ME637 G.

Ahmadi 1
Incompressible Viscous Flows

For an incompressible fluid, the continuity equation and the Navier-Stokes
equation are given as

0 = v , (1)

v v v
v
2
P
1
t
+

= +

. (2)

Using a vector identity, Equation (2) may be restated as

( ) v v v
v
= +

2
v
t
2
(3)

Define vorticity

v = (4)

and taking curl of (3) we find

( ) v

t
. (5)

Noting that curl of gradient is zero and

( ) ( ) ( ) v v v v v + = (6)

Equation (5) may be restated as

v v

2
+ = +

t
(7)

Equation (7) is the vorticity transport equation. It shows that in addition to being
convected and diffused, vorticity is also generated by the first on the right hand side of
Equation (7) by a vortex stretching mechanism.


Two-Dimensional Plane Flows

For two dimensional flows in xy-plane as shown in Figure 1, let


ME637 G. Ahmadi 2
( ) j i
k j i
k v
x y
0 0
z y x

= = , (8)


y v V
u

x


Figure 1. Schematics of plane flows in a Cartesian coordinate system.


That is


y
u

= ,
x
v

= , (9)

and Equation (1) is satisfied.

The nonzero element of is

=

= =
2
Z
y
u
x
v
. (10)

Equation (7) in two-dimensional case reduces to

=

4
y
v
x
u
t
. (11)

Using (10), Equation (11) may be restated as

=

4 2 2 2
y x x y t
. (12)

or


ME637 G. Ahmadi 3

( )
( )
=

4
2
2
y , x
,
t
. (13)

Equation (13) is the equivalent to the Navier-Stokes equation and contains a single
unknown .


Plane Flows in a Cylindrical Geometry


Case (a) 0 v
z
= and
r
v and

v are functions of r and .

y V
v

v
r
r
x




Figure 2. Schematics of plane flows in a polar coordinate system.


For a plane flow in cylindrical geometry as shown in Figure 2, let

( ) ( ) = , r
z
e v . (14)

That is,


r
e e v
r r
1

= , (15)

or




=
r
1
v
r
,
r
v

(16)


ME637 G. Ahmadi 4
The nonzero element of is given by

= =
2
z
(17)

where


2
2
2
2
r
1
r
r
r r
1


+ |
.
|

\
|

= (18)

Equation (7) now becomes

=


+

2
r
r
1
v
r
v
t
. (19)

Using (17), Equation (19) may be restated as

=

4 2 2 2
r
1
r r r
1
t
, (20)

or


( )
( )
=


+

4
2
2
, r
,
r
1
t
. (21)

Equation (21) is the equation governing ( ) , r in plane flows expressed in polar
coordinated system.


Case (b) 0 v =

,
r
v and v
z
are functions of r and z .

z
v
z
V
v
r


r



Figure 3. Schematics of axisymmetric flows in a cylindrical coordinate system.


ME637 G. Ahmadi 5

For an axisymmetric flow in cylindrical coordinates, let

( )
|
.
|

\
|
= z , r
r

e v (22)

That is


z r
e e v
r r
1
z r
1

= , (23)

or


r r
1
v
z

= ,
z r
1
v
r

= . (24)

The vorticity define by Equation (4) now becomes


|
|
.
|

\
|
|
.
|

\
|

= |
.
|

\
|

= =
r r
1
r
r
z r
1
r
v
z
v
2
2
z r

e e v (25)

Thus, the only nonzero component of vorticity is given by

= =

2
E
r
1
, (26)

where


2
2
2
2
2
z r r
1
r
E

= . (27)

Equation (7) may now be restated as

( )
. comp
r
z r
|
r
v
z
v
r
v
t


=

e (28)

Using (26) in (28) we find

( ) ( ) ( ) =

4 2
2
2 2 2
E E
z r
2
E
z r r
1
E
r z r
1
E
t
(29)

or


ME637 G. Ahmadi 6
( )
( )
( )
=

4 2
2
2
2
E E
z r
2
z , r
, E
r
1
E
t
(30)

Equation (30) governs ( ) z , r in axisymmetric cylindrical flows.




Spherical Coordinates

Spherical coordinate system is shown in Figure 4. Here

=
=
=
sin r z
sin cos r y
cos cos r x
(31)


Figure 4. Schematics of spherical coordinate system.


Consider the case when 0 v =

and
r
v and

v are only functions of r and .

