Fluids Mechanics III

Patrick

2020
Chapter 1
Kinematics of fluid element

1
Chapter 2
Compressibility effects in moving fluids/ Compressible flows

2
Chapter 3
Boundary Layer Theory
3.1 Introduction
When a real fluid (viscous fluid) flows pas a stationary solid boundary, a layer of fluid
which comes in contact with the boundary surface, adheres to it (on account of viscos-
ity) and condition of no slip occurs (The no slip condition implies that the velocity of
fluid at a solid boundary must be same as that of boundary itself). Thus the layer of
fluid which cannot slip away from the adjacent layers of the fluid, thereby developing a
small region in the immediate vicinity of the boundary surface in which the velocity of
the flowing fluid increases rapidly from zero at the boundary surface and approaches
the velocity of main stream. The layer adjacent to the boundary is known as boundary
layer. Boundary layer is formed whenever there is relative motion between the bound-
ary and the fluid.
 
Since, τ0 = µ ∂u ∂y
, the fluid exerts a shear stress on the boundary and bound-
y=0
ary exerts an equal and opposite force on fluid known as the Shear resistance.

According to boundary layer theory the extensive fluid medium around bodies mov-
ing in fluids can be divided into following two regions.

i. A thin layer adjoining the boundary called the boundary layer where the viscous
shear takes place.

ii. A region ouside the boundary layer where the flow behavior is quite like that of
an ideal fluid and the potential flow theory is applicable

3.2 Boundary Layer Definitions and Characteristics

Boundary layer is a narrow region near the solid surface over which velocity gradients
and shear stresses are large. Consider the parallel flow of a fluid over a thin stationary
flat plate as shown in Figure 3.1. The x-coordinate is measured along the plate surface
from the leading edge of the plate in the direction of flow and y is measured from the
surface in the normal direction. The fluid approaches the plate in the x-direction with
a uniform upstream velocity u which is nearly equal to the freestream velocity U over
the plate away from the surface. At the leading edge of the plate, the thickness of the
boundary layer is zero, but its thickness increases with distance from the leading edge.
The fluid in contact with the boundary has zero velocity and at some distance d from the
boundary, the velocity is nearly U . Thus, a velocity gradient is set up which develops
shear resistance to the flow and thus, it slows down the motion of the fluid. Due to con-
tinued action of shear resistance, a large group of fluid particles is retarded when this
retarded layer of fluid moves downstream. Thus, the thickness of the boundary layer δ
goes on increasing in the downstream direction. This is also referred to as the growth of

3
 Figure 3.1: Boundary layer for flow over a flat plate and different flow regime

the boundary layer. The shear resistance acting in between the adjacent flowing layers
is responsible for rotational flow within the boundary layer.

Near the leading edge of a flat plate, the boundary layer is wholly laminar. For laminar
boundary layer, the velocity distribution is parabolic.

The thickness of the boundary layer (δ) increases with distance from the leading edge
x, as more and more fluid is slowed down by the viscous boundary, becomes unstable
and breaks into turbulent boundary layer over a transition region.

For a turbulent boundary layer, if the boundary is smooth, the roughness projections
are covered by a very thin layer which remains laminar, called laminar sublayer. The
velocity distribution in the turbulent boundary layer is given by Log law and Prandtl’s
one seventh power law

The characteristics of a boundary layer may be summarised as follows:

i δ (thickness of boundary layer) increases as distance from leading edge x in-

creases.

ii δ decreases as U increases.

iii δ increases as kinematic viscosity (V ) increases.

iv τ0 ≈ µ Uδ ; hence τ0 decreases as x increases. However, when boundary layer

becomes turbulent, it shows a sudden increase and then decreases with increasing
x.

v When U increases in the downward direction, boundary layer growth is reduced.

vi When U decreases in the downward direction, flow near the boundary is further
retarded, boundary layer growth is faster and is susceptible to separation.

vii The various characteristics of the boundary layer on flat plate (e.g. variation of δ,
τ of force F) are governed by inertial and viscous forces; hence they are functions
of either Uvx or UvL .

4
 Ux
viii If v
< 5 × 105 · bounadry layer is laminar (velocity distribution is parabolic).

If Uvx > 5 × 105 · bounadry layer is turbulent (velocity distribution follows Log
law or a power law)
Ux
ix Critical value of v
at which boundary layer changes from laminar to turbulrnts
depends on:

– turbulence in ambient flow

– surface roughness
– pressure gradient
– plate curvature, and
– temperature difference between fluid and boundary.

x Through the velocity distribution would be a parabolic curve in the laminar sub-
layer zone, but in view of the small thickness we can reasonably assume that
velocity distribution is linear and so the velocity gradient can be considered con-
stant.

3.2.1 Boundary Layer Thickness (δ)

It is defined as the distance from the boundary surface in which the velocity (u) reaches
99% of the velocity of the mainstream (U). Thus, δ is defined as the distance y from
the surface at which u = U 0.99. This definition provides the approximate value of the
boundary layer thickness and hence, δ is generally known as nominal thickness.

3.2.2 Displacement Thickness δ ∗

Displacement thickness can be defined as the distance perpendicular to the boundary
surface to which the boundary surface has to be displaced into the flow to compensate
for reduction in the discharge due to the formation of boundary layer.

or

It is an additional wall thickness that would have to be added to compensate for the
reduction in flow rate on account of boundary layer formation. Let fluid of density ρ
flow past a stationary pate with velocity U as shown in the Figure 3.2. Consider an
elementary strip of thickness dy at a distance y from the plate. Assuming unit width,
the mass flow speed per second through the elementary strip

m1 = ρudy

The mass of fluid per second flowing through the elementary strip if the plate were not
there;
m2 = ρU dy

5
 Figure 3.2: Displacement thickness

The reduction in mass flow rate is given by,

m1 − m2 = ρ(U − u)dy

[The difference (U − u) is called velocity of defect]. Total reduction of mass flow rate
due to introduction of plate is given by integrating the above expression.;
Z δ
ρ(U − u)dy
0

When the plate is displaced by an amount δ ∗ , the flow through the displaced volume
must be equal to the reduction in mass flow rate as given below.
Z δ
∗
ρU δ = ρ(U − u)dy
0

Thus Z δ
∗ u
δ = 1− dy (3.1)
0 U

3.2.3 Momentum Thickness (θ)

The momentum thickness may be defined as the perpendicular distance by which the
boundary should be displaced to compensate for the reduction in momentum of the
flowing fluid on account of the boundary layer formation.

or

The distance through which the total loss of momentum per second be equal to if it
were passing a stationary plate.

Refer to Figure 3.2. Mass flow per second through the elementary strip = ρudy

Momentum/sec. of this fluid inside the boundary layer = ρudy × u = ρu2 dy

Momentum/sec. of the same mass of fluid before entering boundary layer = ρuU dy

6
Loss of momentum/sec. = ρuU dy − ρu2 dy = ρu(U − u)dy Total reduction in mo-
mentum per second is given by,
Z δ
= ρu(U − u)dy
0

The reduction of momentum per second of fluid flowing through θ with free-stream
velocity U is given by;

= M ass × V elocity = (ρU θ) × U = ρθU 2

Equating expressions; Z δ
2
ρθU = ρu(U − u)dy
0
Z δ
u u
θ= 1− dy (3.2)
0 U U
shape factor (H)
δ∗
H= (3.3)
θ
The shape factor for laminar flow is approximately 2.6 and for turbulent flow, it is about
1.4. Thus, higher the value of shape factor, smoother is the velocity profile.

3.2.4 Energy Thickness(δe )

The energy thickness may be defined as the perpendicular distance by which the bound-
ary should be displaced to compensate for the reduction in energy of the flowing fluid
on account of the boundary layer formation.

Refer to Figure 3.2. Mass of flow per second through the elementary strip = ρudy

K.E. of this fluid inside the boundary layer

1 1
= mu2 = (ρudy)u2
2 2
K.E. of the same mass of fluid before entering the boundary layer
1
= (ρudy)U 2
2
Loss of K.E. through elementary strip
1 1 1
= (ρudy)U 2 − (ρudy)u2 = ρu(U 2 − u2 )dy
2 2 2
Total loss of K.E. of fluid; Z δ
1
= ρu(U 2 − u2 )dy
0 2

7
Let δe = Distance by which the plate is displaced to compensate for the reduction in
K.E.

Then, Loss of K.E. through δe of fluid flowing with velocity U

1
= (ρU δe )U 2
2
Equating the two equations;
Z δ
1 1
(ρU δe )U 2 = ρu(U 2 − u2 )dy
2 0 2
or; Z δ
1
δe = 3 u(U 2 − u2 )dy
U 0
Z δ 
u2

u
δe = 1 − 2 dy (3.4)
0 U U
Example 1 The velocity distribution in the boundary layer is given by Uu = yδ , where
u is the velocity at a distance y from the plate, δ is the boundary layer thickness and
u = U at y = δ . Determine (i) The displacement thickness, (ii) The momentum thick-
ness, (iii) The energy thickness, and (iv) The shape factor.

u y
Solution: Velocity distribution: U
= δ

i The displacement thickness δ ∗ :

Z δ
∗ u
δ = 1− dy
0 U
Z δ δ
y2

y
= 1− dy = y −
0 δ 2δ 0
δ2
 
δ δ
δ∗ = δ − =δ− =
2δ 2 2

ii The momentum thickness, θ:

Z δ
u u
θ= 1− dy
0 U U
Z δ  Z δ
y y2

y y
= 1− dy = − 2 dy
0 δ δ 0 δ δ
 2 δ
y3 δ2 δ3

y δ δ δ
θ= − 2 = − 2 = − =
2δ 3δ 0 2δ 3δ 2 3 6

8
 iii The energy thickness, δe :
δ
u2
Z  
u
δe = 1 − 2 dy
0 U U
Z δ  Z δ
y2 y y3
 
y
= 1 − 2 dy = − 3 dy
0 δ δ 0 δ δ
 2 δ
y4 δ2 δ4

y δ δ δ
= − 3 = − 3 = − =
2δ 4δ 0 2δ 4δ 2 4 4
δ
δe =
4
∗
iv The shape factor δθ :
δ∗ δ/2
= = 3.0
θ δ/6

3.3 Momentum Equation for Boundary layer(Drag Force

on a Flat Plate; Von Karman)
Consider the parallel flow of a fluid with velocity U over a thin stationary flat plate as
shown in Figure 3.3. Consider a small length dx of the plate at a distance x from the
leading edge as shown in Figure 3.3(a) whose enlarged view is shown in Figure 3.3(b).
Assume unit width of the plate as perpendicular to the direction of flow.

Let ABCD be the control volume of the small element of the boundary layer and u
be the velocity at any point within the boundary layer.

Mass flow rate of fluid entering through AD

Figure 3.3: Drag force on a flat plate

9
 Z δ
mAD = ρudy
0

Mass flow rate of fluid leaving through BC

Z δ Z δ 
d(mAD ) d
mBC = mAD + dx = ρudy + ρudy dx
dx 0 dx 0

mass flow rate of fluid entering through DC;

Z δ Z δ  Z δ Z δ 
d d
mDC = mBC −mAD = ρudy + ρudy dx− ρudy = ρudy dx
0 dx 0 0 dx 0

Momentum rate of fluid entering the control volume through AD is given by,
Z δ
MAD = mAD × u = ρu2 dy
0

Momentum rate of fluid leaving the control volume through BC is given by,
Z δ Z δ 
2 d 2
MBC = mBC × u = ρu dy + ρu dy dx
0 dx 0

The fluid enters through side DC with a uniform velocity U and thus, momentum rate
of fluid entering through DC is given by,
Z δ  Z δ 
d d
MDC = mDC × U = ρudy dx × U = ρuU dy dx
dx 0 dx 0

The rate of change of momentum of control volume ABCD is given by:

···
Continue as it is indicated in the book

Local Co-efficient of Drag

∗ τ0
CD = 1
2
ρU 2
Average Co-efficient of Drag
FD
CD = 1
2
ρAU 2

10
3.4 Laminar Boundary Layer
u ∗
No. U
δ CD CD
u y y 2 5.48x √0.73 √1.46



1 U
=2 δ
− δ
√
Rex Rex ReL
u 3 y 1 y 3 4.64x 0.646 1.292



2 U
= 2 δ
− 2 δ
√
Rex
√
Rex
√
ReL
y

3 y 4
u
− 2 yδ + 5.84x 0.686 1.327



3 U
=2 δ δ
√
Rex
√
Rex
√
ReL
u
= sin π2 yδ 4.795x 0.652 1.31x


4 U
√
Rex
√
Rex
√
ReL
5x 0.664 1.328
5 Blasius results (Re < 3.2 × 105 ) √
Rex
√
Rex
√
ReL

Example 1

Determine the expressions for boundary layer thickness (δ) , shear stress (τ0 ), local
∗
coefficient of drag (CD ) and coefficient of drag (CD ) in terms of Reynolds number for
the velocity profile of laminar boundary layer given by (u/U ) = 2(y/δ) − (y/δ)2 [ht]

11
12
 Example 2

A plate of length 0.5 m and width 0.25 m is placed longitudinally in a fluid of specific
gravity 0.9 and of kinematic viscosity one stoke. If the fluid is moving with a velocity
of 5 m/s, then determine (i) friction drag on the plate, (ii) thickness of boundary layer
and (iii) shear stress at the trailing edge of the plate. Use Blasius solution.

3.5 Turbulent Boundary Layer

As compared to laminar boundary layers, the turbulent boundary layers are thicker.
Further in a turbulent boundary layer the velocity distribution is much more uniform,
than in laminar boundary layer, due to intermingling of fluid particles between different
layers of the fluid. The velocity distribution in a turbulent boundary layer follows a
logarithimic law i.e. u log y, which can also be represented by a power law of the type
u  y n
= (3.5)
U δ

13
 1
where, n = 7
(approx.) for Re < 107 but > 5 × 105

u  y  17
= (3.6)
U δ
This is known as one-seventh power law.

 eqn. 3.6, however, cannot be applied at the boundary itself because at y = 0,

The
δu
δy
= 17 U δ −1/7 y −6/7 = ∞. This difficulty is circumvented by by considering the
velocity in the viscous sub-layer to be linear and tangential to the seventh-root profile at
the point , where the laminar sub-layer merges with the turbulent part of the boundary
layer.

Blausius suggested the following relation for viscous shear stress:

 1/4
2 µ
τ0 = 0.0226ρU (3.7)
ρU δ

(for Re ranging from 5 × 105 to 107 )

Boundary layer thickness, δ

0.371x
δ= (3.8)
(Rex )1/5

Shear Stress τ0
1/4
ρU 2

2 µ 0.576
τ0 = 0.0225ρU = × (3.9)
ρU δ 2 (Rex )1/5

∗
Local co-efficient of drag (CD )
∗ 0.576
CD = (3.10)
(Rex )1/5

Drag force, (FD )

ρU 2 0.072
FD = × ×B×L (3.11)
2 (ReL )1/5

Co-efficient of drag, (CD )

0.072
CD = (3.12)
(ReL )1/5
This is valid for 5 × 105 < ReL < 107 .

14
For Reynolds number between 107 and 109 the following relationship suggested by
Prantl and Schlichting holds good,
0.455
CD = (3.13)
(log10 ReL )2.58

Example

A smooth flat plate of length 5 m and width 2 m moves with a velocity of 5 m/s in
a uniform stationary air stream. Find (i) the thickness of the boundary layer at the trail-
ing edge of the plate if kinematic viscosity and density of air are 15 × 10−6 m2 /s and
1.23kg/m3 , respectively. Also determine (ii) the coefficient of drag, (iii) total drag on
one side of the plate if the boundary layer is turbulent from the very beginning, (iv)
shear stress and (v) the thickness of laminar sublayer.

15
3.6 Total Drag due to Laminar and Turbulent Layers

16
 Example

A barge with bottom surface in rectangular shape is 25 m long and 10 m wide, which
is travelling down a river with a velocity of 0.7 m/s. A laminar boundary layer is ex-
isted up to a Reynolds number equivalent to 5 × 105 and subsequently, abrupt transition
occurs to turbulent boundary layer. If the density and kinematic viscosity of water are
998kg/m3 and 10−6 m2 /s, respectively, then determine (i) the maximum distance from
the leading edge up to which the laminar boundary persists and the maximum boundary
layer thickness at that point, (ii) the total drag force on the flat bottom surface of the
barge and (iii) power required to push the bottom surface through water at the given
velocity.

17
 Example A streamlined train is 240 m long with a typical cross section having a
perimeter of 8.5 m above the wheels. If the boundary layer changes from laminar to tur-
bulent on the train surface and the density and kinematic viscosity of air are 1.24kg/m3
and 1.5×10−5 m2 /s, respectively, then determine the approximate surface drag (friction
drag) of the train when running at 84 km/hr.

3.7 Boundary Layer Separation, its Effects, and Con-

trol
The pressure gradient (∂p/∂x) in the direction of flow greatly affects the boundary
layer thickness. If the pressure gradient is zero, then the boundary layer continues to
grow in thickness along a flat plate. If the pressure gradient is negative (i.e., pressure
decreases in the direction of flow), then the boundary layer reduces in thickness and
is held in place. The transfer of momentum from the main flow to the boundary layer
will sustain the flow in the boundary layer. However, if the pressure gradient is positive
or adverse (i.e., pressure increases in the direction of flow), then the boundary layer
thickens rapidly, the flow decelerates and the velocity pattern reverses or back flow sets
in. The flow near the boundary is continuously retarded and a point is reached when the
flow starts separating from the boundary. The point at which the flow separates from

18
the boundary is called separation point. The positive pressure gradient reduces the mo-
mentum of flow of the fluid within the boundary layer due to higher shear stresses and
causes separation from the solid surface.

Consider the flow past a curved surface (i.e., converging-diverging boundary surface)
as shown in Figure 3.4. As fluid flows in the region of converging boundary ABC, it
is accelerated and the velocity becomes maximum at C where pressure is minimum.
Thus, the pressure decreases in the direction of flow (i.e., pressure gradient becomes
negative) which pushes high pressure region to the low pressure region. Therefore, the
boundary layer remains thin and is held in place on the solid surface.

In the diverging portion CDE, the flow is decelerated and pressure increases in the
direction of flow (i.e., pressure gradient becomes positive). As a result, the net pressure
force in an element of fluid in the boundary layer opposes the forward flow. Thus, at a
certain distance, on the downstream of point C, the fluid near the boundary surface is
soon brought to a standstill. The value of the velocity gradient (∂u/∂y) at the bound-
ary surface is then zero (at point D) and the fluid no longer follows the contour of the
curved surface and it separates from it. This point D is known as separation point. On
downstream of separation point, a further retardation of the fluid close to the boundary
has a reverse or back flow in the separated region known as wake. If all the points be-
low which a reverse flow occurs are joined by a smooth curve, then a line dividing the
forward and reverse flows is obtained which is known as separation streamline. Large
irregular eddies and turbulence developed in the wake causes loss of energy and de-
crease in efficiency.

The flow separation depends on the curvature of the surface, the Reynolds number of
the flow and the roughness of the boundary surface. The following conditions determine
the boundary layer separation.

19
 Figure 3.4: Separation of boundary layer

Separation occurs in the following cases;

(i) Diffusers

(ii) Open channel transitions

(iii) Pumps

(iv) Fans

(v) Aerofoils

(vi) Turbine blades etc.

3.7.1 Effects of Boundary Layer Separation

The boundary layer separation is unstable and it is an inefficient process. For external
boundary layer separation, it leads to increase in pressure drag which is much more
than frictional drag. For internal flows, boundary layer separation causes increase in
flow losses. It occurs in diffusers, turbine blades, fans, pumps, aerofoils, etc.

3.7.2 Methods of Controlling Separation

The methods which are generally adopted to retard the flow separation is as follows.

1. Streamlining of body shapes shifts the point of separation downstream and thereby,
reduces the wake region.

2. Suction of the retarded layers by suction slots.

3. Artificial roughening of the boundary surface produces an early onset of turbu-

lence which resists the separation.

20
4. Providing small divergence in a diffuser.

5. Acceleration of the fluid in the boundary layer.

6. Motion of solid boundary.

7. Guidance of flow in a confined passage.

8. Providing a rotating cylinder near the leading edge which induces Magnus effect
and the fluid remains attached to the upper surface of the body for its full length

21