57:020 Fluid Mechanics Chapter 4

Professor Fred Stern Fall 2013 1

Chapter 4: Fluids Kinematics

4.1 Velocity and Description Methods

Primary dependent variable is fluid velocity vector

V = V ( r ); where r is the position vector

If V is known then pressure r r  xî  yĵ  zk̂

and forces can be
determined using x
techniques to be discussed V (r , t )  uiˆ  vjˆ  wkˆ
in subsequent chapters.

Consideration of the velocity field alone is referred to as

flow field kinematics in distinction from flow field
dynamics (force considerations).

Fluid mechanics and especially flow kinematics is a

geometric subject and if one has a good understanding of
the flow geometry then one knows a great deal about the
solution to a fluid mechanics problem.

Consider a simple flow situation, such as an airfoil in a

wind tunnel:
U = constant
57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 2
Velocity: Lagrangian and Eulerian Viewpoints

There are two approaches to analyzing the velocity field:

Lagrangian and Eulerian

Lagrangian: keep track of individual fluids particles (i.e., solve F

= Ma for each particle)
Say particle p is at position r1(t1)
and at position r2(t2) then,

̂ ̂ ̂
̂ ̂ ̂

Of course the motion of one particle is insufficient to describe the flow

field, so the motion of all particles must be considered simultaneously
which would be a very difficult task. Also, spatial gradients are not
given directly. Thus, the Lagrangian approach is only used in special
circumstances.

Eulerian: focus attention on a fixed point in space

̂ ̂ ̂

In general,
( ) ⏟̂ ̂ ̂

where,
( ), ( ), ( )
57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 3

This approach is by far the most useful since we are usually

interested in the flow field in some region and not the
history of individual particles.

However, must transform F = Ma

from system to CV (recall
Reynolds Transport Theorem Ex. Flow around a car
(RTT) & CV analysis from
thermodynamics)

V can be expressed in any coordinate system; e.g., polar or

spherical coordinates. Recall that such coordinates are
called orthogonal curvilinear coordinates. The coordinate
system is selected such that it is convenient for describing
the problem at hand (boundary geometry or streamlines).
V  v r ê r  v  ê  x  r cos 
y  r sin 
ê r  cos  î  sin  ĵ
ê    sin  î  cos  ĵ

Undoubtedly, the most convenient coordinate system is

streamline coordinates:
V(s, t )  v s (s, t )ês (s, t )

However, usually V not known a priori and even if known

streamlines maybe difficult to generate/determine.
57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 4

4.2 Acceleration Field and Material Derivative

The acceleration of a fluid particle is the rate of change of

its velocity.

In the Lagrangian approach the velocity of a fluid particle

is a function of time only since we have described its
motion in terms of its position vector.

( )̂ ( )̂ ( )̂

̂ ̂ ̂

̂ ̂ ̂

In the Eulerian approach the velocity is a function of both

space and time such that,

( )̂ ( )̂ ( )̂

where ( ) are velocity components in ( )

directions, and ( ) ( ) since we must follow the
particle in evaluating ⁄ .
57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 5

̂ ̂ ̂ ̂ ̂ ̂

where ( ) are not arbitrary but assumed to follow a

fluid particle, i.e.

Similarly for & ,

57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 6
In vector notation this can be written concisely

DV  V
  V  V
Dt t

  
 î  ĵ  k̂ gradient operator
x y z

V
First term, , called local or temporal acceleration results
t
from velocity changes with respect to time at a given point.
Local acceleration results when the flow is unsteady.

Second term, V  V , called convective acceleration

because it is associated with spatial gradients of velocity in
the flow field. Convective acceleration results when the
flow is non-uniform, that is, if the velocity changes along a
streamline.

The convective acceleration terms are nonlinear which

causes mathematical difficulties in flow analysis; also, even
in steady flow the convective acceleration can be large if
spatial gradients of velocity are large.
 57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 7
Example: Flow through a converging nozzle can be
approximated by a one dimensional velocity distribution
u = u(x). For the nozzle shown, assume that the velocity
varies linearly from u = Vo at the entrance to u = 3Vo at the
y exit. Compute the acceleration
DV
as a function of x.
Dt
DV
Evaluate at the entrance
Dt
and exit if Vo = 10 ft/s and
L =1 ft.

u = Vo
u(x) = mx + b
u(0) = b = Vo
Du u u 3Vo  Vo 2Vo
We have V  u ( x )î ,  u  ax m=  
Dt x x L L

x   Vo  Vo  2x  1
2Vo
Assume linear u(x) 
variation L L 
between inlet
and exit u 2V0 2Vo2  2x 
  ax    1
x L L L 

@x=0 ax = 200 ft/s2

@x=L ax = 600 ft/s2

57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 8

Additional considerations: Separation, Vortices,

Turbulence, and Flow Classification

We will take this opportunity and expand on the material

provided in the text to give a general discussion of fluid
flow classifications and terminology.

