ME5304

NATIONAL UNIVERSITY OF SINGAPORE

ME5304 – EXPERIMENTAL FLUID MECHANICS

(Semester I: 2019/2020)

Time Allowed: 2.5 Hours

__________________________________________________________________________________

INSTRUCTIONS TO STUDENTS

1. Please write your student number only. Do not write your name.

2. This assessment paper contains FOUR (4) questions and comprises NINE (9) printed
pages.

3. Students are required to answer ALL questions.

4. Students are to write the answers for each question on a new page.

5. Please answer Questions 1 and 2 on one answer booklet, and Questions 3 and 4 on a
separate answer booklet. Please submit both sets of answer booklets separately.

6. This is a CLOSED BOOK EXAMINATION. Students are not allowed to bring in any
reference materials into the examination room.

7. All questions do not carry equal marks: Question 1 carries 30 marks, Question 2 carries 20
marks, whereas Questions 3 and 4 carry 25 marks each.

8. An information sheet has been provided for the purpose of this examination. Please do not
remove the information sheet from the examination room.

9. Programmable calculators are NOT allowed for this examination.

 Page 2 ME5304

QUESTION 1 (TOTAL: 30 MARKS)

[Important information is provided at the end of the question.]

(a) When air of temperature Ta flows at a velocity U past a heated circular cylinder of
diameter D whose surface is maintained at a constant temperature Tw (Tw > Ta), heat is
lost from the cylinder to the surrounding air due to forced convection. The air density,
viscosity and thermal conductivity are ρ, µ and k, respectively, and are functions of the
air temperature. The forced convection heat transfer coefficient h only depends on D,
U, ρ, µ and k. Use Buckingham’s Π Theorem to determine the relevant dimensionless
parameters which characterize this forced convection heat transfer problem. Your
repeating variables must include D, U and ρ. Explain whether it is sufficient to only use
these three repeating variables and whether it is necessary to include any other repeating
variable(s).
(8 marks)

(b) A student performs a series of experiments using a heated circular cylinder of diameter
D = 10 cm to characterize the forced convection heat transfer characteristics. The
cylinder surface is maintained at a temperature Tw of 80°C and the air temperature Ta is
maintained at 300 K. The heat transfer coefficient h is determined over a wide range of
air flow velocities U. Oblivious of the usefulness of using dimensionless parameters to
present his results, the student summarizes his data using the following correlation:

h = 8.14U 0.81 .

Express the above correlation in the following form:

Π1= M ( Π 2 ) ,
N

where Π1 is the dimensionless parameter which is directly proportional to the heat

transfer coefficient h and Π2 is the dimensionless parameter which is inversely
proportional to the viscosity µ. Determine the values of M and N.

Hence, estimate the rate of heat loss from a heated circular cylinder of 30 cm diameter
and 1 m span maintained at a surface temperature Tw of 65°C. The air flowing past the
cylinder at a velocity U of 20 m/s has a temperature Ta of 300 K.
(11 marks)

(c) One application for the forced convection heat transfer from a circular cylinder involves
the operation of a constant-temperature hot-wire anemometer for velocity
measurements. A hot wire is maintained at a constant elevated temperature Tw (and thus
constant electrical resistance Rw). The hot wire is calibrated by noting the anemometer
output voltage E over a range of flow velocities U. When the ambient air temperature
Ta is varied, a series of calibration curves is obtained, as shown in Figure 1.
 Page 3 ME5304

Figure 1

How should the results be replotted so that the calibration data obtained at different
ambient air temperatures Ta may be collapsed onto one single curve?

A hot wire is maintained at a constant temperature Tw of 250°C. For an ambient air

temperature Ta of 300 K, the anemometer output voltage E is related to the flow velocity
U via the following equation:

E 2 1.794 + 1.680U 0.45 .

=

Determine the anemometer output voltage for a flow velocity of 20 m/s. For the same
anemometer output voltage, determine the corresponding flow velocity at an ambient
air temperature Ta of 60°C. Assume that the hot-wire temperature Tw remains constant
at 250°C.
(11 marks)

Notes:

(i) The rate of heat transfer Q (SI units: W) from a heated surface of area A at
temperature Tw to an ambient fluid at temperature Ta is given by

=Q hA (Tw − Ta ) ,

where h is the heat transfer coefficient.

(ii) The properties of air vary with temperature T according to the following
equations.

Viscosity:

0.85
−5 
T 
µ=
1.846 ×10   Pa ⋅ s
 300 
 Page 4 ME5304

Thermal conductivity:

( )
0.9
 T 
=k 0.0263   W m2 ⋅ K
 300 

Density:

101 325
ρ= kg m3
287 × T
(The ambient air pressure is assumed to remain constant at 101 325 Pa).

In the above equations, the temperature T is given in Kelvins (K). The

temperature in Kelvins (K) is related to the temperature in Degrees Celsius (°C)
by

T ( K ) = T ( °C ) + 273.15 K

(iii) All fluid properties are to be evaluated at the film temperature Tf, which
corresponds to the arithmetic mean of the solid surface or wall temperature Tw
and the ambient air temperature Ta:

T +T
Tf = w a
2
 Page 5 ME5304

QUESTION 2 (TOTAL: 20 MARKS)

(a) What should be the minimum radii of the circular end plates installed at both ends of
the 10 cm diameter circular cylinder described in Question 1(b) so that one obtains the
appropriate two-dimensional flow? If circular end plates of excessively small radii are
used, explain whether it necessarily implies that the flow lacks two-dimensionality.
(4 marks)

