Chapter 3

The Equations of Change

for Isothermal Systems

 In Chapter 2, velocity distributions were determined for

several simple flow systems by the shell momentum
balance method. The resulting velocity distributions were
then used to get other quantities, such as the average
velocity and drag force.
For more complex problems we need a general mass
balance and a general momentum balance that can be
applied to any problem, including problems with
nonrectilinear motion. That is the main point of this
chapter.

 The two equations that we derive are called the equation

of continuity (for the mass balance) and the equation of
motion (for the momentum balance). These equations can
be used as the starting point for studying all problems
involving the isothermal flow of a pure fluid.
Those equations are called as equations of change
because they describe the change of velocity due to the
change of time and position in the fluid system.

3.1. THE EQUATION OF CONTINUITY

(MASS BALANCE)
This equation is developed by writing a mass balance over a
volume element x.y.z , fixed in space, through which a
fluid is flowing (see Fig. 3.1-1):

The rate of mass entering the volume element through the

shaded face at x is (vx)|x y.z, and the rate of mass leaving
through the shaded face at x + x is (vx)|x+x y.z.

 Similar expressions can be written for the other two pairs

of faces. The rate of increase of mass within the volume
element is x.y.z(/t). The mass balance then
becomes

x y z
y z vx x vx x x x z v y y v y y y

x y vz z vz

z z

(3.1-2)

By dividing the entire equation by x.y.z and taking

the limit as x, y, and z go to zero, and then using the
definitions of the partial derivatives, we get

.(3.1-3)

vx
vy
vz
t
x
y
z

This is the equation of continuity, which describes the time

rate of change of the fluid density at a fixed point in space.
This equation can be written more concisely by using
vector notation as follows:.

.v = "divergence of v"

 The vector v is the mass flux, and its divergence has a

simple meaning: it is the net rate of mass efflux per unit
volume. A very important special form of the equation of
continuity is that for a fluid of constant density, for which
Eq. 3.1-4 assumes the particularly simple form
(incompressible fluid)
.
Of course, no fluid is truly incompressible, but frequently
in engineering and biological applications, the assumption
of constant density results in considerable simplification
and very little error.

Example 3.1-1. Normal Stresses at Solid

Surfaces for Incompressible Newtonian
Fluids

Show that for any kind of flow pattern, the normal stresses are
zero at fluid-solid boundaries, for Newtonian fluids with
constant density. This is an important result that we shall use
often.
SOLUTION
We visualize the flow of a fluid near some solid surface, which
may or may not be flat. The flow may be quite general, with all
three velocity components being functions of all three
coordinates and time.
At some point P on the surface we erect a Cartesian coordinate
system with the origin at P.
We now ask what the normal stress zz is at P.
According to Table B.l or Eq. 1.2-6, zz = -2(dvz/dz), because
.v 0 for incompressible fluids. Then at point P on the surface
of the solid

z
y
According to mass balance
for incompressible flow
0
.

First we replaced the derivative dvz/dz by using Eq. 3.1-3 with

constant. However, on the solid surface at z = 0, the velocity
vx is zero at any position of x by the no-slip condition (see
2.1), and therefore the derivative dvx/dx on the surface = 0.
The same is true of dvy/dy on the surface. Therefore zz is zero.
It is also true that xx and yy are zero at the surface because of
the vanishing of the derivatives at z = 0. (Note: The vanishing
of the normal stresses on solid surfaces does not apply to
polymeric fluids, which are viscoelastic).
For compressible fluids, the normal stresses at solid surfaces
are zero if the density is not changing with time (see Problem
3C.2.)

3.2. THE EQUATION OF MOTION

(MOMENTUM BALANCE)
To get the equation of motion we write a momentum balance
over the volume element x.y.z in Fig. 3.2-1 of the form.

Note that Eq. 3.2-1 is an extension of Eq. 2.1-1 to unsteadystate problems.

The fluid is allowed to move through all six faces of the volume
element.
Remember that Eq. 3.2-1 is a vector equation with components
in each of the 3 coordinate directions x, y or z containing shear
stresses, normal stresses and convective momentum fluxes.

 We develop the x-component of each term in Eq. 3.2-1.

The y- and z-components may be treated analogously.
First, we consider the rates of flow of the x-component
of momentum into and out of the volume element shown
in Fig. 3.2-1. (second subscript x: directions of shear and
normal stresses for molecular transport and direction of
convective momentum flux for convective transport)
Momentum enters and leaves x.y.z by two
mechanisms: molecular transport (see 1.2) and
convective transport (see 1.7).

