See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/227412779

Optimization of a central-heating radiator

Article in Applied Energy · November 2006

DOI: 10.1016/j.apenergy.2006.01.009 · Source: RePEc

CITATIONS READS

40 367

2 authors:

Cihat Arslanturk Feridun Ozguc

Ataturk University Istanbul Technical University
22 PUBLICATIONS 513 CITATIONS 6 PUBLICATIONS 174 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Cihat Arslanturk on 26 December 2022.

The user has requested enhancement of the downloaded file.

 APPLIED
Applied Energy 83 (2006) 1190–1197
ENERGY
www.elsevier.com/locate/apenergy

Optimization of a central-heating radiator

a,* b
Cihat Arslanturk , A. Feridun Ozguc
a
Department of Mechanical Engineering, Faculty of Engineering, Ataturk University, 25240 Erzurum, Turkey
b
Faculty of Mechanical Engineering, Istanbul Technical University, Gumussuyu, 80191 Istanbul, Turkey

Accepted 20 January 2006

Available online 5 June 2006

Abstract

An approximate analytical model has been used to evaluate the optimum dimensions of a central-
heating radiator. The radiator problem is divided into three one-dimensional fin problems and then
the temperature distributions within the fins and heat-transfer rate from the radiator are obtained
analytically. The optimum geometry maximizing the heat-transfer rate for a given radiator volume
and the geometrical constraints associated with production techniques, and thermal constraints have
been found. The effects of geometrical and thermal parameters on the radiator’s performance are
presented.
Ó 2006 Elsevier Ltd. All rights reserved.

Keywords: Central heating; Optimization; Radiators

1. Introduction

Radiators are the most popular central-heating emitters. As the radiator is hotter than
the air surrounding it, a certain amount of heat is transferred to the air and thus the water
exists at a lower temperature. Of the various designs available usually equipped with con-
vection fins to improve their heat output, are common in domestic, business and industrial
environments. The use of central-heating radiators is the main form of domestic heating in
the homes. Although radiators are known as radiator, most of their output is by natural
convection [1]. Since the average surface temperature of a central-heating radiator is gen-

*
Corresponding author. Tel.: +90 442 231 4866; fax: +90 442 231 0957.
E-mail address: [email protected] (C. Arslanturk).

0306-2619/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.apenergy.2006.01.009
 C. Arslanturk, A.F. Ozguc / Applied Energy 83 (2006) 1190–1197 1191

Nomenclature

A coefficient defined in Eq. (5)

AF total frontal area, m2
AR frontal area of the radiator, m2
Cj, 1, Cj, 2 temperature distribution coefficients
h convection heat-transfer coefficient (W m2 K1)
k thermal conductivity of radiator material (W m1 K1)
L fin length, m
‘ a square radiator cell half-length, m
mj fin parameters (m1 for j = 1, 2 and dimensionless for j = 3)
q heat-transfer rate per unit depth (W m1)
R radiator’s tube-radius, m
T temperature, K
U average velocity of inner fluid, m/s
xj axial coordinates (m, for j = 1, 2 and dimensionless, for j = 3)
dj thickness of j’s fin, m
hj temperature difference between that of j’s surface and a reference temperature
U volume fraction of radiator material, AR/AF

Subscripts and superscripts

i inner
o outer
* optimum

erally less than 80 °C, the contribution of radiative transfer to the total heat transfer is
smaller than that of the natural convection heat transfer. Because of the low surface tem-
peratures, radiation heat transfer term in the energy balance equations can be linearized in
the thermal analysis of such a radiator [2]. In the present paper, assuming that the predom-
inant modes of heat transfer are conduction and convection, and the effect of the radiation
is ignored, an approximate mathematical model is constructed for finding temperature dis-
tribution and heat transfer rate.
The design and optimization of these radiators or fins and fin assemblies are generally
based on two approaches [3,4]: one is to minimize the volume or mass for a given amount
of heat dissipation, and the other is to maximize the heat dissipation for a given volume or
mass. The optimization problem considered here focuses on finding the optimum dimen-
sions of a central-heating radiator maximizing the heat transfer rate for the given volume
of the radiator material and geometrical and thermal constraints.

