Fluids Heat and Mass Transfer
Fluids Heat and Mass Transfer
Brian D. Storey
Olin College
All
2015 Brian D. Storey
rights reserved
Contents
Introduction
1.1
Scope of the text
1.2
What is a fluid?
1.3
What is a continuum?
1.4
Prerequisites
page 1
3
4
6
7
Dimensional analysis
2.1
Units
2.2
Buckingham Pi Theorem
2.3
Making equations dimensionless
2.4
Summary
9
9
10
22
26
27
27
32
39
43
51
52
59
64
65
69
73
74
iv
Contents
5.2
5.3
5.4
5.5
5.6
76
79
83
84
88
91
92
92
96
103
The
7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
105
106
108
111
112
113
119
120
122
125
126
131
133
134
135
137
140
143
146
149
151
155
157
159
160
163
167
Contents
9.4
9.5
Irrotational flow
Lift on an airfoil
174
175
10
High
10.1
10.2
10.3
10.4
10.5
10.6
10.7
10.8
179
180
182
187
189
190
192
193
195
11
201
202
204
214
12
Thermodynamics
12.1 Equations of state
12.2 Energy
12.3 The First Law of Thermodynamics
12.4 Specific heats
217
218
219
221
223
13
227
228
229
233
240
14
243
244
250
254
258
259
261
15
263
263
267
270
vi
Contents
15.4
Summary
272
273
275
277
280
1
Introduction
The text lays the foundations for the study of fluid dynamics and heat
transfer. The first question a student might ask is why should we be
interested in fluid dynamics and heat transfer? Let me address the
question through some examples of applications where these topics are
relevant.
Biology: Our bodies are mostly water. Fluids are essential to life
as we know it. The understanding of the transport of oxygen and nutrients throughout the body by fluids (at the level of lungs, heart, and
vessels as well a the cellular one) requires a basic understanding of the
topics herein. Locomotion of many species from fish to birds to small
microorganisms is determined by fluid flow.
Geophysical: The atmosphere and oceans are fluids. The currents,
the winds, the gulf stream, and the jet stream carry heat and matter
around the planet. Our basic understanding of the weather, the climate,
and the environment can only begin with a basic understanding of fluid
dynamics. The Earths magnetic field (and its occasional flipping of the
poles) is generated by convection of the mantle deep within the Earth.
Energy: Most of our current energy needs are met by the combustion
of fuel; the basic processes of combustion depend upon the details of
the fluid mixing of fuel and air and and the resulting heat release.
Advances in understanding of the details of these processes has led to a
number of developments over the years which made combustion cleaner
and significantly reduced the number of pollutants emitted from cars.
Many of the renewable sources of energy such as wind, geothermal,
ocean waves and tidal seems obvious to state that fluid flow plays a
Introduction
Introduction
is never quite the way you want it. However, the excellent books that
I use the most and those who influenced my view are,
Fluid Mechanics by Kundu and Cohen (2004).
Fundamentals of Heat and Mass Transfer by Incroprera and DeWitt
(2001).
Transport Phenomena by Bird et al. (1960).
Div grad curl and all that by Schey (1997).
Incompressible flow by Panton (1996).
does not describe Silly Putty. Later we will define fluids based on their
constitutive laws. For now just go with your intuition - if it flows, its a
fluid and not get caught up in a precise definition just yet.
There are two common fluid properties that we will refer to throughout this book, density and viscosity.
Density which typically has the symbol, , is nothing more than
the mass per unit volume. Everything we discuss in the development
of the theory is given on a per unit volume basis. To convert from per
unit volume to total quantities we will always need to integrate over
the volume of interest. For example, to get the total mass of a region
we simply integrate the density over that volume.
Z
Mass =
dV,
Torque
U
.
d
We would find for all things equal, syrup is harder to spin than water
because the syrup has a higher viscosity. Not all fluids would follow the
law above. Only simple fluids such as water, air, syrup, and oil. Fluids
that follow a law such as that above are called Newtonian, a definition
we will be precise about later.
Introduction
Height (z)
Density
1.4 Prerequisites
point we keep shrinking the measuring volume and we keep getting the
same answer. This constant value would be what we would consider
the local, continuum value of the density at that point in space. If we
continued to shrink the measuring volume we run into trouble with the
fact that air is comprised of many molecules. Eventually the measuring volume would be so small that the mass we measure would depend
upon how many molecules we managed to trap. If there are only 10
molecules on average in our volume, then the answer starts to fluctuate
depending on if we randomly had 8, 9, 10, 11, or 12 molecules in the
volume at the time of measurement. The continuum limit is thus the
intermediate asymptotic, where we take the limit such that we can talk
about local quantities (i.e. density, temperature, pressure, velocity) at
a point in space, but we dont take the limit too small that the molecular nature of matter comes into play. As long as the geometry of the
problem is large compared to the molecular scale, then the continuum
assumption is good.
1.4 Prerequisites
I will assume throughout that the reader of the notes is familiar with
Newtons Laws, Calculus (mainly integration and differentiation), and
basic vector calculus. If you know F = ma, can compute some simple
integrals, and have heard of the divergence, gradient, and curl, you
are probably good. If it has been some time since you took vector
calculus, I would recommend the book Div, Grad, Curl and All That
as a complement to these notes. A few common things that we will
make use of are provided below. If all of this seems familiar to you, you
should have no problem. We will review these concepts in context in
later chapters.
We will make use of the Fundamental Theorem of Calculus which
states
Z b
df
dx = f (b) f (a).
a dx
We will use vector calculus notation such as the gradient of a scalar
Introduction
F
F
F
i+
j+
k,
x
y
z
where i, j, k are the unit vectors in the x, y, z direction. We will use the
convention that bold face lower case letters are vectors. The divergence
of a vector field f (x, y, z) is a scalar given as
f =
fx
fy
fz
+
+
.
x
y
z
The primary vector calculus theorem we will use is the divergence theorem, which states,
Z
Z
f dV = f ndS.
This theorem equates the flux of a vector field through a closed surface
to the volume integral of the divergence of the vector field. If the net
flow through a surface is zero, then the divergence is zero. Above, we
use the notation that dS is a unit of surface and dV is a unit of volume.
In some books you will find people use a more explicit notation for these
integrals. For example, you might find a notation such as,
ZZZ
ZZ
f dV = f ndS.
This notation is explicit because it uses two integrals to highlight that
we are integrating over an area and three integral signs where we are
integrating over a volume. The circle in the surface integral is to remind us that we are integrating over a closed area which bounds the
volume. We will follow the simpler notation and will be consistent with
dS meaning a surface integral and dV being the volume integral. The
reason for following the simpler notation is that as our equations get
longer, the more explicit notation, in my opinion, makes the equations
look harder and scarier. In class we will always be explicit and discuss
what we are integrating.
2
Dimensional analysis
The result of any physical equation must remain unchanged regardless of the system of units that we select for the problem. While this
seems like a obvious statement the implications are quite interesting.
The simple fact that units are man-made and thus arbitrary can lead
to an extraordinary simplification in many problems. The number of
free parameters in any physical problem can be reduced by taking the
units out of the problem. This concept is not limited to the topics in
this book, however it seems that the field of fluid dynamics and heat
transfer have adopted this viewpoint to the extent that the discussion
of dimensional analysis is in every book. As we will soon show, the
reasons for adopting a dimensionless world-view are quite compelling.
Books of experimental data can be reduced to a single chart.
However, the dimensionless world-view can be overdone and only
working in a dimensionless world can obscure some intuition about the
physics of a particular problem. We will try and balance the dimensional
world-view and the dimensionless one throughout this book.
2.1 Units
We often refer to fundamental and derived units. Fundamental units in
this course will be mass (M ), length (L), time (T ), and temperature
(). Derived units are things like velocity (L/T ), force (M L/T 2 ), or
density (M/L3 ). The distinction above is arbitrary. Of course we could
say that force is fundamental and mass is derived. If you want to do
10
Dimensional analysis
this, this is your right. I find it easiest to stick with the distinction
above.
We could relate temperature and energy and thus remove temperature as a fundamental unit. In this course, this path would not lead to
simplification. It would require us to bring in a new parameter, Boltzmanns constant. Boltzmanns constant bridges the statistical definition
of kinetic energy of molecules to the macroscopic temperature. If we are
not dealing with statistical mechanics, then introducing Boltzmanns
constant just leads to confusion. In this course, everything is macroscopic and thus treating the temperature as its own fundamental unit
is always going to be the way to go.
2.2.1 Pendulum
The first example on dimensional analysis in every book uses is the
pendulum. I wont be original. Imagine a point mass (m) hanging on a
massless string of length (l) in a gravitational field (g); shown schematically in Figure 2.1. There is no friction. All this is an idealization, but
it is one that we can easily realize in practice to a good degree of approximation. The question is what is the period of oscillation, t, for a
given starting angle and how does it depend upon the parameters in
the problem? There are five independent parameters (m, l, g, and
t). These parameters are made up of 3 independent dimensions (M , L,
and T ). Applying the theorem tells us that N = 5, R = 3 and thus
there are only 2 independent parameters.
There are a number different ways to formally approach the removal
of units from the problem to decide what the dimensionless parameters
are. Once you are comfortable, you can often just remove the units by
inspection. I like the table method as it is a provides a good framework
11
m
Figure 2.1 Schematic of a simple pendulum.
for formally approaching the problem. We make a table with all the
parameters and their units in brackets. We put the primary thing we
want to know, in this case the period, in the first column. For this
problem, our table starts as,
t [T ]
m [M ]
l [L]
g [ TL2 ]
Note that the starting angle has no units. Angles are measured in
radians which is defined as the arc length divided by the radius, thus
the radian is already dimensionless.
Now start adding lines to the table, and in each new line remove
one of the independent dimensions. Lets start by removing mass from
the problem. Since the dimension M only exists in the mass, there
is nothing to do but remove it from the problem. Mass cannot be a
parameter to the problem as there is no way to cancel that fundamental
unit. With each new line to the table it is useful to add yourself a little
note to the right.
t [T ]
t [T ]
m [M ]
l [L]
l [L]
g [ TL2 ]
g [ TL2 ]
remove M
Now lets remove length from the problem by defining a new parameter g/l that does not have length in it. We expand our table by one
line so it reads,
12
Dimensional analysis
t [T ]
t [T ]
t [T ]
m [M ]
l [L]
l [L]
g [ TL2 ]
g [ TL2 ]
g
1
l [T2 ]
remove M
remove L
Finally, we can remove time from the problem by adding another line
to our table as follows,
t [T ]
t [T ]
t [T ]
p
t gl
m [M ]
l [L]
l [L]
g [ TL2 ]
g [ TL2 ]
g
1
l [T2 ]
remove M
remove L
remove T
p
The final result is that there are two parameters, t g/l and . We
can then state that the answer has the following functional form,
r
g
t
= f ()
l
At this point we dont know what the function is, but it says the period
of any pendulum of any length in any gravitational field only depends
p
upon its starting angle. We only need to measure time in units of g/l,
and all pendulums are the same. If we always started the pendulum
from the same angle, the function on the right hand side would then
just be a constant. The result tells if the length of the string is increased
by
p a factor of 4, then the period increases by a factor of 2 to keep
t g/l=constant. You can easily confirm this result experimentally.
13
A2
b
A1
2.2.3 Caveats
There is a nice paper written by Lord Rayleigh on the Principle of
Similitude (Rayleigh (1915)); note that similitude is another name for
dimensional analysis. He starts the paper;
I have often been impressed by the scanty attention paid even by original
workers in physics to the great principle of similitude. It happens infrequently
that results in the form of laws are put forward as novelties on the basis
of elaborate experiments, which might have been predicted a priori after a
minutes consideration.
14
Dimensional analysis
arguing that temperature is not an independent dimension since temperature is related to molecular kinetic energy. Rayleighs reply states
that it would be a paradox if further knowledge of the nature of heat
afforded us by molecular theory put us in a worse position than before.
Pointing to this full exchange, Bridgman says:
The reply of Lord Rayleigh is, I think, likely to leave us a little cold. Of
course we do not question the ability of Lord Rayleigh to obtain the correct
result by the use of dimensional analysis, but must we have the experience
and physical intuition of Lord Rayleigh to obtain the correct result also?
15
harder for the sphere to move through the more viscous fluid all other
things being equal. It also seems reasonable to suspect that the fluids
density, , might matter as well. These are the only two fluid properties
that really matter in this problem. We are starting to already see the
comments of Bridgman come in. We have to know something about the
physics of the problem to know that these are the only fluid properties
that matter. Why doesnt the speed of sound in the fluid matter? Or its
surface tension? In later chapters as we develop the full theory of fluid
flow, we will see that these are the only two properties that emerge for
this problem. For now you will have to take this fact as a given.
Applying the Pi Theorem helps our problem immensely. There are
5 parameters, F , D, U , , and . There are the 3 dimensions of M ,
L, and T . The Pi Theorem says there are only two parameters that
matter. We can proceed by constructing the table and writing all the
parameters and their units.
L
F [M
T2 ]
D [L]
U [ TL ]
[ LM3 ]
M
[ LT
]
Immediately we see some arbitrary choices to make. If we try to eliminate mass, we can divide the drag force by the density or the viscosity.
According to the Pi Theorem, any choice is just as good. Which choice
is better is a matter of experience and even an experienced person
might try several combinations before deciding which choice they prefer. We will proceed to apply the theorem making arbitrary choices
about how to remove the dimensions. (Of course the choices are not
arbitrary, they are ones that I selected.) Here is one possibility;
L
F [M
T2 ]
D [L]
U [ TL ]
F L4
[T2 ]
F
2
U 2 [L ]
F
U 2 D 2
D [L]
U [ TL ]
D [L]
U 1
[L]
U D
[ LM3 ]
M
[ LT
]
[ LT ]
Remove M
Remove T
Remove L
The result says there are two parameters. One parameter is ubiquitous in fluid dynamics and we will discuss it extensively in this course.
It is called the Reynolds number, Re = UD . If you have the chance
to make a grouping that looks like (densityvelocitylength)/viscosity,
you should do so. The other parameter is the dimensionless drag force,
known as the drag coefficient. Typically, the drag coefficient is defined
16
Dimensional analysis
a)
Figure 2.3 Drag coefficient as a function of Reynolds number for
flow around a sphere is shown as the solid curve. The dashed lines
are for the fit Cd = 24/Re and a constant of Cd = 0.4.
17
the terms D, U , and show up). These are flows that are dominated
by viscosity. The fact that the fluid has mass or inertia turns out to
be irrelevant. It takes much more effort to drag the fluid along at low
Reynolds number than it takes to push fluid out of the way. If fluid
inertia is irrelevant to the physics, then we can remove density from
our list of parameters and rework the dimensional analysis.
L
F [M
T2 ]
2
F L
[ T ]
F
U [L]
F
U D
D [L]
U [ TL ]
D [L]
U [ TL ]
M
[ LT
]
D [L]
Remove M
Remove T
Remove L
F
=
= Constant
2
2
U D U D
U D
U D
which states that the drag coefficient will be equal to a constant divided
24
by the Reynolds number. The data supports the function of Cd = Re
.
The factor of 24 comes from a full fluid calculation of the laws of fluid
flow and is also validated by experiments. Constants simply cannot
come from dimensional analysis.
The result I just derived probably seems suspicious to you. If it
doesnt, it should. Why am I allowed to throw out density from my list
of parameters? My derivation again relies on this comment by Bridgman. You have to understand the physics of the problem to successfully
apply dimensional analysis. Since this is just the start of your study,
you should have no reason to accept that density is unimportant at low
Reynolds number. Remember that I have studied fluid mechanics since
graduate school, have read about this problem in numerous books, and
have run the experiment and plotted data myself. This problem is easy
for me at this point; however, faced with a new unfamiliar problem
I have to work hard to get the right answer. With a new problem, I
18
Dimensional analysis
D [L]
U [ TL ]
F L4
[T2 ]
F
2
U 2 [L ]
F
U 2 D 2
D [L]
U [ TL ]
[ LM3 ]
D [L]
Remove M
Remove T
Remove L
Our result now says there is one dimensionless parameter and it must
be a constant,
F
= Constant.
U 2 D2
We see the above equation is the definition of drag coefficient. The
analysis says that drag coefficient is constant if the viscosity of the
fluid is irrelevant. Again, I should not have convinced you that it is
acceptable to ignore the fluids viscosity, however the data between a
Reynolds number of 103 and 105 supports this view. If we look at the
data we see a dip in the drag coefficient that we will later explain - but
it turns out the dip occurs because the fluid has viscosity.
For now, I want to emphasize two points with this example. First,
dimensional analysis lets us represent data that depends on a lot of
parameters in a simpler way. In dimensional terms the drag force is
given as F (U, D, , ). Parameterizing this function experimentally for
all four variables requires a lot of experiments. In dimensionless terms,
19
20
Dimensional analysis
The Pi Theorem says there are only three parameters that matter. This
helps quite a bit. We can proceed by constructing the table and writing
all the parameters and their units.
U [ TL ]
l [L]
D [L]
M
P [ LT
2]
[ LM3 ]
M
[ LT
]
l [L]
D [L]
M
P [ LT
2]
[ LM3 ]
U [ TL ]
l [L]
D [L]
l [L]
D [L]
P
[ T1 ]
P
[ L12 ]
2
[ L1 ]
U D
l
D
P D 2
2
[ LT2 ]
M
[ LT
]
Remove M
Remove T
Remove L
This result is one of many possibilities. My result says there are three
parameters. The first is again the Reynolds number - it will show up
in nearly every fluid mechanics problem. The second parameter is a
geometric parameter, the length to diameter ratio. The final parameter
is the dimensionless pressure drop. Now, conducting experiments for
three parameters is a lot easier than 6. We could fit all the results on a
single graph plotting the dimensionless pressure drop versus Reynolds
number for a family of curves for different l/D. We need a sheet of
paper rather than a book to display all results.
We can actually do even better, but the comment by Bridgman again
applies. We need a little physical reasoning and some knowledge of the
problem. This step might seem a leap now, however, as we learn more
throughout this course we will make this step more easily. The pressure
drop and pipe length are not really independent. It is the parameter
P/l that matters most, the pressure gradient. The assumption we
make here is that when the pipe is long that any section of pipe is
basically the same. If we look at the flow here or there in the pipe, it
is just a length of pipe and the flow doesnt change. If the character
of the flow doesnt change as a function of length, then doubling the
length should not change anything if we double the pressure drop. The
pressure drop per unit length is the parameter that really matters.
We could confirm this assumption experimentally by measuring the
pressure as a function of length and we would find that it is essentially
21
D [L]
U [ TL ]
D [L]
D [L]
[ L1 ]
U D
P
M
l [ L2 T 2 ]
P
1
l [ LT ]
P
1
l2 [ L3 ]
P D
l2
[ LM3 ]
[ LT2 ]
M
[ LT
]
Remove M
Remove T
Remove L
22
a)
Dimensional analysis
b)
Figure 2.4 Dimensional (a) and dimensionless (b) plots of flow as a
function of pressure for the pipe flow problem. This data was taken
in a simple apparatus and consisted of 21 separate configurations
varying fluids (water and oil), pipe diameter, and pipe length. The
experimental dimensional data in (a) is collapsed to a single master
curve in (b).
23
the second so we can change the unit to whatever we feel like. Picking
the right scale for t0 can be a bit of an art form that comes with
experience. However, the equations often guide us to a good answer
more directly than applying the Pi Theorem. Mathematically, t0 could
be anything. Often, the problem might suggest a convenient time scale
for us such such that t spans from 0 to 1.
Lets do a few examples to make this clear.
2.3.1 Pendulum
Newtons law for a simple pendulum with no friction is,
ml
d2
= mg sin(),
dt2
l
d(t t0)
Note how we made the derivative dimensionless. This step seems strange
to most students the first time they see it but we have to remember
that a second derivative in time has units of 1/seconds2 . Thus, the dt2
in the derivative is replaced by d(t t0)2 . All we did was the substitution
t = t t0 . Since t0 is a constant it can be moved in or out of the derivative
operator. This step is no different than changing the time derivative to
be measured in minutes instead of seconds.
Our equation can be rewritten as
d2
gt2
= 0 sin().
2
l
dt
which is a dimensionless equation with one parameter. However, we
have left t0 unspecified. I am allowed to make it anything I want. It
24
Dimensional analysis
seems convenient then to define t20 = l/g and our equation then becomes,
d2
= sin().
(2.1)
dt2
This final equation has no units and no free parameters.
The solution to the equation requires two initial conditions for the
position and the velocity. Expressed mathematically our initial conditions are
d
V
(t = 0) = 0 and
=
dt t=0
l
where V is the velocity in units of L/T . We now must make these initial
conditions dimensionless in the same way as the equations. Under the
change of variables we obtain,
d
V
(2.2)
(t = 0) = 0 and
=
gl
dt t=0
Thus our general problem comprises Equation 2.1 and the initial condition 2.2. The final equation has no parameters and there are two parameters that enter into the initial condition. All pendulums (started
from rest and
p from the same angle) are the same when time is plotted
in units of l/g instead of seconds. The scaling for time is exactly as
we obtained from the Pi Theorem.
2.3.2 Mass-spring-damper
Consider a simple mass-spring-damper system,
d2 x
dx
=
kx
dt2
dt
with the initial conditions that x(t = 0) = x0 and the velocity initially
is zero.
We will make the equations dimensionless by defining x
= x/x0 and
t = t/t0 . Here, t0 is arbitrary, but x0 is the initial position. Using
the initial position seems like a good choice to make x dimensionless
because the solution for x will always be bounded in magnitude by 1.
Applying the change of variables we obtain
m
d2 (
x x0 )
d(
x x0 )
=
kx0 x
d(t t0 )2
d(t t0 )
25
Be careful to note the units of d2 x/dt2 . Its units are length/time2 . Thus
t0 is squared whereas x0 is not. The placement of the squares when we
write d2 x/dt2 are literal; t is squared and x is not. Since x0 appears in
each term it cancels out and we can group t0 together to obtain,
d2 x
x kt20
t0 d
x
.
=
2
m dt
m
dt
Since t0 is arbitrary we select t20 = m/k to obtain,
d2 x
d
x
x
.
=
dt2
mk dt
(2.3)
t [T ]
t [T ]
t [T ]
q
k
t m
x0 [L]
x0 [L]
k [ TM2 ]
1
k
m [T2 ]
k
1
m [T2 ]
[M
T ]
1
[
m T]
1
m [T ]
[M ]
km
Remove M
Remove L
Remove T
=f t
,
,
x0
m
km
or
x
=f
t,
km
.
26
Dimensional analysis
=
.
x
L x
2.4 Summary
Throughout this book we will continue to return to the idea of working
in dimensionless terms. We will often pursue an approach where we
use the fact that the equations for a problem are known, thus we have
mathematics to guide us to the right dimensionless groupings. In
cases where the mathematical description can be stated, but it is too
difficult (or impossible) to solve the problem - we can still obtain insight
to the problem by appealing to dimensional analysis.
3
Conduction heat transfer
We will now begin our study by picking the example of heat conduction
in a solid. Despite the fact that in the introduction we discussed all the
exciting applications which involve fluid flow, we start with the example
of the transport of heat with no fluid flow. The reasons for this are two
fold. First, the problem of heat conduction is important in a number
of applications that we will discuss as we proceed. Second, problems
with no fluid flow are easier. Starting with heat transport in stationary
media allows us to work through the basic mathematics used to describe
flow and transport processes. Later, we will add the complexity of fluid
flow on the transport of heat.
