CHAPTER 3

CONCEPTS OF THERMAL ANALYSIS IN RADIATOR

The thermal performance of an engine is usually studied by heat balance-

sheet. The main components of the heat balance are;
1. Heat equivalent to the effective (break) work of the engine,
2. Heat rejected to the cooling medium,
3. Heat carried away from the engine with the exhaust gases, and
4. Unaccounted losses.
The following table gives the approximate percentage values of various losses
in SI and CI engine [15Moh];
Table 3.1. Approximate Percentage Values of Losses

% Break % Heat to % Heat to exhaust % Unaccounted

Engine
power cooling water gases losses
SI 21-28 12-27 30-55 0-15
CI 29-42 15-35 25-45 10-20

3.1. Determination the Amount of Heat Removed from Radiator

Density of fuel is calculated from
ρf
SG f =
ρw Equation
3.1
Where,
SGf = Specific gravity of fuel (Unit less)

ρf = Density of fuel (kg/m3)

ρw = Density of water (kg/m3)

To find the mass of fuel supplied per hour of engine, it can be considered from

ṁf = Fuel consumption in liter ¿ ρf Equation 3.2

Where,
 28

ṁf = Mass of fuel supplied per hour (kg/hr)

ρf = Density of fuel (kg/m3)

 28

Total amount of heat introduced into the engine can be calculated from
[11Sad]
Qtotal =ṁf ×L .C .V Equation 3.3
Where,
ṁf = Mass of fuel supplied per hour (kg/hr)

L.C .V = Lower calorific value of fuel (MJ/kg)

Heat carried away by cooling water is about 35% of total amount heat
introduced in to the engine [15Moh],
Qcool =35%×Qtotal Equation
3.4
Where,
Qtotal = Amount of heat introduced into the engine (kW)

Qcool = Amount of heat rejected by cooling water (kW)

In order to calculate the heat to be dissipated in the radiator making allowance

for the dirtying process in the radiator surface, the design amount of heat is increased
by 10 percent [10Tor].
Qrad=1. 1×Qcool Equation 3.5
Where,
Qrad= Amount of heat removed from radiator (kW)

Qtotal = Amount of heat introduced into the engine (kW)

3.2. Consideration of Heat Transfer in Radiator

A radiator should be capable of transferring heat at the specified rate in order
to achieve the desired temperature change of the fluid at the specified mass flow rate.
When the wall thickness of the tube is small and the thermal conductivity of the tube
material is high, as is usually the case, the thermal resistance of the tube is negligible.
During the heat transfer process in radiator, two main forms of heat transfer occurs;
heat convection due to the internal flow of fluid passing through non-circular tubes
and heat convection due to the external flow of air across the tube bank. The entire
heat transfer process in radiator is forced convection, therefore, the heat loss to the
surroundings by radiation is negligible.
 29

Two common methods; the logarithmic mean temperature difference (LMTD)

method and effectiveness-NTU method exist for expressing the heat transfer
characteristics of a radiator. The LMTD method is very suitable for determining the
size of a radiator to realize prescribed outlet temperatures when the mass flow rates
 30

and the inlet and outlet temperatures of the hot and cold fluids are specified. The
effectiveness-NTU method in radiator analysis is the determination of the heat
transfer rate, the outlet temperatures of the hot and cold fluids for prescribed fluid,
mass flow rates and inlet temperatures when the type and size of the radiator are
specified.
The following points need to be assumed to calculate the heat transfer
performance in radiator;
1. Steady operating conditions exit.
2. The heat transfer between radiator and surroundings is negligible.
3. Changes in the kinetic and potential energies of fluid streams are negligible.
4. The thermal resistance of the tube is negligible since the tube is thin-walled and
highly conductive.
5. Properties of water and air are used at average temperature in radiator [02Cen].

