Experiment No- 2 (a)

HEAT TRANSFER BY NATURAL CONVECTION

Aim:
To determine the theoretical and experimental heat transfer coefficient for transfer of heat
from a heated vertical cylinder to atmosphere by free or natural convection.

Introduction:
The convective motion is known to enhance the heat transfer coefficient in fluid medium. The
fluid motion can be created either by external agency (forced convection) or by buoyancy of
the medium (free or natural convection). The natural convection phenomenon is attributed to
the flow circulation generated solely by the difference in temperature of the surface causing
change in the density of the fluid and is not created by any external agency. This process is
continuous and the heat transfer takes place due to the relative motion of hot and cold fluid
particles.
The energy transfer by natural convection is the dominant mode of heat transfer in many
industrial devices like steam radiators, power transformers. Natural convection also plays a
significant role in cooling of electronic devices.

Theory:
In convective cooling, as a cool fluid is passed over a heat wall, temperature profile formation
takes place.
Q = h. A.[Ts − T ( y )]

Velocity and temperature profiles for convection

Near the wall the fluid is subjected to no slip condition; that is, there is a stagnant sublayer.
Since there is no fluid motion in this layer, heat transfer is governed by conduction in this
region. The layer above is a region where viscous forces oppose fluid motion. In this region,
some convection may occur but conduction may well predominate. A careful analysis of this
region allows us to use our conductive analysis in analyzing heat transfer. This is the basis
of our convective theory. At the wall heat transfer rate can be expressed as the heat flux.
T
qconv = − kt = h.(Ts − T )
y
Hence,
T
− kt
y
h=
Heat transfer coefficient (Ts − T )

Dimensionless numbers
● Nusselt’s Number (Nu)
h.L
Nu =
k
Heat transfer by convection Conductive resistance
Nu = =
Heat transfer by conduction Convective resistance
A larger value of Nu implies enhanced heat transfer by conduction.

● Prandtl Number (Pr)

 CP
Pr =
k
Momentum diffusivity
Pr =
Thermal diffusivity

Pr represents the relative importance of momentum and energy transfer for diffusion.

● Grashof Number (Gr)

Free stream velocity U∞ is always zero for natural convection so the Reynolds
number is also zero and is no longer suitable to describe the flow in the system
unlike in forced convection. Hence, we use the Grashof number.

L3 . .g.T
(Inertia force).(Buoyant force)
Gr = =
 2
(Viscous force) 2
The tendency of the natural convective system towards turbulence relies on Gr.
Here
L = Characteristic dimension of the surface (Length in this case)
k = Thermal conductivity of fluid
µ = Dynamic viscosity of fluid
 γ = Kinematic viscosity of fluid
Cp = Specific capacity of fluid
β = Coefficient of volumetric expansion of fluid
g = Acceleration due to gravity
 T = Ts -Ta
Ts = Surface temperature
Ta = Ambient temperature

Experimental Setup:
The apparatus consists of a stainless-steel tube fitted in a rectangular duct in a vertical
fashion. The duct is open at the top and bottom and forms an enclosure and serves the
purpose of undisturbed surroundings. One side of the duct is made up of Perspex for
visualization. An electric heating element is kept in the vertical tube which in turn heats
the tube surface. The heat is lost from the tube to the surrounding air by natural
convection. The temperature of the vertical tube is measured by seven thermocouples.
The heat input to the heater is measured by an ammeter and a voltmeter and is varied
by a dimmer stat. The surface of the tube is polished to minimize radiation losses.

Schematic of heat transfer through natural convection

Thermocouple no. 8 reads the temperature of air in the duct (ambient temperature). The
distance between ‘point 1 and point 2’ and ‘point 2 and point 3’ is 4 cm each and the
others are 7 cm each from ‘point 3 to point 7’.