Second subscripts of all

components are the same,
i.e. x (direction of effects)
First subscripts x, y, and z (cause): directions of momentum
transfers due to the change of velocity for molecular transport or
due to convection represented by velocity in x, y and z directions
for convective transport. Cause can be from all directions.
Second subscripts x (effect): directions of shear stress or normal
stress or direction of convective momentum flux represented by
direction of mass flux (in x direction). Effect is only in one
direction in a momentum balance

 The rate at which the x-component of momentum enters

across the shaded face at x by all mechanisms-both
convective and molecular-is xx|x yz and the rate at which
it leaves the shaded face at x + x is xx|x+x yz.
The rates at which x-momentum enters and leaves through
the faces at y and y + y are yx|y zx and yx|y+y zx
respectively.
Similarly, the rates at which x-momentum enters and leaves
through the faces at z and z + z are zx|z xy and zx|z+z
xy

 When these contributions are added we get for the net

rate of addition of x-momentum across all three pairs of
faces.
.
Next there is the external force (typically the gravitational
force) acting on the fluid in the volume element. The xcomponent of this force is
Equations 3.2-2 and 3.2-3 give the x-components of the
three terms on the right side of Eq. 3.2-1.

 The sum of these terms must then be equated to the rate

of increase of x-momentum within the volume element:
x.y.z (vx)/t. When this is done, we have the xcomponent of the momentum balance. When this
equation is divided by x.y.z and the limit is taken as
x, y and z zero, the following equation results:

x-direction

 Here we have made use of the definitions of the partial

derivatives. Similar equations can be developed for the yand z-components of the momentum balance:
.

By using vector-tensor notation, these three equations can

be written as follows:
.
This is a vector equation (vector dot tensor = vector)

 That is, by letting i be successively x, y, and z, Eqs. 3.2-4,5,

and 6 can be reproduced. The quantities vi are the Cartesian
components of the vector v, which is the momentum per unit
volume at a point in the fluid.
Similarly, the quantities gi are the components of the vector
g, which is the external force per unit volume. The term [. ]i is the ith component of the vector -[. ].
When the magnitude of ith component of Eq. 3.2-7 is
multiplied by the unit vector in the ith direction and the three
components are added together vectorially, we get

which is the differential statement of the law of conservation

of momentum. It is the translation of Eq. 3.2-1 into
mathematical symbols.

 In Eq. 1.7-1 it was shown that the combined momentum

flux tensor is the sum of the convective momentum flux
tensor vv and the molecular momentum flux tensor ,
and that the latter can be written as p + . When we
insert = p + vv + into Eq. 3.2-8, we get the
following equation of motion

Vector equation combining 3

directions of 2nd subscripts

 In this equation p is a vector (=vector times scalar)

called the "gradient of (the scalar) p" sometimes written
as "grad p ". The symbol [. ] is a vector (=vector dot
tensor) called the "divergence of (the tensor) " and
[.vv] is a vector (=vector dot tensor) called the
"divergence of vv.
In the next two sections we give some formal results that
are based on the equation of motion. The equations of
change for mechanical energy and angular momentum are
not used for problem solving in this chapter, but will be
referred to in Chapter 7 (this chapter is excluded from
the lecture material!).

3.5. THE EQUATIONS OF CHANGE IN

TERMS OF THE SUBSTANTIAL
DERIVATIVE

The Partial Time Derivative /t (derivative against one variable)

Suppose we stand on a bridge and observe the concentration of
fish just below us as a function of time. We can then record the
time rate of change of the fish concentration at a fixed location.
The result is (c/t)|x,y,z the partial derivative of c with respect to
t, at constant x, y, and z.
The Total Time Derivative d/dt (derivative against all variables)
Now suppose that we jump into a motor boat and speed around
on the river, sometimes going upstream, sometimes downstream,
and sometimes across the current as we wish.

 All the time we are observing fish concentration. At any

instant, the time rate of change of the observed fish
concentration is
dx
dy c

c
dz c
.dc c
(3.5-1)

dt

t
x,y ,z

dt
x

y ,x,t

dt y x,z ,t dt z

x,y ,t

in which dx/dt, dy/dt, and dz/dt are the components of the

velocity of the boat.
The Substantial Time Derivative D/Dt
Next we climb into a canoe and we just float along with
the current to observe the fish concentration.

 In this situation the velocity of the observer = the velocity

v of the stream, which has components vx, vy, and vz.
If at any instant we report the time rate of change of fish
concentration, we are then giving

The special operator D/Dt = /t + v. is called the

substantial derivative (meaning that the time rate of
change is reported as one moves with the "substance").
The terms material derivative, hydrodynamic derivative,
and derivative following the motion are also used.

 Now we need to know how to convert equations expressed in

terms of /t into equations written with D/Dt. For any scalar
function f(x,y,z,t) we can do the following manipulations:

=0

The quantity in the second parentheses

thebalance
second line = 0
According to in
mass
according to the equation of continuity.

