© Natteri M Sudharsan

Lecture Notes

ASSESSMENT QUESTIONS
Part A
1. Write the discretization equation for solving a 1-D unsteady heat
transfer problem using a semi-implicit scheme, assuming that good
stability is obtained if the blend is 80% implicit and 20% explicit.
2. Write the element matrix for 1-D element subjected to convective heat
transfer What is the limitation of a FTCS scheme? How does one
arrive at the maximum possible time stepping for a multi nodal
system?
L

∫N
2
3. Evaluate i N j dx Ni and Nj are the 1-D shape functions.
0

4. Derive the equation for a convection diffusion problem using upwind

finite volume formulation, given that ρU > 0
5. What will be the governing equation for an inviscid, incompressible
fluid whose flow is rotational?
6. Derive the equation for a convection diffusion problem using upwind
finite volume formulation, given that ρU < 0
7. Write the Navier stokes equation and briefly explain the various terms.
8. How do you classify PDE’s as linear and non-linear
9. A hexagonal domain is considered. If the governing equation is
∇ 2φ = 0 , how many boundary conditions are required and why?
10. How are boundary conditions classified
∂φ ∂ 2φ
11. For a parabolic equation = α 2 what are the conditions needed
∂t ∂x
to be specified other than the boundary conditions
12. The rectangular plate has to satisfy the governing equation ∇ 2φ = 0
along with the boundary conditions given in the figure above. The
node ‘i’ is a singular point and has to satisfy both the boundary
conditions. Using central difference scheme and letting ∆x = ∆y write
the discretization for node i.

Assessments 1/10
 © Natteri M Sudharsan
Lecture Notes
13. What are the advantages and disadvantages of stream function
vorticity method
14. Under what circumstances would you choose primitive – variable (u, v,
p etc) instead of stream function vorticity approach
15. What terms (i.e. inertial, viscous etc., ) in the Navier – Stokes equation
become important for the following cases :
16. Water flowing in a distribution channel of uniform cross-sectional area
at a constant and steady flow rate
17. Flow under a bridge with a circular column supporting the bridge
18. Tar/asphalt drum lying horizontally on the road with fluid flowing out
19. Write the advantages and disadvantages of staggered grid
20. Briefly explain how SIMPLE algorithm is employed to solve N-S
equation

Part B
1. A fin of 15mm diameter and 150mm long is used to dissipate heat by
convection. However the heat loss from the fin tip is by both convection and
radiation. Setup the solution matrix taking thermal conductivity as 1.5W/mK, heat
transfer co-efficient 1.0W/m2 K, fin base temperature at 100oC and element
length 50mm. Discuss the solution procedure. (USE FEM)
2. A square plate of size 1m x 1m is initially at ambient temperature of 30oC. The
plate is subjected to the following boundary conditions instantaneously as
follows: the top, bottom and right hand side are maintained at 100oC, the
temperature on the left hand side is given as TLHS = 100 (1 – Sin(πy)). Setup the
solution matrix to estimate the temperature distribution over time using backward
in time, central in space scheme. Discuss the solution procedure. Assume
material properties of Aluminum.
3. A bar of steel is initially at 300oC. Heat is extracted from RHS end of the rod as
given in figure below. Use an explicit scheme to solve for the temperature profile
for the bar of size 15mm in diameter and 150mm in length.

Assessments 2/10
 © Natteri M Sudharsan
Lecture Notes
4. A fin of 12.5mm diameter and 200mm long dissipates heat by convection. At a
location 100mm from the fin base there is a constant loss of 2w. Setup the
solution matrix taking thermal conductivity as 1.5W/mK, heat transfer co-efficient
1.0W/m2 K, fin base temperature at 100oC and element length 50mm. Discuss
the solution procedure. (USE FEM)
5. A square duct of dimension 200 x 200mm and 5m long carries fluid at a
constant temperature of 14oC. The duct is insulated with 150mm thick material.
The insulation is initially at an ambient condition of 30oC. Estimate the
temperature variation along the thickness of the insulation over time. Assume the
temperature variation along the length ∂T = 0 . Take α=1.4m2/s. Choose
∂Z
appropriate value for ∆τ. Use FTCS Scheme.
6. A bar of steel is initially at 30oC. The left hand side of the rod is subjected to a
temperature profile given by T = A Sin ωτ, where, the frequency ω = 1/2π, τ, is
the time, and the amplitude, A =1. The right hand side of the rod is maintained at
a constant temperature of 30oC. Use an implicit scheme to solve for the
temperature profile for the bar of size 15mm in diameter and 150mm in length.