Requirements:
Electric supply: Single phase 220 V AC, 50 Hz, 5-15 Amp socket with earth connection.
Experimental Procedure:
1. Connect the equipment to power supply. Set the voltmeter reading to some value by
the dimmer stat provided and keep it constant.
2. Wait for about one hour before taking the first reading.
3. After one hour, note the temperatures T1 to T7 by the temperature operating switch in
the interval of 30 minutes.
4. Note down the voltmeter and ammeter reading.
5. Note down the ambient temperature (T8).
6. Repeat the experiment for another power input.
7. Tabulate the readings.

Precautions:
1. Never run the apparatus if power is less than 180 volts and above 230 volts.
2. Never switch ON mains power supply before ensuring that all the ON/OFF
switches given on the panel are at OFF position.
3. Switch OFF temperature indicator gently.
4. Always keep the apparatus free from dust.

Observations:
● Data
Diameter of the tube (D) = 45 mm
Length of the tube (L) = 450 mm
● Observation table
S. No. T1 T2 T3 T4 T5 T6 T7 Tavg T8 V I
T1 – Temperature at point 1 (°C)
T2 – Temperature at point 2 (°C)
T3 – Temperature at point 3 (°C)
T4 – Temperature at point 4 (°C)
T5 – Temperature at point 5 (°C)
T6 – Temperature at point 6 (°C)
T7 – Temperature at point 7 (°C)
T8 – Ambient air temperature in the duct (°C)

Calculations:
● Experimental heat transfer coefficient
Q
hexp =
As .(Ts − Ta )
Where

Q = Heat transfer rate (Watt) = V  I

As= Area of the heat transferring surface =  DL

(T1 + T2 + T3 + T4 + T5 + T6 + T7 )
Ts = Average surface temperature (°C) = 7
Ta = Ambient temperature in the duct (°C) = T8

● Theoretical heat transfer coefficient

From a dimensionless analysis as per the Buckingham Pi theorem, a general relationship
between the above dimensionless numbers has been given as:

h.L
Nu = c(Gr.Pr )m =
k

L3 . .g.T
Gr =
2
Ts + Ta
=
2 (Kelvin-1)

Where m and c are constants and they are evaluated as follows:

c = 0.56, m = 0.25 for 104  Gr.Pr  108
1
c = 0.13, m = for 108  Gr.Pr  1012
3

Temperature, T Density, ρ Specific heat Thermal Kinematic

(0C) (kg/m3) capacity, Cp conductivity, k viscosity, γ

(kJ/kg-K) (W/m-K)
(m2/s)
* ×10-6
-150 2.793 1.026 0.0116 3.08

-100 1.98 1.009 0.016 5.95

-50 1.534 1.005 0.0204 9.55

0 1.293 1.005 0.0243 13.3

20 1.205 1.005 0.0257 15.11

40 1.127 1.005 0.0271 16.97

60 1.067 1.009 0.0285 18.9

80 1 1.009 0.0299 20.94

100 0.946 1.009 0.0314 23.06

120 0.898 1.013 0.0328 25.23

140 0.854 1.013 0.0343 27.55

160 0.815 1.017 0.0358 29.85

180 0.779 1.022 0.0372 32.29

200 0.746 1.026 0.0386 34.63

 250 0.675 1.034 0.0421 41.17

300 0.616 1.047 0.0454 47.85

350 0.566 1.055 0.0485 55.05

400 0.524 1.068 0.0515 62.53

Result:
The theoretical and experimental heat transfer coefficients are …..
Sources of errors:
Report the possible sources of error found in the experiment and the observations.
Discuss exactly how these affect the experimental data (i.e. will the readings be larger
or will they be smaller due to the presence of a particular source).

Discussions:
1. Sketch the temperature and velocity boundary layers at different locations.
2. How does the boundary layer thickness affect cooling rate?
3. Give additional inferences of the experiment.
4. Estimate the errors associated with neglecting surface radiation.

References:
1. Heat Transfer by J.P. Holman, 9th edition, McGraw Hill, 2008, page 327-362
2. Heat transfer- a practical approach by Yunus A Cengel, 2nd edition, McGraw Hill