D/Dt = /t + v.

Consequently Eq. 3.5-3 can be written in vector form as

Similarly, for any vector function f(x,y,z,t),
temporal change
Spatial/positional change

These equations can be used to rewrite the equations of

change given in 3.1 to 3.4 in terms of the substantial
derivative as shown in Table 3.5-1.

: in Chapter 7, excluded

in our lecture

 Equation A in Table 3.5-1 tells how the density is decreasing

or increasing as one moves along with the fluid, because of
the compression [(.v) < 0] or expansion of the fluid [(.v)
> 0].
Equation B can be interpreted as (mass) x (acceleration) =
the sum of the pressure forces, viscous forces, and the
external force. In other words, Eq. 3.2-9 is equivalent to
Newton's second law of motion
(density x acceleration = summation of all
forces /volume)

Three most common simplifications of the equation of motion:

For constant and , insertion of the Newtonian expression
for from Eq. 1.2-7 into the equation of motion leads to the
very famous Navier-Stokes equation, first developed from
molecular arguments by Navier, a French engineer, and from
continuum arguments by Stokes, an English mathematician:
.
When the acceleration terms in Navier-Stokes equation are
neglected-that is, when (Dv/Dt) = 0 -we get
which is called the Stokes flow equation. It is sometimes
called the creeping flow equation, because the term (v.v]
0 when the flow is extremely slow and can be approached as
steady flow.

 When

viscous forces in Navier-Stokes equation are

neglected - that is, . = 2v = 0 - the equation of
motion becomes (normal and shear stresses occur due
to viscosity)
which is known as the Euler equation for "inviscid"
fluid in unsteady flow. Of course, there are no truly
"inviscid" fluids, but there are many flows in which
the viscous forces are relatively unimportant (far from
solid surfaces or very high velocity). Examples are
the flow around airplane wings (except near the solid
boundary), flow of rivers around the upstream surfaces
of bridge supports, some problems in compressible gas
dynamics, and flow of ocean current.

Example 3.5-1. The Bernoulli Equation

for the Steady Flow of Inviscid Fluids
The Bernoulli equation for steady flow of inviscid,
incompressible fluids (conditions in Stokes flow and
Euler eqs) is one of the most famous equations in
classical fluid dynamics. Show how it is obtained from
the Euler equation of motion.
SOLUTION
Inviscid Fluids omit the time-derivative term in Eq. 3.59, and then use the vector identity [.vvl = [v.vl =
(v.v) - [v x [ x v]] (Eq. A.4-23) to rewrite the Navier
Stokes equation as

 Next we divide Eq. 3.5-10 by and then form the dot product
with the unit vector s = v/|v| in the flow direction. When the
fluid is inviscid, then there is no vorticity ( x v = 0) and
consequently v x ( x v) = 0, and (s.) can be replaced by d/ds,
where s is the distance along a streamline. Thus we get

When this is integrated along a streamline from point 1 to point

2, we get

which is called the Bernoulli equation. It relates the velocity,

pressure, and elevation of two points along a streamline in a
fluid in steady-state flow of inviscid fluid.

3.6. USE OF THE EQUATIONS OF CHANGE

TO SOLVE FLOW PROBLEMS
To describe the flow of a Newtonian fluid at constant
temperature, we need in general
The equation of continuity Eq. 3.1-4
The equation of motion
Eq. 3.2-9
The components of Eq. 1.2-6
The equation of state = (p)
The equations for the viscosities = (p, T)
These equations, along with the necessary boundary (related to
positions) and initial (related to time) conditions, determine
completely the pressure, density, and velocity distributions in the
fluid.
They are seldom used in their complete form to solve fluid
dynamics problems. Usually restricted forms are used for
convenience, as in this chapter.

 If it is appropriate to assume constant density and

viscosity, then we use
The equation of continuity Eq. 3.1-4 and Table B.4
The Navier-Stokes equation Eq. 3.5-6 and Tables B.5,
6, 7 along with initial and boundary conditions.
From these one determines the pressure and velocity
distributions.

Example 3.6-1. Steady Flow in a Long

Circular Tube
Rework the tube-flow problem of Example 2.3-1 using the
equations of continuity and motion. This illustrates the use
of the tabulated equations for constant viscosity and
density in cylindrical coordinates, given in Appendix B.5.
SOLUTION
We postulate that v = zvz(r, z). This postulate implies that
there is no radial flow (vr = 0) and no tangential flow (v =
0), and that vz f ().
We assume that there is no change of velocity profile in z
direction.
Consequently, we can discard many terms from the
tabulated equations of change, leaving

Datum
plane

 The postulate and Eq. 3.6-1 indicates that vz depends only

on r; hence the partial derivatives in the second term on
the right side of Eq. 3.6-4 can be replaced by ordinary
derivatives.
By using the modified pressure P = p - gh (where h is
the height below some arbitrary datum plane and g is a
constant), we avoid the necessity of calculating the
components of g in r and coordinates, and we obtain a
solution valid for any orientation of the axis of the tube.

