velocity
A+ 1 i ( t kr ) u r (r , t ) = 1 + e 0 cr ik r
Indian Institute of Technology Roorkee
(114)
42/48
By analogy to the definition (75) of the specific impedance of a free plane wave, we define here the complex quantity Z as the ratio of the complex sound pressure to the complex radial particle velocity at a point in a sound field
Z= p ur
(115)
For an outgoing spherical wave, the specific impedance, using equations (112) and (114), becomes
Z= p = 0c ur 1 1 1+ ikr = 0c ikr 1 + ikr
(116)
Multiplying the numerator and denominator by the complex conjugate of the latter, gives
k 2r 2 kr Z = 0c = 0 c +i 1 + k 2r 2 1 + k 2r 2 1 + k 2r 2 k 2 r 2 + ikr
(117)
43/48
Z = R + iX
Some observations that follow from the preceding are:
(118)
(i) Nearfield
In the acoustic near field (kr = 2r/ 1), i.e., when the radius is small in comparison to the wavelength, both the resistance and the reactance approach zero, but the resistance does so more quickly. The reactance therefore dominates, and the impedance approaches
Z(r ) i 0 c kr
(119)
That means that for a given sound pressure, the particle velocity becomes large and its phase shift approaches 90 with respect to the sound pressure. For kr = 1, both the resistance and the reactance are equally large, 0c / 2,
44/48
Z(r ) 0 c
(120)
That means that the phase difference between the sound pressure and the particle velocity approaches zero, as is the case for plane waves. The curvature of the spherical waves in the farfield, with increasing distance to the source, becomes all the less significant and the situation asymptotically approaches that of plane waves.
45/48
Figure 10 Normalized resistance R /0c , and normalized reactance X /0c , for outgoing spherical wave propagation.
46/48
I r (r ) =
(121)
(122)
The time-averaged energy flow through a closed, spherical control surface of radius r is 2 A+ 2 W = I r (r ) 4 r = 2 (123) 0c For a loss-free medium, the sound power is therefore independent of the radius, which is in agreement with the energy principle.
47/48
Thank you