A New Design for an Existing Air-conditioner Heat Exchanger to Improve the Thermal Performance

Full text

A New Design for an Existing Air-conditioner Heat Exchanger to Improve
the Thermal Performance
Aydın Hacı Dönmez1, Yashar Jabbaribehrooz1, S. Can Meydanlı2, Lütfullah Kuddusi1
1Istanbul Technical University, Turkey
2 Arçelik-LG Klima San. ve Tic. A.Ş., Turkey
Corresponding email: donmezayd@hotmail.com
SUMMARY
The capacity of a heat exchanger varies depending on the convection and conduction over its
fin surfaces. The refrigerant passing through the copper pipes, the heat is distributed to the
fins in a heat exchanger of an air conditioner. This study is related with the surface geometry
of the fins to improve the performance of fin and tube heat exchangers.
First of all, the study starts with patent research. Thereby, general information is obtained
about fin and tube heat exchangers and fin forms. After that, the study is supported by a
literature research. Then, solution methods of rival products are examined by benchmarking.
Obtained fin structures and different fin forms are subjected to CFD analysis and comparisons
are made. Finally, design workouts are done by the help of the data obtained from past
studies. The software FLUENT is used in all CFD studies and design workouts.
INTRODUCTION
Performance improvement in every system used, is essential for their efficiencies due to the
importance of the energy. The case, improving the thermal performance of the fin and tube
heat exchangers, is widely used in HVAC applications. In those kinds of fin and tube heat
exchangers, most of the thermal resistance is composed of convective resistance at the airside
(over 85%) [1]. In order to increase the thermal performance, the studies are mainly focused
on issues about convection. Ameliorating the fin profiles and structures is another way to
increase the air side heat transfer [2]. Instead of using flat fins; corrugated, slit and louvered
fins are extensively used for heat transfer enhancement. The air side thermal performance can
reach up two or even three times of that 1960’s flat finned heat exchangers, by doing these
ameliorations [3]. The aim of this study is to analyze and compare the effect of fin types and
structures to the heat transfer so that, more efficient heat exchanger will be obtained by
designing more appropriate fin profile.
This study aims to improve the thermal performance of an existing indoor unit heat exchanger
of a commercial company.
LITERATURE RESEARCH
Literature research started with patent research and continued with article and benchmark
studies. In the literature research it is seen that, in order to increase the heat transfer per unit
area in the heat exchanger, different geometries are modeled on the fins. Briefly, in literature

researches it can be seen that the main aim of modeling this geometries is to implement one of
the below:
1- Passing to the turbulent region as soon as possible: It is obvious that, total heat transfer is
higher in turbulent region, comparatively to the laminar region. The more the Reynolds
number (Re) increases, the more heat transfer rate increases due to the increase in the Nusselt
number (Nu).
 =  
 [4]
 =    (0m1, 0n1, 0a1) [4]
 =  
 [4]
ℎ =  
 ( 
)  [4]
2- Preventing hydraulic boundary layer growth: It is seen from Figure 1 that, heat convection
coefficient is considerably high at the entrance of the laminar region. By taking into account
this issue, in order to have a better heat transfer, fin geometries had to be designed either to
make the flow turbulent as soon as possible or to make new laminar regions to benefit from
their high heat convection coefficients.
Figure 1. Exchange of thermal boundary layer and heat convection coefficient throughout the
flow
3- Minimizing the pressure drop: Geometries on fins can be considered as disturbances
against to the air flow, which causes pressure drop. This causes using more powerful fans
which consume more electricity and engender noise problem. Excessive pressure drop should
be avoided while enhancing the heat transfer.
4- Extending the pathlines of the flow: Extension of the pathlines (Figure 2a) results in
increase in the contact area between air and fins, and heat transfer increases.
5- Reducing the dead zones behind the refrigerant pipes: Reducing the dead zone behind the
refrigerant pipes (Figure 2b) causes using fin area more efficiently, resulting in the increase of
heat transfer between air and fin surface.

