KB7001

Computational Fluid Dynamics and

Heat Transfer
(Lecture 5)
RECALL OF PREVIOUS LECTURE
 LUMPED CAPACITANCE METHOD
The essence of the lumped capacitance method is the assumption that
temperature of the solid is spatially uniform at any instant during the
transient process. This assumption implies that temperature gradients
within the solid are negligible.

 Heat transfer out of the body   The decrease in the energy 

  =  
 during dt   of the body during dt 
− E out = E st (1)

dT
− hAs (T − T ) = ρVc (2)
dt
 CRITERIA FOR LUMPED ANALYSIS
First step in establishing a criterion V – volume of the body
V
for the applicability of the lumped Lc = As – surface area of the body
system analysis is to define a As
characteristic length

kA
(Ts,1 − Ts,2 ) = hA(Ts,2 − T )
L
Biot number
Ts,1 − Ts,2 ( L kA) Rcond hL
= = =  Bi
Ts,2 − T (1 hA) Rconv k

heat convection
Bi =
heat conduction
Criterion to use the lumped
capacitance method
If this condition is satisfied the error
hLc
Bi =  0.1 associated with using the lumped
k capacitance method is small
 FOURIER NUMBER
V T − T   hAs  
With Lc = exponent of = exp−   t  may be expressed as
As Ti − T   ρVc  
hAs t ht hLc k t hLc αt hAs t
= = = = Bi  Fo
ρVc ρcLc k ρc Lc
2
k L2c ρVc

 Bi Fo
Thermal Biot Fourier
diffusivity number number

The rate at which heat is conducted

αt kL2 (1/L) ΔT across L of a body of volume L3
Fourier number = Fo = 2 = =
L ρcLc /t ΔT
3
The rate at which heat is stored
in a body of volume L3
Fourier number is a dimensionless
θ T − T
time, which with the Biot number, = = exp(-Bi  Fo)
characterizes transient conduction θi Ti − T
problems
 FINITE-DIFFERENCE METHODS FOR TRANSIENT PROBLEMS
Analytical solutions to transient problems are restricted to simple
geometries and boundary conditions. However, in many cases the
geometry and/or boundary conditions preclude the use of analytical
techniques, and finite-difference methods must be used.

 T  T 1 T
2 2
Heat Equation + 2 =
x 2
y α t

The Explicit Method

Discretization of heat equation

The Implicit Method

 EN0776/KB7001 MODULE OUTLINE
Lecture 1 - Heat transfer: Physical Lecture 7 - Discretization and Mesh
Origins, Rate Equations, Generation
Heat Transfer by Conduction,
Introduction to Finite- Lecture 8 - Heat Exchangers: Design
Difference Method Considerations

Lecture 2 - Finite-Difference Methods: Lecture 9 - Cavitation

Matrix Inversion Method,
Lecture 10 - Turbulence, Large-Eddy
Gauss-Seidel Method
Simulation
Lecture 3 - Extended Surfaces (Fins)
Lecture 11 - Injection and Spray
Lecture 4 - Transient Conduction, Finite-
Lecture 12 - Revision
Difference Method for
Transient Problems

Lecture 5 - Differential Analysis of Fluid

Motion, Momentum Equation

Lecture 6 - Boundary Layer, Heat

Transfer by Convection,
External Flow, Internal Flow
Differential Analysis of
Fluid Motion
 Main Topics

• Conservation of Mass
• Motion of a Fluid Particle (Kinematics)
• Momentum Equation
 Conservation of Mass
• Rectangular Coordinate System
 Conservation of Mass
Taylor series expansion about point O.

