Anda di halaman 1dari 23

Ghosh - 550 Page 1 12/27/2017

Worked Out Examples


(External Flows)

Example 1 (Velocity Profiles):

Velocity profiles in laminar boundary layers often are approximated by the equations

u y
Linear: 
U 
u  y 
Sinusoidal:  sin  
U 2
2
u  y  y
Parabolic:  2    
U    

Compare the shapes of these velocity profiles by plotting y/ (on the ordinate) versus u/U
(on the abscissa).

1. Statement of the Problem


a) Given
 Three approximated velocity profiles in laminar boundary layers, linear, sinusoidal,
and parabolic.

b) Find
 Compare these three approximated velocity profiles by plotting.

2. System Diagram
It is not necessary for this particular problem.

3. Assumptions
 Laminar boundary layer

4. Governing Equations
None. Just plot them.

5. Detailed Solution
There is no detailed discussion for this problem. Just plot and compare them.
Using MatLab the plots look like:
Ghosh - 550 Page 2 12/27/2017

Approximated velocity profiles in laminar boundary layer


1
Linear
0.9 Sinusoidal
Parabolic
0.8

0.7

0.6
y/

0.5

0.4

0.3

0.2

0.1

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
u/U

By common sense, it can be said that the linear profile is not very close approximation to the
actual shape of boundary layer velocity profile. The plot for the parabolic profile is the
closest approximation to the Blasius solution for velocity profile.

6. Critical Assessment
Note that the velocity profiles are only useful for 0<y<. Although the given equations
hold outside this range, the curves have no meaning for y> in boundary layer theory.

Example 2. (Boundary Layer Thicknesses):

Velocity profiles in laminar boundary layers often are approximated by the equations

u y
Linear: 
U 
u  y 
Sinusoidal:  sin  
U 2
2
u  y  y
Parabolic:  2    
U    

Calculate  * (displacement thickness) and  (momentum thickness) for these velocity


profiles and compare the result for each case.

7. Statement of the Problem


c) Given
Ghosh - 550 Page 3 12/27/2017

Three approximated velocity profiles, linear, sinusoidal, and parabolic, in laminar


boundary layers.
d) Find
 * (displacement thickness) for three approximated velocity profiles
  (momentum thickness) for three approximated velocity profiles
 Compare the result for each case

8. System Diagram
Approximated velocity profiles in laminar boundary layer
1
Linear
0.9 Sinusoidal
Parabolic
0.8

0.7

0.6
y/

0.5

0.4

0.3

0.2

0.1

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
u/U

9. Assumptions
 Steady state condition
 Laminar boundary layer

10. Governing Equations


 Displacement thickness definition
 u    u
* 
0



1  dy   1  dy
U  0
 U 
since u  U at y =  , the integrand is essentially zero for y   .

 Momentum thickness definition


 u  u  u  u
  1  dy  0 1  dy
0 U U U U

Again, the integrand is essentially zero for y   .

11. Detailed Solution

y
Let   , then dy =  d because  = (x).

Ghosh - 550 Page 4 12/27/2017

* (Displacement thickness)

 Linear velocity profile


  y
 linear   1    dy
*
0
 

 1      d
1

0
1
 1 
      2 
 2 0
1
 
2

 Sinusoidal velocity profile


    y 
 sin
*
  1  sin    dy
0
  2  
1   
  1  sin  2      d
0

1
 2   
     cos  
   2  0
 2
 1    
 

 Parabolic velocity profile


  
  y  y 
2

 parabolic   1  2       dy
*

      
0

 1   2       d
1
 2
0
1
 1 
      2   3 
 3 0
1
 
3

 (Momentum thickness)