Figure 2. a) Extending pathlines, b) Dead zones behind the refrigerant pipes.
SOLUTION METHOD:
Modeling the fin structures, generating mesh and CFD analysis are done in turn. In CFD
workouts, fins are simulated in a 2-D tunnel, which has inlet, outlet and two moving walls (up
and down). The length of the tunnel is one third width of the fin longer at inlet and two third
width of the fin longer at outlet. Moreover, the height of the tunnel is one and a half times
longer from the top and bottom of the model. In order to examine the effects of the fins on
each other, three fins are simulated in the tunnel and the data taken from the middle fin is used
for comparisons. Therefore, meshes are concentrated to the center of the model. K- Epsilon
solution method is used in CFD analysis. While generating mesh, convergence of the
solutions is taken into account. It is seen that the results remained unchanged when the mesh
size decreased. In these whole studies at inlet, velocity of air is 4 m/s, temperature of air is
32°C and temperature of the fins are taken 4°C. However, outflow boundary condition is used
at the outlet. Finally; velocity, temperature and pressure fields are obtained by solving the
continuity, momentum and energy equations simultaneously.
CFD
First of all, the flow over the flat fin is simulated, velocity and temperature distribution is
obtained around the model and heat transfer to the fin is calculated. Then, a reference heat
transfer value is defined, dividing heat transfer by fin width. The comparisons are made by
taking into account this unit value. In CFD workouts, eleven units of heat exchanger (five
indoor, six outdoor) of six commercial companies are used. Moreover, six different European
and American patents are subjected to CFD analysis.
Despite the fact that, there have been multitudinous types of fin profiles; wavy, slit and
louvered types are widely used by commercial companies due to their ease of manufacturing.
Generally, thermal performance of the slit and louvered fins are all the better then the wavy
type. As a result, the study is focused on those ones.
Louvered Fins
The first study in louvered fins are about to find the optimum louver angle. As it is mentioned
before, the model consists of overlapping three fins, in order to see the hydrodynamic and
thermal disturbance effects of the ones at the top and bottom.
Flow

Figure 3. Louvered fin with variable louver angle. a) Top view, b) A-A cross-section.
The louver angle (α) started from 10° and increased 5° in every model. The heat transfer rate
led to decrease after 25° slightly but, an instantaneous decrease occurred at 40°. The relation
between heat flux and louver angle is shown in the figure below:
Figure 4. Tri-fin configuration heat flux - louver angle relationship.
Another parameter influencing the heat transfer rate is louver width. This time, by changing
the louver width (L) the variation of heat transfer is observed. Considering the first step of the
study on louvered fins, louver angle is taken as the optimum value, which is 25°.
Figure 5. Louvered fin with variable louver width. a) Top view, b) A-A cross-section.
Φd
Louver Angle [°]
Heat Flux
Φd

Similarly, the heat transfer increased at first, reached a maximum value and started to
decrease. As the width of the louver increased, the area increased so heat transfer is increased.
On the other hand, after a certain value the flow is blocked by the louver so, heat transfer is
decreased.
Figure 6. Tri-fin configuration heat flux - louver width relationship.
Slit Fins
The studies about slits are done in three steps: slit height, slit width, and dividing the slits. As
in the case in louvers, three fins are simulated in the tunnel. Slit height (h) started from 0.2
mm and increased 0.1 mm in every model. Maximum heat transfer rate occurred at 0.6 mm
slit height, which is the middle of two fins. The relation between heat flux and slit height is
shown in the Figure 8:
Figure 7. Slit fin. a) Top view, b) A-A cross-section.
Louver width [mm]
Heat Flux [W/m^2]

Figure 8. Tri-fin configuration heat flux - slit height relationship.
Another parameter influencing the heat transfer rate is slits width. This time, impact of slits
width on heat transfer has been studied. Slits width (a) has been changed from 0,7mm to
2,4mm. Unlike the louvers in slits, slits width do not block the air flow. Thus, total heat
transfer can be increased by increasing the slit width. The results obtained are shown in the
chart below:
Figure 9. Tri-fin configuration heat flux - slit width relationship.
In the final stage, a slit with a constant width has been divided in order to form new boundary
layers and increase the heat transfer by increasing the heat convection coefficient. In these
studies, the heat transfer in a slit with 2.4mm width, 2 slits with 1.2mm width, 3 slits with
0.8mm width, 4 slits with 0.6mm width and 6 slits with 0.4mm width had compared. Thus,
the total slit length of 2.4 mm is kept constant. Despite the total slit width stayed constant,
heat transfer is increased with the increase or the slit number. The data obtained were poured
into the following chart:
Slit Height [mm]
Heat Flux [W/m^2]
Slit width [mm]
Heat Flux [W/m^2]