7. Fluid at 30oC, blows from left to right at 2m/s. It cools a flat plate maintained at
100oC. Set-up the solution matrix by discretiztion the fluid with 5 nodes. Take
ρ = 1.6kg / m3 ,ν = 15.89x107 m2 / s, k = 26.3W / mK ,α = 22.5x107 m2 / s
An incompressible invicid fluid enters the duct at 4m/s. Solve for the velocity
profile inside the domain.

1.

1
1.5

1m 1m

8. Two streams of fluid at different temperature T= 30oC and T= 100oC.enter a

duct at constant velocity. The bottom side of the duct is maintained at T= 30oC
where the cold fluid enters, and top side is maintained at 100oC. The duct can be

Assessments 3/10
 © Natteri M Sudharsan
Lecture Notes
assumed to be of infinite length along Z direction. Write the governing equation
and the finite element discretization for the problem.
9. An incompressible invicid fluid enters the duct at 10m/s. Solve for the velocity
profile inside the domain.
2. .

2m

 1 − 2 0  T1   − 3 
    
10. Given that  − 1 2 − 1 T2  =  0 
 0 − 1 1  T   1 
  3   
What would be the appropriate algorithm to solve the above system of matrix?
11. Assuming that Gauss-Seidel method is used to solve the above, write the first two
iterations assuming the initial temperatures T’s as 1.
12. A square domain of non-dimensional length 1 is assumed. The governing
boundary conditions and boundary conditions are as follows
∇ 2φ = 0
BC ' s
∂φ
X = 0, =0
∂x
Y = 0, φ = 0
∂φ
X = 1, = h(φ∞ − φ ), where φ ∞ = 25
∂x
Y = 1, φ = 100
Draw the domain and discretize with 3 horizontal and vertical nodes. Apply
appropriate boundary conditions and assemble the matrix for solution.

13. A 1m rod is at initial temperature T(t=0) = 500K. The two rods are suddenly
quenched and maintained at T(t) = 300K and there is no other loss to the
surroundings

Assessments 4/10
 © Natteri M Sudharsan
Lecture Notes
(a) State & simplify the given problem
Write the appropriate governing equations and boundary conditions
(a) Discretize the domain with 4 nodes and assemble the matrix for an
Implicit scheme,
(b) Discuss on how the system will be solved

14. The exact solution for a given differential equation

∂ 2φ g ∂φ
+ = 0, = 0 at x = 0, and φ = 0 at x = 1
∂x 2
k ∂x
is given as φ = (
g
2k
)
1− x2

If an approximate solution of the form φ = a1 + a 2 x + a3 x 2 is chosen to solve the

above governing equation, what will be effect of reducing the mesh size? Discuss
the mesh dependency of the approximate solution.
15. Water flows steadily in a channel of size 1m length and 0.1m depth at 1m/s
along the x-direction. The y-velocity component may be assumed to be negligible
and neglected. Assume that there is no frictional effect at the wall surface.
(a) Draw the computational domain and label the boundaries
(b) Write the appropriate governing equation and boundary conditions
(c) Write the discretization for the 4 corner nodes and 2 nodes in the
middle
(d) Discuss the solution procedure
16. A cylindrical fin of uniform cross-sectional area A and length 1m is exposed
to atmosphere at 20C. The fin base is maintained at a constant temperature of
200C. The fin tip is insulated. Assume 1-D heat transfer. Use hP/kA = 25 (kA =
constant)
17. Write the governing differential equation and the boundary conditions. Use
Finite Volume Method to solve the above given problem.
18. Name and write the appropriate discretization scheme for an unsteady 2-D
convection-diffusion equation for the following conditions:-
(a) High Peclet Number flows
(b) Prandtl Number, 0.005 < Pr < 0.05 ( δ << δ t )
19. Write the governing equation for an unsteady 2-D convection-diffusion
equation for Prandtl Number, Pr ≥ 1 ( δ t ≤ δ ) and the Explicit and Implicit
discretization scheme.

