You are on page 1of 7

MSE 321

Engineering Thermodynamics and Heat Transfer



Free and Forced Convection

Peyman Taheri

Determination of heat transfer coefficient h , for free and forced convection in different geometries.

Figure 1 shows the experimental instrument in details. This unit has the following components:

The vertical air duct (1) with a cross-sectional area

of 120 120 mm 2 and a length of 1m is used to
guide the air flow over a heated surface. The
ambient air enters the duct at the bottom and
heated air leaves the duct at the top (11)

The air duct has several measuring glands (2) that

enable one to record the temperature at various
locations by using a type-K thermocouple (3). In
addition, a fan with flow sensor (4) blows air into
the duct and measures its volumetric inlet flow
rate. Two thermistors (5, 6) record the inlet and
outlet temperatures of the air flow.

Assemblies of heaters and heated surfaces with

different geometries (7-9) can be mounted inside
the duct using simple toggle-type fasteners. The
different heated surfaces (flat plate, pipe bundle or
fins) are each heated by four resistive heaters with
maximum heating power of about 170 W. The
applied voltage to the heaters is adjustable for
achieving variable heat output. There are bimetallic strips to interrupt the supply of power,
when certain temperatures are established (also
used to ensure that the temperature does not
exceed 120C).

The control and display unit (10) allow controlling

the power supply and regulate the fan speed and
heaters power. In addition, this unit displays the
electrical power supplied to the heaters, the
volumetric flow rate of air, the inlet and outlet air
temperatures, and the temperatures measured with
the type-K thermocouple. A PC is connected to
the apparatus for data acquisition.

MSE 321 (Fall 2013)

Figure 1: Components of experimental apparatus.

Peyman Taheri

The LabView software is used to display the measured data. Figure 2 shows the system diagram in LabView
with some sample data values. The system diagram shows temperatures and heat transfer parameters at
different locations of the flow path (the blue line). The definitions of the parameters are presented in Table 1.
Table 1: Parameters definition




dm / dt
dQ / dt

Inlet air temperature

Air velocity
Air mass flow rate
Rate of heat transfer
Heater input power
User defined fin surface temperature
Thermocouple (3) temperature
Ambient pressure


Ambient temperature
Ambient relative humidity
Type of the installed fin
Fin efficiency
Heat transfer coefficient
Reynolds number
Nusselt number
Outlet air temperature

Figure 2: System diagram in LabView.

MSE 321 (Fall 2013)

Peyman Taheri
Figure 3 shows different geometries for the heated surface;

Flat plate,

Pipe bundle (needle fins),

Rectangular fins.

Air flows past the heating element and absorbs heat in the process. Sensors record the volumetric flow rate of
the air, the heating power and the temperatures at all relevant points. The measured values can be read on
digital displays. At the same time, the measured values can also be transmitted directly to a PC. Free and
forced convective cooling will be investigated on all three surface configurations. Comparison between the
flat plate results and the results of finned surfaces can be used to find the effects of fins on the heat transfer
coefficient under free and forced convection heat transfer.


Figure 3: (a) Flat plate, (b) pipe bundle, and (3) rectangular fins


There are three modes for heat transfer: convection, conduction, and radiation. The convection heat transfer
plays an important role in many industrial applications. The convection heat transfer is usually subdivided
into free and forced convection. In the forced convection, the fluid is blown or pumped past the heated surface
using a pump or a fan, while in the natural (or free) convection, fluid flow is naturally achieved by buoyancy
effects, i.e., density variation in the fluid.
The heat transfer rate to the fluid Q can be calculated using the fist law of thermodynamics for the heated

Q = m h


where h is the enthalpy variation of the fluid before and after the heated surface and m is the mass flow
rate which is calculated as,

m = wA

MSE 321 (Fall 2013)


Peyman Taheri
here is the air density, w is the averaged velocity and A is the cross-sectional area of the duct which is
equal to 0.0144 m2 in this experiment. The air density can be found from thermodynamics tables. Using ideal
gas assumption for the air, Eq. (1) becomes,

Q = m cP T


The temperature difference T is calculated from the difference between the average inlet and outlet
temperatures. The specific heat capacity of the air cP also depends on the air temperature and should be
found from thermodynamics tables. Since the temperature is varying along the duct, the cP value should be
evaluated in the averaged temperature of air in the duct, TM ,
TM =