 Linear velocity profile


Ghosh - 550 Page 5 12/27/2017

 y y
 linear  
0
1    dy
 

  1       d
1

0
1
      2  d
0
1
1 1 
    2  3
 2 3 0
1
 
6

 Sinusoidal velocity profile


  y     y 
 sin   sin    1  sin   dy
0
2    2  
1      
  sin     1  sin      d
0
2    2 
1    
    sin     sin 2     d
0
2  2 
1
     
  sin 2    
2     2  
   cos       
   2  2  
 4  
  2  0
 1 2
    
 2 

 Parabolic velocity profile


   y  y      y  y 
2 2

 parabolic    2       1  2       dy
         
     
0

1
  
  2   2  1  2   2    d
0

1

      4  4 3  5 2  2  d
0

1
 1 5 
     5   4   3   2 
 5 3 0
2
 
15

Comparison
 */   /
Linear 1/2 = 50% of B.L. 1/6 = 16.7% of B.L.
Sinusoidal 1-2/ = 36.3% of B.L. -1/2+2/ = 13.7% of B.L.
Parabolic 1/3 = 33.3% of B.L. 2/15 = 13.3% of B.L.
Ghosh - 550 Page 6 12/27/2017

*** Note: B.L. means "Boundary Layer."

12. Critical Assessment


Understand the concepts of displacement thickness ( *) and momentum thickness ().
This problem illustrates how to calculate them from velocity profiles. The above table
shows that <*< for most types of velocity profiles.

Example 3. (Use of the displacement thickness):

Air flows in the entrance region of a square duct, as shown. The velocity is uniform, V1 = 30
m/s, and the duct is 80 mm square. At a section 0.3 m downstream from the entrance, the
displacement thickness,  *, on each wall measures 1.0 mm. Determine pressure change
between sections  and .

V1

80 mm

  80 mm
*2 = 1.0 mm
1. Statement of the Problem
a) Given
 Working fluid is air which has air = 1.23 kg/m3 at T = 15C.
 Uniform flow at the entrance, V1 = 30 m/s.
 Duct is H = 80 mm square.
 Displacement thickness, *2 = 1.0 mm, on each wall at a section L = 0.3 m
downstream from the entrance.

b) Find
 Pressure change between sections  and .

2. System Diagram

L = 0.3 m
V1

3. H = 80 mm

Assumptions
  H = 80 mm
*2 = 1.0 mm
Ghosh - 550 Page 7 12/27/2017

 Steady state condition


 Incompressible fluid flow
 No frictional effects in freestream
 Flow uniform at each section outside *2
 Flow along a streamline between sections  and 
 Negligible elevation changes

4. Governing Equations
  
 0
t 
CV
dV  
V  dA … Integral version of mass conservation
CS
 
Incompressible fluid flow problem, the equation above  0   V  dA CS

1 inlet () and 1 outlet ()  0  V1 A1  V2 A2


p V2
 Bernoulli's Equation:   gz  const.
 2
Restrictions:
(1) Steady flow
(2) Incompressible flow
(3) Frictionless flow
(4) Flow along a streamline

5. Detailed Solution

Use the displacement-thickness concept to find the effective flow area for the freestream flow
outside the thin wall boundary layers. Replace the actual boundary-layer velocity profiles
with uniform velocity profiles as sketched in the following figures.

V V

H - 2 *

*
H - 2 *
(a) Actual velocity profile (b) Hypothetical velocity profile (c) Cross section of duct

Apply the continuity and Bernoulli equations to freestream flow outside the boundary-layer
displacement thickness, where viscous effects are negligible.

From Bernoulli's equation, we obtain


p1 V1 p V
  gz1  2  2  gz2
 2  2
Ghosh - 550 Page 8 12/27/2017

 V22  V12 
1
p1  p 2 
2
From the continuity equation, we have
A1
0  V1 A1  V2 A2  V2  V1
A2

Substituting this expression into Bernoulli's equation,


1  A1   1  A 
2 2

p1  p 2    V1   V12   V12  1   1
2  A2   2  A2  

Areas, A1 and A2, are


A1  H 2
A2   H  2 *  … (only effective flow area)
2

Thus,

 H 2  

2
1 2 
p1  p 2  V1    1
2   H  2 *  2  
 

After plugging in values, p1  p 2  58.99 Pa

6. Critical Assessment
To solve this problem, it is critical to understand the meaning and physical
interpretation of displacement thickness concept.