Assessments 5/10
 © Natteri M Sudharsan
Lecture Notes
20. Using finite element formulation discretize using 4 equal elements (5 nodes)
and assemble the matrix for solving the temperature profile in a one-dimensional
fin having the following specifications.
Fin diameter 0.1 sqm, Fin Length 1 m, k=2.5w/mK, h =6w/sqmK, P = 0.2m, fin
tip is exposed to ambient of 20oC and fin base is maintained at 100oC.
21. Using finite element formulation discretize using 4 equal elements (5 nodes)
and assemble the matrix for solving the temperature profile in a one-dimensional
fin having the following specifications.
Fin diameter 0.1 sqm, Fin Length 1 m, k=2.5w/mK, h =6w/sqmK, P = 0.2m, fin
tip is insulated and fin base has a constant flux of 100 w/sqm

22. Given that

∂ 2φ ∂φ
D =λ
∂x 2 ∂t

Assuming D and λ to be 2, and L = 1m, divide the domain into 4 elements (5

nodes) of 0.25m length and assemble the matrix using Crank-Nicholson method
for temperature profile for the next time step. Assume temperature φ =1 at t=0
and take ∆t =0.2. (Use FEM Formulation)

23. Solve the above using Implicit Backward difference method (Use FEM
Formulation).

Assessments 6/10
 © Natteri M Sudharsan
Lecture Notes
PART A (10 X 2 = 20 marks)
1. Name the various types of boundary conditions.
2. If one has to design a thin thermal wear to protect the body from extreme cold
temperature, assuming 1-D heat transfer, how should one discretize the domain?
3. What will be the nature of co-ordinate system for a pure conduction heat transfer,
and for a flow over a cylinder?
∂ 3φ
4. Write the expression for using finite difference approximation.
∂x 2 ∂y
5. The external surface of the wall of a room maintained at 20oC, is exposed to
incident solar radiation. Write the Governing equation and boundary conditions.
6. An ocean wave hits a wall erected on the shore. If its necessary to calculate the
pressure forces on the wall, what would be the appropriate equations to use?
7. What do you understand by the terms, no-slip, slip, non-porous and free surface
condition.
8. What equations and terms are required to solve thermal boundary layer flows
with Prandtl number ≥ 1?
9. Why is staggered grid preferred over co-located grids?
10. What are the various turbulence models available?
PART B (5 X 16 = 80 marks)
11. Fluid at a temperature T∞ and with a velocity U ∞ enters a 2-D channel as given
in the figure below. A wedge at a constant temperature T is placed at a location
downstream of the flow. Assuming Pr ≥ 1, and the flow laminar;
Write the Governing equations for computing the 2-D velocity field,
temperature profile and the boundary conditions that needs to be applied.
Draw a sketch on how the grid should be generated. What are the important
features that must be considered in generating the grid?
Write the finite difference discretization for the problem using implicit (16)
scheme.
Discuss the solution procedure.

12a (i). An incompressible, inviscid fluid flows through a channel of size 1m height and
10m length. The inlet velocity is 3 m/s. Assuming the flow to be irrotational, (12)
solve for the potentials using finite element method.

Assessments 7/10
 © Natteri M Sudharsan
Lecture Notes
12a (ii). What would be the standard Neumann boundary conditions for the boundaries
given in the figure below?
(4)

(OR)
12b (i). A fin of size 12mm diameter and 180mm in length is used to dissipate 10 W/m2.
The heat dissipation to the surroundings at 30oC is by convection. Using finite
element method solve for the fin temperature along the length. Make necessary (12)
assumptions.
12b (ii). Assuming steady state 1-D heat conduction in a composite material having
(4)
different k, A, L and g, write the system of matrix for a 2 element case.

13a. Fluid at a velocity, U and Temperature 30oC picks up heat over a flat plate and
leaves at 120oC. The length of the plate is 1000mm. Assuming 1-D convection-
diffusion and the values for F (ρU) and D ( Γ δx ) as -2 and 0.5 respectively, solve (16)
for the temperature profile using finite volume formulation. Plot the temperature
variation over the distance using analytical solution.

(OR)
13b. A fluid flows inside a duct at constant temperature of 12oC. The duct dimensions
are 300 x 150 mm. The duct is encapsulated in another duct of size 600 x 450
mm. The space inside is filled with a material having a conductivity of k W/mK.
(16)
Left and right walls of the outer duct are maintained at 40oC. The top and bottom
outer walls are assumed to be insulated. Solve for the temperature profile inside
the space between the two ducts using finite volume formulation.

