MOMENTUM TRANSFER

FOR MORE THAN ONE

INDEPENDENT
VARIABLE
 Unsteady viscous flow
• More than one independent variable  PDE partial
• Terms PDE partial  Chem Eng Math
• Example :
– Flow near a wall suddently set in motion
– Newtonion fluid, constant ρ

t = 0, fluid and solid are at rest

t > 0, solid surface is set in motion
 v x  vx
2
  2
t y
• Equation of motion for Cartesian coordinate with
constant ρ in x direction
• From table 3.4-2 D, vz = vy = 0, unsteady state
we get PDE
  2v x
 vx   2
t y
• With condition:
– IC : t = 0 vx = 0
– BC : y = 0 vx = V
y = ∞ vx = 0
 separation variable
• PD partial  DE ordinary
Laplace transform
substitution variable
Substitution variable
a b
- combine variable :   c1 y t

2 2
 v x 2   vx
b  a
t  y 2  2

1 v x 2 2 ( 2 a  2 ) 2b  2v x a = 1 dan b = -1/2
bt  a c y t
  2
 y
 c
t ≈2

 2v x  v x c
1
  0
 2 2c 2  4

 2v x v x
 2 0
 2 
vx
 1  erf ( )
vx 2   2 d V
 1 e
V  0 vx y
 1  erf ( )
V 4t
 Solution
2
 vx v x
 2 0
 2 
 v x 
 
 2vx v x     2
 2
 2
 v x

 v x 
 
     2
 vx 

 v x
ln   2


v x  2
 C1e


v x  C1  e  2
d  C 2
0
0
  0  v x  V  C1  e  2
d  C 2
0

C2  V
 
    v x  0  C1  e  2
d  C 2
0

V V
C1     2

e
 2
d
0

2V
e
 2
vx  V  d
 0

vx 2 vx
e
 2
 1 d  1  erf ( )
V  0
V
 Start up laminar flow inside pipe

Initially the fluid is at rest

P0 PL
P0 > PL

v z p0  pL 1   v z 
   r 
t L r r  r 
Boundary condition
IC t = 0, vz = 0 for 0 < r < R
BC r = 0, vz = finite
r = R, vz = 0
 Dimensionless equation
vz r t
   2
( p0  pL ) R 2 / 4 L R R
New PDE equation
 1    
 4   
     
Boundari condition
IC t = 0, ф = 0 for 0 < ξ < 1
BC ξ = 0, ф = finite
ξ = 1, ф = 0

System will be steady state at τ = ∞

 DE non homogen – BC homogen

 ( , )   ( , )  s ( )
 1    
 4   
     
 s 1     1   s 
  4       
           

Divided into 2 DE
1     
    0
     
1   s 
    4  0
    
 Solution
• Steady state

s ( )  1   2

• Unsteady state
t

J 0 (an )  an2
  8 3 e R 2

n 1 an J 0 ( an )
• Velocity distribution
t
J 0 (an )
  a n2
  s    1    8 3
2
e R 2

n 1 an J 0 ( an )
 Solution: PDE non homogen – BC homogen

 ( , )   ( , )  s ( )
 1    
 4   
     
 s 1     1   s 
  4       
           

Divided into to DE
1     
    0
     
1   s 
    4  0
    
 Unsteadystate
1     
    0
     
1     
   
     
t ( ,0)  s ( )
t (1, t )  0
t (0, t )  finite
t ( , )  0
 Steady state
1   s 
    4  0
    


s ( )  s ( ) Solution:
s (1)  0 s ( )  1   2
s (0)  finite
s ( )  s ( )
 Unsteadystate
1     
   
     
  1    
2
   
 2
    
Separation variable
t ( , )  U ( ).T ( )
 2
d U 1 dU
 a U 0
2

d 2
 d
Bessel function
U ( )  c2 Jo(a )  c3Yo (a )
dT
a 02

d
 a 2
T ( )  c1e
 Solution
• Steady state

s ( )  1   2

• Unsteady state
t

J 0 (an )  an2
  8 3 e R 2

n 1 an J 0 ( an )
• Velocity distribution
t
J 0 (an )
  a n2
  s    1    8 32
e R 2

n 1 an J 0 ( an )
 Stream function
• Momentum transfer in multi dimension has many
equations which are usually very complex and the
solution are difficult to obtain.
• To reduce variables, sometime people can simplify by
combining velocity direction.
• Example : Stream function
– Ψ(x,y) = stream function, is a function that fulfill continuity
equation (for Cartesian coordinate)

vx 
y

vy 
x
 
 v x  
t x

y
  
v y  v z   0
z

0 v z  0
t
v x v y
 0
x y
2 2
  proved
  0
xy xy
Stream function for several coordinate
Can be seen in table 4.2-1
 Potential flow
• Cartesian coordinate