At the right face Neglect higher order terms

 ρ  dx  2 
 ρ 1  dx 
2
ρ x + dx/2 = ρ +   +  2    +    
 x  2  x  2!  2 
 ρ  dx
ρ x + dx/2 = ρ +  
 x  2
 u  dx ρ u
u = u+  ρ, u, and are evaluated at point O.
x + dx/2
 x  2 x x

At the left face

 ρ  dx   ρ  dx 
ρ = ρ +   −  = ρ −   
x − dx/2

 
x 2   x  2 
 u  dx   u  dx 
u x − dx/2 = u +   −  = u −   
 x  2   x  2 
 Taylor Series Expansion
Example
Taylor series is a representation of a function as an infinite sum of terms that are
calculated from the values of the function's derivatives at a single point.
The red line is cos(x), the blue line is the approximation
 Conservation of Mass
Basic Law, and Transport Theorem

Rate of change of mass   Net rate of mass flux out 

inside the control volume  +  =0
   through the control surface
Conservation of Mass
 Conservation of Mass

Rectangular Coordinate System

 Conservation of Mass
Basic Law for a System

Rate of change of mass   Net rate of mass flux out 

inside the control volume  +  =0
   through the control surface

Rectangular Coordinate System

“Continuity Equation”
 Conservation of Mass
Rectangular Coordinate System

“Del” Operator

“Continuity Equation”
 Conservation of Mass
Rectangular Coordinate System
Incompressible Fluid:

Steady Flow:
 Example Problem
A gas-filled pneumatic strut in an automobile suspension system behaves
like a piston-cylinder apparatus. At one instant when the piston is L = 0.15 m
away from the closed end of the cylinder, the gas density is uniform at  =
18 kg/m3 and the piston begins to move away from the closed end at V = 12
m/s. The gas velocity is one-dimensional and proportional to distance from
the closed end; it varies linearly from zero at the end to u = V at the piston.

Find: a) Obtain an expression for the average density as a function of time.

b) Evaluate the rate of change of gas density at this instant.
 Example (cont.)
Basic equation
ρu ρυ ρω ρ
+ + + =0
x y z t
ρu ρ ρ ρu u ρ
Since u = u(x) + =0 =− = −ρ −u
x t t x x x
ρ ρ dρ u
Since  is assumed uniform in the volume = 0 and = = −ρ
x t dt x

x u V dρ V
Since u =V , = , and = −ρ L = L0 + Vt
L x L dt L
ρ t t
dρ V Vdt ρ L0  1 
 ρ L
= − dt = −  L + Vt ln
ρ0
= ln
L0 + Vt
and ρ(t) = ρ0 
+

ρ
0 0 0 0  1 Vt/L0

ρ V kg m 1 kg
At t = 0 = − ρ0 = −18 3  12  = −1440 3
t L m s 0.15m m s
This problem illustrates use of the differential continuity equation to evaluate a density
variation with respect to time.
 Motion of a Fluid Particle
(Kinematics)
• Fluid Translation: Acceleration of a Fluid Particle in a
Velocity Field
• Fluid Rotation
• Fluid Deformation
– Angular Deformation
– Linear Deformation
Motion of a Fluid Particle
(Kinematics)
Motion of a Fluid Particle
(Kinematics)
 Motion of a Fluid Particle
(Kinematics)

   
The particle velocity at time t (position r ) is given by V p = V(x, y, z, t) . Then dV p
  
is the change in the velocity of the particle, in moving from location r to r + dr
 Motion of a Fluid Particle
(Kinematics)
   
 V V V V
dV p = dx p + dy p + dz p + dt
x y z t

Total acceleration of the particle is given by

    
 V dx p V dy p V dz p V
dV p
ap = = + + +
dt x dt y dt z dt t

dx p dy p dz p
=u =υ =ω
dt dt dt
    
 dV p V V V V
ap = =u +υ +ω +
dt x y z t
     
DV  dV p V V V V Substantial derivative, or
= ap = =u +υ +ω +
Dt dt x y z t Total derivative
 Motion of a Fluid Particle
(Kinematics)
• Fluid Translation: Acceleration of a Fluid Particle in a
Velocity Field
 Motion of a Fluid Particle
(Kinematics)
   
DV V V dV
=u +υ + Two - dimensional flow
Dt x y dt
  
DV V dV
=u + One - dimensional flow
Dt x dt
   
DV V V V
=u +υ +ω Three-dimensional steady-state flow
Dt x y z
 Motion of a Fluid Particle
(Kinematics)
Scalar component equations of substantial derivative
 Momentum Equation

