NPTEL Online Certification Courses

Indian Institute of Technology

Kharagpur

Computational Fluid Dynamics

Assignment-Week 4
TYPE OF QUESTION: MCQ

Number of questions: 10 Total mark: 10 X 1 = 10

MCQ/MSQ Question

QUESTION 1:

Which of the statements are true:

(i) FDM may not work for geometries not conforming with the grid.

(ii) FDM does not ensure conservative form of any differential equation.

(iii) Boundedness in FDM can be ensured provided that inner mesh points are resolved with
proper backward or forward difference scheme.

(iv) FEM is more complex and faces same issues as FDM.

Which steps are followed by Finite volume method?

a. (i) only

b. (iii) only

c. (iii) and (iv)

d. None

Correct Answer: b

Detailed Solution:
FDM may not work for geometries not conforming with the grid and does not ensure
conservative form of any differential equation. Boundedness in FDM can be ensured provided
that inner mesh points are resolved with proper backward or forward difference scheme. The
known nodal values are replaced in discretized equation. FEM is complex but it can handle non
conforming geometries.

QUESTION 2:

Consider the following steps:

(i) Integration of the governing differential equation over such sub-domain

(ii) Division of the domain into a number of finite size sub-domain. Each sub-domain is
represented by a finite number of grid points.

(iii) Profile assumption for the dependent variable for evaluating the integral in order to express
the results in terms of algebraic quantities at the grid points.

(iv) Conversion of the governing differential equation into algebraic quantities using Taylor’s
series of expansion

Which steps are followed by Finite volume method?

a. (iv) only

b. (ii) and (iv)

c. (i) ,(ii), and (iii)

d. (i), (ii) and (iv)

Correct Answer: c Steps (i), (ii), and (iii)

Detailed Solution:
Finite volume method involves the following steps-
1. Division of the domain into a number of finite size sub-domain. Each sub-domain is
represented by a finite number of grid points.
2. Integration of the governing differential equation over such sub-domain
3. Profile assumption for the dependent variable for evaluating the integral in order to
express the results in terms of algebraic quantities at the grid points.
QUESTION 3:

Which of the following interpolation function is continuous?

a. Piecewise linear between grid points

b. Piecewise linear between control volume faces
c. Piecewise constant between grid points
d. Piecewise quadratic between control volume faces

Correct Answer: a. Piecewise linear between grid points is a continuous interpolation function.
Detailed Solution:
In case of piecewise linear between grid points the derivative will have same value at the grid
points. For all other case, derivative is not necessarily equal at the intersection point.

QUESTION 4:

Three different profile for a transport variable is shown in the figure. Which of the
following is a valid profile assumption in case of finite volume method?

(a) 1 ( piecewise constant)

(b) 2 (piecewise linear between the control volume faces)
(c) 3(piecewise linear between the grid points)
(d) Profile assumption is not required in case of finite volume method

Correct Answer: c.
Detailed Solution:
Using piecewise constant profile ( profile 1), one cannot calculate the heat flux at the control
volume faces and hence it is not acceptable profile assumption. Similarly, for piecewise
linear between the control volume faces ( profile 2), one cannot calculate the heat flux at the
control volume faces and hence it is not acceptable profile assumption.
QUESTION 5:

Consider a metal rod whose left end is connected to a wall where a constant heat flux is
maintained and right end is maintained at 200oC Consider the following statement regarding the
boundary condition (BC)
(i) The left end BC is Dirichlet boundary condition.
(ii) The right end BC is Neuman boundary condition.
(iii) The right end BC is Dirichlet boundary condition
(iv) The left end BC is Neuman boundary condition

Which of the statement are icorrect?

a. (i) and (ii) are correct
b. (i) and (iii) are correct
c. (iii) and (iv) are correct
d. (i) and (iv) are correct

Correct Answer: c.

Detailed Solution:
𝑑𝑇
Given the condition of the problem, the left end of the rod is =constant, which means a fixed
𝑑𝑥
the derivative of the variable is given. This is called Neumann Boundary condition
The right end is at 𝑇 = 200°𝐶 which is Dirichlet.
QUESTION 6:

Consider the source term in the form 𝑆 = 8 + 5𝑇. The source term can be linearized in the form
S = Sc + S pTp as given below:
(i) 𝑆𝑐 = 8 + 6𝑇𝑝∗ and 𝑆𝑝 = −1
(ii) 𝑆𝑐 = 8 + 9𝑇𝑝∗ and S p = −4
(iii) 𝑆𝑐 = 8 + 5𝑇𝑝∗ and S p = 0

Which of the following source term linearization is/are correct?

a. (i) only
b. (ii) only
c. (ii) and (iii) only
d. all of the above

Correct Answer: d.

Detailed Solution:
S p is negative. S p = 0 , it will work because S p is zero.