2. Mathematical analysis

Consider a radiator which is shown schematically in Fig. 1. Considering steady-state

conditions and neglecting the temperature change across the thickness, we may assume
that the temperature distribution in the radiator is one-dimensional, angular in the tube
1192 C. Arslanturk, A.F. Ozguc / Applied Energy 83 (2006) 1190–1197

To , h o

Ti hi

δ3

L2 x3
R
x1
L1

x2
δ1
δ2

Fig. 1. Schematic of the modelled radiator.

and axial in the fins. Assuming that the predominant modes of heat transfer are conduc-
tion and convection and the effect of radiation is ignored, a linear mathematical model is
considered. Due to the symmetric conditions, it is sufficient that quarter part of the radi-
ator is taken into consideration as shown in Fig. 1. This suggests that the problem be
investigated in terms of three domains for mathematical convenience. Similar mathemat-
ical analyses are seen in the literature [5–7].
For each part of the radiator, energy balance equations are given in the form of
d2 hj
 m2j hj ¼ 0 j ¼ 1; 2; 3 ð1Þ
dx2j
Invoking the continuity of temperature and heat current at the junctions, boundary
conditions of the governing equations can be expressed as:
dh1 ho
 þ h1 ¼ 0 for x1 ¼ 0 ð2aÞ
dx1 k
h1 ðL1 Þ ¼ h2 ð0Þ ð2bÞ
 
dh2  dh1 
¼ ð2cÞ
dx 2 x2 ¼0 dx  1 x1 ¼L1

h2 ðL2 Þ þ T o ¼ h3 ð0Þ þ T i;o ð2dÞ

 
dh2  dh3 
¼ ð2eÞ
dx2 x2 ¼L2 dx3 x3 ¼0

dh3 
¼0 ð2fÞ
dx3 x3 ¼p=2
 C. Arslanturk, A.F. Ozguc / Applied Energy 83 (2006) 1190–1197 1193

where
 1=2  1=2  1=2
2ho 2ho ho þ hi
m1 ¼ m2 ¼ m3 ¼ R ð3aÞ
kd1 kd2 kd3
hi T i þ ho T o
T i;o ¼ h1 ¼ T 1  T o h2 ¼ T 2  T o h3 ¼ T 3  T i;o ð3bÞ
h i þ ho
The solutions of the governing equations given in Eq. (1) are expressed as follows:
hj ðxj Þ ¼ C j;1 cosh ðmj xj Þ þ C j;2 sinh ðmj xj Þ ð4Þ
where j = 1, 2, 3.
Using the well-known Dittus–Boelter correlation, the heat-transfer coefficient, hi, inside
the tube is expressed in terms of pipe radius for selected inner fluid velocity.
0:8 0:2
hi ¼ AðU Þ ðRÞ ð5Þ
The coefficient A in Eq. (5) can be calculated using thermo-physical properties of the
inner fluid [8].
Applying the boundary conditions given in Eqs. (2a)–(2f), the unknown coefficients Cj,1
and Cj,2 in Eq. (4) can be symbolically calculated.

3. Optimization procedure

The objective here is to maximize the heat transfer rate for attaining the radiator vol-
ume fraction and held fix all other thermal parameters. The total heat transfer rate, i.e.
objective function, is readily calculated by applying Newton’s law of cooling between
the tube and the inner fluid as:
Z p=2 Z p=2
qðd1 ; d2 ; d3 ; L1 ; L2 ; RÞ ¼ 4Rhi ðT i  T 3 Þ dx3 ¼ 4Rhi ðT i  h3  T i;o Þ dx3 ð6Þ
x3 ¼0 x3 ¼0

The radiator’s volume-fraction is expressed as an equality constraint.

AR 2L1 d1 þ 2L2 d2 þ pRd3
U¼ ¼ ð7Þ
AF 2L1 ðL2 þ RÞ
Since the radiator’s frontal dimensions, which must be equal or less than ‘ due to the
production technique, are restricted as can be seen in Fig. 1, the following equality con-
straints can be written:
‘ ¼ ðR þ L2 Þ and ‘ ¼ L1 ð8Þ
For simplicity, selecting d1 = d2 = d3 = d and employing the equality constraints given in
Eqs. (7) and (8), an objective function can be found with one independent variable of radi-
ator tube radius, R. The value of R which maximizes the function is obtained by differen-
tiating the function with respect to R and setting the result equal to zero and then solving
the new resulting equation.

4. Results and discussion

It would be of interest to examine the effects of different parameters (such as tube

radius, thicknesses of the fins, heat transfer and coefficients) on the temperature distribu-
1194 C. Arslanturk, A.F. Ozguc / Applied Energy 83 (2006) 1190–1197

tions. However, this would require providing too many examples. Therefore, only one
example is presented that will show the temperature distribution within the fins and the
tube wall. Fig. 2 shows the temperature distributions for a set of given thermal and geo-
metrical parameters.