(b) It is necessary to determine the standard deviation of rapidly-varying pressure

fluctuations at different points along the surface of a circular cylinder. Explain whether
it is possible to use a U-tube manometer for such measurements. A researcher proposes
using an unbonded strain gauge pressure transducer instead. Explain with the aid of a
sketch the operating principles of such a pressure transducer. Due to space constraints
within the wind tunnel, the researcher uses extremely long and narrow flexible PVC
tubings to connect the pressure tappings on the cylinder surface to the pressure
transducer mounted outside the wind tunnel. Explain whether the researcher is able to
accurately determine the standard deviation of the pressure fluctuations.
(9 marks)

(c) The settling chamber of most wind tunnels contain honeycombs and screens. Explain
the main objectives of honeycombs and screens. Two different designs have been
proposed for the settling chamber of a wind tunnel. In the first design, the flow passes
through the honeycomb followed by a series of screens before the flow passes through
a contraction with a relatively small contraction ratio. In the second design, the flow
passes through a series of screens followed by the honeycomb before the flow passes
through a contraction with a relatively small contraction ratio. Explain which is the
preferred design.
(7 marks)
 Page 6 ME5304

QUESTION 3 (TOTAL: 25 MARKS)

A student is going to study the unsteady flow phenomenon in a confined cylindrical container
with one rotating end as shown in the following figure. The flow characteristics are determined
by the Reynolds number Re = ΩR2/ ν and aspect ratio Λ = H/R, where Ω is the rotating speed
of the bottom end wall (rad/s), R (measured at 86.00 ± 0.05 mm) and H are the radius and
height of the cylindrical container. The working fluid is a mixture of glycerin and water with
kinematic viscosity ν = 0.40 ± 0.01 cm2/s at a room temperature of 30.0°C, measured by a
Rheometer.

According to a theoretical study, the flow will experience an oscillating behavior at very low
frequency (less than 1 Hz) at around Re = 2800 and Λ = 1.76. Having considered the available
measurement equipment in the lab, the student chose a constant temperature anemometer
(CTA) and a hot film for the measurement. The hot film has a resistance of R20 = 15 Ohms and
the temperature coefficient of resistance of the hot film α20 = 0.36%/°C.

H
z

Ω θ
r

(a) Please briefly explain the working principle of a CTA. Calculate the hot film operating
temperature if the student sets the overheat ratio to 1.8 under the room temperature of
30.0°C. Please explain if this overheat setting is appropriate and recommend a suitable
overheat ratio if it is not.
(7 marks)

Assume the following equation is true for all test conditions:

(b) Calculate the required rotating speed (Ω) of the bottom end wall to obtain Re = 2800?
If the rotating end wall is controlled by a stepping motor and the accuracy of the rotation
speed is about 1% of the operating value, calculate the uncertainty in the Reynolds
number.
(8 marks)

(c) The student used an NI card for data collection with a sampling frequency of 100 Hz
and the results are shown below. The student observed two dominant frequencies. Can
 Page 7 ME5304

you suggest some steps the student can take to check if the frequencies captured are
correct or not?
(5 marks)

Hot film signal FFT

0.4 0.08

0.2
Hot-film output

Amplitude
0.0 0.04

-0.2

-0.4 0.00
0 20 40 60 0 0.29 0.42 1 2
t (s) f

(d) The lab purchased a 2D PIV system with a frame rate of 10Hz. Describe the main
components of a 2D PIV system. Explain if this PIV system can be used to study the
unsteady flow phenomena in the confined cylindrical container with one rotating end?
(5 marks)
 Page 8 ME5304

QUESTION 4 (TOTAL: 25 MARKS)

(a) In a rectangular wind tunnel with a test section of 0.5m (width) × 2.0m (length) and
freestream velocity adjustable up to 50m/s, the boundary layer starts to develop from the
contraction and grows into the test section as shown in Fig. (a). By the time it reaches the
test section, the boundary layer is fully turbulent. A student measured the boundary layer
profile and its thickness at two locations (x1 and x2) at the bottom wall of the wind tunnel
test section. The following measurements are obtained for a speed setting of V = 10m/s.

δ1 = 3.5cm δ2 = 6.5cm

At both locations (x1 and x2) the boundary layer profile fits to a one-eighth-power law
approximation,
𝑢𝑢 𝑦𝑦 1/8
= � � for 𝑦𝑦 ≤ 𝛿𝛿 and 𝑢𝑢 = 𝑉𝑉 for 𝑦𝑦 > 𝛿𝛿
𝑈𝑈 𝛿𝛿
𝑑𝑑𝑑𝑑
The shear stress can be calculated using the Karman integral equation 𝜏𝜏𝑤𝑤 = 𝜌𝜌𝑈𝑈 2 𝑑𝑑𝑑𝑑
𝛿𝛿 𝑢𝑢 𝑢𝑢
where θ is the momentum thickness, defined as 𝜃𝜃 = ∫0 𝑈𝑈
(1 − 𝑈𝑈)𝑑𝑑𝑑𝑑

Estimate the total skin friction drag force FD acting on the bottom wall of the wind tunnel
between x1 and x2.
(15 marks)

You may assume that the width of the test section b = 0.5m and the flow is completely
two-dimensional, steady and with zero-pressure gradient. The density of air ρ = 1.2
kg/m3.
 Page 9 ME5304

(b) The student wants to further measure the total skin friction drag force acting on the region
between x1 to x2 of the wind tunnel bottom wall to compare the above estimation, using
a six-component balance available in the lab. This balance is connected to a flat plate at
the wind tunnel bottom wall and used to measure the shear stress on this plate, using the
Fx component of the force balance. The specifications of the available balance are as
follows:

Calculate the standard deviation of the force sensor of the Fx component and evaluate if
the force balance can be used to accurately measure the skin friction drag force estimated
in (a). If not, please provide any suggestions to the student to ensure a proper
measurement to be conducted, supposing there is no other balance available in the lab.
(10 marks)

– END OF PAPER –