 Equations 3.6-2 and 3.6-3 show that P is a function of z

alone, and the partial derivative in the first term of Eq.
3.6-4 may be replaced by an ordinary derivative.
For constant change of P against z, by introducing a
constant C0, Eq. 3.6-4 reduces to

 The P equation can be integrated at once. The vz-equation

can be integrated one operation after another on the left
side (do not "work out" the compound derivative there).
This gives
.

The 4 constants of integration can be found from the

boundary conditions:

ln 0 = indefinite, so C2 must not be existence or mathematically

C2= 0 to obtain definite vz.. The resulting solutions are:
.

As pointed out in Example 2.3-1, Eq. 3.6-13 is valid only in the

laminar-flow regime, and at locations not too near the tube entrance
and exit. For Re > about 2100, a turbulent-flow regime exists
downstream of the entrance region, and Eq. 3.6-13 is no longer
valid.

Example 3.6-2. Falling Film

Set up the problem in Example 2.2-2 by using the
equations of Appendix B.5. This illustrates the use of the
equation of motion in terms of .
SOLUTION
As in Example 2.2-2 we postulate a steady-state flow
with constant density.
We postulate, as before, that the x- and y-components of
the velocity are zero (vx and vy = 0) and that vz = vz(x).
With these postulates, the equation of continuity is zero.
According to Table B.l, the only nonzero components of
are xz = zx = -(dvz/dx). The components of the equation
of motion in terms of are, from Table B.5,

Integration of Eq. 3.6-14 gives

.
in which f(y, z) is an arbitrary function. Equation 3.6-15
shows that f cannot be a function of y.

 We next recognize that the pressure in the gas phase is very

nearly constant at the prevailing atmospheric pressure patm.
Therefore, at the gas-liquid interface x = 0, the pressure is
also constant at the value patm. Consequently, f can be set
equal to patm, and we obtain finally from 3.6-14.
.
(p is function of x only). Equation 3.5-16 then becomes
.
which is the same as Eq. 2.2-10. The remainder of the
solution is the same as in 2.2.

Example 3.6-3. Operation of a Couette

Viscometer
The viscosity may also be determined by measuring the
torque required to turn a solid object in contact with a
fluid. The forerunner of all rotational viscometers is the
Couette instrument, which is sketched in Fig. 3.6-1.
Determine velocity distribution and shear stress for the
laminar, tangential flow of an incompressible fluid
between 2 co-axial vertical cylinders. Outer cylinder
rotates with angular velocity o (see Figure 3.6-1). Endeffects is negligible.

Solution
In steady-state laminar flow, fluid moves in circular
direction with velocity components vr = 0 and vz = 0.
There is no pressure gradient in direction (p = p(r,z)). It
is expected that p depends on z due to gravity and on r
due to centrifugal force.
For these postulates all the terms in the equation of
continuity are zero, and the components of the equation
of motion simplify to

p
0
gz
z

The first equation tells how the centrifugal force affects the
pressure.
The second equation gives the velocity distribution.
The third equation gives the effect of gravity on the
pressure (the hydrostatic effect)
For the problem at hand we need only the -component of
the equation of motion for velocity distribution

 Integration of Eq. 3.6-21 results in

.

The boundary conditions are that the fluid does not slip at
the two cylindrical surfaces:
.

 These boundary conditions can be used to get the

constants of integration, which are then inserted in Eq.
3.6-26. This gives
.

From the velocity distribution we can find the momentum

flux by using Table B.1:

 The torque acting on the inner cylinder is then given by

the product of the inward momentum flux (-r), the
surface of the cylinder, and the lever arm, as follows:
.

Therefore, measurement of the angular velocity of the cup

makes it possible to determine the viscosity. The same
kind of analysis is available for other rotational
viscometers.

 For any viscometer it is essential to know when

turbulence will occur. The critical Reynolds number
(oR2/), above which the system becomes turbulent, is
shown in Fig. 3.6-2 as a function of the radius ratio .

 One might ask what happens if we hold the outer cylinder

fixed and cause the inner cylinder to rotate with an angular
velocity i (the subscript "i" stands for inner). Then the
velocity distribution is

This is obtained by making the same postulates (see before

Eq. 3.6-20) and solving the same differential equation (Eq.
3.6-21), but with a different set of boundary conditions.