1. One-, Two-, and Three-dimensional Flow

1D: V = u ( y)î

2D: V = u ( x, y)î  v( x, y) ĵ

3D: V = V(x) = u ( x, y, z)î  v( x, y, z) ĵ  w ( x, y, z)k̂

2. Steady vs. Unsteady Flow

V = V(x,t) unsteady flow

V = V(x) steady flow

3. Incompressible and Compressible Flow

D
 0  incompressible flow
Dt

representative velocity
V
Ma =
c
speed of sound in fluid
57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 9

Ma < .3 incompressible

Ma > .3 compressible

Ma = 1 sonic (commercial aircraft Ma.8)

Ma > 1 supersonic

Ma is the most important nondimensional parameter for

compressible flow (Chapter 7 Dimensional Analysis)

4. Viscous and Inviscid Flows

Inviscid flow: neglect , which simplifies analysis but
( = 0) must decide when this is a good
approximation (D’ Alembert paradox
body in steady motion CD = 0!)
Viscous flow: retain , i.e., “Real-Flow Theory” more
(  0) complex analysis, but often no choice

5. Rotational vs. Irrotational Flow

=V  0 rotational flow

=0 irrotational flow

Generation of vorticity usually is the result of viscosity 

viscous flows are always rotational, whereas inviscid flows
57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 10
are usually irrotational. Inviscid, irrotational,
incompressible flow is referred to as ideal-flow theory.

6. Laminar vs. Turbulent Viscous Flows

Laminar flow = smooth orderly motion composed of
thin sheets (i.e., laminas) gliding smoothly over each
other

Turbulent flow = disorderly high frequency fluctuations

superimposed on main motion. Fluctuations are visible
as eddies which continuously mix, i.e., combine and
disintegrate (average size is referred to as the scale of
turbulence).
Reynolds decomposition
u  u  u ( t )

mean turbulent fluctuation

motion

usually u  (.01-.1) u , but influence is as if  increased

by 100-10,000 or more.
57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 11
Example: Pipe Flow (Chapter 8 = Flow in Conduits)
Laminar flow:

R 2  r 2  dp 
u (r )   
4  dx 
u(y),velocity profile in a paraboloid

Turbulent flow: fuller profile due to turbulent mixing

extremely complex fluid motion that defies closed form
analysis.

Turbulent flow is the most important area of motion fluid

dynamics research.

The most important nondimensional number for describing

fluid motion is the Reynolds number (Chapter 8)

VD  VD V = characteristic velocity
Re = 
  D = characteristic length
57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 12
For pipe flow
V = V = average velocity
D = pipe diameter

Re < 2300 laminar flow

Re > 2300 turbulent flow

Also depends on roughness, free-stream turbulence, etc.

7. Internal vs. External Flows

Internal flows = completely wall bounded;
Usually requires viscous analysis, except near entrance
(Chapter 8)

External flows = unbounded; i.e., at some distance from

body or wall flow is uniform (Chapter 9, Surface
Resistance)

External Flow exhibits flow-field regions such that both

inviscid and viscous analysis can be used depending on
the body shape and Re.
57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 13
Flow Field Regions (high Re flows)

Vc inertia force
Re  
Important features:  viscous force
1) low Re viscous effects important throughout entire
fluid domain: creeping motion
2) high Re flow about streamlined body viscous effects
confined to narrow region: boundary layer and wake
3) high Re flow about bluff bodies: in regions of adverse
pressure gradient flow is susceptible to separation and
viscous-inviscid interaction is important

8. Separated vs. Unseparated Flow

Flow remains attached

Streamlined body w/o separation

Bluff body Flow separates and creates

the region of reverse
flow, i.e. separation
57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 14

4.3 Basic Control-Volume Approach and RTT

̇
57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 15
 57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 16

Reynolds Transport Theorem (RTT)

Need relationship between

d
B 
dt sys
and changes in
Bcv    dm    d .
CV CV

dBcv d
1 = time rate of change of B in CV =    d
dt dt CV

2 = net outflux of B from CV across CS =

∫

dBSYS d
dt
   d   CS V R  n dA
dt CV

General form RTT for moving deforming control volume

57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 17

Special Cases:

1) Non-deforming CV

dBSYS 
     d    V R  n dA
dt CV
t CS

2) Fixed CV

dBSYS 
     d    V  n dA
dt CV
t CS

Gauss’s Theorem:   b d    b  n dA
CV CS

dBSYS  
          V   d 
dt CV 
t 

Since CV fixed and arbitrary lim gives differential eq.

d0


3) Steady Flow: 0
t

4) Uniform flow across discrete CS (steady or

unsteady)

  V  n dA    V  n dA
CS CS
(- inlet, + outlet)
57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 18

Continuity Equation:

B = M = mass of system
β=1

dM
 0 by definition, system = fixed amount of mass
dt

Integral Form:

dM d
dt
0 
dt CV
 d     V R  n dA
CS

d
   d   CS  V R  n dA
dt CV

Rate of decrease of mass in CV = net rate of mass outflow across CS

Note simplifications for non-deforming CV, fixed CV,

steady flow, and uniform flow across discrete CS

Simplifications:
d
1. Steady flow:   dV  0
dt CV

2. V = constant over discrete dA (flow sections):

 V  dA   V  A
CS CS
57:020 Fluid Mechanics Chapter 4
Professor Fred Stern Fall 2013 19

3. Incompressible fluid ( = constant)

d
 V  dA  
CS
 dV
dt CV conservation of volume

4. Steady One-Dimensional Flow in a Conduit:

 V  A  0
CS

1V1A1 + 2V2A2 = 0

for  = constant Q1 = Q2