Let


( )
|
|
.
|

\
|

=

sin r
, r e
v , (32)

That is

r

x
y
z

ME637 G. Ahmadi 7

r
e e v
r sin r
1
sin r
1
2

= (33)

or

=
sin r
1
v
2
r
,
r sin r
1
v

. (34)

The vorticity equation will reduce to

( )
( )
( )
=
|
.
|

\
|

4
2 2
2 2
2
2
E sin
r
1
cos
r sin r
E 2
, r
, E
sin r
1
E
t
, (35)

where

=
2 2
2
2 2
2
2
r
cot
r
1
r
E . (36)


Intrinsic Coordinates

It is sometimes simpler to work with a coordinate system, which is attached to the
surface of revolution. For the body of revolution shown Figure 5, consider the unit
vectors n, s ,

e . For this system the metrics are

1 h h
n 1
= = , = =

h h
2
, 1 h h
s 3
= = (37)

Thus

=

1
s n
e s n (38)

The stream function for axisymmetric flows may now be introduced as


( )
s n e v
n
1
s
1 n , s

=
|
|
.
|

\
|

=

(39)

That is


s
1
v
n

= ,
n
1
v
s

= (40)


ME637 G. Ahmadi 8
z
n
e

s








Figure 5. Schematics of intrinsic coordinate systems.


The vorticity is now given as

=
(

|
|
.
|

\
|

+
|
|
.
|

\
|

= =

2
E
1
s
1
s n
1
n
e v (41)

where


(

|
|
.
|

\
|

+
|
|
.
|

\
|

=
s
1
s n
1
n
E
2
. (42)

Note that in terms of and z ,
2
E is given by (27) as


2
2
2
2
2
z
1
E

= . (43)


ME637 G. Ahmadi 9
Plane Stagnation Flows

Consider a steady plane stagnation flow shown in Figure 1. For steady plane flow
the Navier-Stokes equation reduces to


( )
=

4
2
) y , x (
,
(1)

or

=

4 2 2
y x x y
(2)



Figure 1. Schematics of plane stagnation flow.


Potential stagnation plane flow is described by

ax U = , ay V = , axy = . (3)

We look for a solution of the form

( ) y xf = . (4)

Then

f x u = , f v = , f x
2
= (5)

Using (4) and (5), equation (2) becomes
y
x

ME637 G. Ahmadi 10


( )
x f x f f f f x
4
= (6)

or


( ) 4
f f f f f = . (7)

Integrating (7) we find

c f f f f
2
+ = (8)

The boundary conditions are:

At 0 y = , 0 u = , 0 v = (9)

At large y, Equations in (3) holds . (10)

That is

( ) ( ) 0 0 f 0 f = = , (11)

As y ay f (12)

Using (12) we find


2
a c = (13)

Thus


2 2
a f f f f + = (14)

Introducing a change of variable (Schlichting,1960)

y
a

= , ( ) = a f , (15)

Equation (14) may be restated as

0 1
2
= + + (16)

subject to boundary conditions

( ) ( ) 0 0 0 = = , (17)


ME637 G. Ahmadi 11

as 1 = (18)

Graphical representation of the numerical solution is shown in Figure 2. Additional
details of the solution are discussed by Schlichting (1960). Accordingly,

At 4 . 2 = , 99 . 0 = , ( ) =
U
u
(19)

Hence, the boundary layer thickness is given by


a
4 . 2

= (20)
1.6
1.2
0.8
0.4
0.8
0
1.6 2.4 3.2
y
a

=
"
'=u/U

Figure 2. Schematics of plane stagnation flow solutions.

ME637 G. Ahmadi 12
Axisymmetric Stagnation Flow

Potential axisymmetric stagnation flow is described as

ar V
r
= , az 2 V
z
= , z ar
2
= . (21)

The steady state Navier-Stokes equation for axisymmetric flows is given as

( ) ( ) =

4 2
2
2 2
E E
z r
2
E
z r r
1
E
r z r
1
, (22)

where


r r
1
v
z

= ,
z r
1
v
r

= , (23)


2
2
2
2
2
z r r
1
r
E

= . (24)


Similar to the plane flow case we look for a solution of the form

( ) z f r
2
= , (25)

with

f 2 v
z
= , f r v
r
= , f r E
2 2
= . (26)

Using (25) and (26), Equation (22) reduces to

( ) ( ) ( )
( ) 4 2 2 2
f r f r f 2 f r f 2 f r 2 f r = (27)

or

0 " " f f f 2 = + (28)

The boundary conditions are

0 v v
z r
= = at 0 z = ; z ar
2
= as z (29)

or

( ) ( ) 0 0 f 0 f = = , ( ) a f = or z ar
2
= as z (30)


ME637 G. Ahmadi 13
Integrating (28) we find


1
2
c f f f 2 f + = (31)

For equation (31) to be valid at large z,
2
1
a c =

Introducing a change of variables

z
a

= , ( ) = a f (32)

Equation (31) becomes

0 1 2
2
= + + (33)

subject to

( ) ( ) 0 0 0 = = , ( ) 1 = (34)

The numerical solutions of the flow filed are very similar to the plane stagnation flow
case shown in Figure 2 with a slightly fuller velocity profile. The details of the numerical
solution are discussed by Schlichting (1960).