QUESTION 7:

A non-linear source term is given as: 𝑆 = 3𝑇 − 4𝑇 3 . Then it can be linearized in the form:

S = Sc + S pTp . Then :

(i) 𝑆𝑐 = 8𝑇 ∗ 3 and 𝑆𝑃 = 3 − 12𝑇 ∗ 2 for all 𝑇 ∗

1
(ii) 𝑆𝐶 = 8𝑇 ∗ 3 and 𝑆𝑃 = 3 − 12𝑇 ∗ 2 for 𝑇 ∗ > 4
(iii) 𝑆𝐶 = 3𝑇 ∗ − 4𝑇 ∗ 3 and 𝑆𝑃 = 0
(iv) 𝑆𝐶 = 8𝑇 ∗ − 4𝑇 ∗ 3 and 𝑆𝑃 = −5

Which of the following source term linearization is/are correct?

a. (i) only
b. (ii) only
c. (ii) and (iii) only
d. (ii), (iii) & (iv) only

Correct Answer: d.
Detailed Solution:

Source term linearization: 𝑆 = 𝑆𝐶 + 𝑆𝑃 𝑇𝑃 such that 𝑆𝑃 < 0

𝑑𝑆 ∗
= 3 − 12𝑇 ∗ 2
𝑑𝑡
𝑑𝑆 ∗
Then, 𝑆 − 𝑆 ∗ = (𝑇 − 𝑇 ∗ )
𝑑𝑇
3
Or, 𝑆 = 8𝑇 ∗ + (3 − 12𝑇 ∗ 2 )𝑇
1
𝑆𝐶 = 8𝑇 ∗ 3 , 𝑆𝑃 = (3 − 12𝑇 ∗ 2 ) for 𝑇 ∗ > 4

Also, can be linearized by dumping everything in 𝑆𝐶 :

𝑆𝐶 = 3𝑇 ∗ − 4𝑇 ∗ 3 , 𝑆𝑃 = 0
And by making adjustments, 𝑆𝐶 = 8𝑇 ∗ − 4𝑇 ∗ 3 and 𝑆𝑃 = −5
 QUESTION 8:

For flow through a porous medium, the distributed resistance to flow is expressed by the source term
S = −C u u , for the x – direction momentum equation. Here ‘C’ is a positive constant and ‘u’ is
the velocity component in the x- direction. The source term can be linearized as

(i) S = Cu*2 − 2Cu*u for u  0

(ii) S = −Cu*2 + 2Cu*u for u  0

(iii) S = Cu*2 − 2Cu*u for u  0

(iv) S = −Cu*2 + 2Cu*u for u  0

Which of the following statements is/are correct?

(a) (i) only

(b) (ii) only

(c) (i) and (ii)

(d) (iii) and (iv)

Correct Answer: c.

Detailed Solution:

Case (i) u  0

S = −Cu2

S − S* =
dS
du *
( u − u* )

S − Cu*2 = −2Cu* (u − u* )

S = Cu*2 − 2Cu*u

Since u * is positive,

S p = −2Cu*

It gives negative S p .
Case (ii) u  0

S = Cu2

S − S* =
dS
du *
( u − u* )

S − Cu*2 = 2Cu* (u − u* )

S = −Cu*2 + 2Cu*u

Since u * is negative,

S p = 2Cu*

QUESTION 9:

Consider a 1-dimensional slab of length 3 cm and thermal conductivity 10 W/mK with

uniform heat generation of 1000W/m3. The boundary conditions are: at x=0,T=300K; at
x=L, q=100W/m2. If the domain is discretized into 3 equal control volumes with grid
points as shown in the figure, then the percentage difference in the value of temperature
at node 5 when the boundary condition at node 5 is implemented using finite volume
formulation and finite difference formulation is

(a) 0%

(b) 0.25%

(c) 0.50%

(d) 0.75%

Correct Answer: b.
Detailed Solution:
The governing equation is

d  dT 
+ Qdx = 0
dx  dx 
k

Finite volume method of implementing boundary condition:

Discretizing the governing differential equation, we get

d  dT 
5 5

4 dx  k dx  dx + 4 Qdx = 0
 dT   dT  x
 k  − k  +Q =0
 dx  5  dx  4 2
2k Qx
 −qL" − (T5 − T4 ) + =0
x 2
Q ( x )
2
x
 T5 FVM = − qL" + + T4
2k 4k

Finite difference method of implementing boundary condition:

T5 − T4
qL" = −k
 x 
 2k 
 
x
 T5 FDM = T4 − qL"
2k

% difference =

(T 5 FVM − T5 FDM ) 100%

Q ( x )
2

= 100%
4k
= 0.25%

QUESTION 10:
 Consider a 1-dimensional slab of length 3 cm and thermal conductivity 15 W/mK with uniform
𝑑𝑇
heat generation of 500W/m3. The boundary conditions are: at x=0,T=500K; at x=L, −𝑘 𝑑𝑥 =
ℎ(𝑇 − 𝑇𝑎 ), where ℎ = 50𝑊/𝑚2 𝐾 and ambient temperature 𝑇𝑎 = 300𝐾. If the domain is
discretized into 3 equal control volumes with grid points as shown in the figure, then the value
of temperature at node 5 when the boundary condition at node 5 is implemented using finite
volume formulation is

a. 𝑇5 = 0.984𝑇4 + 4.92

b. 𝑇5 = 0.494𝑇4 + 2.46

c. 𝑇5 = −0.984𝑇4 + 2.46

d. 𝑇5 = 0.984𝑇4 − 4.92

Correct Answer: a.

Detailed Solution:
Finite volume method of implementing boundary condition:

Discretizing the governing differential equation, we get

5
d  dT  5

4 dx  k dx  dx + 4 Qdx = 0
 dT   dT  x
 k  − k  +Q =0
 dx  5  dx  4 2
2k Qx
 −h(T5 − Ta ) − (T5 − T4 ) + =0
x 2

Then,

𝑇5 = 0.984𝑇4 + 4.92
*********END********