• Correlation to stream function

• Potential function f applies for irrotational flow
only
• Stream function y applies for rotational or
irrotational flows
• Potential function f applies for 2D flows [f(x,y)
or f(r,q)] and 3D flows [f(x,y,z) or f(r,q, f)]
 Example
• Viscous flow around a shere

Use stream function

- spherical
R
- µ, ρ constant
- vφ = 0
- Tabel 4.2 – 1 D

 2
 
E  2

1   , E 2  2 E 2  
 2 cos  
1 
sin 
  4
 E
r sin   (r ,  ) 
r sin   r r  
t  

 2
sin    1   1 
E  2 2
2 vr  
  r 2 sin  
r r   sin   
1 
v  
r sin  r
 Viscous flow, the most
important is µ
• Terms without µ can be neglected
E  0
4

  2 sin    1     2 sin    1  
 2 2    2  2    0
 r r   sin     r r   sin   
• Boundary condition:
1 
1. r = R vr   0
r sin  
2

2. r = R 1 
v  0
r sin  r
3. r  ∞ 1
   v r 2 sin 2 
2
 Substitution ψ to DE
 d2 2  d 2 2
 2  2  2  2  f (r )  0
 dr r  dr r 
• Example f(r) = c rn

 d2 2
 2  2 n(n  1)cr n  2  2cr n  2   0
 dr r 
A
f (r )   Br  Cr  Dr
2 4

r
 1 R 3 1 2
    v  v Rr  v r  sin 
2

 4 r 4 2 
 Velocity distribution
1 
vr   2
r sin  

 3 R 1  R 3 
vr  v 1      cos 
 2 r 2  r  

1 
v 
r sin  r

 3 R 1  R 3 
v  v 1      sin 
 4 r 4  r  
 Example potential flow
• Two dimensional flow has equation of
stream function as follows
  2r sin 2
2

• Determine the equation for potential flow

 For cylinder coordinate
1 
vr   4r cos 2
r 


vr    2r 2 cos 2  f1 ( )



v    4r sin 2
r
1 
v    2r 2 cos 2  f 2 (r )
r 
 Homework / exercise
• A viscous fluid flows around a sphere having
stream function equation as follows:

v R 3 v r 2
 sin 2    sin 2 
2r 2
• If it is known that
vz = vr cos - v sin

• Prove that vz = v at very far distance from the

sphere.
   R 3 
vr  v 1     cos 
  r  

 1  R 3 
v  v 1     sin 
 2  r  

v z  v cos   v sin 
2 2


v z  v cos   sin   v
2 2

 Boundary layer theory
• According to Prandtl (1904) the effect of shear
stress at high Re number is only in the thin
layer of the surface which is know as boundary
layer (lapisan batas)
 Boundary layer equation

• Continuity equation
v x v y v x
  0  vy   dy
x y x
• Equation of motion
v x v y  vx 2
vx  vy 
x y y 2
 Unsteady viscous flow
• Example :
– Flow near a wall suddently set in motion
– Newtonion fluid, constant ρ
– The effect of the solid motion at t time is only up to
distance of δ (t)

t = 0, fluid and solid are at rest

t > 0, solid surface is set in motion
• Equation of motion for Cartesian coordinate with
constant ρ in x direction
• From table 3.4-2 D, vz = vy = 0, unsteady state
we get PDE
  2v x
 vx   2
t y y

• With condition:  (t )
– IC : t = 0 vx = 0
– BC : vx
y = 0 vx = V    f ( )
y = ∞ vx = 0 V
 v x V ' 

t  t
 2v x V      V 1
    "
y 2      y  
New PD
d 2d d
 2   
 d dt
1 d 2 d 1 d
  2 d     d
0  dt 0 d
d 2Nt
N   M 
dt M
• N and M can be obtained if: Φ = Φ(η) is known.
For example:
2
  a  b  d
3 1 2
  1   
2 2 21t
• We get M = 5/12 dan N = 1   4.8t
5 / 12
3 y 1 y2
  1 
2 4,8t 2 4,8t
 Solution
v x V ' 

t  t
 2v x V      V 1
    "
y 2      y  
PD baru
d 2 d d
 2   
 d dt
1 d 2 d 1 d
  2 d     d
0  dt 0 d
d
N   M
dt
Ndt  Md
1
Nt  M  c 2