Example 4. (Use of the Momentum Integral Method):

The velocity profile in a laminar boundary-layer flow at zero pressure gradient is


u y
approximated by the linear expression,  . Use the momentum integral equation with
U 
this profile to obtain expressions for  /x and Cf.

1. Statement of the Problem


a) Given
 Laminar boundary-layer flow
 Zero pressure gradient
u y
 Velocity profile is approximated by the linear expression,  
U 
b) Find
Using the momentum integral equation, obtain expression for
  /x
 Cf
Ghosh - 550 Page 9 12/27/2017

2. System Diagram
U

y
u y
x, u 
U 

3. Assumptions
 Steady state condition
 Incompressible fluid flow

4. Governing Equations
 Momentum integral equation

w


d
dx
U 2    *U
dU
dx

where
  u
 *    1   dy … displacement thickness
0
 U
 u u
   1    dy … momentum thickness
0 U U 
5. Detailed Solution

For the special case of flow over a flat plate, U = constant. From Bernoulli's equation, we
see that for this case, p = constant, and thus dp/dx = 0.

The momentum integral equation then reduces to

d d  u u
 w  U 2
dx
 U 2  1    dy
dx U  U 
0

u y
Define     y    dy    d     x  
U 
Ghosh - 550 Page 10 12/27/2017

d  u u
  w  U 2  
dx 0 U 
1    dy
U
d 1
 1       d
dx 0
 U 2

d 1
 U 2
dx 0

   2  d 
1
d 1 2 1 3 
 U 2  2   3  
dx 0

d  1 1  
 U 2      0  0 
dx  2 3  

1 d
 w  U 2
6 dx

u
On the other hand, the shear stress can be calculated by  w  
y .
y 0

u y U
And   u  y , thus
U  

 U 
w    y
y    y 0

U 
   y
 y y 0

U
  1
 y 0

U
 w  

Comparing (equating) this shear stress equation with the previous shear stress expression,

U 1 d
W    U 2
 6 dx

d 6 1
 
dx U 
6
  d   dx
U
6
   d  U  dx
1 2 6
   x  const
2 U

When x = 0,  = 0  const = 0.

12 12 x 2
2  x  
U U x
Ghosh - 550 Page 11 12/27/2017

12 
   x   0 
Ux

 12 12 3.46
  
 x Ux Re x Re x

w
Cf 
Skin friction coefficient is defined as 1 .
U 2
2
Using the result obtained above,

1 d
U 2
Cf  6 dx
1
U 2
2
1 d

3 dx
1 d  12 
   x
3 dx  U 

1 12  1 1
  
3 U 2 x
1 12 

6 Ux
1 12

6 Ux

12 1 0.577
C f   
6 Re x Re x

6. Critical Assessment
This problem dealt with linear velocity profile as an approximate solution. The results
obtained are rough. However the exercise illustrates the use of the momentum integral
method. Practice this method with other types of approximated velocity profile, such as
parabolic, sinusoidal, … etc.

Example 5. (Friction Drag Calculation):


Ghosh - 550 Page 12 12/27/2017

Water at 15  C flows over a flat plate at a speed of 1 m/s. The plate is 0.4 m long and 1 m
wide. The boundary layer on each surface of the plate is laminar. Assume that the velocity
profile may be approximated as linear. Determine the drag force on the plate.