T1 + T2


The heat sources on the testbed consist of electrical resistors; thus, the amount of power that is fed to the
heaters, P1 , can be calculated.
The factor for fin efficiency, , provides information on the losses which occur during heat transfer. This
factor indicates the portion of the input energy that is transferred to the fluid. This can be written as,



The amount of 1 - shows all losses resulted from convection and radiation to the surroundings and not to

the fluid. The heat transfer from a surface to a fluid can be described mathematically,
Q = A Tm


where is the heat transfer coefficient. The heat transfer rate is the same as the amount calculated from
Eq. (3). Determination of Tm is challenging; if one assumes that temperature is varying linearly along the
duct, Tm will be identical to TM calculated from Eq. (4). Another important value introduced in the literature
is Log Mean Temperature Difference (LMTD). It is calculated using the following formula,

Tm =

(T1 - T fin,in ) - (T2 - T fin,out )

(T1 - T fin, in )
(T2 - T fin, out )


Since the surface temperature, T fin , of the heater remains almost constant across the entire area and only the
temperature of the air changes significantly between the inlet and the outlet, one can simplify this equation,
Tm =

MSE 321 (Fall 2013)

T1 - T2
(T1 - T fin )


(T2 - T fin )

Peyman Taheri

Therefore, the heat transfer coefficient, , can be evaluated from Eqs. (3), (6) and (8). The heat transfer
depends not only on the temperature difference and the surface material of the heater, but is also influenced
by the flow regime, i.e., laminar or turbulent flow.
Reynolds number is a criterion for defining whether a flow is turbulent or laminar. For a flow over a flat plate
the transition between these two regimes occurs at approximately Recrit = 1- 5105 . However, there are other

values for pipes and fins. The Reynolds number is defined as,
Re =



where l is a characteristic length scale, which is the plate length for flat surfaces, and = / is the
kinematic viscosity of the fluid (air). The kinematic viscosity of the air is temperature dependant and can be
taken from thermodynamics tables at TM .
The Nusselt number is dimensionless and is used in measuring heat transfer rates,

Nu =



where k is thermal conductivity. The Nusselt number can be calculated once the heat transfer coefficient , is
known. The following correlation (obtained experimentally) offers another way of determining the Nusselt
number for a parallel flow over a smooth surface (plate) for laminar flow,
Nu = 0.664 Re0.5 Pr 0.33


Using the values obtained from Eq. (11), it is possible to check the accuracy of experiments for flat plate

At the beginning of the experiment the heater element with the flat plate is connected to the control and
display unit prior to being fixed in the air duct. Once the power supply has been connected, the potentiometer
on the control unit is set to 100% and the surface temperature T fin is measured using the thermocouple.
Once a steady-state condition has been reached, there is no noticeable temperature change at the surface of the
heater element, the temperature is saved. The heater is then placed in the duct. Once again it is necessary to
wait until a steady-state condition is established. The following values are measured;

Temperature T1 at the inlet of the air duct,

Volumetric flow rate ( wA ) at the inlet of the air duct,

Temperature T2 at the outlet of the air duct.

This experimental sequence is applicable to all experiments as tabulated in Table 2. In the case of the pipe
bundle and the fins, the surface temperature must be measured prior to fitting in the air duct.
Turn on the fan and repeat the above-mentioned sequence. Increase the fan power to 100% of its rated
capacity with the heater at full power.

MSE 321 (Fall 2013)

Peyman Taheri

The temperature drop within the heated pipe bundle is measured on the heater insert for the free and forced
convection cases. The heater is operated at 100% power output and the thermocouple pushed into the air duct
through the measuring gland such that the tip is in the centre of the cylinder bore. The thermocouple is then
removed from the first measuring gland and is inserted in the other three holes in the cylinder.
Table 1: Measurement data.
Flat Plate


Pipe bundle


Rectangular fins


w [m/s]

After completing the measurements, find the efficiency of each system at natural and forced convection
modes. Determine effects of flow regime on the heat transfer rate. For the flat plate heater calculate the
accuracy of the experimental data through comparison of the measured values to the one obtained from
Eq. (11).

1) What are the differences between laminar and turbulent flows? Which one has the higher heat transfer
2) What is the difference in the measured values of heat transfer coefficient if one uses linear average
temperature instead of LMTD?

MSE 321 (Fall 2013)