28
Hot side
29
Cold side
Figure 3.1 Schematic of heat conduction in a gas. When a fast moving hot molecule collides with a slow moving cold molecule then the
cold molecule speeds up and the hot one slows done. Eventually
all the molecules inside the box undergo enough collisions that the
temperature is uniform throughout the box.
k
T
L
(3.1)
30
Thot
q=
K
T -T
L hot cold
Tcold
Figure 3.2 Schematic of 1D heat conduction and the use of Fouriers
law at steady state. The heat flux is proportional to the temperature
difference across the material.
(3.2)
31
q = h(Thot - T )
Thot
Figure 3.3 Schematic for heat flow from a solid to a fluid by convection. The convection coefficient, h, is usually taken as an empirical
parameter.
(3.3)
32
(3.4)
(3.5)
The rate equation says we simply need to look at the net rate that
energy is coming in and out of our volume and that must be equivalent
to the time rate of change of the energy stored inside the volume. I
should comment on the use of E in to denote the rate of energy coming
into our volume. It is common notation to use a dot above a variable
to denote a rate. This dot should not be confused with the one often
used in mathematics books where df /dt f 0 (t) f(t).
Energy has units of Joules and rate that energy crosses the surface,
has units of Watts. Remember that a Joule is the same as NewtonE,
meter.
34
N
X
i=1
N
mC X
mdx
C(T (x, t) Tref ) =
(T (xi , t) Tref )dx
L
L i=1
The sum of the energy of all the slices is the same as the integral in the
continuous limit,
Z
mC L
E =
(T (x, t) Tref )dx.
L 0
If we wanted to be a little more general, we could assume the case
where A is the walls cross sectional area. You should confirm the units
of this expression and convince yourself the units are Joules. The above
form is the one we will always use, where our quantity is give on a per
unit volume basis and then we integrate over a volume.
Inside the wall, there is an internal heat flux q(x, t) (W/m2 ) that
flows from hot to cold within the wall. The local heat flux is different
at every location and can evolve with time. The balance of energy for
the entire wall simply states that the rate of change of the energy stored
in the wall is equal to the net rate of heat passing in from the left at
x = 0 minus what goes out at x = L. Since heat flux has direction,
positive heat flux points from left to right. The balance of energy in
rate form is,
dE
= A(q(0, t) q(L, t)),
(3.7)
dt
where A is the cross sectional area of the wall. Using our expression for
the energy stored we obtain,
" Z
#
L
d
A
C(T (x, t) Tref )dx = A(q(0, t) q(L, t)).
(3.8)
dt
0
Now lets assume the case which is common for solids where is the
density and specific heat are just constants. Since the reference temperature is a constant with time we can remove it from the equation
when we take the time derivative. We also cancel the cross sectional
area from both sides,
Z
d L
CT (x, t)dx = q(0, t) q(L, t).
(3.9)
dt 0
We can now use the Fundamental Theorem of Calculus to rewrite the
right hand side of the equation as,
Z
Z L
d L
q
CT (x, t)dx =
dx.
dt 0
0 x
Note that we used the Fundamental Theorem of Calculus backward
36
from how we are probably used to thinking about it. Since the region
of integration is not changing with time and we have constant material
properties the order of integration and differentiation on the left side
of the equation doesnt matter,
Z
C
q
T
+
t
x
dx = 0.
q
T
+
= 0.
t
x
(3.10)
k
T.
L
(3.11)
T
.
x
(3.12)
The law states there is a linear relationship between heat flux and temperature gradient. The constant, k, is called the thermal conductivity.
It is a material property that we can look up in a book (or wikipedia).
Also consider the sign in Fouriers law. If the temperature is going from
hot to cold as you move left to right, then the derivative of the temperature with respect to x is a decreasing function. Therefore, the heat
flux vector points from left to right.
As with the specific heat, the thermal conductivity is related to the
molecular nature of the material. For some simple materials such as
gases, there are molecular theories which allow us to compute the thermal conductivity from essentially first principles. It is also a property
that can be measured. In this course we will treat it as a material constant which we know. You might be familiar with the property already.
If I placed a copper spoon and a plastic spoon (of the same size) in a
pot of boiling water and told you you had to grab one of them - if you
picked the plastic spoon then you know about thermal conductivity.
38
T
T
=
k
.
(3.13)
C
t
x
x
Assuming that the density, thermal conductivity, and specific heat are
constants and using the definition of thermal diffusivity as = k/C
we obtain a final equation as,
T
2T
= 2.
t
x
(3.14)
39
(3.15)
T (x = l, t > 0) = Tc .
(3.16)
40
3.3.1 Pi Theorem
If we didnt know the equation but only knew the parameters of the
problem we would say the parameters are the temperature T , the spatial coordinate x, the time t, the initial temperature Th , the temperature at the boundaries, Tc , the size of the object l, the density , the
thermal conductivity k, the specific heat C. The units of the problem
are mass [M ], length [L], time [T ], and temperature []. Thus we have
9 parameters and 4 independent dimensions, and 5 dimensionless parameters. Since temperature difference is the only thing that usually
matters, we could probably argue that there are two temperatures that
matter T Tc and Th Tc , though we are still left with 4 dimensionless
parameters. If we wrote out the table we would have
T Tc []
x [L]
Th Tc []
[T]
l [L]
[ M
]
L3
2
C [ L2 ]
T
k [ M3L ]
T
[L]
[T]
Th Tc []
l [L]
T Tc []
[L]
[T]
Th Tc []
l [L]
[ M
]
L3
2
C [ L2 ]
T
2
C [ L2 ]
T
k [ M3L ]
T
k [ L4 ]
T 3
[L]
[T]
Th Tc []
[T]
Th Tc []
T Tc []
x
l
T Tc []
x
l
t
l2
T Tc
Th Tc
x
l
t
l2
Th Tc []
l [L]
2
[L ]
T
[ 1 ]
T
l2
remove L
remove T
remove
The Pi Theorem, and using the fact that the equation tells us that
matters rather than the three parameters independently yields the
41
(3.17)
(3.18)
(3.19)
42
T Tc
.
Th Tc
(3.20)
Think about this change of variables for the temperature for a minute,
it has a really nice feature of bounding the temperature in problem
between 0 and 1. This scaling changes the boundary conditions to
(
x = 0, t > 0) = 0;
(
x, t = 0) = 1;
(
x = 1, t > 0) = 0.
(3.21)
The scaling for temperature does not effect the governing partial differential equation, i.e.
2
=
.
x
2
t
(3.22)
= f (
x, t) or
,
=f
Th Tc
l l2
The functional form is exactly that obtained via the Pi Theorem.
While it may not seem so, these results are very powerful and useful;
it says all diffusion problems (with this boundary condition) are the
same. We simply scale the geometry to have a length that ranges from
zero to one. The combination of the size and the thermal diffusivity
gives us the appropriate time unit. On the scaled domain and in the
proper time units, problems of different size and material property will
have the same solution. Regardless of the size, material or temperatures
involved, if you solve the diffusion equation for this problem you have
solved it once and for all. Once we have our solution we simply use the
definitions above to move from the dimensionless solution to physical
units by multiplying by the appropriate constants.
Suppose we have a slab of copper ( = 1.17cm2 /s) that is one centimeter thick, the appropriate time scale is t0 = L2 / = 12 /1.17 seconds. If we then double the thickness to 2 centimeters the time scale
changes to t0 = 22 /1.17 seconds. This means that temperature field
will be the same in the second case at a physical time that is four times
longer. If you are cooking two steaks and one is twice as thick as the
other, the thicker one will need to cook four times longer to achieve the
43
same center temperature. The scaling also shows that the cooling of an
object from 200 C to 25 C is the same as cooling from 30 C to 25 C.
This result is truly remarkable!
This scaling for time in the heat conduction equation is very important. It provides us with an estimate for how long things take to heat
and cool. If we know the material and the length scale, then thermal
equilibrium will be reached on a time scale of the order of L2 /. This
is not a precise estimate, but a useful one nonetheless. The estimate
tells us the order of magnitude. If the estimate tells us the time scale
is one minute, we cant be sure if it will take 1 minute or 3 minutes.
However, we can be confident that heating will not occur in one second
or take hours. It is an order of magnitude estimate - thats all. These
types of estimates based on arguments of units are useful because they
are so simple and can give some immediate intuition.
44
45
46
T increasing
T decreasing
q<0
q>0
47
q
T constant
q=0
Figure 3.6 Schematic of the divergence of the heat flux. The temperature at a point rises, falls, or stays the same depending on the
local divergence of the heat flux vectors.
T
+ q = 0,
t
(3.27)
There is a linear relationship between heat flux and temperature gradient. Remember that temperature is a scalar field. It has a single value
at each location in space. The gradient operator creates a vector from
the scalar field. The negative sign in the equation is there so that heat
flux will point from hot to cold.
48
T
= (kT ).
t
(3.28)
(3.29)
49
4
Analysis with the heat equation
(4.1)
with appropriate boundary and initial conditions can provide the temperature as a function of time and all three spatial dimensions. The
equation is linear and well-behaved such that we can usually obtain a
solution to the problem of interest. For transient heat conduction in
one dimension, we can often solve the problem analytically. By analytically, I mean I can write an equation on a piece of paper which is the
solution to the equation and satisfies the boundary condition. We can
also make some progress analytically in higher dimensions in simple
geometries with simple boundary conditions.
We can also make analytical progress when we are interested in the
steady state, equilibrium behavior. In particular, 1D steady state behavior is extremely easy to calculate. While this assumption might seem
limited, such an analysis often allows us to make simple estimates and
obtain approximate or order of magnitude results.
In complicated geometries or with complicated boundary conditions,
we can use finite element software to solve the problem numerically.
There are a number of free and commercial codes that can be used
to solve such problems. Numerically, the heat equation is generally a
well behaved and well understood equation, thus numerical solutions
are usually (but not always) obtained without much difficulty with
modern software.
52
4.1.1 Quenching
Lets consider the transient behavior of the quenching problem that
we stated earlier, a hot 1D object of length L suddenly cooled at the
boundaries. Mathematically the problem is stated as,
2T
T
= 2
t
x
T (x, t = 0) = Th ;
T (x = 0, t > 0) = Tc ;
(4.2)
T (x = L, t > 0) = Tc .
(4.3)
Read the above equations carefully. Dont just scan them. Spend some
time to make sure that the equations speak to you. These equations
are encoding all the physical understanding we have about this problem.
Pay careful attention when we state a problem that we have properly
defined the equation, boundary conditions, and initial condition.
In the previous chapter, we worked through how this problem could
be stated in dimensionless form as,
2
=
,
x
2
t
(
x, t = 0) = 1;
(
x = 0, t > 0) = 0;
(
x = 1, t > 0) = 0,
53
4 n2 2 t
e
sin(n
x),
n
n=1,3,5...
T (x, t) = Tc + (Th Tc )
4 n2 2 t/L2 nx
e
sin
.
n
L
n=1,3,5...
54
T
k
= h(T T ).
x x=L
=
,
x
2
t
(t = 0, x) = 1;
hL
=
;
x
x=0
k
hL
= .
x
x=1
k
(4.4)
a)
b)
c)
d)
55
56
1
L
RL
0
T dx and by
dT
= 2q(0, t).
dt
(4.5)
This expression for the rate of change of the mean temperature is always
true for this problem - we have not yet made any assumptions.
Under some circumstances (i.e. a high thermal conductivity) it seems
reasonable to assume that the object is at approximately a uniform
temperature. In Figure 4.2 we saw this assumption was true if the
Biot number was small. Under the approximation of uniform spatial
temperature, the boundary temperature and the mean temperature are
about the same. In this limit we would obtain an ordinary differential
equation for the objects temperature, T (t),
d T (t)
2h
=
(T (t) T ).
dt
CL
(4.6)
2k
(T (L/2, t) T (0, t)).
L
57
T (0, t) T )
T from the boundary to the fluid
2k
In the limit of small Biot number, Bi = hL
k << 1, then the temperature difference inside the object is small relative to the temperature
difference between the object and its surroundings. This is the case
where the uniform temperature approximation is valid.
The lumped analysis gives rise to a new time scale. The cooling time
scale is given by t0 = CL/2h. The time scale is set by how quickly
convection can remove heat. This is in contrast to the high Biot number
limit where the time scale is given by how long it takes heat to diffuse
through the object.
If we take the lumped cooling equation and define a dimensionless
temperature as = (T T )/(Th T ) and an arbitrary time scale
t = t0 t, then we have
2ht0
d
=
.
CL
dt
We are led to the conclusion that the natural unit of time is t0 =
such that we obtain,
d
= .
dt
CL
2h ,
= et
The dimensionless formulation shows us that under the lumped assumption, all objects cool along the same dimensionless curve. An example of some experimental data collapsed using the analysis presented
here is shown in Figure 4.3.
The general problem of quenching by convection has two thermal
time scales. One scale is how long it takes for heat to get from the
center of our object out to the edge. This time scale is the conduction
time, L2 / = L2 C/k. The other is how long it takes heat to leave the
solid and be carried away by the fluid which is CL/2h. The ratio of
2
2h
these two time scales is L kC CL
= 2hL
k , or the Biot number! When
the Biot number is small it takes a very short time for heat to get from
58
the center of the solid to the boundary relative to how long it takes
heat to make its way into the fluid. Thus in the limit of small Biot
number we expect the solid to have approximately uniform temperature
differences. Changes inside the solid equilibrate instantly compared
to the rate that overall temperature is dropping. In the other limit
of large Biot number, it means that it takes a long time for heat to
make its way from the center of the solid to the boundary. In this
limit we expect the boundary to reach the fluid temperature. When
approached with a quenching problem, even in complicated geometry,
the first thing to do is estimate the conduction and convection time
scales and decide if your problem is convection or conduction dominated
and what the approximate time scale of the cooling will be. The Biot
number is important in heat transfer problems as it provides a reliable
and simple estimate about the basic physical picture.
59
(4.7)
(4.8)
where a and b are constants which are determined from the two boundary conditions. The temperature profile at steady state is linear. In
terms of the heat flux, when the temperature is linear the heat flux is
constant everywhere.
If we consider a domain 0 < x < L with boundary conditions of
holding the temperature at the two ends fixed, we obtain,
Lx
+ T (L).
L
The heat flux, obtained from q = kT is,
T (x) = (T (0) T (L))
q=k
T (0) T (L)
k
= T.
L
L
(4.9)
(4.10)
60
modeled as two resistors in series. Two materials side by side with the
same temperature difference imposed across both can be modeled as
resistors in parallel. However, we need to be careful as we move to
resistors in parallel because the heat flux is a per unit area quantity,
thus if the two materials in parallel have different areas we need to be
careful. The heat flux through a material with a temperature gradient
is,
k
(4.11)
q = T.
L
The total heat flow, Q, in Watts is
Q=
kA
T,
L
(4.12)
where A is the area through which the heat flux passes. Thus, when
working with resistor analogies we need to be a little bit careful if we are
talking about total heat flow or heat flux. When working with resistor
networks we will usually include the area to keep things clear and thus
we use,
T = QR
(4.13)
61
Cross-Section
frame
glass
frame
glass
inside
outside
glass
frame
Figure 4.4 Window glass surrounded by a metal frame. The resistance to heat flow through the window and the frame act as resistors
in parallel.
(4.15)
which follows the same resistor analogy where the resistance due to
convection is R = h1 . Likewise, for the exterior of the wall we have heat
62
qconv, in
qcond
q conv, out
Tw, in
Tin
1
R= h
Tw, out
R = LK
Tout
R = 1h
Figure 4.5 Heat flow through a 1D wall at steady state with convection on the interior and exterior of the wall. The three resistances
to heat flow act in series.
flow by convection,
qconv,out = hout (Tw,out Tout ).
(4.16)
Since the heat flow from the internal air to the wall must equal the
heat flow through the wall and equal to the heat flow from the wall to
the exterior air, we have
qcond = qconv,in = qconv,out = q
At steady state, all the heat flows in this example are equal and there is
a single q. The problem is the equivalent to resistors in series as shown
in Figure 4.5
We have three resistors in series; indoor convection, wall conduction,
and outdoor convection. The total resistance is the sum of the three
resistances,
1
L
1
R=
+ +
.
hin
k
hout
The heat flux from indoors to out is then the total temperature difference divided by this resistance,
q=
Tin Tout
=
R
Tin Tout
.
1
+ Lk + hout
1
hin
The temperature of the wall indoors can be found in the same way
we find voltages at intermediate nodes in a set of resistors in series,
namely by equating the heat flow through different resistors. Here, we
63
equate the heat flow through the whole system to the heat flow from
the indoor air to the wall by convection,
q=
Tin Tout
Tin Tw,in
=
,
1
L
1
+ k + hout
hin
1
hin
1
hout
1
hin
1
+ Lk + hout
In the case (for simplicity) where hin = hout , the equation can be
rewritten as
Tin Tw,in
1
.
=
Tin Tout
2 + hL
k
In the limit where convection is vigorous and the wall is relatively
resistive, the Biot number, hL/k, is large and
Tin Tw,in
0,
Tin Tout
meaning Tw,in Tin . A thick, well insulated wall feels the same temperature as the room you are in. In the case where conduction resistance
is very small compared to convection, the Biot number is small and
1
Tin Tw,in
,
Tin Tout
2
meaning Tw,in (Tin + Tout )/2. If the wall is thin and poorly insulted
it will be a temperature halfway between the indoor and and outdoor
temperature.
Note that we were able to extract some important things with this
simple analysis. First, we were able to easily get the answer for how
much heat flow is found in our system. Even if the geometry was more
complicated such that our resistor model is not that accurate, remember that our knowledge of h is not that accurate in the first place. A
more accurate geometric model doesnt mean a more accurate result
overall. Second, we easily able to extract some insight into the limiting behavior to provide some insight to the physics and very little cost.
We even found the single controlling dimensionless parameter. It would
be harder to see these kind of trends with a computer simulation, for
example.
64
(4.17)
65
T
q=h(T(x)-T )
Th
66
0 < x < L.
Notice the the extra factors of r that do not appear in cartesian coordinates.
While it is possible to numerically solve this equation for the 2D
temperature distribution, let us consider the case where the cylinder
radius is somewhat small relative to the length and the material is
rather conductive. In this case, it seems logical to assume that the
temperature distribution will mostly change in the axial direction.
As with the lumped analysis, if you want to try and construct a simpler model for your problem the first thing to try is taking an integral or
average. Lets take the integral of this equation over the cross sectional
area, i.e.
Z R 2
T
T
1
k
r
= 0 2rdr.
+
x2
r r
r
0
which is equivalent to
d2
k 2
dx
T 2rdr
0
Z
+k
0
2
r
T
r
r
dr = 0
T (x) =
T 2rdr
R2
0
67
we have
d2 T
kR2 2
dx
T
+ 2kR
=0
r r=R
T (x, 0) T (x, R)
hR
T (x, R) T
k
If the parameter hR/k, which is the Biot number based on the cylinder
radius, is small then the temperature gradients in the radial direction
are small compared to the temperature difference with the fluid. When
the radial Biot number is small then the 1D model should be appropriate. As with the lumped model in a previous section, the Biot number
is representing the strength of convection relative to conduction.
68
T T
.
Th T
2hL2
kR
on the domain
0<x
< 1.
The general solution to the equation is
= Acosh(m
x) + Bsinh(m
x).
You can confirm by inspection that the solution works. Applying the
boundary condition at x
= 0 we get A = 1. Applying the boundary
condition at x
= 1 we get B and the final result is
= cosh(m
x)
cosh(m) + 2L
R sinh(m)
sinh(m
x)
2L
sinh(m) + R cosh(m)
Some sample solutions are shown in Figure 4.8. In the limit of large
values of the parameter m, the solution limits a simpler form of =
exp(mx), which is shown compared to the actual solution.
While this might all seem like a mathematical exercise, the result is
important. The result represents a way to compute and predict heat
4.5 Simulations
69
Figure 4.8 Solutions to the dimensionless problem for heat conduction in a rod where one end is held at a fixed temperature and all
other surfaces are exposed to convection. For all curves the ratio
of length to radius is fixed at L/R = 10, while the solid curves
at for values of Biot number, Bi = hL/k = 0.01, 0.1, 0.2, 1, 10.
The dashed curves show the approximation = exp(mx); this
approximation is valid for large values of m.
4.5 Simulations
The heat equation in any 3D geometry can readily be solved with modern simulation software. Since the heat equation is generally well behaved both commercial or open-source software solutions work without
much effort or training. Many mechanical design and CAD softwares
have thermal analysis as a built in feature. In using these kinds of
software, it will be important for you to keep the basic mathematical
70
4.5 Simulations
71
your simple analysis tools. Paper and pencil analysis can also sometimes show you that some effect is small and relatively unimportant
and thus can be safely ignored.
In problems where convection is involved as a boundary condition it
is important to realize that the value of h is never known precisely. As
we will discuss later, it is common for error bars on h to be 25 percent or
greater even in well controlled cases. In less controlled cases the error
bars can be even greater. While a 3D simulation may be extremely
accurate with respect to the conduction model, the final result must be
suspect due to uncertainties in h. Thus, a simple model that lumps a
complex geometry as a 1D line, for example, might be just as accurate
overall as a full simulation.
Finally, it is important to note the ease with which one can make mistakes with software. It is easy to mix up your units, put in the wrong
parameters, or make some other seemingly minor software mistake. It
is good to have some order of magnitude estimates and qualitative understanding of your problem before resorting to numerical simulation.
5
Diffusive and convective mass transfer
Now we will consider the case of mass transfer, which we will discover
is quite similar to heat transfer. However, we will use this example
to introduce how fluid flow can dramatically change the behavior and
analysis. Fluid flow also greatly impacts heat transport in liquids and
gases. However, there are some details with energy transport that can
be a little subtle so we will postpone that discussion until we have
discussed thermodynamics in a little more detail.
In this chapter we consider an inert and dilute chemical species mixed
with a fluid. We may think of this as a little cream mixed with your
coffee, food coloring added to water, or transport of pollutants in the
atmosphere. We will consider a species which is at such a low concentration that it does not influence the properties of the bulk fluid. Analysis
of concentrated fluid mixtures can be quite complex and interesting but that is a topic for another day. When we talk of our dilute chemical
species, we will think of it as the transport and mixing of dye molecules,
such as food coloring, which have been added to the fluid. The notion
of dye implies they are passive (no chemical reactions) and we have a
way to measure their concentration.
Even though we are considering dilute dye, it is important not to
forget that our analysis is still in the continuum realm. Over any length
scale of interest, there are many many many dye molecules in that
volume - there are just far fewer dye molecules than fluid ones. While
we work in the continuum realm, we will refer to the dye as comprised of
molecules; the microscopic physical picture is helpful in our discussion.
74
cdV.
(5.1)
Since the dye molecules do not chemically react their number must
be conserved. If we consider an arbitrary volume then the time rate of
change of the number of molecules stored inside the volume must equal
the rate that dye molecules are entering and leaving the volume.
dN
= N in N out .
dt
(5.2)
Z
j ndS.
(5.3)
The minus sign is there because the convention is to define the normal
vector to point outwards from a closed surface.
75
(5.5)
76
(5.6)
The equation is exactly the same as we derived in chapter on conduction. It is a single equation for the concentration at every point in space
and time. It can (in principle) be solved when provided an initial condition and appropriate boundary conditions. If you have a solution to
the heat equation, you can use it for this mass diffusion equation. This
equation is appropriate only for stationary media.