1. Statement of the Problem


a) Given

Working fluid is water at T = 15 C   = 999 kg/m3 &  = 1.14  10-3 Ns/m2

U = 1 m/s

L = 0.4 m

W=1m

The boundary layer on each surface of the plate is laminar

Velocity profile is linear (assuming approximately)
b) Find
 Drag force on the plate

2. System Diagram

U = 1 m/s

L = 0.4 m
3. Assumptions
 Steady state condition
 Incompressible fluid flow
 Laminar boundary layer

4. Governing Equations
w
Cf 
 Skin friction coefficient definition: 1
U 2
2
Ux
 Reynolds number definition for a flat plate: Re x 

5. Detailed Solution
u y
We know that for a linear velocity profile   ,
U 
0.577
Cf 
Re x

Equating this result and the definition of skin friction coefficient,


Ghosh - 550 Page 13 12/27/2017

0.577 w 1 0.577
Cf     U 2 
Re x 1 2  w
U 2 Re x
2
1 0.577
 w  U 2 
2 Ux

Drag force on one side of the plate is given by FD   Ap


 w  dA .
L
Since dA = w  dx and 0  x  L, FD  Ap
 w  dA  
0
 w  w  dx .
L 1 0.577
FD   0 2
U 2 
Ux
 w  dx


0.577 1 L 1

2
U 2 w
x 
0
x
 dx


L
0.577 1  1
 U 2 w  2 x 2 
2 x  0

0.577 1
 FD  U 2 w 2 L
2 x

Plug in values into this expression obtained above for FD,  FD  0.3894 N.
For both sides of the plate  FD ,Total  2 FD  0.779 N.

6. Critical Assessment
Problem says, "the boundary layer on each surface of the plate is laminar." Let us
double check that this is true.
UL
Re L   592.05 << 500,000 = Recr  Obviously it is a laminar flow.

(Note: This problem could be solved by first obtaining the Overall Skin Friction
Coefficient, C f . In that case, the calculation will proceed by obtaining
1
C f   C f ( x)dx , where the integration limits will be set at x = 0 and x = L.
L
1
Then C f   w  U .C f  FD  2 w . Aw , where, Aw (=W.L) indicates the wet
2

2
area on each face of the plate.)
Ghosh - 550 Page 14 12/27/2017

Example 6. (Power Calculation using Friction Drag):

A flat-bottomed barge, 25 m long and 10 m wide, submerged to a depth of 1.5 m, is to be


pushed up a river at the rate of 8 km/hr. Estimate the power required to overcome skin
friction if the water temperature is 15  C.

1. Statement of the Problem


a) Given

L = 25 m

W = 10 m

D = 1.5 m

V = 8 km/hr = 2.222 m/s

Working fluid is water at T = 15 C   = 999 kg/m3 &  = 1.14  10-3 Ns/m2
b) Find
 Power required to overcome skin friction

2. System Diagram

D
Water line

3. Assumptions
 Model a flat-bottomed barge as a flat plate
 Steady state condition
 Incompressible fluid flow
 Neglect separation

4. Governing Equations
Ghosh - 550 Page 15 12/27/2017

FD
CD 
 Drag Coefficient Definition: 1
V 2 A
2
5. Detailed Solution

First of all, calculate Reynolds number:


VL  999kg / m 3  2.222m / s  25m 
Re L    4.87  10 7
 1.14  10 N  s / m 
3 2

Transition of laminar to turbulent flow occurs at


Vxcr
Re cr  5  10 5 

 xcr 
5  10    5  10 1.14  10 N  s / m   0.25678m << 25 m
5 5 3 2

V 999kg / m  2.222m / s 
3

This xcr shows that the effect of laminar flow is negligible. It can be said that the flow is
turbulent from the leading edge.