2
t  0;   0  c  0
1 2Nt
Nt  M 2

2 M
• N dan M dapat dicari bila : Φ = Φ(η) diketahui.
Misal dipilih :
2
  a  b  d
3 1 2
  1   
2 2
 d
1 1 1
M    d    d   0   d
1

0
d 0 0
1
 3 1 2
M  0   1     d
0
2 2 
1
3 2 1 3 5
M      
4 6 0 12
1 1
d  d
1 2
 3 
N   2 d        1
0
 d 0  2 0
Didapat M = 5/12 dan N = 1

21t
  4.8t
5 / 12
2
3 y 1 y
  1 
2 4,8t 2 4,8t
 Flow near the leading edge

• Continuty equation
v x v y v x
  0  vy   dy
x y x
• Equation of motion
v x v y  vx 2
vx  vy  2
x y y
 v x  v x  v y  2v x
vx  dy  
x  x  y y 2

BC : x=0 vx = v∞
y=0 vx = 0
y=δ vx = v ∞ δ = f(x) η = y/ δ
v x V  '   2v x 1
   V "
x  x y 2 2

              1 
v  v '        v '      v '   2 v "

    x       x    
 d 
( B  A)  C
dx v
Φ for 0 < η < 1  1 1
A    'd C    " d
in boundary layer 0 0
3 1 3 1  
   B    '    'd d
 
2 2 0 0 

We get : A = 9/35 x
B = 33/28  ( x)  4,64
v
C = - 3/2
3
   
   
3 y  1 y 
v x  v     
 4,64 x 
2 2  4,64 x 
 v   v 
 Homework / Test
• A student in Fluid Flow Laboratory is doing a
research using non Newtonion fluid follows
Ostwald De Waele model which is put in two
parallel plates.

• Flow near a wall suddenly set in motion

• t = 0, fluid and solid are at rest
• t > 0, solid surface is set in motion

• Derive the velocity distribution using boundary

layer theory
 Aliran viskus unsteady
Non Newtonion

– Aliran dekat dinding yang digerakkan tiba – tiba

– Fluida newton, ρ tetap
– Pengaruh gerakan plat pada saat t hanya terasa sampai
jarak δ (t)

t = 0, plat diam
t > 0, bidang datar diberi kecepatan v
secara tiba - tiba
• Unsteady state didapat persamaan
differensial
n 1
  v x v x
 vx   ( yx )  yx  m
t y y y
• Dengan kondisi batas :
– IC : t=0 vx = 0
– BC : y = 0 vx = V
y = ∞ vx = 0
vx y
   f ( ) 
V  (t )
 m  v x n
vx   ( )
t  y y
n 1 2 
v x mn  vx    v x 
   
t   y   y 2 
 

vx V '     2v x V
    "
t   t  y 2
 2

PD berubah menjadi
n 1
V ' d mn  V '  V 
     2  "
 dt      
 B A
1  n d mn n 1 1 n 1 
  ' d   V     '  " d 
0  dt  0 
n
3 1 3 3 1
 ( )  1     3  B   A   
2 2 8 2 n
n
3 n d mn n 1 3  1
  V  
8 dt  2 n
n
1 n 1 8 m n 1 3 
  V   t
n 1 3 2
 1
 8  3 n m(n  1)V n 1  n 1
    t
 3  2   
  
3
   
   
 3 y 1  y  
vx  v 1     
 2 1 2 1 
 8  3 n m(n  1)V n 1  n 1 
 8  3 n m(n  1)V n 1  n 1 
  
    t    t 
 3 2      3 2    
       

 PR / Latihan
• Seorang mahasiswa di laboratorium Aliran Fluida ingin
membandingkan teori dari Von Karman dan teori
lapisan batas. Mahasiswa tersebut mengalirkan suatu
fluida viskus yang mengikuti model Ostwald De Waele
secara ajeg diujung suatu plat datar.
• Dengan teori lapisan batas, susunlah persamaan
differensial untuk distribusi aliran tersebut diatas dan
carilah distribusi kecepatannya. Dari data diketahui
bahwa suatu fluida yang mengalir diatas suatu ujung
plat datar, menurut neraca momentum dari Von
Karman persamaan geraknya dapat dinyatakan
sebagai berikut : 
 1  2 d vx  vx 
   yx   v  1 
dx 0 v   v
dy
  y 0 
  1  v  vx 
2 d x
   yx   v   1
dx 0 v   v 
dy
  y 0 

Prove that :
1
 7,18m(n  1)v n  2  3  n  n 1
 ( x)   
  x
  2 