77
dye particle to change direction and speed often. The path of any particle is very random and cannot be predicted. We can however, speak
of a statistical average of the random motion.
The random walk is often called the drunken sailor problem. The
drunk sailor walks out of the bar has no idea where he is going and
does not remember where he has been. Every step is taken at complete
random. I do not know why we are so quick to stereotype sailors, but
that is beside the point. Some of his steps will be small, some long;
some to the left, some to the right. A simulation of this situation is
easy to do. Simply start at known x, y coordinate location and with
each time step, move a random distance in both x and y. In doing so
we use the Gaussian, or normal, distribution to determine the direction
and magnitude of each step. Every time we run the simulation we will
get something different, a sample is shown in Figure 5.1.
Since each simulation is different we can repeat the experiment many
times and ask the question, on average how far away from the bar is
the sailor? If we plot the total distance squared from the bar versus
time for one situation and the average of many simulations on log-log
coordinates we obtain a result as shown in Figure 5.2. We find that
in this case that an individual sailor gets further from the bar in a
random way, but that on average r2 increases linearly with time. To
go twice the distance takes 4 times as long. This is the same result
78
79
c
dx.
x
where c(x) is the concentration on the left and c(x + dx) is the concentration on the right. If the net flux is proportional to the difference in
concentration on the left and right side of the line, the net flux would
proportional to the concentration gradient.
80
chapter we will assume by some magic that the fluids velocity field is
known. In the next chapter and for much of the rest of this book we
will discuss how to calculate that velocity field - a non-trivial matter.
81
dumping some pollutants and you are tasked with figuring what is
going on. Lets imagine a case where the river water flow is constant
in time. In principle, if you had enough sensors you could describe the
concentration field of the pollutant at every point in space and time;
i.e. c(x, y, z, t). This is difficult to do, so you set about to do some more
localized sampling.
First, you decide to fix your reference frame by standing in one spot
on a river bank and we measure the rate of change of pollutant by
simply taking the time derivative of the measured concentration, c/t.
Since the rivers flow is constant, if you measure the concentration
changing in time we would assume this must have something to do with
changes in the rate that pollutants are dumped in the river upstream.
Now imagine that we decide to do some sampling of the river on a
small boat. We get on a small boat and drift down the river, just going
with the flow. If we monitored the concentration of pollutant from the
perspective of the flow, we would sense something different. If we are
upstream of the source, we sense no pollutant. As we pass the source,
we would presumably measure an increase in concentration. Now that
we have become a fluid particle the rate of change of concentration
with respect to time from the perspective of the fluid, is different than
in the fixed frame. We will find it useful to describe the rate of change
of something with respect to time from the perspective of the fluid.
To more formally describe change from the perspective of the fluid,
we take the variable a, to be the location of a fluid particle at time t = 0
and r(a, t) is the location of the same particle at some later time t; see
Figure 5.3. If we knew the function r(a, t) we would know everything
about the flow. Think about this for a minute, the notation above is
perhaps not intuitive at first glance. We are defining everything about
the flow based on tracking where each and every blob of fluid goes.
In this frame, the velocity of a fluid particle is defined as
r(a, t)
=v
t a=const
The notation above says we take the derivative of r and hold a constant.
It is perfectly valid for us to define the velocity as a function of a,
material or as a function of r space. Note that this expression is a
vector equation as denoted by the boldface notation.
To demonstrate the material derivative, consider the concentration
82
t=0
y
particle path
a
r(a,t)
x
Dc
c(r(a, t), t)
=
+
v
c
=
+
v
c=
(5.7)
t
t
t
Dt
a=const
The notation D/Dt = /t + v is commonly used for compactness
and is called the material derivative. The material derivatives purpose
83
in life is to tell us the rate of change of something in time from the perspective of the fluid. It captures the two effects; the field itself changing
with time and a fluid particle moving about in space where the field
changes from point to point. If we expand into component form we have
c
c
c
c
c
Dc
=
+ v c =
+u
+v
+w .
Dt
t
t
x
y
z
84
where (x, t) is the density field. The subscript V (t) reminds us that the
volume of integration is not fixed in time and may be changing shape as
the flow deforms the material. Our volume V (t) is going with the flow
and following the material for all times. Since our volume of integration
was taken to be a material volume (i.e. it encloses the same amount of
material at all times) and since mass is conserved then dM
dt = 0. Using
the Transport Theorem (Equation 5.9),
Z
Z
dM
d
=
dV =
+ (v) dV = 0
dt
dt V (t)
t
V (t)
Since the volume in our analysis was arbitrary and since the integral is
always equal to zero then the the integrand must be zero,
+ (v) = 0.
t
(5.10)
This equation must be true point-wise in the fluid and is the differential
statement of mass conservation. By rearrangement,
D
+ v = 0.
Dt
(5.11)
85
This equation says that the density of a fluid particle can only change
if the velocity field has non-zero divergence. Divergence of the velocity
field is a measure of the net rate of fluid flowing into/out of a point.
If the velocity vectors are all pointing to one location, then the flow
must be accumulating mass at that point and the density should be
increasing. If the velocity vectors all point away from a point in the
flow, the density should be decreasing.
For an incompressible flow with constant density we have the simplified relation,
v = 0.
This assumption of incompressibility is one that we will use throughout
the course. For a liquid the assumption of incompressibility is a good
one. I challenge you to compress some water to any significant degree.
For a gas the assumption is questionable. It turns out that a rule of
thumb people use is that if the flow velocity is less than about 1/3
of the speed of sound in a gas, then the assumption of incompressible
flow is basically fine. For air this means we can safely treat cases with
flow velocities on the order of 100 m/s or less as an incompressible flow.
Flow around a car is incompressible, but flow around an airplane is not.
Compressible flows have many interesting features that will be beyond
the scope of our first course.
It is also important that we refer to a flow as incompressible and not
the fluid. For the flow of air, we have many examples of incompressible
flow despite the fact that air is quite compressible.
86
the equation and the divergence theorem for the right side we obtain,
Z
Z
c
+ (cv) dV =
jdV
t
V (t)
V (t)
We can group all the integrands together and since the volume in our
analysis was arbitrary the integrand must always be zero,
c
+ (cv) = j.
t
Substituting in Ficks law as before and assuming a constant diffusivity,
c
+ (cv) = D2 c.
(5.12)
t
Now, a new term has appeared in the equation (cv) which represents
convection of the dye molecules by fluid motion.
Lets think about this equation for a minute. In the case where there
is no fluid motion, v = 0, we are back to the classic diffusion equation.
In another limit, lets assume there is no molecular diffusion. As we
discussed earlier in the chapter, molecular diffusion is very slow and so
in many cases ignoring it is not an unreasonable assumption. In that
case the equation would be,
c
= (cv).
t
The product cv is a vector which gives the convective flux or rate
that concentration is carried by the fluid per unit time, per unit area
at a point. If you recall that c has units of number of molecules per
unit volume, you can check the units of cv yourself. Now recall that
the divergence operator gives us a measure of whether a vector field
is converging or diverging at a point. If the convective flux vectors
are converging at a point, then the concentration field must be locally
increasing in time. If the convective flux is flowing outward from a point
in space, the local concentration should be decreasing.
By rearrangement, Equation 5.12 could be written as,
c
+ v c + c v = D2 c.
t
Substituting in the conservation of mass from the previous section we
can replace the divergence of the velocity field as,
c
c D
+ v c
= D2 c,
t
Dt
87
= D2 c.
Dt Dt
(5.13)
This last equation has an interesting interpretation. The material derivative D/Dt describes the rate of change of concentration from the perspective of the fluid. The operator is accounting for convection by fluid
motion. The right hand side is the diffusion term. The term on the
left c D
Dt describes the concentration of dye (number per unit volume
of fluid) changing because the carrier fluid itself is being compressed
or expanded. If I have a container containing a fixed number of dye
molecules then I compress the closed contained to decrease the total
volume the dye concentration has increased even though the number
of molecules has not.
For incompressible flow the density doesnt change and thus the mass
transfer equation is simplified to
Dc
= D2 c,
Dt
(5.14)
88
c(r = ) = 0.
c
D 2
+ c v
=
c.
UR
t
(5.15)
The chosen time scale was the time it takes to convect something the
distance of the cylinder radius. This time scale was a convenient choice
as it made the coefficient in the equation go to 1.
The parameter U R/D is called the Peclet number (pronounced, pecklay) and is a dimensionless measure of the strength of convection to
the strength of diffusion. A high Peclet number occurs when the flow is
fast around a large object. A high Peclet number means that the fluid
carries the dye downstream. We might expect at high Peclet number
the dye is mostly concentrated in a wake downstream of the cylinder.
For a small Peclet number diffusion dominates, and in this limit we
would expect the dye to work its way upstream against the incoming
89
6
Conservation of momentum in a fluid
Z
f (x, t)dV =
V (t)
V (t)
f
+ (f v) dV.
t
(6.1)
(6.2)
92
b
d
bdV =
b
+ (v) +
+ v b dV.
dt V (t)
t
t
V (t)
This form is convenient since the first term in parenthesis is zero from
mass conservation, Equation 6.2. Our final result is then,
Z
Z
Z
d
b
Db
bdV =
+ v b dV =
dV,
(6.3)
dt V (t)
t
V (t)
V (t) Dt
which is useful for subsequent analysis. Note that we made no assumptions about incompressibility here, so this relationship is one that is
always true. This relationship is convenient because it already separates out conservation of mass.
where (x, t) is the density field and v(x, t) is the velocity vector field.
Note that this is a vector expression as the momentum has an x, y, and
z component. We can use Equation 6.3, simply replacing b with v,
Z
Z
dp
d
Dv
=
vdV =
dV
(6.4)
dt
dt V (t)
V (t) Dt
93
This form might make you uncomfortable since this is a vector equation and we only discussed the transport theorem with scalar functions.
If it does you could proceed with only the x component of the momentum instead,
d
dpx
=
dt
dt
Z
udV =
V (t)
V (t)
Du
dV.
Dt
+ v y
+ w z
)u =
For example, the x component of (v )v is (u x
u
u
u
(u x +v y +w z ). The entire operator expanded into component form
is a vector,
u
u
u u
x + v y + w z
v
v
v v =
u x + v y + w z ,
w
w
u w
x + v y + w z
where the first, second, and third rows are the x, y, and z components
of the vector.
The expression 6.4 allows us to compute the rate of change of momentum of a material volume. Newtons laws requires that this vector
must equal the sum of the forces acting on the control volume. Determining the forces acting on the fluid is not a simple matter and will
take some work. Before we turn to computing forces, lets make sure
we really understand the idea of acceleration in fluid flows.
94
95
x
U(t=0)
U(t=dt)
dx
=Udt
x=0
x = dx
Figure 6.1 Schematic of acceleration in a fluid flow from the perspective of the fluid. A person riding a raft through a narrowing
river feels acceleration even if the fluid flow is constant when observed from the banks of the river.
Now you are observing from the river bank and want to know the
fluid acceleration at the same point. If you watch me go past you could
measure my acceleration the same way. However, assume there is no
tracer for you to observe in the flow and you are stuck taking point
measurements of the rivers flow velocity. The previous formula for
acceleration is still valid, however from the bank you can only measure
the river velocity as a function of position. The equivalent measurement
of the fluid acceleration would would be,
Acceleration =
The distance dx that I would have moved on the raft over time dt is
related to the flow velocity as dx = u dt. Therefore substituting for dt,
Acceleration = u
Now the fraction in this equation is the definition of the spatial derivative of the flow velocity in the limit that dx goes to zero,
Acceleration = u
u
.
x
96
6.3 F = m a
In particle mechanics, we usually think of Newtons Law expressed as
the equation F = m a. However, this is for a constant mass particle
and a more general statement would be
dp
= F,
dt
where p is the linear momentum and F is the sum of the forces acting
on the particle. So we have developed a way to describe the rate of
change of the momentum for a fluid particle. Now we need to work
on the forces. We will account for body forces acting on the whole of
the material volume, fb and surface forces fs acting on the boundary
separately. Generally then, conservation of momentum can be stated
as,
Z
Dv
dV = fb + fs .
(6.5)
V (t) Dt
We could also consider electrostatic forces (if the fluid were charged
and subjected to an electric field) or Lorenz forces (if the fluid were
passing a current and subject to magnetic fields). These are interesting
topics for another day, however their inclusion into our theory would
not be difficult. If we know how to calculate the force per unit volume,
we only need to integrate this force density over our volume.
6.3 F = m a
97
s
s
dS
S
s
s
Figure 6.2 The local stress vector, s, acts everywhere on the surface,
S, of an arbitrary region. Integrating that stress over the surface area
gives the net force acting on the surface of the volume.
98
6.3 F = m a
99
Tyy
Tyx
Tyz
Txy
Tzy
Tzz
Txx
Tzx Txz
vector s is,
s=nT
The dot product projects the tensor onto the surface with a normal
vector n to get the stress vector at that point on the surface. The dot
product of two vectors gives a scalar, while the dot product of vector
and a tensor gives a vector. To remember what n T means, it is like
the vector-matrix operation,
Tzy
Tzz
You can test this yourself by taking a normal vector pointing in the xdirection, n = [1 0 0]. Taking n T with this normal vector returns the
first row of T which are the three forces acting on that face. Fortunately,
conservation of angular momentum will require the stress tensor to be
symmetric, thus if you forget the order of multiplying the tensor and
vector, you will get the right answer anyway. I still to this day confuse
(without looking it up) the what components go in the rows and which
go in the columns of T but since the tensor is symmetric I am saved.
Also, you might be tempted to confuse n T as being Tn (i.e. the usual
matrix vector multiply), and again symmetry of the tensor will give you
the right answer. Since we wont use the tensor notation much outside
100
of deriving equations and since this is likely the first class where you
have seen tensors, Ill cut you some slack on keeping everything straight!
The total force exerted on an arbitrary volume element by the surface
forces, fs , is given as simply the integral of the stress vector around the
closed surface, i.e.
Z
Z
fs = sdS = n TdS.
By the divergence theorem, which also works for tensors just as it does
for matrices,
Z
Z
fs = n TdS = TdV.
(6.7)
In component form, the divergence of the stress tensor is,
T = x
Tyx Tyy Tyz =
y z
Tzx
Txx
x
Tyx
y
Tzx
z ,
Txy
x
Tyy
y
Tzy
Tzy
z ,
Tzz
Txz
x
Tyz
y
Tzz
z
6.3.3 Meaning of T
Above we used the definition of the stress tensor and our vector calculus
to show that the net surface force acting on an arbitrary volume is the
divergence of the stress tensor. We can get the same result using a free
body diagram for a infinitesimal element of fluid. Consider a small cube
that is dx, dy, and dz on each side. Consider the point P in the middle
of the cube. To find the forces on the face of the cube in the x-direction,
we must consider that the stress tensor varies in space.
The value of the stress at one point is related to the value at a
6.3 F = m a
101
Tyx+ Tyx dy
x 2
dy
y
x
Txx- Txx dx
x 2
Tyx- Tyx dy
x 2
Txx+ Txx dx
x 2
dx
dx Txx
.
2 x
Here Txx is the component of the stress tensor at the center of the cube.
Likewise, the stress on the left hand face of the cube in the x-direction
is,
Txx
dx Txx
.
2 x
102
Which reduces to
dfsx =
Txx
Tyx
Tzx
+
+
x
y
z
dxdydz.
TdV.
and we get the same result as the previous section. The net surface
force at a point in space is given by the divergence of the stress tensor.
If the stress tensor is constant in space, then there is no net force on
the fluid.
V (t)
V (t)
V (t) Dt
Since all the integrals are volume integrals over any arbitrary volume
in the fluid, the equation holds at every point,
Dv
= g + T.
(6.9)
Dt
While this looks like a beautifully simple equation, there is a serious problem. There are too many unknowns. The equation is actually
three equations for the 3 unknown components of the velocity vector.
However, there are 9 components of the stress tensor. If we considered
conservation of angular momentum (we will skip the derivation) we
would find that the stress tensor must be symmetric. Thus there are
103
7
The Navier Stokes equations
Dv
= g + T.
Dt
These equations say nothing because we have not defined the stress tensor, T. Defining the stress tensor requires knowledge of the material.
Its definition is essentially empirical or comes from a molecular model
of the material. The stress tensor is not a fundamental law of nature.
The situation is just like the case of heat conduction where our fundamental conservation law was written in terms of the heat flux vector
and Fouriers law was needed to close the problem. The relationship
which defines the stress tensor is our constitutive law.
There are some constraints on what form the tensor must take. For
example, we stated in the previous chapter (without proof) that it is
symmetric due to conservation of angular momentum. It also can only
take limited forms if we want its definition to remain constant under
a change of coordinate systems (we always want this) or if we want it
to be isotropic (we often, but not always want this). Well save these
topics for a more advanced course on continuum mechanics and not
discuss them further.
106
P
0
0
T= 0
P
0
0
With this stress tensor and the definition of the divergence we write all
the terms out,
P
0
0
, y
, z
T = x
P
0
0
0
=
P P P
,
,
x
y
z
= P
Substituting this expression into our conservation law would yield conservation of momentum to be,
Dv
= P + g.
Dt
Dv
= P + g.
Dt
(7.1)
(7.2)
107
108
U
d
109
The terms with the velocity gradients is a little more complicated and
will be the subject of much discussion later in this chapter. Since velocity is a vector, its gradient is a tensor. The velocity gradient tensor
is written out as,
u v w
v =
u
y
v
y
w
y
u
z
v
z
w
z
(7.5)
Coupled with mass conservation for an incompressible flow, our problem is closed;
v = 0.
(7.6)
Equations 7.5 and 7.6 represent four equations for four unknowns; the
three components of velocity and the pressure. These equations are
known as the incompressible Navier Stokes equations and comprise the
mathematical foundation for describing fluid flow.
Just to be clear on the notation, if we expand the momentum equation in component form for a Cartesian coordinate system we obtain,
2
u
u
u
u
P
u 2u 2u
+u
+v
+w
= gx
+
+
+
t
x
y
z
x
x2
y 2
z 2
v
v
v
v
+u
+v
+w
t
x
y
z
w
w
w
w
+u
+v
+w
t
x
y
z
P
= gy
+
y
P
= gz
+
z
2v
2v
2v
+
+
x2
y 2
z 2
2w 2w 2w
+
+
x2
y 2
z 2
The great thing about expressing the formulation in our vector calculus
notation is that we can change coordinate systems and simply look
110
+ v v = (P + gz) + 2 v.
t
where (without loss of generality) we aligned the coordinate system
such that gravity is pointing in the negative z direction. We can introduce the dynamic pressure as Pdyn = P + gz and rewrite the equation
as
v
+ v v = Pdyn + 2 v,
t
The dynamic pressure is the pressure that results from fluid motion
and subtracts out the hydrostatic component.
This form is convenient as it allows us to essentially ignore the effect
of gravity when computing fluid motion in a constant density fluid.
From the above equation we can see that if there is no fluid velocity
and we are at equilibrium then Pdyn = 0. At equilibrium the dynamic
pressure (pressure due to fluid flow) is constant and does not impact the
flow. At equilibrium the measured pressure is given as (P + gz) = 0,
meaning the pressure is only hydrostatic. In constant density flows,
gravity does not play a role in determining the flow field, but it does
effect the pressure. If the density is constant then gravity acts equally
on the whole fluid and thus there is no flow driven by gravity. If the
density of the fluid is not constant, then we could have buoyancy driven
flows.
111
curvature in the velocity field. While the full Navier Stokes equations
have other more complicated terms, we can think that viscosity will
always act to smooth the velocity field and pull the velocity to some
uniform or simple state - just like the heat equation.
The ratio of / appears often that it is given its own symbol,
= /. This parameter is called the kinematic viscosity and has units
of m2 /s. The material property, , is properly called the dynamic viscosity. The kinematic viscosity plays a role like the thermal diffusivity,
, or the diffusion coefficient D. In thermal problems, , sets the rate
that heat propagated over some distance; the kinematic viscosity plays
a similar role. The kinematic viscosity sets the rate that momentum is
transferred from one sliding fluid layer to another.
112
u
n = [1 ]
0
y
x
P
0
T=
0
P
+
2 u
x
u
v
y + x
u
y
v
+ x
v
2 y
!
.
01
"
P
0
P
+
2 u
x
u
v
+
y
x
u
y
v
+ x
v
2 y
!#
.
113
+
u
0
0 P
y
Thus the stress vector at our point simplifies to,
P
s = u
y ,
The result says that the x component of the stress is given by the
horizontal velocity gradient, just as we described when defining a Newtonian fluid. This is the force that would cause a fluids viscosity to drag
a solid along with the flow. The y component of the stress is given by
the pressure.
114
XA(dt)
t=dt
115
XB(dt)
U(XA)
U(XB)
t=0
XA(0)
XB(0)
x
Figure 7.2 Schematic of a material line in a 1D flow. The points A
and B are subjected to different velocities and thus move different
distances in one instant of time, dt.
116
dt UA
A
dt VA
A
o
t=0
o
t=dt
dt Vo
dt VB
B
dt Uo
dt UB
(vB v0 )dt
,
dx + (uB u0 )dt
tan() =
(uA u0 )dt
.
dx + (vA v0 )dt
You can then use the Taylor Series to relate the velocities at point
A and B to that of point O, i.e. uB = u0 + u
x dx. Substituting the
117
Taylor Series results and taking the limit, these geometric relationships
become rate equations for the rate of change of the angles,
v
d
=
;
dt
x
d
u
=
.
dt
y
Note that in the limit of small angles which occur over small times,
tan() . Thus the sum of these two rates,
d
v
u
d
+
=
+
,
dt
dt
x y
is the rate the angle between the two lines OA and OB close or open
up. This is the shear strain rate.
This relationship makes sense. Hold your hands together in front of
you, pointing upwards. Sliding your hands up and down relative to each
other creates shear. The right hand has a different vertical velocity than
v
the left hand. Therefore x
is not zero. Holding your hands horizontal
creates velocity gradient as u/y. The shearing motion of your hands
would act to deform a material held between them.
7.5.3 Rotation
Using the results from the prior section we can also ask what is the
difference of the rate of change of the two angles. This difference is
twice the rate of solid body rotation of the line elements at that point,
d
v
u
d
.
dt
dt
x y
In two dimensions, you may recognize this quantity is the same as the
curl of the velocity, which is referred to in the fluid mechanics world
as vorticity, . The curl is denoted by = v. In component form,
the vorticity vector is
w
v
y z
w
= v = u
.
z x
v
u
x
y
The curl is the vector cross product of the gradient operator and the
velocity vector v. In two dimensions where the velocity is constrained
to the x-y plane, the vorticity only has a component normal to this
118
v
u
.
x y
7.5.4 Generalized to 3D
In three dimensions, the results of this section are summarized as follows. The strain rate tensor is defined as
1
v + vT .
(7.8)
S=
2
In component form this tensor is
u
v
u
w
2 u
x
y + x
z + x
1
v
v
v
w
2 y
S = u
y + x
z + y
2
u
w
v
w
2 w
z + x
z + y
z
The strain rate tensor is symmetric and provides the rate of strain
of fluid elements in the different directions. The diagonal components
of the tensor are the linear strain rates in the three dimensions. The
off-diagonal components are the shear strain rates in the three planes;
x y, x z, and y z.