• Momentum Equation for

Inertial Control Volume
 Momentum Equation
Newton’s Second Law


dF = dFx + dF y + dFz
dFx = dFS x + dFBx
dFy = dFS y + dFB y
dFz = dFS z + dFBz
 Momentum Equation
• Infinitesimal fluid particle of mass dm
• Forces acting on a fluid particle

dV = dxdydz
 Stress Field
➢ Surface forces act on the boundaries of a medium through direct contact.
➢ Body forces are forces developed without physical contact, and distributed over
the volume of the fluid. Gravitational and electromagnetic forces are examples
of body forces.
 
➢ The force δF acting on δA may be resolved into two components, one
normal to and the other tangent to the area.
Normal stress
δF
 n = lim n
δ An →0 δA
n

Shear stress
δF
 n = lim t
δ An →0 δA
n
 Fluid Viscosity
• Newtonian Fluids
Fluids in which shear stress is directly proportional to the rate of
deformation (shear rate) are called Newtonian fluids

– Most of the common fluids (water, air, oil, etc.)

– “Linear” fluids
 Fluid Viscosity
• Non-Newtonian Fluids
Fluids in which shear stress is not directly proportional to shear
rate are called non-Newtonian fluids
- Special fluids (e.g., most biological fluids, toothpaste, some paints, etc.)
- “Non-linear” fluids

Apparent viscosity

Most non-Newtonian fluids have apparent viscosities that are relatively

high compared with the viscosity of water.
 Fluid Viscosity
• Non-Newtonian Fluids
➢ Fluids in which the apparent viscosity decreases with increasing deformation rate
(n < 1) are called pseudoplastic fluids (ex. polymer solutions, paper pulp in water)

➢ If the apparent viscosity increases with increasing deformation rate (n > 1) the
fluid is called dilatant (ex. suspensions of starch and sand)
➢ A fluid that behaves as a solid until a minimum yield stress is exceeded and
subsequently exhibits a linear relation between stress and rate of deformation is
referred to as an ideal or Bingham plastic (ex. tooth paste, drilling muds, etc.)

Viscosity Non-Newtonian fluid

 Momentum Equation
 σ dx   σ dx 
dFS x =  σ xx + xx dydz −  σ xx − xx dydz
  x 2   x 2 
 τ yx dy   τ yx dy 

+  τ yx +  
dxdz −  τ yx − dxdz
y 2   y 2 
 
 τ dz   τ dz 
+  τ zx + zx dxdy −  τ zx − zx dxdy
 z 2   z 2 

 σ xx τ yx τ zx 
dFS x =  + + dxdydz

 x y z 

 σ τ yx τ zx 
dFx = dFB x + dFS x =  ρg x + xx + + dxdydz

 x y z 

 τ xy σ yy τ zy 
dFy = dFB y + dFS y 
=  ρg y + + + dxdydz
  z 
 x y 
 τ xz τ yz σ zz 
dFz = dFB z + dFS z 
=  ρg z + + + dxdydz
  z 
 x y 
 Momentum Equation
Forces Acting on a Fluid Particle

Substitute these equations into

 Momentum Equation
Differential Equations of Motion or Momentum Equations

Before the equations can be used to solve for u, v, and w, suitable

expressions for the stresses must be obtained in terms of the velocity
and pressure fields.
 Momentum Equation
For a Newtonian fluid the viscous stress is proportional to the rate of shearing
strain. The stresses may be expressed in terms of velocity gradients and fluid
properties in rectangular coordinates as follows:
 υ u 
τ xy = τ yx = μ +  p – the local thermodynamic pressure.
 x y  Thermodynamic pressure is related to the
density and temperature by the
 ω υ 
τ yz = τ zy = μ +  thermodynamic relation usually called the
 y z  equation of state.