Figure 10. Tri-fin configuration heat flux - slit number relationship.
RESULTS
Design workouts are done in the final step of the study. The heat transfer of the flat fin per
width is taken unity and existing fin profiles chosen as reference value to be improved. The
new fin design (Figure 11) is the combination of slit and louvered fins.
Figure 11. The new fin design.
Heat transfer performances of the existing and new designed fin are obtained from CFD
studies and showed in the table below:
Table 1. Heat transfer performances of the existing and new designed fin.
FIN TYPE
HEAT FLUX
FIN
WIDTH
FLUX/WIDTH
HEAT TRANSFER
FACTOR [W/m^2] [mm]
Flat fin 114,64 25,4 4,51 1
Existing fin 1 130,62 19,045 6,86 1,52
Existing fin 2 114,06 12,7 8,98 1,98
Existing fin 3 96,52 18,19 5,31 1,17
New Design 111,32 11,9 9,35 2,07
Slit Number
Heat Flux [W/m^2]

REFERENCES
1. T’Joen C., Steeman H.-J., Willockx A., and De Paepe M., 2005: Determination of heat transfer
and friction characteristics of an adapted inclined louvered fin. Experimental Thermal and Fluid
Science, 30, 319-327. Retrieved February 10, 2009, from http://0-
www.sciencedirect.com.divit.library.itu.edu.tr
2. Wongwises S., and Chokeman Y., 2004: Effect of fin pitch and number of tube rows on the air
side performance of herringbone wavy fin and tube heat exchangers. Energy Conversion and
Management, 46, 2216-2231. Retrieved February 10, 2009, from http://0-
www.sciencedirect.com.divit.library.itu.edu.tr
3. Jabardo J. M. S., Filho J. R. B. Z., and Salamanca A., 2006: Experimental study of the air side
performance of louver and wave fin-and-tube coils. Experimental Thermal and Fluid Science,
30, 621-631. Retrieved February 10, 2009, from http://0-
www.sciencedirect.com.divit.library.itu.edu.tr
4. Incropera F. P., and DeWitt D. P., 2007: Fundamentals of Heat and Mass Transfer, 4th edition,
John Wiley & Sons
5. Baek Y., Kim Y., 1999. Samsung Electronics Co. Ltd., United States Patent, No: 5947194,
Date: 07.09.1999
6. Kim Y., 1998. Samsung Electronics Co. Ltd., United States Patent, No: 5853047, Date:
29.12.1998
7. Park I., Choi H., Kim W., 2003. Mando Climate Control Co., United States Patent, No:
20030150601, Date: 14.08.2003
8. Kim Y., 1999. Samsung Electronics Co. Ltd., United States Patent, No: 5887649, Date:
30.03.1999
9. B., Kim Y., 1999. Samsung Electronics Co. Ltd., United States Patent, No: 5927392, Date:
27.07.1999
10. Fujino H., Kim H., Kamada T., Kasai K., 2008. Daikin Industries Ltd., European Patent, No:
1906129, Date: 02.04.2008
11. Abrahamian D., Conroy K., 2007. Cameron International Co., European Patent, No: 1977180,
Date: 09.08.2007
12. Bemisderfer C., 2004. York international Co., United States Patent,No: 6786274, Date:
07.09.2004
13. Yu S., 2008. Samsung Electronics Co. Ltd., United States Patent, No: 5553663, Date:
10.09.1996
14. Maziers E., 2008. LGL France, United States Patent, No: 2008164013, Date: 10.07.2008
15. Jun H., 1999. Samsung Electronics Co. Ltd., United States Patent, No: 5896920, Date:
27.04.1999
16. Komori K., 2008. Matsushita Electric Industrial Co. Ltd., European Patent, No: 1985958, Date:
29.10.2008
17. Hu Z., Le Guayer P., Fang L., 2002. VALEO INC., World Intellectual Property Organization,
No: 03062731, Date: 31.07.2003