To rotation tensor is defined as
0
z y
1
(7.9)
R = z
0
x
2
y x 0
where are three components of the vorticity vector, = v.
The off diagonal elements of the rotation tensor provide the amount
of rotation in the three planes x y, x z, and y z. The tensor is
anti-symmetric. It is fairly easy to see that the velocity gradient is the
sum of the strain rate and the rotation tensors,
v = S + R.
Velocity gradients lead to both strain of material elements and rotation
of those elements. The strain rate tensor provides the rate of strain of
the material. The rotation tensor provides the amount of rotation.
7.6 Non-dimensionalization
119
Using the above tensors, the Newtonian constitutive law for an incompressible flow states that
T = P I + 2S.
The forces exerted on the fluid are proportional the strain rate, but not
the rotation of the fluid.
Please note that this discussion of kinematics is quite brief and more
detailed derivations can be found in the references. The main point is
to demonstrate that the form of the Newtonian stress tensor is connected to the kinematics of rate of deformation of the material. The
stress is related to the velocity gradients. However, the use of the symmetric strain rate tensor subtracts our solid body rotation which does
not generate stress. In a Newtonian fluid only the rate of deformation
generates stress.
7.6 Non-dimensionalization
Consider the Navier-Stokes equations for incompressible constant density flow,
v
+ v v = P + 2 v,
t
v = 0.
Lets say the geometry of the problem provides some characteristic
length, L, and some characteristic flow velocity, U0 . The equations can
be made dimensionless following the same procedure as in previous
chapters. Lets define v
= v/U0 , [
x, y, z] = [x, y, z]/L, t = tU0 /L,
and P = P/P0 . Our time scale is picked so that it takes one time
unit for a fluid particle to travel the length L; i.e. our dimensionless
time is t = tU/L. When making equations dimensionless, it is always
important to recall that derivatives have units. The derivative, /x has
units of 1/Length, for example. Therefore, when we make our operator
= L. The dimensionless
dimensionless, we make the substitution
2
,
+v
v =
P+ 2 v
L
L
L
t
120
= 0.
v
v =
P + 2 v
+v
,
U0 L
t
v
= 0.
Thus the problem has only one dimensionless parameter that matters,
the Reynolds number,
U0 L
Re
.
The Reynolds number came out in several of our problems early in this
book when we discussed dimensional analysis.
121
is, lets consider a few hypothetical cases. For water, the density is =
1000 kg/m3 and the viscosity is approximately = 0.001 Ns/m2 . So
lets say a person 2m tall who can swim at 1 m/s would have a Reynolds
number of 2 106 . A small organism that is 100 m and can swim one
body length per second would have a Reynolds number of 0.01. For
you to match the Reynolds number of this organism you would need to
swim at about the same speed (100 m/s) in molasses!
In a pipe flow, a standard rule of thumb is that the flow will transition from laminar to turbulent at a Reynolds number of 2000. Here
the Reynolds number is defined on the diameter of the pipe. To get a
sense of where a Reynolds number of 2000 is - water flowing in a 1/4
inch pipe (6.35 mm) at a velocity of 1 ft/s (0.3 m/s) we would be at
a Reynolds number of about 2000. The transition number of 2000 is
only a rule of thumb. If you try this experimentally you will find the
transition number changes based on how carefully you set up the experiment. If you just hook up some tubing and push water through it
without worrying about keeping everything smooth, straight, and well
conditioned, you will probably observe a transition at a much lower
Reynolds number, probably under 1000. If you are extremely careful
and take great pains to do a perfect experiment the flow will transition
at a much high Reynolds number. In 2011, the best theoretical determination of a true transition Reynolds number to sustained turbulence
was determined to be Re = 2040. While a seemingly simple problem,
the details are quite complex.
Reynolds original 1885 experimental apparatus survived at the University of Manchester, where he was a professor. A century later, his
exact experiments were recreated with his apparatus. The experiment
showed a transition Reynolds number much lower than he reported.
The difference was found to be due to cars in modern day Manchester,
the small amount of shaking from the streets perturbed the simple but
delicate experiments.
122
7.8 Summary
The starting point for much of our discussion in future chapters are the
Navier-Stokes equations for incompressible constant density flow,
v
+ v v = g P + 2 v,
t
v = 0.
These equations are the formal statement of conservation of mass and
momentum (i.e. Newtons laws of motion). The equations in this form
assume a Newtonian fluid constitutive law which relate the instantaneous rate of strain to stresses. Not all fluids follow the Newtonian
assumption but it is a good model of simple fluids such as air, water,
and oil. To achieve Eulers equations simply take the limit that the
viscosity is zero and remove that term from the equations. This zero
viscosity limit is a funny one which we will address in a later chapter.
In the past two chapters we worked through the derivation of these
equations. The discussion of the past two chapters was not as complete
or detailed as can be found in many excellent textbooks. The aim of my
discussion was to try to give you a sense of where the equations come
from and how to think about the physical meaning of the different
terms. At this point in your study, I would expect that many of the
details seem at least somewhat confusing.
Of course the equations on their own dont mean much. We will need
to understand the solution of these equations in order to connect to
physical phenomena. Even though you now have a complete mathematical formulation for fluid flow in hand, your study (and mine) of
the subject is just beginning. To move forward in our understanding of
fluid flow we need to learn how to use these equations to explain phenomena that we observe everyday. As explained previously, the Navier
Stokes equations are very difficult to solve, so we will need to use simple solutions to give us insight to the fundamentals and learn to use
the equations to guide our qualitative thinking. There is no one better
to make this point than Richard Feynman in his famous Lectures on
Physics. Italicized words are ones I changed to switch the context from
electromagnetism to fluid dynamics;
Mathematicians, or people who have very mathematical minds, are often led
7.8 Summary
123
astray when studying physics because they lose sight of the physics. They
say: Look, these differential equations - the Navier-Stokes equations - are
all there is to fluid dynamics; it is admitted by the physicists and engineers
that there is nothing which is not contained in the equations. The equations
are complicated, but after all they are only mathematical equations and if
I understand them mathematically inside out, I will understand the physics
inside out. Only it doesnt work that way. ... They fail because the actual
physical situations in the real world are so complicated that it is necessary to
have a much broader understanding of the equations. What it means to really
understand an equation - that is, in more than a strictly mathematical sense
- was described by Dirac. He said: I understand what an equation means if
I have a way of figuring out the characteristics of the solution without actually solving it. A physical understanding is a completely unmathematical,
imprecise, and inexact thing, but absolutely necessary for a physicist and
engineer.
My hope with the subsequent chapters is to begin to help you understand the Navier-Stokes equations in this imprecise and inexact
sense.
8
Solutions to the Navier-Stokes equations
126
PHi
u(y)
PLow
y
x
Figure 8.1 Schematic for Poiseuille flow between two parallel plates.
solve some simple problems. In all the examples that follow we will
study flows where there is no acceleration and force due to viscosity
will balance some driving force.
+u
+v
=
+
+ 2
2
S
@
x
y
x
xSS
y
@
t
S
+ 2
+u
+v
=
g +
2
AA
@
x
y
y
xSS
y
t
@
127
Su
+ v = 0.
S
x
S y
Note that we keep the axial pressure gradient as the pressure drives the
flow and must drop continuously as we move down the channel.
From conservation of mass, v/y = 0; an equation which states that
the vertical velocity is a constant in the y direction. Since v = 0 at the
wall, v = 0 everywhere. We can now cancel out all terms that have the
vertical velocity, v in them;
!
2
2
u
u
P
u
u
S
S
@
S
+u
+
+v =
+
2
SS
@
S
x
x
xSS
y 2
@ y
t
S
!
2
2
v
v
v
P
v
S
S
S
A
@
+u
+
+v =
g +
2
S2
SS
@
AA
x
y
xSS
y
t
@ y
S
It makes sense that the left side of the Navier Stokes equations is zero.
The left side represents the acceleration of a fluid particle. There is
no acceleration in this system. There is neither acceleration due to
unsteadiness (change in velocity with respect to time) nor from the
perspective of a fluid particle. The channel is like a calm river whose
width doesnt change; if you were on a raft you would translate at
constant velocity. Neither the observer on the raft or on the banks
would see any acceleration in this flow.
Thus, the momentum equations are simplified significantly to
2u
P
= 2.
x
y
(8.1)
P
= g.
y
(8.2)
128
the axial flow. Since the flow was assumed invariant in x, the only consistent solution would be that the axial pressure gradient, dP/dx, is a
constant. Since the left side of Equation 8.1 is a constant the equation
can be easily integrated twice with respect to y,
u(y) =
dP 1 2
y + C1 y + C2 .
dx 2
dP 1
y(H y).
dx 2
dP H 3
dx 12
Usually we use the fact that dP/dx is the same as the total applied
pressure difference divided by the total length of channel. We need to
be careful with the signs. Convention would usually define a positive
pressure difference, P , as going from high pressure on the inlet to low
pressure on the outlet. With this convention, dP/dx = P/L. Using
the total pressure drop notation gives,
12L
.
H3
The expression says that pressure and flow rate are linearly related
- double the pressure and you double the flow. This expression is a
hydraulic version of Ohms law (V = iR) where pressure difference
P = Q
129
acts like voltage and volumetric flow rate acts like current. The number
12L/H 3 is the hydraulic resistance.
There are a few things we should check with our resistance formula.
First, we can check the units to make sure they agree with what we
expect and that we did not make an error. The other check is that
the trends all go in the right way. It makes physical sense that the
resistance increases with viscosity, channel length, and with a decrease
in the channel height. It is always good to look at a result and see if
you believe the basic facts about that result before proceeding.
Note that the mean flow velocity is u
= Q/H,
u
=
P H 2
12L
and the maximum flow velocity (along the centerline) is found by evaluating the velocity profile at y = H/2,
P H 2
3
H
=
= u
.
u y=
2
8L
2
Since there is no acceleration in this flow we can check our result
with a simple force balance If we draw a box (control volume) the net
force (per unit depth into the page) exerted on the box due to pressure
is
FP = P H.
This force must be balanced by the shear stress exerted at the two
walls. Recall the method of finding the stress on a surface discussed
in the chapter on the Navier-Stokes equations. The shear stress acting
tangential to the wall is given by nT where T is the total stress tensor.
The normal vector for the lower wall is n = [0 1]. The stress tensor at
the wall for this simplified flow is
!
P u
T
y
.
T = P I + v + v =
P
u
y
Thus the stress vector is,
s=nT=
u
y
P
130
u(y)
131
y
x
Figure 8.2 Schematic for Couette flow between two parallel plates.
132
example and cross out all the time derivatives and x derivatives. We can
then apply conservation of mass to tell us there is no vertical velocity,
v. Crossing out all these terms again yields the x and y momentum
equations as in Equations 8.1 and 8.2,
2u
P
= 2.
x
y
P
= 0.
y
However, in this case the x momentum equation simplifies further as
there is no applied pressure gradient. In the example of the concentric
cylinders there could be no pressure gradient around the closed loop
of the fluid gap, otherwise the pressure would be discontinuous as we
went around the circle. For Couette flow, the x momentum equation
simplifies to,
2u
.
0=
y 2
The curvature of the velocity field is zero and thus we expect a linear velocity profile. Integrating this expression twice with respect to y
yields,
u(y) = C1 y + C2 .
The two boundary conditions used to determine the two constants of
integration are u(y = 0) = U and u(y = H) = 0. Applying these
conditions to get C1 and C2 , the full velocity field is,
H y
.
H
The velocity varies linearly between the two plates. The total flow rate
between the plates is computed as
Z H
UH
Q=
u(y)dy =
.
2
0
u(y) = U
133
dP 1 2
y + C1 y + C2 .
dx 2
To find the constants of integration we would need to apply the boundary conditions, u(y = 0) = U and u(y = H) = 0. Applying the first
condition yields,
u(y = 0) = U = C2 .
Applying the second condition yields,
u(y = H) = 0 =
dP 1 2
U
dP 1
H + C1 H + U C1 =
H.
dx 2
H
dx 2
dP 1
yH
y(H y) + U
.
dx 2
H
u(y) =
Q=
0
UH
dP H 3
+
.
dx 12
2
If you look at the velocity profile and total flow rate derived for
Poiseuille and Couette flow independently, you will see the above result
is just the superposition of the two solutions. We could have guessed
this result from the start since under our assumptions the equations are
linear. When you have linear equations you can compose a solution to
a problem as the sum of other solutions. Superposition is a useful technique for solving more complex flow problems as we will demonstrate
in a few of the coming examples.
134
y
x
135
The pressure does not depend on x, thus the x momentum equation is,
gsin() =
2u
.
y 2
gsin() 2
y + C1 y + C2 .
2
The two constants of integration are found from the boundary condition. At the surface of the solid ramp, the no-slip condition holds,
u(y = 0) = 0. Applying this boundary condition shows that C2 = 0.
Therefore,
gsin() 2
u(y) =
y + C1 y.
2
The second boundary condition is that there is no shear stress at the
free surface of the liquid. Nothing is there (other than a little air) to
exert a force on the upper surface of the free fluid film. Thus we need
to set u/y = 0 at the upper surface y = H. Applying this condition
yields,
gsin()
y
u=
y H
.
2
You should always confirm at the end that the solution satisfies the
equation and the boundary conditions.
Also note that the solution here is half a parabola - or essentially one
half of the flow between two solid plates. The driving force in Poiseuille
flow is pressure whereas here it is gravity. Otherwise, the solutions look
very similar.
136
Since the plate is infinite in x there is no reason to think that the flow
at any x location should be different than any other. Thus all terms
with /x are set to zero. Applying this reduction to conservation of
mass yields,
Su
+ v = 0
S
xS y
Therefore, v/y = 0, and the vertical velocity is a constant in the y
direction. Since v = 0 at the plate, v = 0 everywhere. Even though the
flow is transient in time, the argument from conservation of mass about
why the vertical velocity is zero remains the same since conservation of
mass is satisfied instantaneously in an incompressible flow.
We can now turn to the x component of the momentum in the NavierStokes equations and we can discard terms that are multiplied by v or
have x gradients. The complete x-momentum equation with discarded
terms is,
!
2
u 2u
u @u Su
P
S
S
H
+u
+v =
+
+ 2 .
g
x
H
2
SS
@
S
t
x
xSS
y
@ y
x
S
The Navier-Stokes is simplified significantly to
u
2u
= 2.
t
y
137
= 2 2.
t
So clearly the right choice is to set to H 2 / which yields the
problem,
2u
=
.
y2
t
with initial condition u
(t = 0, y) = 0, and boundary conditions
u
(t, y = 0) = 1,
and u
(t, y = 1) = 0.
138
Pressure = 0
U1(y)
H1
U2(y)
H2
L1
L2
Figure 8.4 Schematic for sliding block moving to the left over a
stationary surface. In the reference frame of the block the lower
wall appears to move to the right at constant velocity. While the
gap size is exaggerated here, the solution assumes the gap is very
thin relative to the length of the block.
profile is not plausible since the total flow rate would be different in
the two regions. Since the flow is incompressible, the total volumetric
flow rate in each section under the slider must be the same. Thus, the
manner in which the problem rectifies itself is that a pressure develops
in the thin gap to push fluid out of the gap as shown schematically in
Figure 8.4. If the pressure is high at the step then the flow rate will be
enhanced in region 2 and decreased in region 1.
The velocity profile in each region can be thought of as the superposition of Couette and Poiseuille flow. Using the results derived previously,
the total flow rate in region 1 will be
U H1
dP H13
+
,
Q1 =
dx 1 12
2
and similarly for region 2. Overall conservation of mass would state
that Q1 = Q2 ,
U H1
dP H23
U H2
dP H13
+
=
+
.
dx 1 12
2
dx 2 12
2
The expression gives us the relationship between the two pressure gradients in regions 1 and 2. To close the problem, we need additional
information. We need to know what is the overall applied pressure
across the entire fluid gap. Lets assume the upper slider is open to the
fluid through which it moves. Therefore, the pressure at the two ends
would be the same and there is no overall applied pressure. Since only
pressure differences matter, we can take the far left and right ends of
139
the slider to be zero pressure. If P is the pressure at the step then the
pressure gradient in region 1 would be P/L1 and the pressure gradient
in region two is P/L2 . The difference in sign is because the pressure
rises from 0 to P in region 1 and falls from P to 0 in region 2. We now
have enough information to solve for the pressure at the step, P ,
1
6U
1 H13
P
=
(H1 H2 ).
+
L2
L1 H23
H23
Notice that when H1 = H2 the pressure is zero and we are back to
Couette flow. Also notice that the pressure under the slider is positive
when H1 > H2 and negative when the situation is reversed.
In order to understand the result a little easier, lets explore the case
where L1 = L2 = L/2 and H2 = H1 /2. Substituting in these parameters yields a pressure at the step of,
P =
4U L
.
3H 2
The pressure grows as the gap gets smaller. The positive pressure in
the gap (when H1 > H2 ) provides a lift force on the sliding object.
Since the pressure grows as the layer is squeezed, the fluid can prevent
the two solids from coming into contact. The basic effect is used in
hydraulic bearings where the viscous fluid forces replaces those of solidsolid contact.
Recall from the Poiseuille flow example that the stress acting on the
solid wall is,
!
u
y
.
s=nT=
P
The total normal load that could be supported by the pressure in the
lubrication layer is the integral of the pressure (the y component of
the stress) under the slider. Since the pressure gradient is a constant,
the pressure varies linearly with x and the total normal force per unit
width would be,
FN =
4U L2
.
3H 2
The tangential force is that required to drag the block through the
fluid. The tangential force is found by evaluating the shear stress =
u/y at the wall and then integrating over the length. The shear
140
,
y y=0
L
H
Substituting for P gives,
1 =
4U
U
7U
=
.
3H
H
3H
=
,
y y=0
2L
H
3H
The total tangential force per unit width is then (1 L1 + 2 L2 ), or
FT =
8U L
.
3H
u 2(y)
h2
141
x
y2
Uint
H
h1
u 1(y)
y1
x
Figure 8.5 Poiseuille flow with two fluids of different viscosity in the
channel.
fluid is that now this equation must be solved for both fluids and the
solutions must match at the interface. To solve the axial momentum
equation for each fluid region we will need 4 total boundary conditions.
We will have no slip conditions at the top and bottom walls, namely
u1 (0) = 0 and u2 (H) = 0. The other two conditions come from the
interface between the two fluids. When we have two different fluids in
contact, the velocity at the interface must match; u1 (h1 ) = u2 (h1 ). The
fourth and final condition must be that there is no net force exerted on
the interface; the shear stress in the fluid must be continuous.
We have seen in the Poiseuille flow example how to compute the
shear stress at the interface. The continuity of stress condition at the
interface is,
1
u2
u1
= 2
.
y
y
The statement above is that the stress exerted on the interface from
both fluids must balance. There is a discontinuity of the slope of the
velocity field and the ratio of the slopes will equal the ratio of the
viscosity.
There are two ways forward in this problem. One is to integrate
the two equations and apply the four boundary conditions to get all
constants of integration. A second way is to use the combined PoiseuilleCouette flow solution for each fluid. We can attach a coordinate system
for each fluid problem at the respective solid wall and leave the velocity
of the interface between the two fluids as a variable, Uint . For the height
of the two regions we will use h2 = H h1 . The solution for each region
142
must be,
u1 (y1 ) =
dP 1
y1
y1 (h1 y1 ) + Uint .
dx 21
h1
u2 (y2 ) =
dP 1
y2
y2 (h2 y2 ) + Uint .
dx 22
h2
P H
.
2L h11 + h22
2
1 H
y
P
1.
y1 (h1 y1 ) +
u1 (y1 ) =
1 H
2 H h1
21 L
+
h1
h2
u2 (y2 ) =
P
2 H
y
2.
y2 (h2 y2 ) +
1 H
2 H h2
22 L
+
h1
h2
The flow rates of the two fluids are determined from integration
of the
Rh
velocity profiles across their respective regions, Q1 = 0 1 u1 (y)dy and
Rh
Q2 = 0 2 u2 (y)dy, which yields,
P 3
31 h1 H 2
.
Q1 =
h1 +
1 H
121 L
+ 2 H
h1
h2
P 3
32 h2 H 2
.
Q2 =
h2 +
1 H
122 L
+ 2 H
h1
h2
143
Note that the solution is backwards from how one might do an experiment. In an experiment it would be easy to take two pumps for the
two separate fluids and set their flow rates and then we could measure
the location of the interface. Our equations are derived assuming that
interface location is the known and then we solve for the flows.
t
r
r
r
r
1 rur
r r
!
1 2
u
2 u
r
+ 2 2 2
,
r
r
144
Ri
Ro
Figure 8.6 Schematic for Couette flow between two concentric cylinders. The outer cylinder with radius Ro rotates with angular velocity
.
momentum,
v
ur v
v
v v
1 P
= +
+ ur
+ +
t
r
r
r
r
1 rv
r r
1 2
v
2 ur
+ 2 2 + 2
r
r
!
,
+ ur +
=
+
t
r
r
r
r
!
1 2
1
ru
u
2 u
r
r
+ 2 2 2
,
r r r
r
r
145
momentum,
v
v
v v
ur v
1 P
+ ur + +
= +
r
r
r
t
r
!
1 2
v
2 u
1 rv
r
,
+ 2 2 + 2
r r r
r
r
The equations are significantly simplified just as in the other examples of this chapter. One difference however, is that there is one acceleration term that shows up in the radial momentum balance. This is
the centripetal acceleration. This acceleration is balanced by the radial
pressure gradient. The radial pressure is analogous to the case where
you swing an object tied to the string in a circle above your head and
the tension in the string balances the objects acceleration. The equations we will solve are the radial momentum balance,
P
v2
=
r
r
1 rv
r r
r
Ri
v = Ro R
o
Ri
Ri
r
Ri
Ro
146
U( r )
+ +
= 0.
r r
r
z
147
t
r
z
z
r
1
r r
!
uz
u
1 2
2 uz
z
r
+ 2 2 +
,
r
r
z 2
Since there is only one component of the velocity we drop the subscript
on the uz and just use u to denote the axial velocity for simplicity.
The final reduced momentum equation is analogous to Equation 8.1 in
cylindrical coordinates,
1
u
dP
=
r
.
(8.3)
dz
r r
r
Integrating twice gives
u(r) =
1 dP 2
r + C1 ln(r) + C2
4 dz
dP 1 2
(r R2 ).
dz 4
P R4
,
8L
148
128L
D4
149
150
cylindrical Couette flow problem; we are making the same approximations in the two problems. The final equations we will solve are
conservation of r momentum,
P
v2
=
+
r
r
and momentum,
0=
1 rv
r r
.
v =
C2
.
r
v2
P
=
.
r
r
2 R 4
P
=
.
r3
r
Integrating yields,
2 R 4
+ C3 (z)
2r2
The pressure is a function of z due to gravity and thus the constant of
integration above is a function of z. This expression says the pressure
is lower at the surface of the cylinder than off far away.
P =
151
152
r=R
r=a
inner fluid, i
outer fluid, 0
L
Figure 8.9 Poiseuille flow in a pipe with two fluids of different viscosity. The more viscous fluid surrounds the less viscous fluid.
and the less viscous fluid occupies the annular region around the core.
The flow is axi-symmetric and we will assume the tube is relatively
long compared to its diameter, then fully developed conditions occur.
The velocity profile no longer evolves with distance down the tube and
becomes constant in the axial direction. Under these assumptions, the
axial velocity, u, is a function of r only. This example is the cylindrical
coordinate version of the problem of two fluid layers in contact that we
solved previously.