14a. Derive the Navier-Stokes equation for an incompressible Newtonian fluid. (16)

(OR)
14b (i). Derive the methodology to obtain pressure for a viscous flow, assuming that the (10)
velocity distribution is computed from the stream function.

14b (ii). How is vorticity at the moving wall surface estimated from ψ? (6)
15a. Write short notes on the advantages and disadvantages of the various
(16)
turbulence models.
(OR)
15b. Write notes on the following:
(4)
Advantages and disadvantages of Staggered grid.
(4)
The method of solution and the advantages and disadvantages of ξ - ψ method
(4)
Discuss the solution procedure of Navier – Stokes equation using SIMPLE
algorithm.
Write the algorithm and solution procedure for TDMA (4)

Assessments 8/10
 © Natteri M Sudharsan
Lecture Notes
PART A (10 X 2 = 20 marks)

11. Give on example for a non-linear governing equation and a non-linear boundary
condition.
12. What is meant by a mesh independent solution?
13. Why does the exponential scheme give results that match the exact solution in a
convection-diffusion equation?
14. A plane wall is subjected to the same boundary conditions on both the sides.
How can the problem be simplified?
15. When should one choose between a central & upwind scheme?
16. For a 1-D heat transfer, write the FEM conduction and convection matrix.
17. What are the advantages and disadvantages of the stream function-vorticity
formulation?
18. Momentum equation can be neglected for flows with Prandtl number ≥ 1, True or
False.
19. What are the disadvantages of co-located grids?
20. What are the various turbulence models available?

PART B (5 X 16 = 80 marks)
11. Three sides of a 1m x 1m plate initially at 100oC, is exposed to ambient at 30oC
and heat is convected out. One side is maintained at 1000C. Sketch the domain,
write the governing equation and the respective boundary conditions. Assemble
(16)
the system of matrices using backward difference in time to track the spatial as
well as the temporal variation in temperature. Take α = 99.75m2/s, k = 237
W/mK, h = 24 W/m2K and ρ =2700Kg/m3.
A cylindrical fin having an uniform diameter of 10mm and length 150mm is
12a
exposed to ambient condition of 20oC is used to dissipate a constant heat flux of
50 W/m2 from the base. Assuming hP/kA = 25, Write the governing differential (16)
equation and the boundary conditions. Discretize the solution domain using 5
nodes and solve using finite difference method.
(OR)
12b. Three materials having thicknesses 0.1m, 0.2m, and 0.1m respectively are
sandwiched together. The materials can be assumed to be semi-infinite slab
having thermal conductivity of 100, 200 and 50 W/mK respectively. The material
(16)
in the centre generates 20W/m3. If one end receives a flux of 12W/m2 and the
other end is exposed to ambient at 17oC, solve for the temperature profile. Use
Finite element formulation. Take h as 25W/m2 K.

13a. Assuming exponential scheme for interpolating the face temperature from the
nodal temperature derive the generalized equation for a 1-D convection-diffusion (16)
problem using finite volume formulation.

Assessments 9/10
 © Natteri M Sudharsan
Lecture Notes

(OR)
13b. Name and write the appropriate governing equations and discretization scheme
for an unsteady 2-D convection-diffusion equation for the following conditions:-

(a) 0.005 < Pr < 0.05 (8)

(b) Pr ≥ 1 . (8)

14a. A fluid flows in a duct of height 100mm and length 4m at 2m/s. Fluid can be
assumed to be inviscid and incompressible and flow irrotational. Disctretize the
(16)
domain and set-up the solution matrix to solve for stream functions and for
estimating the velocity profile inside the duct. Use finite difference formulation.

(OR)
14b. For the problem specified in 14a, solve for velocity potentials using finite (16)
element formulation.

Write notes on the following:

15a.
(i) Advantages and disadvantages of using of ξ - ψ formulation over
(4)
primitive variables.
(4)
(ii) Write and classify the generalized Navier – Stokes equation.
(4)
(iii) Turbulence
(4)
(iv) Reynolds transport theorem
(OR)
15b. Write notes on the following:
(i) Prandtl mixing length model. (4)
(ii) K - ε model (4)
(iii) Discuss about the linear sub layer, log layer and velocity defect (4)
layers. (4)
(iv) Various iterative algorithms for solving non-linear equations.

Assessments 10/10