 u ω 
τ zx = τ xz = μ + 
 z x 
2  u
σ xx = − p − μ  V + 2μ
3 x
2  υ
σ yy = − p − μ  V + 2μ
3 y We substitute these equations into the
differential equations of motion introduced
2  ω
σ zz = − p − μ  V + 2μ earlier.
3 z
 Momentum Equation
Du p    u 2      u υ     ω u 
ρ = ρg x − +  μ 2 −   V  +  μ +  +  μ + 
Dt x x   x 3   y   y x  z    x z 
Dυ p    u υ     υ 2      υ ω 
ρ = ρg y − +  μ +  +  μ 2 −   V  +  μ + 
Dt y x   y x  y   y 3  z   z y 
Dω p    ω u     υ ω     ω 2  
ρ = ρg z − +  μ +  +  μ +  +  μ 2 −   V 
Dt z x   x z  y   z y  z   z 3 

These equations of motion are called the Navier-Stokes equations.

 Momentum Equation
The equations are greatly simplified when applied to incompressible flow with
constant viscosity. 0 0 0
Du p    u 2      u υ     ω u 
ρ = ρg x − +  μ 2 −   V  +  μ +  +  μ + 
Dt x x   x 3   y   y x  z   x z 
0 0 0
Dυ p    u υ      υ 2       υ ω 
ρ = ρg y − + μ
   + 
  + μ
   2 −   V 
  + 
  +
μ 
Dt y x   y x  y   y 3  z   z y 
0 0 0
Dω p    ω u     υ ω     ω 2  
ρ = ρg z − +  μ +  +  μ +  +  μ 2 −   V 
Dt z x   x z  y   z y  z   z 3 

 u u u u  p   2u  2u  2u  Navier-Stokes
ρ +u +υ + ω  = ρg x − + μ 2 + 2 + 2 
 x z 
equations for
 t x y z  x  y incompressible
flow with constant
 υ υ υ υ  p   2υ  2υ  2υ 
ρ + u + υ + ω  = ρg y − + μ 2 + 2 + 2  viscosity
 x z 
 t x y z  y  y
 ω ω ω ω  p   2ω  2ω  2ω 
ρ +u +υ +ω  = ρg z − + μ 2 + 2 + 2 
 x z 
 t x y z  z  y
Claude-Louis Navier (10 February 1785 – 21 August 1836)
was a French engineer and physicist who specialized in
mechanics. Navier formulated the general theory of elasticity in
a mathematically usable form (1821), making it available to the
field of construction with sufficient accuracy for the first time. In
1819 he succeeded in determining the zero line of mechanical
stress, finally correcting Galileo Galilei's incorrect results, and
in 1826 he established the elastic modulus as a property of
materials independent of the second moment of area. Navier is
therefore often considered to be the founder of
modern structural analysis. His major contribution however
remains the Navier-Stokes equations (1822), central to fluid
mechanics.

George Gabriel Stokes (13 August 1819 – 1 February

1903), was a mathematician and physicist. Born in Ireland,
Stokes spent all of his career at University of Cambridge,
where he served as the Lucasian Professor of Mathematics
from 1849 until his death in 1903. Stokes made seminal
contributions to fluid dynamics (including the Navier-Stokes),
optics, and mathematical physics (including the first version
of what is now known as Stoke’s theorem).
 Navier-Stokes Equations
➢ Navier–Stokes equations, named after Claude-Louis Navier and
George Gabriel Stokes, describe the motion of fluid substances. These
equations arise from applying Newton’s second law to fluid motion,
together with the assumption that the fluid stress is the sum of a
diffusing viscous term (proportional to the gradient of velocity) and a
pressure term.

➢ The Navier–Stokes equations are also of great interest in a purely

mathematical sense. Somewhat surprisingly, given their wide range of
practical uses, mathematicians have not yet proven that in three
dimensions solutions always exist (existence), or that if they do
exist, then they do not contain any singularity (smoothness). These are
called the Navier–Stokes existence and smoothness problems.

➢ The Clay Mathematics Institute has called this one of the seven most
important open problems in mathematics and has offered a
US$1,000,000 prize for a solution or a counter-example.
 Example Problem
A liquid flows down an inclined plane surface in a steady, fully developed
laminar film of thickness h.