Following the procedure outlined in the previous examples for Poiseuille
flow, the Navier-Stokes equations simplifies significantly to become the
same as Equation 8.3,
1
u
P
=
r
L
r r
r
where r is the radial coordinate which is zero at the pipe center. The
pressure gradient down the pipe is a constant as in our previous examples. The constant pressure gradient is the same as P/L, where L
is the length of pipe and P is the total pressure drop. We take R to
be the radius of the tube and a to be the radius of the inner core of
viscous fluid.
The two fluids are denoted by subscripts i for inner core region and
o for outer sheath of fluid and where i > o . Thus for the outer region
a < r < R,
P
1
uo
=
r
,
Lo
r r
r
153
=
r
.
Li
r r
r
The equation is the same except for the different values of viscosity.
Integrating the expression for the inner core with respect to r twice
gives,
P 2
ui =
r + C1 ln(r) + C2 ,
4Li
with two constants of integration which are found through the boundary conditions. Since the velocity of the fluid at r = 0 must not be
infinite, C1 = 0. Setting the fluid velocity at r = a to that of the
interface, Uint , we have the velocity field for the inner core,
P
a2 r2 .
ui = Uint +
4Li
For the outer fluid we have the same form upon integration, but use
the boundary condition of no-slip at r = R and the fluid must match
the interface velocity at r = a. The velocity profile becomes,
ln(r/R)
P
P
2
2
2
2
R r +
R a
.
Uint
uo =
4Lo
ln(a/R)
4Lo
In order to find the velocity at the interface we need one additional
boundary condition. At the interface between the two fluids the stress
must be equal; o uo /r = i ui /r. This boundary condition is the
same as in the two fluid problem in plane 2D flow in cartesian coordinates. Using this condition allows us to find,
P
Uint =
R 2 a2 .
4Lo
Finally, we have the full velocity profile of the inner fluid as,
o 2
P
(R2 a2 ) +
(a r2 )
ui =
4Lo
i
and the outer fluid as,
uo =
P
R2 r 2 .
4Lo
You can check that these formula satisfy the equations and boundary
conditions. If you take the limit of the viscous inner core having an
154
infinite viscosity relative to the outer layer, the velocity profile in the
inner core becomes a constant which does not vary with r- it is as
though you are pushing a solid rod down the pipe. It is also interesting
that when the viscosity of the inner core is infinite, only the properties
of the outer fluid determine the flow velocity.
The total volumetric flow rates can be found by integrating the velocity profile over the region of interest in cylindrical coordinates, namely,
Z a
Z R
Qo = 2
uo rdr, and Qi = 2
ui rdr
0
The total volumetric flow is the sum of the these, which simplifies to
P
o
Q=
R4 a4 + a4
.
8o L
i
The expression can be re-arranged to the usual way Poiseuille flow in
a pipe is written.
!
1
128o L
P = Q
o a4
a4
D4
1 R
4 + R4
i
The final relationship is equivalent to Poiseuilles law where the extra
term in parenthesis can be thought of as the effective relative viscosity of the mixture. The expression for the effective viscosity has the
right limits. When a = 0 we limit to the viscosity of the outer fluid
and when a = R we limit to the viscosity of the inner fluid. Example
velocity profiles and a sample of the effective resistance are plotted in
Figure 8.10. It is clear that at high viscosity that the core limits to a
uniform velocity profile. Also, from the effective resistance you can see
that a small amount of water significantly reduces the overall hydraulic
resistance.
As with flow in a pipe of a single fluid, this solution is not always
observable as the predicted flow field can be unstable. Experiments
show complicated flow regimes which can emerge including flows where
155
Figure 8.10 a) Core annular flow velocity profiles for viscosity ratios
of 1, 10, 100 and 1000. b) Effective viscosity as a function of the ratio
of the inner viscous core flow to the total.
the interface between the fluids is wavy, the two fluids separate and go
down the pipe as slugs, and droplet flow where one of phases breaks
apart into small droplets and disperses in the other. While all this
behavior is encoded in the Navier-Stokes equations - it can be very
hard to predict or analyze.
156
The most straightforward method of stability analysis is to take a situation where a simple solution to the Navier-Stokes equations is known.
We can then consider very small perturbations to this solution and then
decide if those perturbations grow or decay in time. Mathematically we
make a substitution where we would say the the velocity is u = u0 + u
157
linear analysis. As you exceed the threshold non-linear effects come into
play and the patterns become ever more complicated and beautiful.
While the topic of linear stability analysis is beyond what we will do
in this class, it is not that difficult. It is a topic found in a number of
textbooks.
Flow in a pipe is a case where the linear stability analysis works
remarkably bad. In this case, the analysis predicts the flow is stable
at all Reynolds numbers, though we know in practice that is not true.
The mechanism for transition to turbulence in simple pipe flow is a
problem which is still worked on extensively to this day. While a lot is
understood about the laminar to turbulent transition in this problem,
it is still an area of research 125 years after Reynolds first made his
observations.
There are many complications with the Navier-Stokes equations but
the point I want to emphasize is simple. It is not sufficient to simply
solve the Navier Stokes equations. We always need to ask whether those
solutions are stable. Predicting stability is sometimes straightforward
and sometimes not. Just be aware that asking the question of stability
is a big deal in fluid mechanics and the complications of stability lead a
number of interesting and beautiful phenomenon. In the end, however,
any analysis or simulations you conduct must be brought before the
ultimate judge - experiments.
158
also estimate about a 10 year lag between when a cutting edge supercomputer problem becomes feasible with a typical low-cost computer.
These two forces, improving hardware and software, continually lower
the financial cost and required expertise to using CFD.
There are different numerical methods which all approximate the
solution of the Navier Stokes equations as discrete chunks. The most
common for commercial CFD packages are based on either finite element, finite volume, or finite difference methods. I suggest that if you
ever work with CFD for a job or in graduate school you learn the
difference. For linear problems such as heat conduction, a commercial
package can accurately solve the heat equation for any arbitrary 3D
geometry with very little effort. A novice user can quickly generate a
good simulation.
Due to the non-linear nature of the Navier-Stokes, we are not yet at
a state of plug and play for all problems. Laminar single phase flows
are at that point today. If we have a 2D or 3D flow where the Reynolds
number is low enough that the non-linear terms are not too strong,
then a modern CFD package can usually do a good job without the user
getting too involved in the details. This does not mean that laminar
flow is always easy, that there arent sometimes difficulties, that you
should run to the computer and shy away from analysis or that you
are always going to get a physical answer. But practically speaking, if
you are interested in a single phase laminar flow you could probably
simulate everything you need without too much trouble or expertise.
All the examples we worked in this section could be set up and run
very quickly using a commercial CFD package. However, as we discussed earlier in the heat conduction chapters, when you just do a
simulation you can miss simple scaling laws, and simple formulas that
are good for basic understanding of trends and also very useful in design problems. Simulation is another tool that can work side by side
with good analysis, but is rarely the substitute.
We will discuss turbulent flows in a later chapter, but simulating
turbulent flows is a much trickier situation and one where you should
know what you are doing. There are fundamental limitations that make
high Reynolds number flows challenging that we will address later. For
now, it is worth noting that if you use a CFD package and get good
results, turn up the Reynolds number and it is guaranteed to stop
working.
9
Inviscid flow, Eulers equation and Bernoulli
(9.1)
Dv
= P + g.
(9.2)
Dt
We previously found that for a human swimming at a reasonable pace,
the Reynolds number was on the order of a million. We can imagine for a car going down the highway it is even larger. For weather
related phenomena a Reynolds number of 1010 to 1015 would not be
uncommon. Since the Reynolds number is a measure of the ratio of
inertia to viscosity, it seems that perhaps dropping the viscous terms
and turning to Eulers equation might be a good approximation for
high Reynolds numbers. Eulers equation looks simpler but its behavior can be quite different than the Navier-Stokes even in the limit of
high Reynolds number. The mathematical difference is in the number
of boundary conditions required. For flow over a solid surface, Eulers
equation would only require (and could only enforce) the boundary
condition that flow does not penetrate the surface. Fluid would slip
over the surface in Eulers equation. Since the Navier Stokes equations
has higher spatial derivatives (in the viscous terms), it requires more
boundary conditions. The Navier Stokes requires that flow does not
penetrate the surface and that the flow does not slip at the surface.
This subtle difference can lead to quite different behavior. Mathematically, the zero viscosity limit is called a singular perturbation. This
160
means that we take the limit of the small parameter (viscosity) going
to zero and we fundamentally change the problem at hand.
In the last chapter on Navier-Stokes solutions we studied examples
where the balance of forces was between viscosity and some driving
force such as applied pressure. In all the problems in the last chapter,
the fluid had no acceleration. When using Eulers equation we are ignoring viscosity and the balance of forces is between acceleration and
pressure (or gravity). A good strategy for cases of high Reynolds number flows where fluid acceleration is important is to start a problem
by assuming that Eulers equations work. We could then check what
the predictions and trends are versus what experiments and experience
say to decide if the approximation was reasonable. We will begin to
understand the breakdown of Eulers equations better after the next
chapter when we discuss boundary layers.
2
which, when density is constant, is equivalent to
1 2
v + P + gz .
v ( v) =
2
We discussed this previously, the curl of the velocity vector v is also
known as the vorticity . Vorticity is a measure of the local solid body
rotation of a fluid particle. It can be useful in a number of settings, but
it is a little hard to wrap your head around what it means. Basically if
you measured a flow and plotted an image of the vorticity field, it would
161
be highest where the fluid was spinning (i.e. in tornadoes, toilets, and
whirlpools). It is a property of the curl operator, the vorticity always
points perpendicular to the velocity vector. If you have a 2D flow in the
plane of a sheet of paper, the vorticity will point out of the page. We
will discuss vorticity more in a coming section, for now you can consider
it a mathematical thing. Rewriting conservation of momentum with the
vorticity yields,
1 2
v + P + gz .
v =
2
Note that at this point we really havent done anything, just rewritten the same equation in a different form using some vector calculus
identities.
Lets take the whole equation and take the dot product with the
velocity vector itself,
1 2
v + P + gz
.
v v =
2
Notice the term on the left side. The curl of the velocity and the vorticity vectors will always be perpendicular to those two vector. Thus
the dot product with the velocity vector itself will be zero. This fact
can also be easily proven by carrying out all the terms in component
form. Since the left side is always zero, we obtain
1 2
v + P + gz .
0=v
2
This fact means that the quantity in parenthesis does not change in the
direction of the velocity vector. In a fluid flow we define a streamline
as a line that follows the velocity vectors. In a steady flow a streamline
will correspond to the path a blob of injected dye would follow. Thus
in a steady flow,
1 2
v + P + gz = a constant along a streamline.
2
(9.3)
162
1
2
Figure 9.1 Venturi meter. The cross sectional area at point 2 is less
than at point 1, thus the flow accelerates. Since the velocity at the
throat is faster than the upstream, Bernoullis equation predicts the
pressure is lower at the throat.
per unit volume is 12 v2 , and the potential energy is gz. These expressions should look familiar from particle mechanics only here we use the
mass density rather than the total mass. Pressure exerts a force per
unit area, thus a change in pressure between two locations indicates
that the pressure is doing work on the fluid. This work is related to the
familiar expression that work is equivalent to the integral of force over
distance. Bernoullis equation is saying that the work done by pressure
is equal to the change in energy (kinetic plus potential).
While simple and powerful, Bernoullis equation can also be misleading. There is a tendency to want to use it under conditions that are
not appropriate. We must always ask if the assumptions are met when
using Bernoulli, and ultimately determine whether this simple equation
provides predictions which match reality. The most common mistake is
using Bernoullis equation when viscosity is not negligible.
163
from conservation of mass. Here v1 and v2 are average velocities. Applying Bernoulli from point 1 to 2 yields
1 2
1 2
=
v + P
v + P .
2
2
1
2
Combining these expression yields,
1
1
1
P1 P2 = v2 2 v1 2 = v12
2
2
2
A1
A2
2
1
164
V
r
=
.
r
r
165
H(r)
z=0
z
r
Figure 9.3 Schematic for the example of computing the free surface
shape of a cylindrical tank of water in solid body rotation.
P
v2
=
r
r
2 r =
P
.
r
1 2 2
r
2
The pressure is higher at the wall of the container than the center. The
vertical momentum equation is simple since there is no vertical flow,
namely, at a constant radial location (r = R),
P (z, R) P (z = 0, R) = gz.
If we set z = 0 to the surface of the liquid at the center (r = 0) of
166
1 2 2
r .
2
1 2 r 2
.
2 g
You can confirm the result that the surface of the fluid is a parabola
by spinning a cup of fluid. You will see the shape of the surface looks
parabolic. We could have obtained this result using dimensional analysis quite readily. We would not have known the factor of 1/2 if we had
done dimensional analysis.
9.3 Vorticity
167
Center
pressure
Wall pressure
X
Figure 9.4 Schematic of the pressure distribution in a venturi meter
comparing the center pressure to the wall pressure.
9.3 Vorticity
This section is going to rely on a fair amount of vector calculus and
identities. If you do not remember all these identities, then dont get
too worried. Try to follow the derivation, but focus on understanding
the consequences. First lets restate that vorticity is a vector defined as
the curl of the velocity,
= v.
(9.5)
168
9.3 Vorticity
169
+ v v = 0
t
(9.8)
grad v term.
9.3.1 2D flows
Lets restrict our analysis to 2D flows to make things a little easier.
First of all in a 2D flow, velocity only has components in u and v and
gradients in x and y. Vorticity in a 2D flow only has a component in
the z direction; v = [u v 0]; = [0 0 z ]. Thus the term in Equation
9.9 on the right hand side is zero since the vectors are not in the same
plane. In a 2D flow,
D
= 0 in 2D flows.
Dt
This equation states that vorticity is a material property. Whatever
vorticity the fluid particle has initially, it retains that amount and then
simply goes with the flow. Since the vorticity for a fluid particle is
conserved, if we draw a loop which is a material loop (a massless,
strechable string with no elasticity that can go with the flow), the
vorticity inside that loop is the same for all time. Thus the integral of
the vorticity or the circulation is constant,
D
= 0 around a closed material loop.
Dt
This statement is known as Kelvins circulation theorem. It is an interesting theorem and is more a general theorem than the simple argument
170
Figure 9.5 Drag a spoon slowly through a bowl of water. You will
see two vortices form behind the behind of opposite sign but equal
strength. The two vortices will cancel each other such that the total
circulation around the outer dashed contour is zero.
I made here. A formal proof of the theorem would show that the theorem is valid if Eulers equations hold along the contour around which
the circulation is calculated. Thus, there can be viscosity inside the
closed loop. The theorem also holds in three dimensions.
So how does this theorem manifest itself? Try the following experiment. Take a bowl of water and sprinkle some tracer particles in the
water; fine black pepper works well. Take a spoon and give the surface of the fluid a little flick. You should see two vortices generated.
They will be of opposite spin and propagate themselves for a bit before
decaying away. Now, imagine you draw your loop to calculate the circulation around the spoon and far away from it. Since everything is at
rest intially, the circulation is zero. Now you flick the spoon and create
your vortices. So even though viscosity has done something inside the
loop (we could not create the vortices without it), the circulation theorem still applies far away from the loop. So the circulation is still zero
even though there is vorticity inside the loop. The circulation theorem
states that the circulation is zero and remains so. This means that the
positive and negative vorticity you generated must cancel each other
out via the definition of circulation as the vorticity integrated over the
area (via Stokess theorem). In this experiment you get two vortices of
equal strength and equal size, such that the total integrated vorticity
is zero. The theorem tells you nothing of the vortices decay, just that
the total circulation should always be zero.
You may have noticed that the equal sized but oppositely signed
vortices propagate themselves in a straight line. This behavior can be
9.3 Vorticity
171
understood by using an ideal vortex which is a solution to Eulers equation. The ideal vortex has a velocity field given as
v (r) =
2r
rd =
d = .
v (r)rd =
2r
2
0
0
0
The circulation is a constant regardless of the radius of the circle we
choose to perform our calculation along. The vorticity, which only has
a component in the z direction is,
z =
1
1 rv
=
=0
r r
r r 2
172
Solid
wall
Image vortices
Figure 9.6 Motion of two ideal vortices of opposite sign approaching a wall. The velocity field of each vortex propagates the other.
As the vortices approach the wall, the introduction of an imagined
image vortex can meet the boundary condition at the wall of no flow
penetrating the surface. As the four vortices approach the wall, the
induced velocity fields cause then to spread apart.
push each other outward if you consider the action of each vortex on
the other.
9.3.2 3D flows
In 3D, the above analysis just gets a little more complicated. First of all,
the circulation theorem remains, so without proof we will simply state
that in 3D the circulation is constant around a material loop. However,
simple applications of the circulation theorem (at least in this course)
are usually simple 2D approximations to give us some qualitative understanding of the flow.
The vorticity equation in 3D is,
D
= v.
Dt
and has a non-zero term on the right hand side. Vorticity for a material
point is not constant in 3D as there is an extra term in the equation.
This extra term accounts for things which cannot occur in a 2D flow.
One effect is vortex stretching. If a vortex is stretched out by a flow,
it intensifies, just like an ice skater speeding up as they pull their arms
in. If a vortex is squashed, the vorticity decreases. The stretching of
the vortex can be see in any bathtub drain. The vortex gets intensified
9.3 Vorticity
173
174
tubes which extend far across the sky. If you observe these streaks
closely you will notice that some distance away from the plane, the
individual tubes start developing a wavy character. These waves are
due to a fluid instability known as the Crow instability. Notice that
as the amplitude of the waviness increases the two tubes will interact,
cross, reconnect, and form a series of vortex loops. These circles are now
closed vortex tubes. How well you can make these observations depends
on the weather as the level of atmospheric turbulence can disrupt the
vortex visualization.
175
176
a)
b)
Figure 9.7 A flat plate airfoil in potential flow. In (a) there is no
circulation and streamlines wrap around the rear of the airfoil. In
(b) there is circulation and the rear streamline leaves the airfoil at
the trailing edge.
The flow experimentally, shows that the streamlines leave the body
at the trailing edge. The only way we can make the calculated and
observed flow match is that if we add a net circulation to the flow. This
circulation effectively moves the point where the streamlines connect
with the solid airfoil. The behavior with and without circulation is
shown in Figure 9.7 for a flat plate airfoil.
This seems arbitrary? Why would there be net circulation? The
source of the circulation is subtle. Imagine we start an airfoil from
rest. We draw loop around the airfoil and since everything is at rest,
the circulation is zero. If we impulsively start an airfoil, we would see
that a vortex gets shed from the airfoil at the instant that it starts.
This is just like moving the spoon in the water. However, the circulation theorem would tell us that the circulation around the loop is still
zero. Therefore, this starting vortex would need to be compensated by
some other circulation of opposite sign. This is the circulation around
the airfoil. When we impulsively start the airfoil, viscosity acts to move
the point where the streamline separates from the airfoil. In doing so,
vorticity is created and the counteracting circulation around the airfoil
must keep the net circulation at zero.
The idea that the circulation must be such that it moves the streamline to the trailing edge is not something that we can prove. It is just
a useful rule of thumb that turns out to work as long as the airfoil is
not at too steep of an angle of attack. This rule of thumb is known as
the Kutta condition. Once you know the flow field from the solution
177
Figure 9.8 Schematic of the starting vortex for an airfoil. The circulation calculated around the outer oval must be zero, so the circulation around the two inner loops must cancel each other out. The
airfoil must have circulation of the opposite sign as the starting vortex. If the airfoil is suddenly stopped a vortex of the opposite sign as
the starting vortex is shed in order to maintain no net circulation.
of Eulers equation, you can find the pressure from the fact that in
the absence of gravity effects, P + 12 ()2 = 12 U02 . Once you know
the pressure, you you can find the lift on the airfoil by integrating the
pressure around the surface of the airfoil and it turns out to actually
agree pretty well with experiments.
10
High Reynolds number flows
The analysis of fluid flow via Eulers equations seems relatively nice.
While we presented only a few highlights in this book you should have
gotten the impression that a number of problems could be readily solved
mathematically or computationally using Eulers equations for incompressible flow. So while we know viscosity is always with us, it is a
small force compared to the others in the problem in a number of applications. At the scale of humans, everything is high Reynolds number.
However, it has been observed for a long time that Eulers equations do
an unusually bad job at making useful predictions. It was not until the
early 1900s when Prandtl introduced the idea of the boundary layer,
that people began to appreciate why Eulers equations were so poor
(Anderson (2005)).
If we have a high Reynolds number flow around an object such as
an airfoil, what is observed is that the fluid velocity just away from
the surface of the airfoil very closely matches the solution to Eulers
equations. This part of the flow is not influenced by viscosity. As we
approach the surface, the flow velocity goes to zero because the no-slip
condition is obeyed. However, the size of this layer is extraordinarily
thin. Flow over an airplane wing could have boundary layers measured
in the units of millimeters. Why should this thin region play such a
major role in determining flow? The answer is the topic of this chapter.
We need to keep in mind all the forces at play in our problem. When
we looked at solutions to the Navier-Stokes equations we found we
could readily solve problems where there was no fluid acceleration - the
balance was between viscosity and some driving force such as applied
pressure. When we discussed Bernoullis and Eulers equation, we con-
180
t0 2 u
= 2
.
y2
t
This scaling indicates that the thickness of the region where the flow
181
transitions from the plate velocity, to being at rest grows with the
square root of time and depends on the kinematic viscosity of the fluid,
= t.
This scaling argument is confirmed by the analytical solution to this
problem, which we will not derive but has the form,
y
u(y, t) = U 1 erf
.
2 t
where erf is known as the error function. If you are unfamiliar with
the error function, you can look up its definition and you will find
that most mathematical software will understand the definition and
generate a plot of the function for you. The solution is shown in Figure
10.1. Since = 106 m2 /s in water, you can see that the length scale
of the boundary layer is pretty small. If we pulled the plate for an hour
in water, the thickness of the boundary layer would be 6 cm.
What is interesting about the analytical solution is that the solution
182
Boundary
Layer
t, which
= f (Re)
L
where the Reynolds number based on the plate length is ReL = U L/.
183
The boundary layer thickness can only depend upon the plate Reynolds
number.
We could estimate the dependence by a simple
argument using the
solution to the impulsively started plate, t. In the boundary
layer problem, there is no time. At a given downstream location, the
time is the time it takes fluid to reach that point, i.e. t L/U .
r
L
t
.
U
In terms of the Reynolds number this equation can be rewritten as
r
r
.
L
U L
Re
Since the flow progresses downstream, we the argument above works
for any x location. Thus we might expect,
r
r
,
x
U x
Rex
where we use the Reynolds number based on the x location. This scaling
suggests that the boundary layer thickness grows as the square root
of the distance down the plate. To get a sense of scale, lets take air
= 106 flowing at U = 1 m/s over a 1 m length plate. The boundary
layer would be 1 mm. If the velocity were increased to 100 m/s (an
airplane taking off), the boundary layer thickness would be 100 microns.
The full boundary layer problem over a flat plate has a mathematical
solution. This solution makes use of the fact that /x is usually a pretty
small number. The basic idea is that when there is a vast difference in
scales, one can make the equations dimensionless using the different
scales. So we will scale the x coordinate
q by the length of the plate, L,
. We will also use different
and the y direction is scaled by = UL
y
This equation implies that V0 = U /L would be a good choice. This
scaling implies that the vertical velocities are less than the horizontal
velocities by a factor of /L. If you think about this comment for a
184
u
2u
+ v
=
+
+
u
y
x
U L L2 x
2
y2
and
2
L2
v
u
+ v
x
P
2
=
+ 2
y
L U L
2 2 v 2 v
+ 2
L2 x
2
y
.