3.2.1. Determination of Convective Heat Transfer Coefficient for Internal Flow

(Water)
To determine the heat transfer coefficient, we first need to find properties of
water at average or mean temperature, volume flow rate and velocity of water flow in
radiator tubes. The properties of water are density, specific heat, thermal conductivity,
dynamic viscosity and Prandtl number. The average water temperature in the radiator
is computed as [12Car]
ΔT w
T w,avg =T w,in−
2 Equation 3.6

Where,
T w,avg = Average temperature of water flowing through the radiator (°C)

T w,in = Inlet water temperature of radiator (°C)

ΔT w = Temperature drop of water (°C)

The volume flow rate of water through the radiator is calculated from

Qrad
V w=
ρw C P ΔTw
w Equation 3.7
Where,
 31

V w = Volume flow rate of water (m3/s)

Qrad= Amount of heat removed from radiator (kW)

ρw = Density of water (kg/m3)

CP =
w Specific heat of water (kJ/kgK)

ΔT w = Temperature drop of water (°C)

The velocity of water flow in tube is [12Car]

Vw
v w=
At ×N t Equation 3.8
Where,
v w= Velocity of water flow in tube (m/s)

V w = Volume flow rate of water (m3/s)

A t = Cross-sectional area of tube (m2)

N t = Number of tubes in radiator

Once the Nusselt number is available, the convection heat transfer coefficient
is determined from [05Kut, 02Cen]

Nu×k w
hw =
Dh Equation
3.9
Where,
hw = Convective heat transfer coefficient of internal flow (W/m2K)

Nu= Nusselt number (Unit less)

k w = Thermal conductivity of water (W/mK)

Dh = Hydraulic diameter (m)

Depending on the flow condition, flow in a tube can be laminar or turbulent.

For flow in tube Reynolds number is defined as [05Kut, 02Cen]
 32

ρ w v w Dh
Re=
μw Equation 3.10
Where,
Re= Reynolds number (Unit less)

ρw = Density of water (kg/m3)

v w= Velocity of water flow in tube (m/s)

Dh = Hydraulic diameter (m)

μw = Dynamic viscosity of water (kg/ms)

For fully developed turbulent flow in smooth tube, friction factor and Nusselt
number correlation include the range (0.5≤ Pr ≤ 2000, 2300 < Re <10000) where most
automotive radiators operate [02Cen].

f =(1 .58lnRe−3 . 28)−2 Equation 3.11

f/2(Re−1000 )×Pr
Nu= 1 2
1+12 .7 (f/2) 2 ×(Pr 3 −1 ) Equation 3.12
Where,
f = Friction factor
Nu= Nusselt number
Re= Reynolds number
Pr= Prandtl number
For flow through noncircular tubes, the Reynolds number as well as Nusselt
number and the friction factor are based on the hydraulic diameter defined as [02Cen]
At
D h =4×
Pt Equation 3.13
Where,
Dh = Hydraulic diameter (m)

A t = Cross-sectional area of tube (m2)

Pt = Wetted perimeter of tube (m)

For a rectangular tube or pipe, the cross-sectional area of tube and wetted
parameter is [12Car]
 33

A t =W t ×H t Equation 3.14
Pt =2W t +2Ht Equation 3.15
Where,
At = Cross-sectional area of tube (m2)
Pt = Wetted parameter of tube (m)
Wt = Width of tube (m)
Ht = Height of tube (m)

3.2.2. Determination of Convective Heat Transfer Coefficient for External Flow (Air)
To determine the heat transfer coefficient, we first need to find properties of
air at average or mean temperature, volume flow rate and velocity of air flow through
the radiator core. The properties of air are density, specific heat, thermal conductivity,
dynamic viscosity, kinematic viscosity and Prandtl number. The average air
temperature flow through the radiator is computed as [12Car]
ΔT a
T a,avg=T a,in +
2 Equation 3.16
Where,
T a,avg= Average temperature of air flow through the radiator (°C)

T a,in= Inlet air temperature of radiator (° C)

ΔT a = Temperature rise of air, (° C)

The volumetric air flow through the radiator core is given by

Qrad
V a=
ρa C P ΔT a
a Equation 3.17
Where,
V a = Volume flow rate of air (m3/s)

Qrad= Amount of heat removed from radiator (kW)

ρa = Density of air (kg/m3)
CP =
a Specific heat of air (kJ/kgK)