Find:
(a) Continuity and Navier-Stokes equations simplified to model this flow field
(b) Velocity profile
(c) Shear stress distribution
(d) Volume flow rate per unit depth of surface normal to diagram
(e) Average flow velocity
(f) Film thickness in terms of volume flow rate per unit depth of surface normal to
diagram
(g) Volume flow rate in a film of water 1 mm thick on a surface 1 m wide, inclined at 15C
to the horizontal
 Example Problem (cont.)
Basic equations written for incompressible flow with constant viscosity are

4 3
u υ ω
+ + =0 - Continuity
x y z

1 4 5 3 4 4 3
 u u u u  p   2u  2u  2u 
ρ +u +υ + ω  = ρg x − + μ 2 + 2 + 2 
 x 
 t x y z  x  4 y 5 z 
1 4 5 3 3
 υ υ υ υ  p   2υ  2υ  2υ  - Momentum
ρ + u + υ + ω  = ρg y − + μ 2 + 2 + 2 
 x z 
 t x y z  y  y
1 3 3 3 3 3 3 3 3
 ω ω ω ω  p  ω  ω  ω
2 2 2
ρ +u +υ +ω  = ρg z − + μ 2 + 2 + 2 
 x z 
 t x y z  z  y
Assumptions:
(1) Steady flow
(2) Incompressible flow,  = const
(3) No flow or variation of properties in z direction; ω = 0 and /z = 0
(4) Fully developed flow, so no properties vary in the x direction; /x = 0
(5) is constant. Since  is zero at the solid surface, then  must be zero everywhere
 Example Problem (cont.)
 2u
0 = ρg x + μ
y 2
(a)
p
0 = ρg y −
y
d 2u ρg x sin θ
=− = − ρg
dy 2 μ μ
Integrating Integrating again
du sin θ sin θ y 2
= − ρg y + c1 u = − ρg + c1 y +c 2
dy μ μ 2

Boundary conditions
u=0 at y=0 (no-slip condition at the solid surface)
du
= 0 at y = h (zero-shear-stress condition at the liquid free surface)
dy
 Example Problem (cont.)
c2 = 0

sin θ
0 = − ρg h + c1
μ

sin θ sin θ y 2
c1 = ρg h substituting into u = − ρg + c1 y +c 2
μ μ 2

sin θ y 2 sin θ sin θ  y 2 

u = − ρg + ρg hy or u = ρg hy − (b)
μ 2 μ μ   2 

The shear stress distribution

du
τ yx = μ = ρg sin θ(h − y) (c)
dy

The volume flow rate

h
Q =  udA =  ubdy
A 0
 Example Problem (cont.)
h
h
ρg sin θ  y 2  sin θ  b  hy 2 y 3 
Q=  μ 
hy -
2  
bdy = ρg
μ  2
 − 
6 
0  0

ρg sin θ  b h 3
Q= (d)
μ 3
Average velocity

Q ρg sin θ h 2 (e)
V = Q/A = Q/bh thus V = =
bh μ 3
Film thickness
13
 3μ Q 
h=  (f)
 ρg sin θb  A film of water h=1 mm thick on a plane
b=1 m wide, inclined at  =15 would carry

kg m ms (0.001)3 m 3 L L
Q = 999 3  9.81 2  sin(15 )  1m 
0
  1000 = 0.846 (g)
m s 1.00  10 −3 kg 3 m3 s
 Momentum Equation
Newtonian Fluid: Navier–Stokes Equations 0
Momentum Equation for Frictionless Flow

• Special Case: Euler’s Equation

• Rectangular Coordinates
 Euler’s Equations in Streamline
Coordinates (Frictionless flow)

 p ds   p ds 
p− dndx −  p + dndx − ρg  sin β  dsdndx = ρas  dsdndx
 s 2   s 2 
 Euler’s Equations in Streamline
Coordinates (Frictionless flow)
Simplifying the equation
p z
− − ρg sin β = ρas sin β =
s s
1 p z
− −g = as
ρ s s
Along any streamline V = V(s,t) and the total acceleration