Since /L is a small number and we have scaled all the partial derivative
terms in our equation to be on the order of magnitude of 1, we can take
the perhaps somewhat uncomfortable step of removing all terms with
small numbers out in front. This step seems odd the first time you see
it. However, it is pretty standard practice. It is a logical thing to do.
We have supposedly scaled all the partial derivatives in the equation
to be on the order of 1. Thus the terms with (/L)2 simply go away
because they are small. Taking this leap of faith provides the boundary
layer equations,
u
P
1 2u
+ v
=
+
x
y
x
Re y2
(10.1)
185
and
0=
P
.
y
(10.2)
The second equation simply states that the pressure across the boundary layer is a constant. The fact that the pressure across the boundary
is constant is really useful. What you can do, in practice, is solve Eulers equations for flow around your object. This solution gives you the
pressure at the surface of the object. You can assume that since the
boundary layer is so thin that the pressure at the surface of the object and the pressure at the edge of the boundary layer would be the
same. Thus, you get the pressure and Px from the solution to Eulers
equation, and then use the now known pressure gradient to solve the
boundary layer equations, Equation 10.1.
In the flat plate problem, the pressure far away from the plate (in
units of ) is just a constant and there is no pressure gradient. Thus,
our final boundary layer equations for the two components of velocity
are,
u
1 2u
+ v
=
u
y
Re y2
u
v
+
=0
x
y
The boundary layer equations describe a balance between viscosity and
fluid acceleration.
There is a mathematical solution to this flat-plate problem, and we
will skip the details and only show the result. The analytical solution
is not really a closed form answer. The mathematical solution simply
transforms this partial differential equation into an equivalent ordinary
differential equation. This ODE still needs to be solved numerically.
The solution to the velocity profile is shown in Figure 10.4.
Once the boundary layer equations are solved numerically, you can
calculate everything you want to know about the boundary layer. The
thickness, defined as the distance when the velocity field is 99 % of the
free stream is,
5
=
.
x
Rex
The factor of 5 comes from the solution, but the form was acquired
186
U
.
x
(x)
0.664
,
=
2 /2
U
Rex
f
1.328
=
.
2 /2
LU
ReL
The numerical values come from the calculated numerical solution. The
really important point is that the basic scaling of the answer comes from
consideration of dimensional analysis and some physical arguments.
187
= f (Rex ) .
x
Thus a small number of experiments can provide an empirical solution to the boundary layer thickness. Empirical fits to the data given
boundary layer thickness as,
0.37
=
,
x
Re1/5
x
188
Y
laminar
Turbulent
a)
b)
Cf =
0.0592
Re1/5
x
These are not fundamental laws and there is no mathematical derivation, the expressions just fit the data well.
The turbulent velocity profile is much blunter than the laminar one.
The random turbulent eddies pull high momentum fluid from the free
stream and bring it close to the solid surface. The drag is much higher
in turbulent flow. The high momentum fluid sits closer to the plate as
shown schematically in Figure 10.5a. The boundary layer thickness and
coefficient of friction as a function of distance down the plate are shown
in Figure 10.5b. There is a jump at Re = 105 in the two solutions. In
reality, there is not a jump but a transition regime from one solution to
the other. The thickness of the boundary layer rapidly expands when
the boundary layer becomes turbulent. Because the velocity gradient on
the blunt turbulent velocity profile is greater than the laminar one, the
shear stress at the solid surface is greater for the turbulent boundary
layer.
189
190
b)
a)
Figure 10.6 a) Streamlines for irrotational, inviscid flow over a cylinder. b) The pressure along the surface of the cylinder as a function
of angle. The data are plotted as the coefficient of pressure which is
defined as Cp = P1 P
. The pressure at the forward and rear stagU 2
2
nation points is higher than the ambient pressure while the pressure
at the top of the sphere is the lowest. The forward and rear stagnation points are at 0 and .
191
v
v =
P + 1 2 v
+v
,
Re
t
v
= 0.
In theory we can solve these equations to obtain the velocity field and
pressure at every point. However, we now should know that the dimensionless velocity field will only depends upon the Reynolds number.
To compute the net force on the sphere we would need to integrate the
stress vector around the surface of the sphere. The stress vector is the
dot product of the normal vector and the stress tensor. Mathematically
the force is found,
Z
F = n (P + (v + (v)T ))dS
Computing this force from our dimensionless solution (and remembering to make dS dimensionless) yields,
Z
U0
2
v)T ))dS
F=D
n (P0 P +
(
v + (
D
where we are now integrating around the surface of the unit sphere.
This can be written as,
Z
U0
v)T ))dS
F = D2 P0 n (P +
(
v + (
DP0
U0
1
where, DP
. Recall that in our dimensionless equations, P0 =
= Re
0
2
U0 , pressure scales with the kinetic energy of the flow. Thus,
F
= f (Re).
U02 D2
The drag force normalized by U02 D2 (known as the drag coefficient)
is only a function of the Reynolds number. Both the solution to the
whole velocity field and the calculation of the drag force only depends
on this one parameter.
We could have achieved the same result from doing a dimensional
analysis with the table method and using the Pi Theorem. As long as
we know the parameters that matter and pick the right ones we know
the table is easy to generate. In the example above, we used a mathematical statement of the problem to guide us to a precise statement of
192
193
top or bottom of the sphere, they would be straight and thus the pressure would be the ambient. Thus the pressure in the wake should be
something close to the ambient pressure. Therefore, very crudely, the
force acting on the front half of the sphere due to pressure is 21 U 2 r2 .
The force in the back half is zero. Thus the net force due to pressure
drag is on the order of 21 U 2 r2 . Thus the drag coefficient is about 1.
Pressure drag dominates over friction at the surface in this regime.
However, we said that separation occurs because momentum is deprived from the fluid in the boundary layer. When the boundary layer
becomes turbulent, the velocity profile gets blunter. The transition from
zero velocity at the wall to the free stream value occurs over an even
smaller distance than the laminar case. Thus, high momentum fluid is
even closer to the wall. The high momentum fluid near the wall allows
the turbulent boundary layer to push into the adverse pressure gradient
further than the laminar counterpart. Separation is delayed and occurs
further down the back of the sphere. The low pressure wake acts over
a smaller area when the boundary layer is turbulent rather than laminar. Pressure drag is decreased when the boundary layer is turbulent.
So even though the steeper velocity gradients in a turbulent flow increase the friction at the surface, the net drag is dominated by pressure
drag and thus the overall drag force (coefficient) decreased when the
boundary layer becomes turbulent.
The role of boundary turbulence in reducing drag is the reason for
dimples on golf balls. The irregular surface induces turbulence to occur
at much lower Reynolds number. The turbulent boundary layer delays
separation and the golf ball goes further. Imperfections on the surface
of a sphere and its effect on tripping the boundary layer into a turbulent
state can help explain why curveballs curve in baseball.
194
195
196
197
where L is the length scale of the largest eddies and the start of the
cascade. For an atmospheric flow we can easily reach Reynolds numbers
of 1012 or higher; there are indeed many orders of magnitude of length
over which the turbulence cascade exists.
P
We can describe the flow via a spatial Fourier series u(x) = Ak eikx ,
where the wavenumber k is like a spatial frequency. Kolmogorov argued
from physical arguments and dimensional analysis that the energy per
wavenumber follows a scaling,
E(k) k
5
3
198
where
Z
Total kinetic energy =
E(k)dk.
199
range of current computing capability. So now lets say that we decide that we only care about the big swirls so we decide to make our
grid spacing 1 mm. Certainly this will suffice? Now we have 300 grid
points in each direction; still a lot but not outside of what a modern
computer and software could handle in 3D. However, 1 mm is 20 times
greater than the length scale where viscosity can remove energy from
the system. What happens is that energy will be put in at the large
scale, that energy will cascade to smaller scales, and when the energy
in the simulation reaches 1 mm (the grid spacing) it will just sit there.
There is no mechanism to remove energy from the simulation. Energy
will build up and the simulation will die a sad numerical death.
There are fixes to this problem by adding models which act as numerical sponges to soak up all the grid scale energy from the simulation.
Methods called Large Eddy Simulations (LES) have this feature of being able to resolve our stirred bucket by modeling, but not resolving,
the smallest scale motions. The LES models are often called turbulence
closure models. There are many different models that have advantages
and disadvantages in different situations. There is no single model that
always works the best and it requires training and experience to do LES
of turbulent flows well. You can easily see that this turbulence closure
problem will be with us for a while. For an atmospheric flow which
effect the climate, the Reynolds number can be 1015 the large scales
structures are the sizes of continents. We are going to be continue to
need closure models in weather and climate modeling in my lifetime.
Most modern CFD packages have a number of different turbulence
closure models built in. The variety and character of these models is
too extensive to cover here. In the software, it can be just as easy as
clicking from one model to the next. However the different models have
different limitations and cases where they work better or worse. Just
be warned that you should be skeptical of the results unless you have
had more training and experience than you are getting from this brief
chapter.
Many people think that CFD is the answer to everything fluids. The
impulse for many people when confronted with a fluids problem is to
run to CFD. CFD can be powerful, but it is a tool that can be easily
misused. If you are ever working somewhere and your boss wants to
do CFD of a turbulent flow, be cautious going down this road unless
you have learned more about fluid dynamics than this course. It is very
200
11
Control volume analysis
202
problems (including the ones Ill demonstrate) that tend to gloss over
the assumptions that are made as though you just believe they are
true. Like dimensional analysis, the procedure of control volume analysis is straightforward, it is the proper use of it that takes some physical
insight that only comes with experience. The control volume is an excellent tool, but keep in mind that there are also problems where doesnt
really help us much.
Control volume analysis is an excellent tool for getting order of magnitude estimates or simple scaling laws. Often in engineering situations
we care about the approximate magnitude of the total force or the pressure, for example, and a simple control volume analysis can be more
useful that a complex simulation. Often we can get a useful and reasonable answer with just a little effort. Control volume analysis is also
a quick way to check other more detailed calculations or simulations.
If you do a full analysis of the Navier-Stokes equations and solve for
the full velocity field, but that field does not satisfy the overall control
volume balance then you have done something wrong.
Many textbooks on fluid dynamics put the control volume analysis very early in the development. A good reason for doing so is that
the analysis is much easier than analysis with the Navier-Stokes equations. However, when confronted with a real world problem that we
dont find in a textbook executing a useful control volume analysis requires some physical intuition. Similar to dimensional analysis, control
volume analysis is usually most powerful when coupled with good physical understanding and intuition about the important effects. Similar to
dimensional analysis, it can be straightforward to learn the technique
of control volume analysis, but using it wisely comes with experience.
Dv
= g + T dV
Dt
203
We can now move the integral on the left hand side to a different form
using the Reynolds Transport Theorem,
Z
Z
Z
Dv
(v)
dV =
dV + (v)v ndS.
Dt
t
Using the divergence theorem we can also convert the stress tensor
term on the right side of the original equation back to integration over
a surface,
Z
Z
Z
Z
(v)
dV + (v)v ndS = gdV + n TdS.
t
Substituting our constitutive law for a Newtonian fluid in incompressible flow (T = P I + 2S), we have,
Z
Z
Z
Z
Z
(v)
dV + (v)v ndS = gdV P ndS + 2 n SdS.
t
Since we are now interested in a control volume fixed in space, the
derivative on the left hand side can be moved outside the integral, so
Z
Z
Z
Z
Z
d
vdV + (v)vndS = gdV P ndS+2 nSdS. (11.1)
dt
The above expression holds for the fluid only. A more useful form includes an external force acting on an object in the flow. In this case,
you can draw a control volume around whatever you like, as long as
you remember to include the force which is holding the object in place.
Z
Z
Z
Z
Z
d
vdV + (v)vndS = gdV P ndS +2 nSdS +Fext .
dt
(11.2)
Following the same ideas, the integral form of conservation of mass for
a fixed control volume is
Z
Z
d
dV + v ndS = 0.
(11.3)
dt
The integral form of the momentum and mass conservation equations
for a fixed region can be very useful in a number of cases. The primary
use is in making estimations/calculations of net forces acting on an
object in a flow.
You should first note that conservation of momentum is a vector
equation - 3 components for x, y and z. On the left side of Equation
11.1 we have the rate of change of momentum inside the control volume
204
(first term) and the net flux of momentum coming into/out of the
control volume (2nd term). In the second term on the left side, note
that the momentum v is a vector quantity, while v n is the rate
that the momentum crosses the surface. On the right side, the first
term is the net body force which acts on the whole volume. The second
term is the force due to pressure and the third term is the force due to
viscosity - both of these exert forces only on the surface of our volume.
What is nice about this expression is there are many cases where we
are only interested in the net force on an object, and in those cases we
can often use the integral form of the equation without worrying about
the details of the flow.
R
Getting the sign right of the term (v)v ndS can be a little tricky.
You have be be careful that the velocity vector has a sign, as well as
the term v n. Since we take normal vectors to point out of the volume,
then v n is positive for outflow and negative for inflow.
11.2 Examples
The only way we can make sense of the complex looking equation is to
work through a few examples. The basic approach I usually take with
these problems is as follows. I start by drawing a sketch and picking
a control volume that seems convenient. A convenient control volume
often (but not always) has surfaces where the velocity inlets and outlets
are normal to the surface. I like to write out the entire control volume
equation and then systematically cross out terms that I have good
reason to ignore (say unsteady effects in a steady flow). I then would see
where that takes me. If I can get an interesting result without making
many assumptions, then great. If I dont get much then I might start
making assumptions that I am less confident are true (say ignoring
viscosity in a case where it is not clear I can). I like to try and keep
track of my assumptions so when I get to the end of a problem, it is
easy for me to review what I had to do to get there. One of the most
common assumption you will tend to make in these problems is that
the velocity field is uniform over some region, an assumption we often
make because we have little choice.
11.2 Examples
205
A2 n
Fx
A1
Fy
Figure 11.1 Control volume for finding the force a water jet exerts
on an inclined block. The control volume is shown as the dashed
line.
206
evaluate the surface integral without knowing the velocity field of the
inlet and outlet jet. Lets assume for now that the jet enters and exits
with a uniform velocity field, U1 and U2 respectively. On the inlet the
normal vector and the velocity are not aligned so the dot product is
negative and on the exit, the velocity and the normal vector are aligned
so the dot product is positive. Since the velocity at the inlet and exit
is assumed constant, then the surface integral is easy to do,
Z
v ndS = U1 A1 + U2 A2 = 0.
This statement simply says that the total mass flow rate in equals
the total mass flow out. Note that when we write the surface integral
above, it means that we have to evaluate v n at every point along the
control volume surface and integrate it up. In this example it is only
the two little patches where fluid comes in and goes out that has any
fluid velocity.
Now lets consider the momentum equation. Lets start by writing
out the full integral equation and cross out the unsteady terms and
gravity which we previously stated we would neglect,
Z
Z
Z
H
Z
Z
Z
d H
P ndS +2 nSdS +Fext .
Z
H
vdV
+
(v)vndS
=
gdV
Z
dt HH
H
Z
Notice that the surface integral terms for pressure and viscosity are
evaluated only on the surface of the control volume. So while viscous
forces and pressure may do something inside the control volume on the
surface we have drawn it would be reasonable to assume that they have
no contribution. Since the control volume is open to atmosphere everywhere, the pressure around the surface of the control volume would be
atmospheric. If we integrate a constant pressure around a closed loop,
then the integral of P n around the loop results in zero net force. While
viscosity certainly acts along the surface of the block, along the surface we have drawn the control volume boundary viscosity would not
seem to play a role since we are assuming uniform flow outward. Recall
that viscous stresses only occur when we have velocity gradients, so
a uniform free jet with uniform flow has no viscous stresses. You are
starting to see that in order for us to make progress we have to start
to use some physical insight. By making these physical arguments to
neglect the pressure and viscous terms on the control volume boundary,
11.2 Examples
207
Note that the dot product of v n = U1 since the velocity vector and
normal vector are not aligned. Since the velocity at this surface is all
in the x-direction, then v = U1 . On the outlet at surface 2,
Z
Z
(v)v ndS = (U2 cos())(U2 )dS = U22 A2 cos().
2
Note that the dot product of v n = U2 since the velocity vector and
normal vector are aligned. Since the velocity at this surface is inclined
at angle the x-component of the velocity is U2 cos(). Putting this
together gives us the external force as,
U22 A2 cos() U12 A1 = Fx .
Substituting in conservation of mass (U1 A1 = U2 A2 ) yields,
U1 A1 (U1 U2 cos()) = Fx .
So our final result still doesnt tell us everything we want since U2 is
not yet determined. We presumably know U1 and A1 as that is given
by the size and flow rate of the hose. Without knowing U2 or A2 we
have no way to get the force. If we measured these quantities we could
proceed. We could also make an assumption that A1 A2 . While this
seems like a reasonable thing to assume, it need not be true. If the area
is the same then mass conservation says U1 = U2 and the x component
of the force would be,
U12 A1 (1 cos()) = Fx .
In this case, the force is related to the deflection of x momentum out
of the control volume. The rate that momentum is carried into the
control volume is U12 A1 . The term (1 cos()) represents the fraction
of x momentum that is deflected. If the angle were zero, a flat plate,
then the force would be zero.
208
Note that the dot product of v n = U2 since the velocity vector and
normal vector are aligned. The dot product term is the same in the x
and y momentum equations. Since the velocity at this surface is inclined
the y-component is v = U2 sin(). The total y component of the force
is,
U22 A2 sin() = Fy .
When we make our assumption that the area of the jet does not change,
U12 A1 sin() = Fy .
Thus the final result for the case where the area of the jet does not
change is
cos() 1
Fext = U12 A1
sin()
11.2 Examples
209
While this result assumes the constant velocity profile and the fact that
the jet area remains the same from inlet to exit, it seems reasonable
to think that scaling, order of magnitude and trends might be quite
reasonable.
210
U0
Fx
U0
v(x)
u(y)
U0
Figure 11.2 Control volume for finding the drag on an object from
an experimentally measured wake.
a very viscous fluid coming from our hose, we would expect there to
be a more dramatic change in the increased area of the jet. Using the
control volume approach we can only compute the force on the plate if
we had a measurement of the exit area.
If the fluid had very low viscosity (high Reynolds number), we could
apply Benoullis equation. If we applied Bernoullis equation from the
inlet to the exit, we would see that just like in the last example the
velocity should not change. Since Bernoullis equation is only valid for
inviscid flow, our conclusion is consistent with the fact that at high
Reynolds number, we expect the inlet and outlet areas of the jet to be
approximately equal.
11.2 Examples
211
The order of the terms are left, right, and top/bottom (hence the factor
of 2). Note that on the inlet (left boundary) the velocity and normal
vector point in opposite directions hence the negative sign on that term.
Conservation of mass simply states that since there is a velocity deficit
in the wake of the object, there must be some flow out the top and
bottom surface.
Conservation of momentum is,
Z
Z
Z
H
Z
Z
Z
XX
d H
HH
X SdS +Fext .
Z
H
+
(v)vndS
=
gdV
P
ndS
+
2
vdV
XX
Z HH nX
dt HH
H
Z
We have crossed out terms since 1) the flow is steady, 2) we will ignore the effect of gravity (or assume perpendicular to the page), 3) the
object is in an external flow field and thus the pressure is atmospheric
everywhere, 4) that along the boundary of control volume which is far
from the object there are no significant viscous stresses. Under our assumptions, the net force on the object is given by the imbalance in the
momentum coming in and out of the control volume.
Z
(v)v ndS = Fext .
Now, lets evaluate our terms for each surface for the x component of
the momentum,
Z H
Z H
Z L
2
2
U0 dy +
u(y) dy + 2
U0 v(x)dx = Fx .
0
The first term is the inlet. Since the flow and the normal vector are in
opposite directions, v n = U0 and the x component of the velocity
is U0 . The exit integral is computed similarly, only the sign is positive
since the velocity and flow direction are aligned. The top and bottom
terms should be treated with care. For the top and bottom, the outflow
is v n = v(x). The x component of velocity is U0 since the surface is
far from the object.
212
A2
U1
A1
U2
Figure 11.3 Control volume for finding the net force required to
hold the nozzle on a pipe.
U0 dy+
u(y) dy+U0
U0 dy
0
u(y)dy
= Fext .
11.2.4 Nozzle
Consider the tapered nozzle as shown in Figure 11.3. Lets consider a
steady flow rate and we want to estimate the force required to hold the
nozzle in place at the flange. We draw the control volume as shown.
We will assume steady flow, ignore gravity, assume the inlet and outlet
flow fields are uniform, and assume a high Reynolds number flow.
At this point, we can probably do the conservation of mass pretty
easily,
Z
H
Z
dH
dV + v ndS = U1 A1 + U2 A2 = 0.
H
dt H
H
11.2 Examples
213
214
Reynolds number and apply Bernoullis equation from the inlet to the
outlet. Bernoulli would yield,
1
1
P1 + U12 = P + U22
2
2
or
P1 P =
1 2 1 2
U U1 .
2 2
2
1
= U12
2
A21
1
A22
Using the result from Bernoulli and going back to our momentum
balance would give the nozzle force as
2
1 2
A1
Fx = U1 A1
1
2
A2
When A1 > A2 then the force acting on the nozzle is negative - meaning
without a bolt holding the nozzle on, there is a tendency of the nozzle
to pop off. The final result is only valid for high Reynolds number where
Bernoullis equation is reasonable.
Inverse sprinkler
215
Control volume
analysis
Which
way?
n
F
water intakes into the sprinkler, the question is which way does the
sprinkler spin? It seems the control volume analysis may be useful for
this problem because we just want to know which way it spins, not how
fast.
To analyze the problem, lets look at one half of the sprinkler, imagine
that we hold the sprinkler fixed and calculate the force, F , needed to
hold the normal sprinkler in place. Lets assume the water comes out
as a jet of uniform velocity u and area A, the flow is steady, the density
is constant, and viscosity is not important. Under these assumptions,
we have,
Z
Z
(v)v ndS = P ndS + F
The only tricky part of control volume analysis is getting the sign correct and since we are trying to get the direction correct, we need to treat
our signs with care. There are two signs to keep straight on the left side
of the equation, v n and v. At the exit as drawn,
R both v n and v are
positive and thus we can estimate the term as (v)v ndS Au2 .
Normally, we would assume that the pressure at the exit is the same
as the pressure in the environment. The force is positive and estimated
to have a magnitude of F = Au2 and be positive as drawn in the
schematic of figure 11.4. If we remove the force, the sprinkler spins.
Now consider the inverse sprinkler. The same integral expression
must hold true. Momentum enters the control volume at the same rate
as the sprinkler in forward motion. Since v n and v are both negative, their product is positive; the momentum across the surface is the
same in magnitude and sign as the sprinkler in normal operation. If this
216
were all, the sprinkler should spin the same direction whether the fluid
comes in or out of the sprinkler. If one does the experiment (you can
try yourself with a bendable drinking straw) you find that the sprinkler
does not spin. If you did the experiment carefully you would actually
find the sprinkler spins very slowly in the reverse direction.