For ReL < 109, the empirical equation given by Schlichting


0.455
CD 
 log Re L  2.58
fits experimental data very well.

Friction force is (from the definition of drag coefficient)


1 1 0.455
FD  V 2 A  C D  V 2 A 
2 2  log Re L  2.58
1 0.455
  FD  V  V 2 A  V
2  log Re L  2.58 , where A is the wetted area, A = LW+2(LD).
Finally,
1 0.455
  V 2  L  W  2 L  D    V
2  log Re L  2.58
Plug in values into this expression   = 4200.8 W = 4.20 kW

6. Critical Assessment
Drag coefficient must be chosen depending upon the value of Reynolds number for a
particular flow condition. Some of CD expressions are derived by analytical calculation,
and others are empirical formulas.
Ghosh - 550 Page 16 12/27/2017

Example 7. (Flow Separation Characteristics):

Two hypothetical boundary-layer velocity profiles are shown. Obtain an expression for the
momentum flux of each profile. If the two profiles were subjected to the same pressure
gradient conditions, which would be more likely to separate first? Why?

U U

2
u  y  y
   2    
u U    

1. (a) (b) Statement of the Problem


a) Given
 Two hypothetical boundary-layer velocity profiles
b) Find
 Expression for the momentum flux of each profile
 Which would be more likely to separate first if the two profiles were subjected to the
same pressure gradient conditions? And why?

2. System Diagram

U U

2
u  y  y
u y    2    
 u U    
U 

3. Assumptions
(a) (b)
 Steady state condition
 Incompressible fluid flow

4. Governing Equations
 Definition of Momentum Flux (mf)
Ghosh - 550 Page 17 12/27/2017

  
d  mf   V  V  d A
Ghosh - 550 Page 18 12/27/2017

5. Detailed Solution

Since the flow is 1 - D (positive x direction) and dA = w  dy, the momentum equation can be
written as
d  mf   u    u  dA  u    u  w  dy

 
 mf  0 u    u  w  dy  0 u    u  w  dy
 The integrand is essentially zero for y  .
Linear Velocity Profile
u y U
 u y
U  

 U  U 
 mf linear    y     y   w  dy
0 
   
U2 
 w y 2  dy
 2 0


U 2 1 3 
 w y
 2  3  0

U 2 w 
Finally, mf linear  .
3

Parabolic Velocity Profile


2
u  y  y
 2    
U    

y u 1
Let   . Then  2   2 and d  dy     x .
 U 

Now,

mf parabolic   u    u  w  dy
0
1
  u    u  w    d
0
1
   w     u 2  d
0
1
 
   w     U 2   2
0
 2
d
1
   w    U 2   4 2  4 3   4 d
0
Ghosh - 550 Page 19 12/27/2017

1
4 1 
   w   U 2    3   4   5 
3 5 0

8 U 2 w 
Finally, mf parabolic  .
15

Which separates first?


Separation occurs when the momentum of fluid layers near the surface is reduced to zero by
the combined action of pressure and viscous forces.

As shown in this figure below, the momentum of the fluid near the surface is greater for the
parabolic velocity profile.

Velocity Profiles Momentun-Flux Profiles


1 1

0.8 0.8
Linear
0.6 Linear 0.6
y/ 

y/ 

0.4 0.4 Parabolic


Parabolic
0.2 0.2

0 0
0 0.5 1 0 0.5 1
u/U 2
(u/U)
U w  2
8 U 2 w
Our previous calculation also shows mf linear  < mf parabolic  .
3 15
Consequently, the parabolic velocity profile is better able to resist separation in the same
pressure gradient condition.

 Linear velocity profile would separate first.

6. Critical Assessment
Review and understand how the flow separation occurs. Flow separation occurs only
when there exists an adverse pressure gradient.

Example 8. (Terminal Velocity Calculation):

A small sphere (D = 6 mm) is observed to fall through caster oil at a terminal speed of 60
mm/s. The temperature is 20  C. Compute the drag coefficient for the sphere. Determine the
density of the sphere. If dropped in water, would the sphere fall slower or faster? Why?
Ghosh - 550 Page 20 12/27/2017

1. Statement of the Problem


a) Given

D = 6 mm = 0.006 m

Working fluid is caster oil at T = 20 C  S.G.oil = 0.969 & oil = 0.9 Ns/m2

Vt = 60 mm/s = 0.06 m/s
b) Find
 Drag coefficient for the sphere
 Density of the sphere
 If dropped in water, would the sphere fall slower or faster? Why?