DV V V
as = = +V
Dt t s
1 p z V V
− −g = +V
ρ s s t s
1 p V
For steady flow and neglecting body forces = −V
ρ s s
 Euler’s Equations in Streamline
Coordinates (Frictionless flow)

 p dn   p dn 
p− dsdx −  p + dsdx − ρg  cos β  dndxds = ρan  dndxds
 n 2   n 2 
 Euler’s Equations in Streamline
Coordinates (Frictionless flow)
Simplifying the equation
p z
− − ρg cos β = ρan cos β =
n n

1 p z
− −g = an The normal acceleration of the fluid element is
ρ n n toward the center of curvature of the streamline in
the minus n direction.
−V 2 Then Euler’s equation normal to the streamline is
an =
1 p z
R written for steady flow as 2
V
− −g =−
For steady flow in a horizontal plane ρ n n R
Euler’s equation normal to a streamline
becomes
In regions where the streamlines are straight, the radius
1 p V 2
= of curvature, R, is infinite and there is no pressure
ρ n R variation normal to the streamlines.
Euler’s Equations in Streamline
Coordinates
• Along a Streamline
(Steady Flow, ignoring body forces)

• Normal to the Streamline

(Steady Flow, ignoring body forces)
 Airfoil
n


 Bernoulli Equation
integration of Euler’s equation along a streamline for steady flow

(1)

If a fluid particle moves a distance, ds along a

streamline, then
p
ds = dp (the change in pressure along s)
s
z
ds = dz (the change in elev ation alon g s)
s
V
ds = dV (the change in speed along s)
s
Daniel Bernoulli (8 February 1700–
17 March 1782) was a Swiss
Thus, after multiplying (1) by ds, we can write mathematician and physicist

p dp
− − gdz = VdV or + VdV + gdz = 0
ρ ρ
dp V 2
Integrating this equation  + + gz = const (along s)
ρ 2
 Bernoulli Equation
integration of Euler’s equation along a streamline for steady flow

We must specify relation between pressure and density.

For incompressible flow  = const
2
p V
+ + gz = const
ρ 2
Units:  m   m2   m2 
2
 2  2  2
 s   s   s 
or
J  J  J 
     
 kg   kg   kg 

Bernoulli equation can be viewed as an expression of

mechanical energy balance
 Bernoulli Equation
integration of Euler’s equation along a streamline for steady flow

• Bernoulli Equation

Restrictions:
1. Steady Flow
2. No Friction
3. Flow Along a Streamline
4. Incompressible Flow
 Bernoulli Equation
integration of Euler’s equation along a streamline for steady flow

• Static, Stagnation, and Dynamic Pressures

(Ignore Gravity)

Stagnation Dynamic
Static
 Example
A pitot tube is inserted in an air flow to measure the
flow speed. The tube is inserted so that it points
upstream into the flow and the pressure sensed by
the tube is the stagnation pressure. The static
pressure is measured at the same location in the flow,
using a wall pressure tap. If the pressure difference is
30mm of mercury, determine the flow speed.

Basic equation

Assumptions
(1) Steady flow
(2) Incompressible flow
(3) Flow along a streamline
(4) Frictionless deceleration along stagnation streamline
 Example (cont.)
Writing the Bernoulli’s equation along the stagnation streamline (with z = 0)

p0 p V2 p0 is stagnation pressure at the tube opening where

= + the speed has been reduced, without friction, to zero.
ρ ρ 2

Solving for V gives

2(p0 − p) p0 − p = ρHg gh = ρH2O gh( SG Hg )

V =
ρair

2ρH 2O gh(SGHg ) kg m m m3
V = = 2  1000 3  9.81 2  30mm   13.6 
ρair m s 1000mm 1.23kg

V = 80.8 m/s
Airspeed measurement
Airspeed measurement
Airspeed measurement
THE END OF LECTURE 5