The reason the control volume approach does not work is that we
have not gotten the exit pressure right. In the sprinkler operating in
normal mode the pressure at the exit must be at least a little higher
at the exit than the environment since pressure gradients would be responsible for pushing the flow. When the sprinkler is operating in
reverse mode, the pressure at the exit must be lower than the environment for flow to go in that direction. It turns out that if viscosity is
not important then the low pressure at the sprinkler exit (which wants
to make the sprinkler spin counterclockwise) exactly cancels the momentum coming in (which wants to make the sprinkler spin clockwise).
I cannot derive the experimental result using control volumes. Control
volume analysis gives you no way to calculate the pressure in this problem. While a number of people have written about this problem I have
yet to find a solution that I find truly satisfying, nor have I been able
to work one out myself.
12
Thermodynamics
Before we start the topic in the next chapter, which is the development
of the conservation of energy for fluid flow, we should probably discuss
a little thermodynamics. Thermodynamics is a course in itself. The concepts in thermodynamics are deep and far reaching. While the concepts
can describe many phenomena of our physical world, thermodynamics
can also be obscure, abstract and difficult to grasp. I think the best
quote about thermodynamics is credited to Arnold Sommerfeld, one of
the pioneers of quantum mechanics:
Thermodynamics is a funny subject. The first time you go through it, you
dont understand it at all. The second time you go through it, you think
you understand it, except for one or two small points. The third time you go
through it, you know you dont understand it, but by that time you are so
used to it, it doesnt bother you any more.
In this chapter I will introduce you to a few key ideas that will
hopefully make the subsequent chapter clear. If you have already taken
a course in thermodynamics, this should all seem familiar. If you have
not, you should do so. In this chapter we consider only a single phase
simple fluid; we are not going to worry about mixtures or phase change
(i.e. from liquid to gas). I will try to keep the ideas simple but if they
seem complicated you can use Sommerfelds quote as comfort that you
are not alone.
218
Thermodynamics
12.2 Energy
219
12.2 Energy
There are a two important laws of thermodynamics, which you may
already be familiar with. These laws are not provable, but they have
stood the test of time and there are no exceptions. If you find an exception you are wrong or you will win a Nobel prize (smart money is on
the former). The First Law of Thermodynamics is a statement of conservation of energy. The Second Law of Thermodynamics deals with a
concept called entropy, but more intuitively the second law states that
there are many processes that are not reversible in time. You cannot
always put things back to the way they were. You and I will get older,
you cannot unscramble an egg, work lost to friction cannot be recovered, and heat cannot flow from cold to hot. We will not deal with the
second law in this course.
So the First Law has to do with conservation of energy. So what
is energy? You probably know that energy comes in different forms.
From particle mechanics you are likely familiar with kinetic ( 12 mv 2 )
and potential (mgh) energy. You probably also are familiar from physics
with the conservation of energy concept. If you drop something from
rest and want to know how fast it is when it hits the ground, just equate
the potential and kinetic energy to get v = 2gh. You also know that
energy can be thermal. We discussed the idea of thermal energy in the
conduction chapter. Something hot has more energy than something
cold.
You probably also know that we can also convert from thermal to
mechanical energy and vice verse. If I drop a rock from rest with potential energy mgh from the surface of a pond to the bottom, it seems
reasonable to say that due to drag, the velocity when the rock hits
the bottom will not be v = 2gh, but something much less. The lost
mechanical energy has gone into motion of water in the wake of the
falling rock and into viscous drag along the surface of the rock. When
the water in the pond has finally come to rest, it is viscosity which has
220
Thermodynamics
removed energy from the system and thus the lost mechanical energy
has gone into thermal energy by heating up the water ever so slightly.
This is the concept that we discussed in the turbulence chapter. Heating by viscosity is analogous to warming your hands by rubbing them
together rapidly.
You should also be familiar with the idea of converting thermal energy to mechanical energy. This process is essentially how we derive
most of our energy needs throughout the world. We burn fuel, converting chemical energy locked up in fuel into thermal energy, this thermal
energy then drives the engine. This is, in at least a simplistic view, how
power plants and cars work.
So I gave some examples, but you may notice I never answered the
question, what is energy? When I was a student I found the idea of
conservation of energy to be intuitive. I thought I understood energy.
It turns out I was wrong in thinking I understood. It was not until I
read the essay on energy found in Feynmans Lectures on Physics, that
I really appreciated that most physics books never tell you what energy
is. While the concept of energy seems quite real and physical, nothing
could be further from the truth. Energy isnt really anything concrete
but it is an abstract, mathematical concept.
The conservation of energy statement says only that there exists a
number that we can compute. We happen to call that number energy.
We have rules for computing this number such as mgh or 1/2mv 2 .
Conservation of energy is a mathematical statement that if something
happens in the world, and we compute the energy number again it will
be the same. Energy is a nothing more than a number, a mathematical
function.
If we want to calculate the energy, we must be diligent. We have to
be careful that we account for any energy that has come in or out of the
system we are watching. Energy can change forms so we must be sure
we have accounted for all the different types. If we calculate our number properly, the number will always remain the same. Conservation
of energy seems intuitive, but it is actually quite abstract. You should
consult the Feynman Lectures on Physics for a more eloquent discussion, or the book Understanding Thermodynamics (VanNess (1969))
which also has excellent discussions.
221
222
Thermodynamics
223
calculate the internal energy function. Once we know the internal energy function for a substance, we can use the First Law to calculate the
work or the heat for some other process. If this circular logic bothers
you (which came first, the First Law or the internal energy function),
I suggest you just go with it until it doesnt bother you anymore.
224
Thermodynamics
The ratio q/dT is called the constant volume specific heat. It is the
amount of heat added to a substance to change its temperature a degree
while holding the volume constant. We have discussed the specific heat
before, but now we are being more precise. The specific heat must be
measured holding something constant, in this case the volume. The
specific heat is something we can measure and it is equivalent to the
derivative of the internal energy function with respect to temperature.
We can also measure the specific heat at constant pressure. The analysis here is only a little bit trickier. Since the first law tells us that
de = q P dv, when the pressure is constant we can rewrite as
d(e + P v) = q.
The quantity e + P v shows up so often in thermodynamics, it gets its
own name enthalpy, h. It is perfectly legitimate to combine properties in any way we please to get a new property. While this seems a
mathematical trick, remember that energy is just a mathematical function, thus it is no more abstract to define a new function, the enthalpy.
Defining the enthalpy, we can write for a constant pressure process,
dh = q.
Since h is a property we can write h(T, P ) and use calculus to get,
dh =
h
T
dT +
P
h
P
dP.
T
The last term drops for a constant pressure process. Equating the last
two equation we get an expression to calculate the constant pressure
specific heat.
q
h
=
cp .
dT
T P
The difference between the constant volume and constant pressure specific heat is that in the constant pressure case some work is done in the
expansion/contraction of the material.
In the case of liquids and solids, cp and cv are essentially equal. For
gases they are not.
225
cp cv = R
where R is the gas constant specific for the material.
Calculating the change in energy between two temperatures is then
given by,
Z T2
cv dT .
(12.2)
e2 e1 =
T1
In practice, the specific heat is approximately constant if the temperature changes are not too great and we have a very simple integral,
e2 e1 = cv (T2 T1 ).
(12.3)
13
Energy conservation in fluid flows
T
= k2 T.
t
228
Dt
S(t)
S(t)
V (t)
V (t)
We can convert the surface integrals to volume integrals on the right
hand side using the divergence theorem to obtain,
Z
D(e + 12 v2 )
= q + (T v) + v fb dV.
Dt
V (t)
As before, since the integral equation is true for any arbitrary material
volume, the integrand is zero at all points in space. Thus we arrive at
conservation of energy in differential form,
D(e + 12 v2 )
= q + (T v) + v fb .
Dt
229
Recall that the total stress tensor can be written, for a Newtonian,
incompressible flow, as T = P I+2S, where S is the strain rate tensor.
Thus, for a Newtonian fluid where gravity is the only body force,
D(e + 12 v2 )
= q (P v) + v g + 2 (S v) .
Dt
(13.1)
d mv
= mgv.
dt
We can re-write the left side and noting that velocity is the time rate
of change of position, dz/dt = v,
d 21 mv 2
dz
= mg
.
dt
dt
which we can rearrange to obtain,
d 1
mv 2 + mgz = 0.
dt 2
This statement should be familiar to you and says that the kinetic plus
potential energy of the system is a constant. We often describe kinetic
and potential energy to be forms of mechanical energy.
230
Dt
which can be expanded out to obtain,
D 21 v2
= P v + g v + 2( S) v.
Dt
The mechanical energy can be subtracted from the total energy equation, Equation 13.1. The result is an equation for only the thermal
energy. Its fine to subtract equations from each other since we are removing a piece of the equation that is already satisfied (i.e. momentum).
Skipping quite a few details of expanding out the vector operators, the
resulting thermal energy equation is,
De
= q P v + 2S : v.
Dt
(13.2)
The operator : is a new one for you. It is like a tensor dot product.
You simply multiply each entry of the tensor by the respective entry of
the other tensor and sum them all up. The equation reads as follows.
The left side is the material derivative of the internal energy of a fluid
particle as it moves about. This internal energy goes up or down based
on the right hand side. The first term on the right side is the divergence
of the heat flux. This term is just like we found in the conduction problem. The second term is energy going up or down due to compression.
Recall that v is a measure of the fluid being locally compressed. If
you compress something, it heats up. The last term is heat generated
by viscous dissipation, or friction. It turns out that you can prove (via
the Second Law of Thermodynamics) the last term always increases
the thermal energy. If you shook up a bottle of fluid and let it come to
rest by viscosity, the kinetic energy of the fluid in the bottle has to go
somewhere, it goes into heating up the fluid.
In practice, it is more convenient to work with temperature as a
variable rather than internal energy. We can get the equation in terms
of temperature by 1) applying Fouriers law for the heat flux q = kT ,
and 2) using thermodynamics to relate internal energy to temperature.
This is the point where the subtleties really come in. However, for
231
232
Dt
Dt
T
Substituting into our original energy equation yields,
e
DT
= kT P
v + 2S : v,
cv
Dt
T
There are a variety of thermodynamic relationships that are derived
based on multi-variable calculus and the First Law. Using one of these
relationships (see Bird et al. (1960) for details) we can rewrite the
equation as,
P
DT
= (kT ) T
v + 2S : v.
(13.3)
cv
Dt
T V
Which is a general statement of conservation of energy
for a Newtonian
fluid in terms of temperature. The quantity, P
T V comes from the
equation of state This form is useful as all the variables are measurable
thermodynamic properties. Dont worry that the steps which make use
of thermodynamic property manipulations seem mysterious; it is just
a trick that was worked out.
Using some more relationships from thermodynamics and conservation of mass, we can transform the equation to another equivalent,
cp
DP
DT
= (kT ) + T
+ 2S : v.
Dt
Dt
(13.4)
Here cp is the constant pressure specific heat and = 1 T
is called
P
the bulk expansion coefficient. The bulk expansion coefficient comes
from the equation of state and is a property that can be looked up.
Equations 13.2, 13.3, and 13.4 are all formally equivalent statements of
conservation of energy. Even though the details of each step might seem
mysterious, we are doing nothing other than using thermodynamics to
change variables to a more useful and equivalent form of the equation.
Again, dont get caught up in the details of the manipulation.
For an ideal gas, T = 1, thus,
cp
DP
DT
= (kT ) +
+ 2S : v;
Dt
Dt
In many cases of gas flows the term DP/Dt is relatively small and can
233
13.2.2 Summary
All of the detail in this section is a subtlety that is not too important
for our class. It is important for people who work closely with these
equations and use these equations in modeling. While the detail here
is not intuitively obvious, but hopefully the final result makes some
sense. The main point to remember at this time is that there are some
assumptions built into the energy equation that we will use and you
need to keep these assumptions in mind.
The final result is that we had the same equation for the flow of
liquids or a gas which is flowing at low speeds. The energy equation we
use for the remainder of this book is,
cp
DT
= (kT ) + 2S : v.
Dt
(13.5)
234
direction and there are no gradients along the channel. If you recall,
when we assume steady flow that is uniform in x, conservation of mass
states that there is no vertical velocity. The axial velocity is a function
of y only; u(y).
Lets write out the the 2D energy equation in component form,
2
T
T
T
2T
T
+u
+v
=k
+
+ 2S : v
cp
t
x
y
x2
y 2
Under these assumptions our equation simplifies because the convection
terms are removed from the equation
!
2T
2T
T @T @T
@
+u @ +v
+
+ 2S : v
cp
=k
2
@
@
t
x
y
x
y 2
@
@
@
giving,
cp
2T
T
= k 2 + 2S : v.
t
y
2 u
x
u
v
+
y
x
u
y
v
+ x
v
2 y
!
,
but in our simple flow where v = 0 and u(y) this tensor reduces to,
!
0 u
1
y
S=
.
u
0
2
y
Thus the viscous heating term is
2S : v =
u
y
2
We are left with the simplified equation for 2D plane parallel flows,
2
T
2T
u
cp
=k 2 +
.
(13.6)
t
y
y
Couette flow - Insulated walls
For the case of Couette flow the velocity field is linear, u/y = U/H
and the viscous heating term is
2
u
U2
= 2.
y
H
235
To solve equation 13.6 we need to know the thermal boundary condition. If we take the walls to be thermal insulators, there is no temperature gradient at the wall, T /y = 0. The heating term is constant
throughout the domain, so we expect the temperature of the fluid to
be constant throughout the volume. Thus, we can integrate the entire
energy equation across the domain from y = 0 to y = H,
Z H 2
T
U2
T
=k
dy + H 2
cp H
2
t
y
H
0
or
!
k
U2
T
T
T
cp
=
+
t
H y y=H
y y=0
H2
and using the insulated boundary condition where the temperature
gradient is zero at the walls,
U2
dT
= 2.
dt
H
The equation states that fluid temperature increases at a constant rate,
cp
dT
U2
=
.
dt
cp H 2
The mechanical power delivered to the fluid goes into heating the fluid
up. For water with = 106 m2 /s and cp = 4186 J/kgK, and if we
assumed the upper plate was moving at 1 m/s, you would have to have
a geometry with a 1 mm gap to achieve 1 degree Celsius heating per
hour. A 1 mm gap for a plate moving at 1 m/s is pretty fast. You
typically have to have pretty extreme conditions to obtain high rates
of viscous heating.
We could also have derived this result in a simple way. The stress the
fluid exerts on the upper plate is proportional to the velocity gradient,
= U/H. The force on the plate is the viscous stress multiplied
by the area of the plate, F = AU/H. The mechanical power that
the plate is giving to the fluid is the force multiplied by velocity, i.e.
Power = AU 2 /H. The power per unit volume is,
Power
U 2
=
,
AH
H2
where AH is the volume of fluid between the two plates. The quantity
U 2 /H 2 is the number of Watts per unit volume which must be taken
236
by the fluid and converted into thermal energy when the system is
insulated. This result is equivalent to the viscous heating term above.
Couette flow - Constant temperature walls
We can also integrate Equation 13.6 for constant temperature boundary
conditions in Couette flow. We would expect if the walls were held
at constant temperature then the system would reach a steady state
in time where the mechanical power put into the fluid balanced the
amount of heat flowing out. At steady state, the /t term in the
equation would be zero. Therefore we would need to solve,
k
U2
2T
= 2
2
y
H
U 2 2
y + C1 y + C2 .
2kH 2
Using the boundary conditions to obtain the two constants of integration we have,
T (y) =
U 2
y(H y) + Twall .
2kH 2
237
u
dy.
dt
y 0
y 2
0
The first term on the right side is zero because of the no-slip condition
on the velocity at the walls. Therefore we are left with,
Z H
d
2u
Energy stored + total heat loss = L
u 2 dy.
dt
y
0
Recalling that in this problem conservation of momentum states that
2
P = L yu2 . The second derivative of the velocity is a constant,
238
uP dy = P Q,
0
RH
recalling the total volumetric flow rate was calculated as, Q = 0 udy.
In fluid mechanics the pressure difference times volumetric flow rate
has units of power, energy per unit time. Its nice that in this simple
problem we can obtain a nice intuitive result. The mechanical power
(measured in Watts) added to the fluid is equal to the number of Watts
of heat loss that leaves the fluid plus the rate you add heat to the fluid.
We have essentially gone back to a fundamental statement of the first
law of thermodynamics
d
Energy stored = Mechanical work in total heat out.
dt
At steady state where the temperature of the fluid is a constant in
time, then the mechanical power in equals the heat flow out. Of course
since our original equations were stated in terms of these kind of overall
energy balances, it is not surprising that we obtain the result we did.
The only point of this section on Poiseuille flow was to attempt to
connect back to our original ideas of energy and the first law, and make
connections between the mathematical formulations that we have been
discussing in this chapter.
The power relationship is analogous to electric circuits where voltage
drop times current equals power. In a resistor, voltage times current
gives power where that amount of lost electric power is dissipated as
heat. In our Poiseuille flow, we find that at steady state the power
delivered to the fluid equals the amount of heat that leaves the fluid.
The analogy of pressure driven flow in a pipe to a voltage applied across
a resistor is a very good analogy.
239
DT
= k2 T.
Dt
(13.7)
This equation is the one we put forth at the beginning of the chapter,
with the exception that we have been careful about which specific heat
to use. We should also note at this time that the above equation is
identical that put forth in the chapter on mass transfer. Heat and mass
transport share many of the same mathematical solutions and physical
pictures.
Despite the restriction of constant pressure in gas flows, we will use
the simplified energy equation as it is reasonably accurate for incompressible flow. The equation for a liquid and a gas at constant pressure
are the same. The restrictions on the above equation may not be fully
clear at this point given the looseness of the derivation. A more rigorous
and detailed treatment requires us to delve in little deeper into ther-
240
Dt
DT
cp
= k2 T.
Dt
This is a well-posed coupled equation set for conservation of mass, momentum, and energy. With appropriate boundary and initial conditions,
the equations can be solved and then we know everything about the
velocity and temperature fields. However, we have already learned that
solving fluid flow problems on their own are very complicated especially
when flows are turbulent.
We should take care to note the coupling between these equations.
First, the coupling of the momentum equation into the thermal energy
equation is direct. The velocity field convects thermal energy (through
the D/Dt operator), thus the flow determined by the Navier-Stokes directly feeds into the thermal energy equation by carrying energy around
the system.
The primary coupling of the temperature equation on the velocity
field is that the fluid density changes with temperature. These density
changes can give rise to fluid flow through a buoyancy force, an effect
called natural convection. The thermal equation feeds back into the
241
14
Convective heat transfer: external flows
In the previous chapter we derived the equation for conservation of energy in a fluid flow. In principle the coupled equations for conservation
of mass, momentum, and energy can be solved, except we have already
learned that solving fluid flow problems on their own are complicated.
Coupling in the transport of heat makes things even more complicated.
In many applications, the Reynolds number is high and thus analytical
and even numerical simulations may not be feasible. The problem of
convective heat transfer seems even more difficult than the fluid motion
that has already troubled us so much.
In this chapter we will see that we can use the equations to guide
a dimensionless formulation. This formulation will allow us to express
the dimensionless heat transfer coefficient from an empirical correlation
of two parameters. By this, I mean that someone for the geometry of
interest has conducted a number of experiments and found an approximate function that collapses data over a wide range of conditions. These
empirical correlations can then be readily used to calculate the dimensional heat transfer coefficient and make real predictions. This approach
may seem an easy way out. We are developing all these mathematical
equations and then we are going to give up and go to a handbook? Yes.
However, this approach is very powerful as we will provide a simple
way to get real engineering numbers.
244
Ts
fluid; , p, K, Cp
Figure 14.1 Schematic for the forced convection problem with external flow.
Dt
DT
= k2 T.
cp
Dt
The coupling in this case is only one way. You need to solve the mass
and momentum equation first for the velocity field. This now known
velocity field would then be used to solve the temperature field. In order
to solve the equations we need boundary conditions. For the velocity
field we use no-slip at the solid surface and for temperature lets consider that the object is held at a fixed constant temperature which is
hotter than the surrounding fluid.
We can make these equations dimensionless as we have done before,
considering a constant density flow. Lets say the geometry of the object
provides some characteristic length, L, and some characteristic flow
velocity, U0 . Lets define v
= v/U0 , [
x, y, z] = [x, y, z]/L, t = tU0 /L,
v
2v
v =
P + 1
,
+v
Re
t
v
= 0.
245
(14.1)
(14.2)
RePr
Dt
(14.3)
246
object into the fluid. In dimensional terms, the local heat flux out of
the surface is,
q = n kT,
where the temperature gradient is the temperature gradient of the fluid
at the surface and k is the conductivity of the fluid. In terms of the
dimensionless temperature,
q=
k
(Ts T )n T
L
hL
= n T
k
247
q = hA(T
s T ),
where A is the surface area and Ts is the average surface temperature. We define the average Nusselt number through this averaged heat
transfer coefficient,
hL
Nu =
.
k
The average dimensionless heat transfer coefficient, Nu is a function of
only two parameters,
Nu = f (Re, Pr).
The above expression is great. Without even solving the problem,
we now know that if we wish to compute the average heat transfer
coefficient for flow over a particular shaped object, we need to solve
the problem (numerically or experimentally) as a function of two parameters. We can then make a graph of Nusselt number vs. Reynolds
number for different values of the Prandtl number. This graph (or empirical equation) is difficult to come by and requires a set of careful
experiments. However, once the experiments are conducted, they are
done for all time. It is much more reasonable to deal with a system
which only depends on two parameters than the dimensional formu = f (U , , , D, k, cp , Ts , T ). Fortunately,
lation which would say, h
in textbooks on heat transfer, online or in a number of handbooks,
you can find correlations or graphs of Nu = f (Re, Pr) for a number of
common geometries. Even if you have a problem where your geometry
is different than anything that was done before, you can usually find
something close to give you a reasonable approximation.
Typically we will find that we can collapse the data with some function of the form,
Nu = CRem Prn ,
where the constants C,m, and n are found from experiments. There
often not really any deep meaning into this functional form, it just
works well and is simple to use.
It should also be noted that the error bars on heat transfer correlations can be reasonably large. If you use a correlation in a handbook
or take data from a graph, you should look at what the stated error is.
248
Nu = CRem Pr 3 .
The details of this correlation can be found in the heat transfer references such as (Incroprera and DeWitt (2001)). In their book the provided values for C and m depend upon the Reynolds number. The
ranges for the experimentally determined constants are for 4 < Re <
40, C = 0.911, m = 0.385; 40 < Re < 4, 000, C = 0.683, m = 0.466;
4000 < Re < 40, 000, C = 0.193, m = 0.618.
Lets use concrete example and take a cylinder held at 350K in air
flow at 300K. The flow velocity will be 10 m/s and the diameter of the
sphere is 2 cm. To evaluate the fluid properties, I will use the average
temperature of 325 K to evaluate the properties. Note that this choice
(as opposed to just using ambient conditions) will change the exact
answer we get, but the impact is not very dramatic. We first compute
the Reynolds number,
kg
1.07
10 m
0.02 (m)
m3
s
U D
Re =
=
10, 900.
0.0000196 kg
ms
249
hD
= 53
k
W
53 0.0282 mK
W
= 75 2
0.02 m
m K
Note the simplicity of the methodology here. Once we have an empirical formula for the Nusselt number - which generating the first time
is the real hard work - it is just a matter of computing the appropriate dimensionless numbers and carefully calculating the convection
coefficient.
h=
250
t t.