21.6
21.5
21.4
21.3
T1 -To , [ C]
o

21.2
21.1
21.0
20.9
20.8
20.7
0.00 0.01 0.02 0.03 0.04
a x1, [m]

23.0

22.8

22.6
T2 -To , [ C]

22.4
o

22.2

22.0

21.8

21.6
0.000 0.005 0.010 0.015 0.020 0.025
b x2, [m]

60

55
50

45
T3 -To, [ C]
o

40

35
30

25
20
0 π/4 π/2
c x3, [rad]

Fig. 2. Temperature distribution of the three radiator sections.

 C. Arslanturk, A.F. Ozguc / Applied Energy 83 (2006) 1190–1197 1195

The optimization procedure can be conducted by locating the geometrical and thermal
conditions that yield the total heat transfer rate. However, the existence of such a value
should first be checked. In Fig. 3, the dimensionless heat transfer rate is plotted versus tube
radius for three inner fluid velocities. A clear maximum heat transfer rate is shown in Fig. 3.
Note also that for higher inner fluid velocities, the maximum appears at higher radiuses.
The variations of maximum heat transfer rate and optimum dimensions as a function of
radiator volume fraction are shown in Fig. 4 for three different environment temperatures.

7000

U=0.4
6000
U=0.3
U=0.2
5000
Heat transfer rate, q [W]

4000

3000
=0.04 m
ho=10 W/m2K
2000
k =200 W/mK
T i=353 K
1000 T o=293 K
Φ- =0.2
0
0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035 0.040
Tube radius, R [m]

Fig. 3. Heat transfer rate versus radiator tube radius for the given volume fraction and inner fluid velocity.

Fig. 4. The effects the inner fluid temperature on optimum dimensions and maximum heat transfer rate.
1196 C. Arslanturk, A.F. Ozguc / Applied Energy 83 (2006) 1190–1197

Fig. 5. The effects the room temperature on optimum dimensions and maximum heat transfer rate.

As expected, the increase of radiator volume fraction increases the maximum heat transfer
rate and optimum tube diameter. Assuming that the flow inside the tube is turbulent flow,
the well-known Dittus–Boelter equation was used in optimization calculations. The values
of the optimum tube diameters presented in Fig. 4 have been used for checking the validity
of the aforesaid assumption. It has been seen that the inner fluid flow is turbulent for all
cases in Fig. 4.
The variations of maximum heat transfer rate and optimum tube radius as a function of
radiator volume fraction for three different ambient fluid temperatures are shown in Fig. 5.
It can be seen that the curves shown in Fig. 5 approach an asymptotic value of the heat
transfer rate at the large values of radiator volume fraction. That the maximum heat trans-
fer rate depends on ambient fluid temperature but the corresponding optimum tube radius
does not depend on this parameter is also shown in Fig. 5.

5. Conclusions

An approximate analytical model has been proposed for the optimum design of central-
heating radiators in the present paper. The radiator problem has been divided into three
one-dimensional fin problems. The problems have been solved to evaluate the temperature
distributions within the fins using the boundary conditions of the radiator and the conti-
nuity of temperature and heat current at the junctions of the fins. The temperature differ-
ences have been used within the heat transfer rate from the radiator to the environment.
The optimum radiator geometry maximizing the heat transfer rate has been obtained by
using the approximate analytical model. The present optimization technique can be
extended to central-heating radiator with more complex geometry.

References

[1] Beck SMB, Grinsted SC, Blakey SG, Worden K. A novel design for panel radiators. Appl Therm Eng
2004;24(8–9):1291–300.
 C. Arslanturk, A.F. Ozguc / Applied Energy 83 (2006) 1190–1197 1197

[2] Incropera FP, DeWitt DP. Introduction to heat transfer. 3rd ed. New York: John Wiley & Sons; 1996. p. 8–
12.
[3] Gardner KA. Efficiency of extended surfaces. Trans ASME 1945;67:621–31.
[4] Mikk I. Convective fin of minimum mass. Int J Heat Mass Transf 1980;23:707–11.
[5] Arpaci V. Conduction heat transfer. MA: Addison-Wesley Publishing Co.; 1966. p. 151–152.
[6] Bejan A, Almogbel M. T-shaped fins. Int J Heat Mass Transf 2000;43:2101–15.
[7] Braunshtein O, Kalman H, Ullmann A. Performance and optimization of composed fin arrays. Heat Transf
Eng 2004;25(4):4–12.
[8] Kakac S, Yener Y. Convective heat transfer. 2nd ed. Boca Raton: CRC Press; 1995. p. 292.

View publication stats