The 2012 International Conference on Green Technology and Sustainable Development (GTSD2012)
A STUDY ON ENHANCING HEAT TRANSFER EFFICIENCY OF LED
LAMPS
Thanhtrung Dang
1
, Vanmanh Nguyen
1
, Nhatlinh Nguyen
1
, Tansa Nguyen
1
, Quocdat Vu
1
, Dinhvu Tran
1
,
Vanchung Ha
1
, Jyh-tong Teng
2
, and Ngoctan Tran
2

1
Ho Chi Minh University of Technical Education, Vietnam
2
Chung Yuan Christian University, Taiwan
ABSTRACT
This paper presented investigations for enhancing heat transfer efficiency of LED Lamp,

contact with the environment which is at
lower temperature. One effective way to
increase the contact area is by attaching a
heat sink to the heat source; in this case, the
heat source is the LED lamp [1]. Heat sinks
are devices which enhance heat dissipation
from a hot surface, usually for the case of a
heat generating component, to a cooler
ambient. Alvin et al. [1] studied thermal
resistance of extruded fin heat sink on LED
lamp. In their study, the most significant
factor affecting the thermal resistance value
between LED and heat sink is the heat sink
mounting pressure, followed by thermal
interface material and heat sink materials.
However, the study did not compare the
influence of heat sink configurations on the
overall thermal resistance for the LED
system. Luo et al. [2] presented temperature
estimation of high-power light emitting
diode street lamp by a multi-chip analytical
solution. In their study, the fin-heat-sink is
still the predominant method used in the
lighting industry due to its highest
reliability and lowest cost. Heat pipe [3, 4]
is becoming a good option for emerging
high power LEDs. Thermal analysis of
high power LED array packaging with
microchannel cooler was done by Yuan et
al. [5]. Weng [6] studied advance thermal

The governing equations in this system
consist of the continuity equation,
momentum equations, and energy equation
[10-12]. The equations can be expressed by
u/t+(u)u=[-pI+(u+(u)T)]+F (1)
u = 0 (2)
CpT/t+(-T)=Qi-CpuT (3)
where

is dynamic viscosity,

is density,
u is velocity field, p is pressure, I is the unit
diagonal matrix, F is body force per unit
volume (F
x
= F
y
= F
z
= 0 N/m
3
), Q
i
is
internal heat generation, T is temperature,
C
p
is specific heat at constant pressure, and


The 2012 International Conference on Green Technology and Sustainable Development (GTSD2012)
2.2 Experimental setup
The experimental system includes a power
supply, a temperature measurement unit, a
fan, and a velocity measurement unit, as
shown in Fig. 2. The heat dissipation
patterns – fin aluminum heat sink - were
tested under different heat transfer modes:
natural convection and forced convection.
The LED with a power supply of 7W was
used in this study. Accuracies and ranges of
testing apparatuses are listed in Table 1.
Table 1. Accuracies and ranges of testing
apparatuses
Testing apparatus
Accuracy
Range
Thermocouples
 0.1 C
0 100 C
Velocity meter
 1 %
0  50 m/s

Figure 2. A photo of the experimental system

Experimental data obtained from the LED
heat sinks were under the constant room
temperature condition of 30 ºC. For the
case with natural convection, air velocity

Fin length of LED, mm
Numerical
Experimental

Figure 4. Comparison between numerical and
experimental results for Model 1 heat sink
configuration
Comparison between numerical and
experimental results for Model 1 heat sink
configuration is shown in Fig. 4. It is
observed that the results obtained from the
numerical simulation are in good
agreement with those obtained from the
experimental data, with the maximum
diffrence of 4.6%. The difference is due to
the error in temperature measurements
caused by temperature sensors which were
soldered at the outer rims of the fins while
The 2012 International Conference on Green Technology and Sustainable Development (GTSD2012)
the numerical results indicated more exact
phenomena taken place in the air
surrounding the heat sink.
b. For Model 2 Heat Sink Configuration
For the same experimental condition above,
with air velocity of 0.1 m/s, the bottom
temperature of heat sink was measured to
be 50.4 ºC. Temperature profiles of heat
sink and air for Model 2 heat sink
configuration are shown in Fig. 5. Fig. 6
shows the comparison between numerical


Figure 7. Temperature profiles of heat sink and
air Model 3 heat sink configuration
0
10
20
30
40
50
0 20 40 60 80
Temperature of the middle fin,
o
C
Fin length of LED, mm
Numerical
Experimental

Figure 8. Comparison between numerical and
experimental results for Model 3 heat sink
configuration
Comparison between numerical and
experimental results for Model 3 heat sink
configuration is shown in Fig. 8. It is also
indicated that the numerical and
experimental results are in good agreement.
From Figs. 3-8, for the natural convection
case, it is observed that the bottom
temperature of heat sink for Model 1 heat
sink configuration was the lowest. It is due
to the fact that Model 1 heat sink

40
50
0 20 40 60 80
Temperature of the middle fin,
o
C
Fin length of LED, mm
Numerical
Experimental

Figure 9. Comparison between numerical and
experimental results for Model 3 heat sink
configuration with forced convection case
4. CONCLUSION
Numerical and experimental studies have
been performed on three LED heat sinks
with different configurations. In natural
convection case, the heat transfer capability
obtained from the heat sink without crevice
was higher than those obtained from the
heat sinks with crevice or crevices. The
heat transfer capability obtained from the
case with forced convection is higher than
that obtained from the case with natural
convection case: at the same room
temperature condition and LED power
supply capacity, the bottom temperature of
LED heat sink is reduced from 49.7 to 38.5
ºC. Furthermore, the results obtained from
the experiments were in good agreement

(IEDMS2011), Taipei, Taiwan, Nov
17-18, 2011, P-C-19, pp. 1-4
[2] X.B. Luo, W. Xiong, T. Cheng, and S.
Liu, Temperature estimation of
high-power light emitting diode street
lamp by a multi-chip analytical
solution, IET Optoelectron, 3, 2009, pp.
225–232
[3] L. Kim, J.H. Choi, S.H. Jang, and M.W.
The 2012 International Conference on Green Technology and Sustainable Development (GTSD2012)
Shin, Thermal analysis of LED array
system with heat pipe, 6th Symposium
of the Korean Society of
Thermophysical Properties, Seoul,
2006, pp. 21–25
[4] Zirong Lin, Shuangfeng Wang, Jiepeng
Huo, Yanxin Hu, Jinjian Chen,
Winston Zhang, and Eton Lee, Heat
transfer characteristics and LED heat
sink application of aluminum plate
oscillating heat pipes, Applied Thermal
Engineering, 31, 2011, pp. 2221-2229
[5] L.L. Yuan, S. Liu, M.X. Chen, and X.B.
Luo, Thermal analysis of high power
LED array packaging with
microchannel cooler, 7th International
Conference on Electronics Packaging
Technology, Shanghai, 2006, pp.
574–577.
[6] C.J. Weng, Advanced thermal

microchannel heat sink, 2012 IEEE
International Symposium on Computer,
Consumer and
Control (IS3C2012), June 4-6, 2012,
Taichung City, Taiwan, pp. 252-257
[12] Thanhtrung Dang and Jyh-tong Teng,
Comparison on the heat transfer and
pressure drop of the microchannel and
minichannel heat exchangers, Heat and
Mass Transfer, 47, 2011, pp.
1311-1322.
[13] J.P. Holman, Heat transfer, Ninth
Edition, McGraw-Hill, New York,
2002
[14] COMSOL Multiphysics version 3.5
(2008) – Documentation.
Contact
Thanhtrung Dang, Ph.D.
Tel: +84913606261
Email: