Introduction

Fluid Mechanics II

2-D Potential Flows Theory

AAiT School of Mechanical and Industrial Engineering - SMiE

 Integral Relations for Control Volume

Conservation of Mass

The Linear Momentum Equation

The Energy Equation

Simplified assumption: - Steady flow

- Incompressible flow
- Friction forces ignored

AAiT School of Mechanical and Industrial Engineering - SMiE

 Differential Relations for a Fluid Particle

 The viscous solutions for the partial deferential equation is limited to simple
geometries and unidirectional flows, where the difficult nonlinear convective terms
were neglected.

 For more complicated viscous flows, no theory or solutions are found, except
experimental data or computer solutions.

 The starting point for solving a potential flow problem is the set of eqns governing
the flow.

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory

 Potential flow is

 An inviscid, incompressible, irrotational, steady flow

 It uses mainly the simplified version of continuity and Bernoulli
equation

 Potential Flow Theory is

a mathematical method developed to solve flow problems that can

be closely approximated as a potential flow

 Question
What do we mean by ‘solve’ ?

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory
air

body

• Consider an air craft moving in air:

• Questions?
– Why doesn't it fall down ? (What keeps it floating in air?
– Why/How does it move forward?

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory

-p

+p
Surface pressure and shear stress distribution on the surface of airfoil

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory
• Known:
– The forces exerted on bodies moving through a fluid are caused only by two
reasons:

1. Pressure distribution ( p ) over the body surface,

2. Shear-stress distribution ( ) over the body surface.

– Pressure and Shear-stress are the only two mechanisms nature has for
‘communication’ of force & moment between the body and the moving fluid.
• In general,
– The net effect of the p and  distributions integrated over the entire body surface
constitutes a resultant force, R, b/n the fluid and the body; and is given by:

• In particular, R  pdA  dA

entire entire
surface surface

– For a potential flow, there is no ; we have only p

– A good e.g. for ‘solve’ is determining the velocity and pressure distribution, from
which R could be determined (i.e., solving an Engineering problem).

AAiT School of Mechanical and Industrial Engineering - SMiE

 2- D Potential Flow Theory

• Recall the set of Differential Eqns

for a (general) fluid flow

• Consider simplified assumption

of 2-potential flow theory

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory

• When a flow is both frictionless and irrotational, pleasant

things happen (Frank M. White).

• First, the momentum equation reduces to Euler’s equation:

• Second, there is a great simplification in the acceleration

term.

• A beautiful vector identity exists for the second term

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory

• Now combine the above two equation and divide by ρ , and

rearrange on the left-hand side.

• Dot-product the entire equation into an arbitrary vector

displacement dr :

• Nothing works right unless we can get rid of the third term.

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory

• This will be true under various conditions:

1. V is zero; trivial, no flow (hydrostatics).
2. ζ is zero; irrotational flow.
3. dr is perpendicular to ζ x V ; this is rather specialized
and rare.
4. dr is parallel to V ; we integrate along a streamline

• Condition 4 is the common assumption. If we integrate

along a streamline in frictionless compressible flow
and take, g = -gk for convenience, the previous equation
reduces to

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory : Irrotational Flow

• A potential flow is characterized by an irrotational velocity field,

which is a valid approximation for several applications

• The irrotationality of a potential flow is due to the curl of the gradient

of a scalar always being equal to zero.

• If viscous effects are neglected, low-speed flows are irrotational

• One of the most important applications of potential-flow theory is to

aerodynamics and marine hydrodynamics

• Applications of potential flow are for instance: the outer flow field for
aerofoils, water waves, and groundwater flow.

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory : Irrotational Flow

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory: Irrotational Flow

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory: Irrotational Flow

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory: Irrotational Flow-Examples
Flow fields involving real fluids often include both regions of negligible shearing stresses and regions of
significant shearing stresses

Uniform flow in x direction

Various regions of flow: (a) around bodies; (b) through channels
AAiT School of Mechanical and Industrial Engineering - SMiE
 2-D Potential Flow Theory : Velocity Potential

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory: Laplace Equation

 2  2  2 Laplace Equation
  0
2
or  2  2 0
x 2
y z

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory: Velocity Components

 From velocity potential, velocity components can be expressed in

term of scalar function
  
u v w
x y z
 In cylindrical polar coordinates, velocity components

 1  
vr  v  vz 
r r  z

 Laplace’s equation for Cylindrical coordinate

1     1  2  2
r  2 + 2 0
r r  r  r  2
z

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory : Stream functions
• Streamlines of the flow defines lines of constant stream function ,ψ.

• The stream function works only if the continuity equation can be

reduced to two terms. In general, we have four terms:

• For steady flow in x-y plane,

• This equation is satisfied identically if a function ψ ( x , y ) is

defined such that

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory : Stream functions

• It is just a mathematical trick of replacing two

variables (u and υ ) by a single higher-order function
ψ.

AAiT School of Mechanical and Industrial Engineering - SMiE

2-D Potential Flow Theory : Stream functions

• The definition of a streamline in two-dimensional flow is

• Introducing the stream function from, we have

• Thus the change in ψ is zero along a streamline, or

ψ = constant along a streamline

• Having found a given solution ψ ( x , y ), we can plot lines of

constant ψ to give the streamlines of the flow.

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory : Stream functions for polar coordinate

AAiT School of Mechanical and Industrial Engineering - SMiE

 Geometric interpretation of velocity potential function and stream function

AAiT School of Mechanical and Industrial Engineering - SMiE

 Geometric interpretation of velocity potential function and stream function

 For stream function

 
 = (x, y)  d  dx  dy  vdx  udy
x y

     vdx   udy  const

 For velocity potential

 
   ( x, y )  d  dx  dy  udx  vdy
x y

    udx   vdy  const

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory

Elementary Potential Flow Models

Solving the Potential Flow problem

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow Theory: Solving the Potential Flow problem

AAiT School of Mechanical and Industrial Engineering - SMiE

 2-D Potential Flow: Concept behind the superposition principle

AAiT School of Mechanical and Industrial Engineering - SMiE

 Elementary Potential Flow Models
Steps to apply towards the superposition principle

1. Start with known elementary potential flow models like:-

 Uniform flow,
 Source/sink flow,
 Doublet,
 Vortex, etc.

 Note: The velocity potential / stream functions of these flow models

are known.

2. Superimpose/combine linearly a number of the elementary

flow models that together simulate the actual flow pattern, and
then,

a. Determine the stream function of the combined flow from the sum of
the stream functions of the individual elementary flows.

AAiT School of Mechanical and Industrial Engineering - SMiE

 Elementary Potential Flow Models
Steps to apply towards the superposition principle

a. Determine the velocity components of the combined flow from the

combined stream/velocity potential function

b. Determine the pressure distribution of the flow field from Bernoulli

eqn.

c. Determine the engineering parameter (e.g. aerodynamic force

/moment) by integrating the differential pressure force over the entire
surface

AAiT School of Mechanical and Industrial Engineering - SMiE

 Elementary Potential Flow Models: Uniform Flow

   udx   vdy  const     vdx   udy  const

AAiT School of Mechanical and Industrial Engineering - SMiE

 Elementary Potential Flow Models: Uniform Flow

Uniform flow: (a) in the x direction; (b) in arbitrary direction

Uniform flow can be described by either a stream function of a velocity potential.

For case (a)   Ux   Uy

for case (b)   U  x cos   y sin     U  y cos   x sin  

AAiT School of Mechanical and Industrial Engineering - SMiE

 Elementary Potential Flow Models: Uniform Flow (zero divergence & zero curl)

AAiT School of Mechanical and Industrial Engineering - SMiE

 .
Elementary Potential Flow Models: Uniform Flow (zero divergence & zero curl)

 The equivalent irrotationality condition is that ψ(x, y) satisfies

Laplace’s equation.

AAiT School of Mechanical and Industrial Engineering - SMiE

 Elementary Potential Flow Models: Source and Sink

AAiT School of Mechanical and Industrial Engineering - SMiE

 Elementary Potential Flow Models: Source and Sink

AAiT School of Mechanical and Industrial Engineering - SMiE

 Elementary Potential Flow Models: Source and Sink

AAiT School of Mechanical and Industrial Engineering - SMiE

 Elementary Potential Flow Models - Line Vortex

AAiT School of Mechanical and Industrial Engineering - SMiE

 Elementary Potential Flow Models - Line Vortex
 Tangential velocity varies inversely with the distance from the origin, with a
singularity occurring at r = 0

 Since the velocity field is known, one can determine the velocity potential and
stream functions of the model by integration, using:

AAiT School of Mechanical and Industrial Engineering - SMiE

 Elementary Potential Flow Models - Line Vortex

• Vortex motion can be either rotational or irrotational

• Rotation refers to the orientation of fluid element and not the path followed by the
element

• Free vortex is a potential flow

Irrotational (free) vortex Rotational (forced) vortex

AAiT School of Mechanical and Industrial Engineering - SMiE
 Elementary Potential Flow Models - Forced Vortex and combined Vortex

• If fluid rotates as a rigid body, vortex flow is rotational (forced vortex)

• Forced vortex is not potential flow

• Combined vortex is one with a forced vortex as a central core and a

velocity distribution corresponding to that of a free vortex outside the core.
• For a combined vortex
v   r r  r0
K
v  r  r0
r
• where K and  are constants and r0 corresponds to the radius of the central
core

AAiT School of Mechanical and Industrial Engineering - SMiE

 Circulation

• The circulation is defined as the counterclockwise line integral,

around a closed curve C, of arc length ds times the velocity
component tangent to the curve:

AAiT School of Mechanical and Industrial Engineering - SMiE

 Circulation

• From the definition of ϕ, for an irrotational flow

=

•  in an irrotational flow would equal the final value of ϕ minus the

initial value of ϕ

• With ϕ = Kθ for line vortex , there is a change in ϕ of amount 2πK as

for making one complete circle

• Path enclosing a vortex:  = 2πK

 
     ln( r )
2 2

AAiT School of Mechanical and Industrial Engineering - SMiE

 Circulation

• The line vortex flow is irrotational every where except at the origin, where
the vorticity is infinite.

• This means that a certain line integral called the fluid circulation Γ does not
vanish when taken around a vortex center.

• Velocity potential and stream function for the free vortex are commonly
expressed in terms of the circulation as

 
  and    ln r
2 2

AAiT School of Mechanical and Industrial Engineering - SMiE

 Superposition of the Elementary Flows

• We can now form a variety of interesting potential flows by summing the

velocity potential and stream functions of a uniform stream, source or sink,
and vortex

• The streamlines of the combined flow could be obtained, e.g., graphically (i.e.,
by summing up the constant  lines of the elementary flow models).

• Potential flows are governed by Laplace’s equation, which is linear partial

differential equation.

• Any streamline in an inviscid flow field can be considered as a solid boundary

(no flow through the boundary or the streamline)

• If we combine some basic velocity potentials or stream functions to yield a

streamline that corresponds to a particular body shape, that combination can
be used to describe the flow around that body.

• This method is called the method of superposition

AAiT School of Mechanical and Industrial Engineering - SMiE
 Superposition of the Elementary Flows

• A simple means of accomplishing -tot = i graphically is to plot the

individual stream functions separately and then look at their intersections.

AAiT School of Mechanical and Industrial Engineering - SMiE

 Superposition of the Elementary Flows

• Each of the three elementary flow patterns such as uniform flow,

source/sink and line vortex is an incompressible irrotational flow and
therefore satisfies both plane “potential flow” equations:

• These are linear partial differential equations, any sum of such basic
solutions is also a solution

AAiT School of Mechanical and Industrial Engineering - SMiE

 Superposition of the Elementary Flows: Source and sink

• Consider a source +m at (x, y) = (-a, 0), combined with a sink of

equal strength -m, placed at (+a, 0)

AAiT School of Mechanical and Industrial Engineering - SMiE

 Superposition of the Elementary Flows: Doublet

Doublet is formed by a source and sink approach one another (a →0)

K sin  K cos 
  
r r

K is a strength of the doublet

ma
K


AAiT School of Mechanical and Industrial Engineering - SMiE

 Superposition of the Elementary Flows: Doublet

Streamlines for a doublet are circles through the origin tangent to the x axis.

AAiT School of Mechanical and Industrial Engineering - SMiE

 Summary of Basic Potential Flow

AAiT School of Mechanical and Industrial Engineering - SMiE

 Superposition of the Elementary Flows: The Rankine Half Body

• Superimposing a uniform x-directed stream against an isolated source, a

half-body shape appears.

• If the source is at the origin, the combined stream function is, in polar
coordinates,
m m
  Ur sin      Ur cos  ln r
2 2
• A curved, roughly elliptical, half-body shape appears, which separates the
source flow from the stream flow.

• Stagnation point will occur at x = -b (r = b ), where

m
b
2 U

AAiT School of Mechanical and Industrial Engineering - SMiE

 Superposition of the Elementary Flows: The Rankine Half Body

Value of stream function at stagnation point (r = b)

m
   bU
2
Equation of the streamline passing through stagnation point

b    
 bU  Ur sin   bU or r
sin 

AAiT School of Mechanical and Industrial Engineering - SMiE

 Superposition of the Elementary Flows: The Rankine Half Body

 Replace streamline passing through stagnation point with solid boundary

and get the flow around streamlined body placed in uniform stream.
 The body is open at downstream end, and is called a half-body.
 Singularity (source) occurs inside the body, and there are no singularity in
the flow field of interest (outside the body)
 Width of the half-body asymptotically approaches 2 b

AAiT School of Mechanical and Industrial Engineering - SMiE

 Superposition of the Elementary Flows: The Rankine Half Body

For half-body

1  m 
vr   U cos   and v    U sin 
r  2 r r

Square of magnitude of velocity, V, at any point

Um cos  m
V 2  vr2  v2  U 2  
r 2 r
and since b  m 2U

 b b2 
V  U 1  2 cos   2 
2 2

 r r 

With velocity known, the pressure at any point can be determined from Bernoulli
equation
AAiT School of Mechanical and Industrial Engineering - SMiE
 Flow Around of Circular Cylinder

Doublet combined with a uniform flow

can be used to represent flow around a
circular cylinder.
Stream function, velocity potential and
velocity components:
 a2   a2 
  Ur 1  2  sin  ;   Ur 1  2  cos 
 r   r 
 a2   a2 
vr  U 1  2  cos  ; v  U 1  2  sin 
 r   r 
On the surface of the cylinder (r = a)

vr  0 v s  2U sin 

Maximum velocity, 2U, occurs at the top

and bottom of the cylinder and reduces
to U far away from cylinder

AAiT School of Mechanical and Industrial Engineering - SMiE

 Pressure Distribution
Pressure distribution on the cylinder
surface is obtained from Bernoulli
equation
ps  p0  U 2 1  4sin 2  
1
2
Only on the upstream part of the
cylinder is there approximate
agreement between the potential flow
and the experimental results

Because to the viscous boundary

layer that develops on the cylinder,
the main flow separates from the
surface of the cylinder, leading to the
large difference between theoretical,
frictionless fluid solution ant the
experimental results on the
downstream side of the cylinder

AAiT School of Mechanical and Industrial Engineering - SMiE

 Rankine Ovals

Rankine ovals are formed by combining a source and sink with a uniform flow.
m m
  Ur sin   1  2    Ur cos   ln r1  ln r2 
2 2
m  2ay 
alternatively   Uy  tan 1  2 2 
2  x  y a 
2

Flow around a Rankine oval: (a) superposition of source-sink pair and a uniform flow;
(b) replacement of streamline  = 0 with solid boundary to form Rankine oval
AAiT School of Mechanical and Industrial Engineering - SMiE
 Rankine Ovals

Streamline with  = 0 forms closed body which is called Rankine oval

Flow around a Rankine oval: (a) superposition of source-sink pair and a uniform flow;
(b) replacement of streamline  = 0 with solid boundary to form Rankine oval

AAiT School of Mechanical and Industrial Engineering - SMiE

 Rankine Ovals

Stagnation points occur at the upstream and downstream ends of the body
Location of stagnation points depend on a, m and U
1 1
 ma  2 l  m  2
l   a2  or   1
 U  a   Ua 

Flow around a Rankine oval: (a) superposition of source-sink pair and a uniform flow;
(b) replacement of streamline  = 0 with solid boundary to form Rankine oval
AAiT School of Mechanical and Industrial Engineering - SMiE
 Rankine Ovals

Body half-width, h, equals value of y where y axis intersect  = 0 streamline

h2  a 2 2 Uh h 1  h     Ua  h 
2

h tan or     1 tan 2   
2a m a 2  a     m  a

Flow around a Rankine oval: (a) superposition of source-sink pair and a uniform flow;
(b) replacement of streamline  = 0 with solid boundary to form Rankine oval

AAiT School of Mechanical and Industrial Engineering - SMiE

 Rankine Ovals

 Parameter Ua/m determines body shape

 Large value of Ua/m – flow around long slender body

 Small value of Ua/m – flow around blunt body

Flow around a Rankine oval: (a) superposition of source-sink pair and a uniform flow;
(b) replacement of streamline  = 0 with solid boundary to form Rankine oval

AAiT School of Mechanical and Industrial Engineering - SMiE

 Flow Past a Circular Cylinder with Circulation

• At large source strength, the Rankine oval becomes a large circle, much
greater in diameter than the source–sink spacing 2a

• Thus the stream function for flow past a circular cylinder with circulation,
centered at the origin, is a uniform stream plus a doublet plus a vortex:

For all cases the line ψ = 0

corresponds to the circle r = a,
that is, the shape of the cylindrical
body.

Flow past a circular cylinder with circulation for values of K/(U∞a) of 0

AAiT School of Mechanical and Industrial Engineering - SMiE
 Flow Past a Circular Cylinder with Circulation

Flow past a circular cylinder with circulation for values of K/(U∞a) of (a) 0, (b) 1.0, (c) 2.0, and (d) 3.0.

AAiT School of Mechanical and Industrial Engineering - SMiE

 Flow Past a Circular Cylinder with Circulation

• The velocity at the cylinder surface r = a is purely tangential, as expected:

• For small K, two stagnation points appear on the surface at angles θs where
Vθ = 0;

• Figure above is for K = 0, θs = 0 and 180 degree, or doubly symmetric

inviscid flow past a cylinder with no circulation

AAiT School of Mechanical and Industrial Engineering - SMiE

 The Kutta-Joukowski Lift Theorem

• The cylinder flows with circulation, b to d above, develop an inviscid

downward lift normal to the free stream.

• From the streamline patterns that the velocity on top of the cylinder is less,
and, thus, from Bernoulli’s equation, the pressure is higher.

• On the bottom, we see tightly packed streamlines, high velocity, and low
pressure; viscosity is neglected.

• The surface velocity is given by

• From Bernoulli’s equation neglecting gravity, the surface pressure ps is

given by

AAiT School of Mechanical and Industrial Engineering - SMiE

 The Kutta-Joukowski Lift Theorem

 Resultant force (per unit length) developed on the cylinder can be

determined by integrating the pressure over the surface.
2
Fx   ps cos  ad
0

 Drag (force parallel to direction of the uniform flow)

2
Fy    ps sin  ad
0

 Lift (force perpendicular to direction of the uniform flow)

 But the integral of cos θ times any power of sin θ over a full cycle 2π is
identically zero.

 Substitution for ps and integration gives Fx = 0

AAiT School of Mechanical and Industrial Engineering - SMiE

 The Kutta-Joukowski Lift Theorem -d’Alembert Paradox

 Thus, drag as predicted by potential theory (incompressible and inviscid) for a fixed
cylinder in a uniform flow is zero

 However, we know from experience that there is a significant drag developed on a

cylinder when it is placed in a moving fluid

 This discrepancy is known as d’Alembert paradox

 Prandtl first pointed out the profound effect of the thin viscous boundary layer on
the flow pattern in the rear, which .causes drag force.

AAiT School of Mechanical and Industrial Engineering - SMiE

 The Kutta-Joukowski Lift Theorem
• The lift force L normal to the stream, taken positive upward, is given by
summation of vertical pressure forces: where b is the length of the cylinder.

• According to inviscid theory, the lift per unit depth of any cylinder of any shape
immersed in a uniform stream equals ρu∞, where  is the total net circulation
contained within the body shape. The direction of the lift is 90 degree from the stream
direction, rotating opposite to the circulation.
N. Joukowski theorem

AAiT School of Mechanical and Industrial Engineering - SMiE

 Airfoil Theory

• The problem in airfoil theory is thus to determine the net circulation  as a

function of airfoil shape and free-stream angle of attack .

• Even if the airfoil shape and free-stream angle of attack are specified, the
potential flow theory solution is nonunique

• An infinite family of solutions can be found corresponding to different values

of circulation 

• All airfoil theories use the Kutta/Joukowski condition, which is in good

agreement with experiment.

• It turns out that the correct circulation Kutta depends on flow velocity, angle
of attack, and airfoil shape.

AAiT School of Mechanical and Industrial Engineering - SMiE

 Airfoil Theory

The Kutta condition properly simulates the flow

about an airfoil; (a) too little circulation, stagnation
point on rear upper surface; (b) too much,
stagnation point on rear lower surface; (c) just
right, Kutta condition requires smooth flow at
trailing edge.

Case c simulates a real airfoil where the

upper and lower flows meet and leave the
trailing edge smoothly.

If the trailing edge is rounded slightly,

there will be a stagnation point there.

If the trailing edge is sharp, approximating

most airfoil designs, the upper- and lower-
surface flow velocities will be equal as
they meet and leave the airfoil.

AAiT School of Mechanical and Industrial Engineering - SMiE

 Flow Around Rotating Cylinder

Flow around the rotating cylinder is obtained by adding free vortex to flow around a
cylinder. Stream function and velocity potential:

 a2    a2  
  Ur 1  2  sin   ln r   Ur 1  2  cos   
 r  2  r  2

Tangential velocity on the surface of the cylinder (r = a)


v s  2U sin  
2 a

Variety of streamline patterns can be developed, depending on the vortex strength

Location of stagnation point is obtained from expression


sin stag 
4 Ua

AAiT School of Mechanical and Industrial Engineering - SMiE

 Flow Around Rotating Cylinder


sin stag 
4 Ua

The location of stagnation points on a circular cylinder:

(a) without circulation; (b, c, d) with circulation.

AAiT School of Mechanical and Industrial Engineering - SMiE

 Flow Around Rotating Cylinder-Magnus Effect

For cylinder with circulation:

surface pressure 1 2 2 sin  2 

ps  p0  U 1  4sin  
2
 2 2 2
2   aU 4 a U 

drag
Fx  0

lift
Fy   U 

Thus, for the cylinder with circulation lift is developed

If U is positive (in the positive x direction) and  is positive (free vortex with
counterclockwise rotation) lift is downward

Development of lift on rotating bodies is called the Magnus effect

AAiT School of Mechanical and Industrial Engineering - SMiE