ΔT a = Temperature rise of air¿C)

 34

Va
va=
Ac −( N t ×H t ×Lt ) Equation 3.18
Where,
v a= Velocity of air flow through the radiator core (m/s)

V a = Volumetric flow rate of air (m3/s)

A c = Area of core (m2)

N t = Number of tubes

H t = Height of tube (m)

Lt = Length of tube (m)

Area of core is
A c =H c ×Lc Equation 3.19
Where,
Lc= Length of core (m)

H c = Height of core (m)

The Stanton number is another dimensionless representation of the heat
transfer coefficient. It is defined as [03Sha]
ha
St=
ρa v a C p a
Equation 3.20
Where,
ha = Convective heat transfer coefficient for external flow of air (W/m2K)
St= Stanton number (Unit less)

ρa = Density of air (kg/m3)

v a = Velocity of air flow through the radiator core (m/s)

CP =
a Specific heat of air (kJ/kgK)

The Colburn factor is a modified Stanton number to take into account the
moderate variations in the fluid Prandtl number. It is defined as [03Sha]
 35

2
3
j=St×Pr Equation 3.21
Where,
St= Stanton number

j= Colburn factor
Pr= Prandtl number

As the Stanton number depended on the fluid Prandtl number, The Colburn
factor is nearly independent of the flowing fluid for (0.5≤ Pr ≤ 10 ) from laminar to
turbulent flow condition. Colburn factor can be defined as a function of Reynolds
number as follows. For best accuracy select data for a coolant flow rate where
ℜ> 4000 [10San],
−0 . 277
j=0.5457Re Equation 3.22
Reynolds number is defined as [10San]
v a ×W f
Re=
υa Equation 3.23
Where,
Re= Reynolds number (Unit less)

va= Velocity of air flow through the radiator core (m/s)

W f = Width of fin (m)

υa= Kinematic viscosity of air (m2/s)

3.2.3. Determination of Fin Efficiency

The important variable for convection heat transfer through the fin surface is
the fin surface area. The extended surface efficiency of air-side surface is related to
the fin efficiency of the extended surface as follow [12Car];
N f × Af
ηo =1− ( 1−η fin )
A fin,base Equation 3.24
Where,
ηo = Overall fin surface efficiency (%)
N f = Number of fins per tube

A f = Single fin surface area (m2)

 36

A fin,base= Total fin per base surface area of a single tube (m2)
ηfin = Fin Efficiency (%)

tanh(mlc )
ηfin =
ml c Equation
3.25
Where,
-1
m = Fin parameter (m )
lc= Fin conduction length (m)

2h a

Where,
m=
√ k copper ×H f Equation 3.26

ha= Convective heat transfer coefficient of air (W/m2K)

k copper = Thermal conductivity of copper (W/mK)

H f = Thickness of fins (m)
Hf
l c=L f +
2 Equation 3.27
Where,
Lf = Length of fin (m)

H f = Thickness of fin (m)

A typical automobile radiator has a fin efficiency of over 95% at 80 to 90

km/hr automobile speed. If fin efficiency is maintained high, as pointed out above,
most of the assumptions made for the fin heat transfer analysis are quiet adequate
[10San].
Single fin surface area is [12Car]
A f =2×W f ×l c Equation 3.28
Where,
Wf = Width of fin (m)
l c= Fin conduction length (m)
For three tubes per fin row, un-finned area on tube surface is [12Car]
 37

A b =3Lt W t −H f W f N f Equation 3.29

Where,
Lt = Length of tube (m)
W t = Width of tube (m)
H f = Thickness of fin (m)

W f = Width of fin (m)

N f = Number of fins per tube

Total fin per base surface area of a single tube is [12Car]

A fin,base =N f A f + A b Equation 3.30

Where,
N f = Number of fins per tube

A f = Single fin surface area (m2)

A b = Un-finned area on tube surface (m2)

3.2.4. Determination of Heat Transfer Rate in Radiator

The effectiveness-NTU method is used to determine the heat transfer
performance of a specified heat exchanger. If effectiveness of the heat exchanger is
known, the actual heat transfer rate of radiator can be determined from [02Cen]