This scaling is the same as momentum diffusing due to the kinematic
viscosity and is what we discussed in the conduction chapters earlier in
this book. In the boundary layer problem, the time for heat to diffuse
251
fluid; , p, K, Cp
Pr > 1
Th
Th
Figure 14.2 Schematic for the thermal boundary layer problem. The
thickness of the boundary layers are exaggerated.
1
t
.
x
xU
Rex Pr
where Rex = U x/ is the local Reynolds number. Recallp
from previous
discussion that the momentum boundary layer that 1/Rex . Thus
our argument would suggest that the ratio of the momentum to thermal
boundary layer scales as,
Pr.
t
This result indicates an important idea. When the Prandtl number is
1, then the thickness of the thermal boundary layer and the momentum
boundary layer are the same. Air has a Prandtl number of about 0.7,
so the two boundary layers are of comparable size. When the Prandtl
number is large (as in an oil), momentum diffuses quicker than heat.
In this case the thermal boundary layer is much thinner than the momentum boundary layer. In the other extreme of low Prandtl number
(a liquid metal) heat diffuses more readily than momentum and the
thermal boundary layer is relatively thick compared to the momentum
one.
252
Nux =
1
1
hx
= 0.332 Rex2 Pr 3 .
k
From the solution we obtain the ratio of two boundary layers as,
Nux =
Pr 3 .
t
253
U
x
12
Pr 3
[Laminar].
1
1
hL
= 0.664 ReL2 Pr 3
k
[Laminar]
4
1
hx
= 0.0296Rex5 Pr 3 .
k
[Turbulent]
254
xc
hlam dx +
0
!
htur dx ,
xc
255
Dt
DT
cp
= k2 T.
Dt
Note that I am just providing the final result and intentionally not
deriving these equations in detail. If you need more detail, a derivation
is easily found in a number of references. The important thing is that
now there is a force in the momentum equation which is proportional to
the temperature. This is the buoyant force that will cause less dense air
to rise. The coupling between the equation is now two way, the velocity
field convects the fluid temperature and the temperature drives the
flow.
Making the equations dimensionless just as in the previous section
yields,
v
=0
1 2
g(Ts T )L
D
v
T+
= Pd +
v
U
Re
Dt
0
1
DT
=
2 T.
RePr
Dt
(14.4)
(14.5)
(14.6)
Re
Dt
(14.7)
where Re = Gr where in problems with natural convection it is common to define the Grashof number as,
2
g(Ts T )L U0 L
g(Ts T )L3
Gr =
=
.
U02
2
The Grashof number is the ratio of the buoyant force to the viscous
256
force. While we can think of the Grashof number as appearing like the
Reynolds number in the equation we give it a new name by tradition
because the definition of the Grashof number has no driven velocity
field as part of the definition.
If we repeated the analysis of computing heat transfer from an object due to natural convection we would find that the average Nusselt
number is now a function of the dimensionless numbers;
Nu = f (Gr, Pr).
In practice, it turns out to be useful to define the Rayleigh number as
Ra GrPr.
You will see the Rayleigh number appear in a number of correlations
for Nusselt number.
As a rule of thumb, the average dimensional natural convection coefficient in air is usually about 10 W/m2 K. By about I mean, maybe it
might really be 5 or 20 for your application. The point is that it is not
usually 100 or 1. I find keeping this order of magnitude of 10 W/m2 K
in my head very useful.
Finally some problems have a forced and natural convection component. Our dimensional analysis lets us quickly estimate which is
more important or if both matter. For a given mixed problem we can
compute a Reynolds number based on the driven velocity field and a
Grashof number. We can then compute the number, Gr/Re2 , and decide whether forced or free convection is dominant. If Gr/Re2 is a small
number, free convection can be neglected. We are back to the equations
used for forced convection. If Gr/Re2 is a big number it means that
natural convection dominates over any externally imposed flow field. If
Gr/Re2 1 then both forced and natural convection will play a role
in the problem and we would find the proper dimensionless expression
would be,
Nu = f (Re, Gr, Pr).
We will not consider any examples here where both natural and forced
convection matter.
257
9.8
m
s2
1.84
1
325
105
g(Ts T )D3
1
3
K 50 (K)0.02
m2
2.62 105
s
(m3 )
25, 000
m2
s
We then look at the textbook (Incroprera and DeWitt (2001)) and see
that for this range of Rayleigh number that we should use C = 0.48
and n = .25. Using these values gives
Nu = 0.48Ra0.25 = 6
Finally, we find the heat transfer coefficient by,
Nu = 6 =
h=
6 0.0282
0.02 m
W
mK
hD
k
= 8.5
W
m2 K
258
4
Nux
3
For the vertical plate in natural convection the experimentally observed transition to turbulence occurs around
Ra < 109 for laminar flow.
When the Rayleigh number is too high, then we expect the flow to be
turbulent. With turbulent flow, as always, we cannot calculate anything
directly. Thus one empirical fit to the data has the form,
2
1
Nu = 0.825 +
0.387Ra 6
9
1 + (0.492/Pr) 16
8
27
Dont read too much into the powers of 8/27 and the constants. This
is just an empirical expression that describes the data with reasonable
accuracy.
259
between. The bounding plates are held at fixed temperature with the
bottom plate hotter than the upper one. Gravity points downward.
When the temperature difference is small, heat flows from hot to cold
by conduction. The fluid near the lower hot plate is less dense and
wants to rise. However the geometry of the solid plates constrains the
flow since for some fluid to move up, some other fluid must move down.
If the temperature difference is small then the conduction state is stable
and the fluid stays at rest. If the temperature difference is increased,
then at some point the situation becomes unstable and convection rolls
can set in. Some fluid rises and other fluid sinks. The critical parameter
is the Rayleigh number and convection sets in when Ra > 1708. In this
case the critical Reyleigh number can be determined through analysis
and the result matches experiments quite well. The critical parameter is
computed via a linear stability analysis, meaning one takes the simple
conduction state with no flow and assumes small linear perturbations
from that state.
This is a well studied problem with a lot of interesting experimental
behavior. Interesting large scale patterns can emerge with one seeing
regimes with stripes, hexagons, nearly perfect spirals, spirals with defects, or completely chaotic flow. The system is one of the most widely
studied pattern forming systems where beautiful large scale regular
patterns form spontaneously. I highly recommend a brief search on the
images associated with this problem to get a sense for the richness
associated with such a simple physical system.
260
good you can expect your predictions to be. It is also important that
you understand any assumptions behind the correlation.
The typical methodology for convection problems is straightforward.
Decide if the problem is one of natural convection, forced convection,
or mixed convection. If the problem is not clearly natural or forced
convection, you can just guess and proceed. At the end we can check
the assumptions and repeat.
Identify the geometry and find an appropriate correlation in a text or
handbook. Depending on how good of an answer you require, you can
often make the spherical cow approximation for quick estimates.
Read the details of the correlation; How accurate is it? Is it good
for a limited range of Re or Pr? Make sure you understand if it is
appropriate for the problem at hand.
Make sure you understand what length scale is used in the correlation for the calculation of the parameters. The length scale can be
arbitrary so you need to know. For example, for a sphere one could
base things off the radius or diameter.
Decide if you need local heat transfer coefficients or an average.
Evaluate all the fluid properties at the appropriate average temperature, (Ts + T )/2.
Calculate Re, Pr, and Gr. Double check your calculations. For mixed
convection, double check whether you are in natural or forced convection regime by calculating Gr/Re2 . Double check that you are in
a valid regime for the correlation.
Use the correlation to calculate Nu.
Use the fluid properties and geometry to calculate h.
Do it again from scratch to make sure you believe your number. It is
easy to make mistakes. If you have a complicated problem you may
want to make a dramatically simplified model first just so you know
what order of magnitude to expect.
Despite the fact that we spent a lot of time developing the thermal
energy equation, in the end we are not even using it. We are using experiments to obtain data. However, all the work with the equations was
not lost. The equations guided us to tell us that the Reynolds number,
the Nusselt number, the Grashof number and the Prantl number were
the right dimensionless parameters. Recall how in the first chapter
that we could proceed with dimensional analysis without even knowing
14.6 Simulation
261
the equations. We saw in the first chapter though, that in a complicated problem such as we consider here, how to proceed with a pure
dimensional analysis is not always clear. Our approach to convection is
to use the equations as a guide to select the most natural way to make
our data dimensionless. This approach is similar to the idea of the drag
coefficient that we introduced in an earlier chapter.
14.6 Simulation
The same simulation tools we discussed which are available for computing fluid flow can also be used for convection calculations. Generally the
more details we add to a problem the more complex the flows and numerical solutions become. For forced convection with no gravitational
effects, if the Reynolds number is not too large and the geometry is
not too complex, then there is a good chance that we can get a good
numerical result without investing in too much expertise using modern
simulation software. A problem such as the 2D laminar boundary layer
or flow around a cylinder at moderate Reynolds number, can be easily
computed. Once the velocity and temperature fields are computed, it is
straightforward to extract the average convection coefficient. As with
flow problems, the higher the Reynolds number the more complex the
numerical solution will be and the more care, compute time, and expertise is required. I am being purposely vague in what I mean by high
Reynolds number since what is consider high continually increases as
computing power increases. Many of the details of what is consider a
high Reynolds number depends upon the problem at hand.
Natural convection calculations are also quite amenable to solution
with modern simulation software. However, with increased problem
complexity comes the increasing need to really understand the physics
of the problem at hand and know what we are doing. The current state
of the available software is that it is much more difficult for a novice
to generate good solutions to natural convection problems. While it requires a little more care it is such solutions are certainly possible with
a little physical insight and experience with the particular software. It
was not that long ago that solving natural convection problems with
even simple geometries was a very challenging undertaking and one
that was the topic of a PhD thesis. Today, we can solve these problems
262
15
Energy balance in internal flows
Imagine hot water flowing through the pipes in your house. As the water moves from the heater to your shower, the water cools due to heat
transfer. Heat is transfered from the hot fluid to the solid walls of the
pipe by convection in the internal flow. This heat flows through the
solid pipe walls by conduction and then to the external environment.
The problem of convection in a pipe is different than in external flow
because the fluid temperature in the pipe can be constantly changing with distance down the pipe. In the previous chapter we assumed
that there was enough fluid that the far field was always at a constant
temperature. When we consider flow inside a pipe or enclosure, then
the fluid temperature is changing with distance. This subtle difference
leads to some new behavior than we saw in the previous chapter. While
the fundamental equations are the same, we need to be aware of the
differences in internal flow.
DT
= (kT ) = q.
Dt
(15.1)
264
Integrating this equation over a fixed volume and using some of our
integral rules we discussed earlier, the energy balance can be stated as,
Z
Z
Z
d
cp T dV + cp T v ndS = q ndS.
dt
For this section let us consider only the steady state behavior and the
temperature equation in integral form becomes,
Z
Z
cp T v ndS = q ndS.
The integral on the left hand side represents thermal energy being
convected into the domain, and the integral on the right is the net heat
flux into the domain.
Now, lets take a length L of cylindrical pipe and define that to be
the volume of interest. There are three surfaces which we must evaluate
the integrals, the inlet, the outlet and the pipe wall. The integral on
the left hand side is zero along the solid wall since v n = 0 there; no
fluid and thus no energy is convected across the walls. There are only
convection contributions where fluid flows in or out of the domain,
Z
Z
Z
cp T v ndS +
cp T v ndS = q ndS.
x=0
x=L
v ndS and
(15.2)
Convince yourself that the signs are correct. Since the normal vector
points out from the surface, when the heat flux at the wall points
into the pipe, then the outlet temperature is greater than the inlet
temperature.
In many cases, especially Rwhen the pipe is long relative to the diameter, then the evaluation of q ndS over the inlet and outlet are small
relative to what comes through the wall. Thus, it is a reasonable and a
common approximation to only consider heat flux through the wall and
not axial conduction through the fluid. To proceed any further with our
265
Dq
x.
m
The mean temperature of the fluid increases linearly with distance down
the pipe. The fluid heats up continuously since energy is added as we
move down the pipe. What is interesting is that fluid temperature as
a function of distance is known simply from the energy balance. There
is no need to solve for the flow or know the convection coefficient. The
only reason you would need the convection coefficient is if you wanted
to know the temperature of the wall.
266
Using the Fundamental Theorem of Calculus, the left side of the equation becomes,
Z
mc
p (T (L) T (0)) =
mc
p
0
dT (x)
dx.
dx
dT
= Dh(x)(Ts T (x)).
dx
Dh
mc
p x
DhL
mc
p
We can see two limits that make some physical sense. When the parameter,
DhL
mc
p
is greater than about 3, then you can consider the exit temperature
of the pipe to be that of the externally held wall temperature Ts . The
parameter is larger when convection inside the pipe is vigorous, the
pipe is long, and the mass flow rate is relatively small. Our result makes
intuitive sense that the exit temperature would be close to the ambient
conditions. When the above parameter is much less than one, then the
exit temperature will be close to the inlet temperature.
267
hD
48
=
constant heat flux,
k
11
hD
= 3.66 constant surface temperature.
k
If you dont believe this result, it is one that is easy to confirm using
268
Nu =
hD
= 0.023Re4/5 Prn ,
k
where n = 0.4 for heating and 0.3 for cooling. There are other correlations available in the literature that have either more accuracy or
account for fluids with large property variations. In the end, the basic message is the same. A correlation is used to calculate the Nusselt
number, which is a function of the Reynolds number and the Prandtl
number. One can then use the Nusselt number to obtain the convection
coefficient and the heat transfer rates can be calculated.
269
m
s
10, 000.
At this high Reynolds number the flow is certainly turbulent. Using the
correlation for turbulent flow to find h, we use,
Nu =
hD
= 0.023Re4/5 Pr0.3 = 0.023 100004/5 70.3 = 65.
k
Therefore,
Nu =
hD
W
= 65 h 3000 2 .
k
m K
270
numbers depend on the details of the house, the physical values I assumed were not unrealistic for a realistic application. However, usually
we expect to have water much hotter than we computed.
The analysis we just completed assumed that the wall of the pipe was
held at 20 C. Even if the pipe is not insulated, there is still a resistance
to heat transfer between the outer surface of the pipe and the ambient
air that was not accounted for in the original analysis. In reality, the
pipe wall is not held at 20 C but instead the pipe is immersed in air at
20 C. In a previous chapter I noted that for external flows driven by
natural convection, h 10 mW
2 K was typical. Assuming this convection
coefficient outside the pipe would yield and effective h which would
be the two in resistances in series; internal and external convection.
Since the external resistance to heat transfer is so much larger than the
internal resistance, the effect value for h will be very close to the value
from the exterior. Therefore, the dimensionless parameter becomes,
5(m)
0.0127 (m) 10 mW
DhL
2K
= 0.005.
=
J
mC
p
0.1 ( kg
s ) 4186
kg K
If the pipe is cooled by natural convection, then the temperature would
barely change from inlet to outlet
T (L) = Ts + (T (0) Ts ) e0.005 = 20 + 30 e0.005 = 49.85C.
In this case you can see that any more detailed calculation of the interior and exterior values of h is likely pointless. If we have the order of
magnitude correct for the exterior value then little will change. Would
we care if we found through a more detailed calculate that the exit temperature was 49 C? As far as we are concerned, the outlet temperature
is the same as the supply temperature.
271
2
The integral on the left represents the rate that the total energy
is convected into the domain, thus there are only contributions where
fluid flows in or out of the domain. These convectionRterms are zero
along the walls of the pipe. On the right side, the term 2(n S) vdS
is also zero along the walls where the velocity is zero. Thus for a closed
pipe system,
Z
Z
Z
1
2(n S) vdS
(e + P + v2 )v ndS = q ndS +
2
in/out
in/out
We can also neglect (either exactly or approximately) the second term
on the right side. This term is the net difference in normal viscous
stress at the inlet and the outlet. You can see this is identically zero
if we considered 2D Poiseuille flow. To demonstrate the term is zero
you only need to expand all terms for the known flow solution; an
exercise we will leave for the reader. Thus, for internal pipe flow the
total conservation of energy states,
Z
Z
P
1
v e + + v2 + gz ndS = q ndS
(15.3)
2
in/out
This equation says that the difference in the total energy (internal,
kinetic, potential) over the inlets and outlets, plus the flow work (pressure), must equal the total heat which has crossed the pipe surface. If
we assume the quantity in parenthesis on the left side can be approximated as uniform across the inlet and outlet of a section of pipe we
would obtain,
Z
P
1 2
P
1 2
1
e + + v + gz
e + + v + gz
=
q ndS,
2
m
out
in
where m
is the total mass flow rate in the pipe.
272
=
+ v + gz
+ v + gz
q ndS,
2
m
out
in
The left side of the equation is the same as Bernoullis equation. For
pipe flow, the difference in the Bernoulli constant along the pipe is
related to the heat dissipated in order to maintain a constant temperature. Heat is generated by the fluid internally due to viscosity and
that heat is lost through the pipe walls if the fluid temperature does
not change. The form of Bernoullis equation for pipe flow is often used
and called the modified Bernoulli equation. The loss term on the right
side is not computed, but estimated from experiments and charts.
We already discussed this equation in the chapter on high Reynolds
number flow. When we introduced the modified Bernoulli equation we
simply stated that the extra term was related to the frictional losses.
Here, we now see that losses are quantified as the total amount of heat
lost to the surroundings.
15.4 Summary
The situation for heat transfer problems in internal flows is much the
same as in external flows. While there are some real differences, the big
ideas are the same. In the end we have three primary tools at hand;
mathematical analysis, numerical simulation, or experiments. Mathematical analysis is difficult other than for the simplest geometries and
cases. Laminar flow in a pipe is one case we can calculate the heat
transfer performance by hand. Numerical simulation, as we have discussed throughout this book, is useful for many flows but caution and
expertise are required when the fluid flow is unstable or turbulent. Experiments and empirical correlations are useful since the dimensional
analysis gives that the average Nusselt number is a function of Reynolds
and Prandtl number only. As with external flows, we can conduct a
limit set of experiments and then use those dimensionless results to
solve new problems.
Appendix
Solution to the 1D transient heat equation
(A.2)
(A.3)
1 d2 X(x)
1 d(t)
=
.
(t) dt
X(x) dx2
(A.4)
which is rewritten as
274
(A.5)
The minus sign and the power of two are put in the equation arbitrarily.
This will not effect the answer but they will make the answer come out a
little nicer in the end. The reason for the negative sign will be apparent
shortly. Since I know the answer ahead of time I will include these
factors now. We can now solve each of the problems independently;
1 d(t)
= 2 .
(t) dt
(A.6)
d(t)
= 2 dt,
(A.7)
log() = 2 t + C,
(A.8)
= e
t+C
= eC e
(A.9)
= Ce t .
(A.10)
It is clear why the negative sign was needed when choosing the constant.
If the constant was positive, the solution would grow out of control with
positive feedback. The negative sign allows a nice gentle decay as we
would expect from physical reasoning.
Now we turn to the equation that was a function of x;
1 d2 X(x)
= 2 .
X(x) dx2
(A.11)
We have seen this equation several times and can integrate it directly
or simply go straight to the general solution,
X(x) = Asin(x) + Bcos(x)
(A.12)
275
where A and B are constants of integration. You may now see why we
chose 2 as our constant. Combining our expressions for X and we
obtain,
2
(A.13)
T (x = 1, t > 0) = 0;
When x = 0,
2
T (x = 0, t) = e t B = 0,
(A.15)
T (x = 0, t) = e t Asin() = 0.
(A.16)
X
n=0
An e(n) t sin(nx).
(A.17)
276
An sin(nx) = 1.
(A.18)
n=0
We now need a way to compute the An so that the above expression holds true. This method will seem odd at first, however, just bear
through the solution and convince yourself that it works. To find the
coefficients An you can use the following integral properties. These
properties can be derived or you can look them up in a calculus textbook, but we will just state the result here:
Z 1
sin(nx)sin(mx)dx = 0;
n and m are integers, n 6=(A.19)
m
Z
0
1
sin(nx)sin(nx)dx = 1/2;
n is an integer
(A.20)
Using these properties of integrals allows us to extract the coefficients. We multiply A.18 by sin(mx) and integrate the function across
the domain;
Z 1X
Z 1
An sin(nx)sin(mx) =
sin(mx)
(A.21)
0 n=0
The sum results in many terms, however they are all zero except for
the one where n = m. Using the fact that most terms in the sum are
zero under this operation, we are left with,
Z 1
Z 1
An sin(nx)sin(nx) =
sin(nx).
(A.22)
0
4
n
when n = 1, 3, 5, 7
(A.23)
when n = 0, 2, 4,
(A.24)
and
An = 0
4 (n)2 t
e
sin(nx).
n
n=1,3,5,..
(A.25)
277
sum is taken to infinity, however in practice it need not be. The smoother
the function, the fewer terms are needed to represent the solution as a
sum of sines. The convenient aspect of this solution is that we can find
the temperature profile at any instant in time without finding the solution at any earlier times as we must do when we numerically integrate.
An sin(nx) = f (x)
(A.26)
n=0
(A.27)
(A.28)
Since the inner product of any two of the different unit vectors is zero,
and the inner product of like unit vectors is one we obtain
Ax = F i.
(A.29)
This example may seem rather trivial and obvious given that it is so
familiar. The procedure is so simple since the original basis vector set
278
had the nice property that the inner product of any two unlike units
vectors is zero.
Now consider our functional expression
A1 sin(x)+A2 sin(2x)+A3 sin(3x)+A4 sin(4x)+... = f (x) (A.30)
and let us find the coefficients in the exact same way. Think of each funcsin(nx) as a basis vector. In this function space the operation,
Rtion
1
f
(x)sin(n2x)dx
will serve as the inner product of our vector f
0
th
with the n basis vector sin(nx). The inner product operation satisfies the same conditions as our usual operation with three dimensional
basis vectors. We get zero for the inner product of any combination of
basis functions, except when we take the inner product with itself. For
example,
Z 1
sin(x)sin(2x)dx = 0;
(A.31)
Z
0
1
sin(x)sin(x)dx = 1/2;
(A.32)
(A.33)
Since the basis functions are orthogonal, the inner product is zero for
any combination of unlike basis functions. Our analogous expression to
A.29
Z 1
f (x)sin(x)dx.
(A.34)
A1 = 2
0
(A.35)
Bibliography
Anderson, J.D. 2005. Lundwig Prandtls boundary layer. Physics Today, 58,
4248.
Bird, R.B., Stewart, W.E., and Lightfoot, E.N. 1960. Transport Phenomena.
Wiley.
Bridgman, P.W. 1922. Dimensional Analysis. Yale University Press.
Incroprera, F.P., and DeWitt, D.P. 2001. Fundamentals of Heat and Mass
Transfer. Wiley.
Kundu, P.K., and Cohen, I.M. 2004. Fluid Mechanics. Elsevier Academic
Press.
Panton, R.L. 1996. Incompressible Flow. Wiley.
Rayleigh. 1915. The principle of similitude. Nature, 95, 6668.
Reynolds, O. 1883. An experimental investigation of the circumstances which
determine whether the motion of water shall be direct or sinuous and
the law of resistance in parallel channels. Phil. Trans. R. Soc., 174,
935982.
Saffman, P.G. 1992. Vortex Dynamics. Cambridge University Press.
Schey, H.M. 1997. div grad curl and all that. W.W. Norton and Co.
VanNess, H.C. 1969. Understanding Thermodynamics. Dover.