2. System Diagram

Caster oil at T = 20C

Vt

3. Assumptions
 Steady state condition
 Incompressible fluid flow

4. Governing Equations
FD
CD 
 Drag Coefficient Definition: 1
V 2 A
 2
dP  
 Newton's Second Law:
dt
  F , where P is momentum.

When the mass is constant, ma   F  1 - D in y direction  ma y   Fy

VD
 Reynolds Number for Sphere: Re D 

Ghosh - 550 Page 21 12/27/2017

5. Detailed Solution

Drag Coefficient
First of all, calculate Reynolds number:
 oilVt D  S .G.oil   water   Vt D  0.969 1000kg / m 3  0.06m / s  0.006m 
Re D   
   0.9 N  s / m 2 
ReD = 0.3876 < 1  There is no flow separation from a sphere. The wake is laminar and the
drag is predominantly friction drag.

Stokes has shown analytically, for very low Reynolds number flows where inertia forces may
be neglected, that drag force on a sphere of diameter D, moving speed V, through a fluid of
viscosity , is given by
FD  3VD
The drag coefficient, CD, is then
FD 3VD 24 24
CD    
1 1    VD Re D
V 2 A V 2  D 2 
2 2 4  
 2
(Note: For sphere, the area, A, is just a cross-sectional area, which is D .)
4
Thus,
24 24
CD    61.92
Re D 0.3876

Density of the Sphere

Free Body Diagram


W  mg   sVs g
FD FB

FD  3 oilVt D

y FB  Woil   oilVs g
displaced

The sphere reached the terminal speed  ay = 0


ma y   Fy

0  FD  FB  W
0  3 oil Vt D   oilV s g   sVs g
3 oil Vt D   oil Vs g
s 
Vs g
Ghosh - 550 Page 22 12/27/2017

 4  D 3 
3 oil Vt D   S .G.oil   water         g
 3  2  
s 
4  D   3

    g
 3  2  

After plug in values into this expression, s = 3721 kg/m3.

If dropped in water …

If the working fluid is water at T = 20 C  w = 998 kg/m3 & w = 1 10-3 Ns/m2

Because w = 1 10-3 Ns/m2 << oil = 0.9 Ns/m2, the author guesses the sphere drops faster
in water than in caster oil.

If it's faster and w << oil , Re D  Re D  0.3876 .


water oil

 FD = 3VD cannot be used because the equation works only for very low Reynolds
number which we don't know whether this is appropriate or not any more for this case.

Now, guess a value of CD from Figure 9.11 (Drag coefficient of a smooth sphere as a function
of Reynolds number) and calculate VtW. Then calculate ReD and verify the chosen CD was
appropriate or not.

1
Guess CD = 0.4  FD   wVtW2 A  C D
2

Free Body Diagram (again)


W  mg   sVs g
FD FB

1 2
FD   wVtW A  C D
2
y

FB  Wwater   wVs g
displaced
W

The sphere reaches a new terminal speed, VtW  ay = 0


ma y   Fy

0  FD  FB  W
1
0   wVtW2 A  C D   wVs g   sVs g
2
Ghosh - 550 Page 23 12/27/2017

 4  D 3 
 s  w         g
  s   w Vs g  3  2  
VtW  
1 1  2 
 w AC D  w D   CD
2 2  4 

After plugging values into this expression, VtW = 0.732 m/s.

 wVtW D
With this new terminal speed, Reynolds number is Re D   4383 .
w
Figure 9.11 says when CD = 0.4, ReD  4  103, which is about right for this case. This shows
the new terminal speed is a valid number.

VtW = 0.732 m/s > Vt = 0.06 m/s  The sphere drops faster in water than in caster oil.

6. Critical Assessment
Drag coefficient depends upon the value of Reynolds number. Be careful with choosing a
right CD depending on a particular flow condition.

Anda mungkin juga menyukai