Q rad=εQ max Equation 3.31

Where,
Qrad= Heat transfer rate of radiator (kW)
ε= Heat exchanger effectiveness (%)
Qmax= Maximum heat transfer rate (kW)

Qmax =C min (T w,in −T a,in ) Equation 3.32

Where,
C min = Minimum heat capacity rate of water or air ( C w or C a )

T w,in = Inlet water temperature of radiator (°C)

 38

T a,in= Inlet air temperature of radiator (°C)

C w =ṁw ×C P
w Equation 3.33
Where,
C w = Heat capacity rate of water (W/K)
ṁw = Mass flow rate of water (kg/s)
CP
w = Specific heat of water (kJ/kgK)
C a= ṁ a×C P
a Equation 3.34
Where,
C a= Heat capacity rate of air (W/K)

ṁa = Mass flow rate of air (kg/s)

CP =
a Specific heat of air (kJ/kgK)

The radiator utilizes a cross-flow single pass design where both fluids remain
unmixed. The effectiveness relation for this type of heat exchanger can be determined
from [02Cen]
0 . 22
NTU 0.78

{ [ e(−CrNTU )
−1 ]}
Cr
ε =1−e Equation
3.35
The effectiveness of a heat exchanger is a function of the number of transfer
units and the heat capacity ratio. The number of transfer units is expressed as [02Cen]
UA
NTU=
C min Equation 3.36
Where,
NTU= Number of transfer units (Unit less)
U= Overall heat transfer coefficient (W/m2K)
A= Total heat transfer surface area of the heat exchanger (m2)
C min= Minimum heat capacity rate (W/K)
 39

1
UA=
1 1
+
ηo h a Aex h w A in Equation 3.37
Where,
ηo = Overall fin surface efficiency (%)

ha = Convective heat transfer coefficient of air (W/m2K)

A ex = Total external surface area of radiator core (m2)
hw = Convective heat transfer coefficient of water (W/m2K)
A in = Total internal surface area of radiator core (m2)

A ex = Afin,base×N t Equation 3.38

Where,
A fin,base= Total fin per base surface area of a single tube (m2)
N t = Number of tubes
A in =( 2W t +2Ht ) Lt N t
Equation 3.39
Where,
W t = Width of tube (m)
H t = Height of tube (m)

Lt = Length of tube (m)

N t = Number of tubes
Heat capacity rate ratio is [02Cen]
C min
Cr=
Cmax Equation 3.40
Where,
Cr= Heat capacity rate ratio

C min= Minimum heat capacity rate (W/K)

C max = Maximum heat capacity rate (W/K)
The foregoing derivation illustrates how the effectiveness for a given flow
arrangement can be expressed in terms of two dimensionless parameters, the heat
capacity rate ratio Cmin/Cmax and the ratio of the overall conductance to the smaller heat
 40

capacity rate, UA/Cmin. The effectiveness can be determined for given values of NTU
and Cmin/Cmax in Fig. B.6 [02Cen].

3.3. Determination of Radiator Outlet Temperature of Water and Air

The outlet temperatures must be iterated with the initially guessed outlet
temperatures until the numbers are equivalent. The iterated outlet temperatures for
both air and water are the theoretical values, which are used to compare with the
calculated results [12Car].
Outlet water temperature of radiator can be calculated from
Q rad
T w,out =T w,in −
Cw Equation 3.41
Temperature drop of water is
ΔT w =T w,in −T w,out Equation 3.42
Outlet air temperature of radiator can be calculated from
Q rad
T a,out=T a,in +
Ca Equation 3.43
Temperature rise of air is
ΔT a =T a,out −T a,in Equation 3.44
Heat lost by water is
Q w =ṁ w C P ( T w,in −T w,out )
w Equation 3.45
Heat gain by air is
Qa =ṁa C P (T a,out −T a,in )
a Equation 3.46
By the first law of thermodynamics, the rate of heat transfer from the hot fluid
Qw is equal to the rate of heat transfer to the cold fluid Qa [02Cen].
Qw ≅ Qa Equation 3.47