Department of Mechanical Engineering, National Chiao Tung University, Hsinchu 300, Taiwan
Green Energy & Environment Research Laboratories, Industrial Technology Research Institute, Hsinchu 310, Taiwan
a r t i c l e
i n f o
Article history:
Received 17 April 2014
Received in revised form 14 September 2014
Accepted 15 September 2014
Available online 2 October 2014
Keywords:
Fin-and-tube heat exchanger
Louver n
Vortex generator
a b s t r a c t
In this study, a comparative study of the airside performance of n-and-tube heat exchangers having
plain, louver, and semi-dimple vortex generator (VG) are made. A total of eighteen samples are made
and tested with the corresponding n pitch (Fp) being 1.6 mm and 2.0 mm and the number of tube
row (N) are 1, 2 and 4, respectively. Test results indicate that the heat transfer coefcient for N = 1 for
louver n geometry with a smaller n pitch of 1.6 mm is higher than that of semi-dimple VG and plain
n geometry. For N = 1 with a larger n pitch of 2.0 mm, the semi-dimple VG is marginally higher than
that of louver n geometry when the frontal velocity is lower than 2 m s 1. However, the trend is
reversed where the heat transfer coefcient for louver n outperforms that of semi-dimple VG when
the velocity is increased further. For the airside performance for N = 2 or N = 4, the heat transfer coefcients for louver n geometry is about 215% higher than those of semi-dimple geometry. The difference
is increased with the rising velocity and the results prevail for both n pitches. However, the difference is
smaller at a larger n pitch due to the comparatively effectively swirled motion of the semi-dimple VG.
The effect of the number of tube row on the heat transfer coefcients is negligible for louver n geometry
and is also rather small for semi-dimple VG. For the plain n geometry, the effect of tube row is also small
when N > 1. The heat transfer performance for N = 1 is different from N = 2 or N = 4 due to its inline
conguration.
2014 Elsevier Ltd. All rights reserved.
1. Introduction
Extended surfaces or ns are employed in heat exchangers for
effectively improving the overall heat transfer performance. This
is especially imperative for air-cooled heat exchanger since the
dominant thermal resistance is usually on the air side. The surfaces
can be in the form of continuous surfaces (e.g. plain, wavy) or
interrupted (louver, slit, offset, and the like). Several review articles
by Wang [12] had reported the patents of enhanced surfaces relevant to the n-and-tube heat exchangers. He reported that 90%
out of the surveyed patents were related to the interrupted surfaces. However, the interrupted surfaces showed appreciable pressure drops in association with heat transfer performance. In this
connection, one of the recent designs is via introduction of the
so-called vortex generator through which the heat transfer
Corresponding author at: EE474, 1001 University Rd., National Chiao Tung
University, Taiwan. Tel.: +886 3 5712121x55105; fax: +886 3 5720634.
E-mail address: ccwang@mail.nctu.edu.tw (C.-C. Wang).
http://dx.doi.org/10.1016/j.ijheatmasstransfer.2014.09.030
0017-9310/ 2014 Elsevier Ltd. All rights reserved.
282
C.-C. Wang et al. / International Journal of Heat and Mass Transfer 80 (2015) 281287
Nomenclature
A
Dc
Fp
h
hi
N
NTU
Pl
surface area, m2
collar diameter, m
n pitch, m
heat transfer coefcient, W m-2 K-1
inside heat transfer coefcient, W m 2 K
number of tube row
number of transfer unit, dimensionless
longitudinal tube pitch, m
Pt
DP
U
Vfr
e
Df
2. Experimental setup
As tabulated in Table 1, a total of eighteen sample coils which
includes plain, louver, and semi-dimple VG. The detailed dimension and the photo of the semi-dimple VG is schematically shown
in Fig. 1(a) whereas the louver n is depicted in Fig. 1(b). Notice
that the n thickness (Df), collar diameter (Dc), transverse pitch
(Pt), and longitudinal pitch (Pl) for all the test samples are
0.11 mm, 7.5 mm, 21 mm, and 18.2 mm, respectively. The corresponding n pitch (Fp) ranges from 1.62.0 mm and the number
of tube row (N) spans from 1 to 4 as shown in Table 1. Detailed
construction of the circuitry arrangement is identical to those by
Wang et al. [27]. The experiments are conducted in an open wind
tunnel as shown in Fig. 2. The ambient air ow was forced across
the test section by means of a 5.6 kW centrifugal fan with an inverter. To avoid and minimize the effect of ow maldistribution in the
experiments, an air straightener-equalizer and a mixer were provided. The inlet and the exit temperatures across the sample coil
were measured by two T-type thermocouple meshes. The inlet
measuring mesh consists of twelve thermocouples while the outlet
mesh contains 36 thermocouples. These data signals were individually recorded and then averaged. During the isothermal test, the
variance of these thermocouples was within 0.2 C. In addition,
all the thermocouples were pre-calibrated by a quartz thermometer with 0.01 C precision.
The pressure drop of the test coil was detected by a precision
differential pressure transducer, reading to 0.1 Pa. The air ow
measuring station was a multiple nozzle code tester based on the
ASHRAE 41.2 standard [28]. The working medium in the tube side
was hot water. The inlet water temperature was controlled by a
thermostat reservoir having an adjustable capacity up to 25 kW.
Table 1
Detailed geometric parameters of the test samples.
No.
Fp (mm)
N, row
Geometry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
1.6
1.6
1.6
2.0
2.0
2.0
1.6
1.6
1.6
2.0
2.0
2.0
1.6
1.6
1.6
2.0
2.0
2.0
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
Plain
VG
Louver
Plain
VG
Louver
Plain
VG
Louver
Plain
VG
Louver
Plain
VG
Louver
Plain
VG
Louver
283
C.-C. Wang et al. / International Journal of Heat and Mass Transfer 80 (2015) 281287
P H.X
Air Flow
3-phase, 220V
P SAT
P NOZZLE
T
Hot Water
Thermostat
Reservoir
PID
GPIB
Hybrid
Recorder
SCR
PC
Computer
1 Air inlet
2 Straightener
3 Developing section
4 Pressure tap(inlet)
5 T/C inlet temperature measuring station
6 Test unit(Heat Exchanger)
7 T/C outlet temperature measuring station
8 Pressure tap(outlet)
9 Mixer
10 Settling Devices (Flow Straightener)
15 Centrifugal fan
16 Air discharge
17 Water pump
18 Water flow meter
19 RTD inlet temperature of water side
20 RTD outlet temperature of water side
21 Temperature sensor
22 Inverter
C.-C. Wang et al. / International Journal of Heat and Mass Transfer 80 (2015) 281287
140
120
Plain h
VG
Louver
Plain
P
VG
Louver
140
120
100
100
80
80
P (Pa)
Both the inlet and outlet temperatures were measured by two precalibrated RTDs (Resistance temperature device, Pt-100). Their
accuracy was within 0.05 C. The water volumetric ow rate is
detected by a magnetic ow meter with 0.002 L/s resolution.
All the data signals are collected and converted by a data acquisition system (a hybrid recorder). The data acquisition system then
transmitted the converted signals through GPIB interface to the
host computer for further operation. During the experiments, the
water inlet temperature was held constant at 60.0 0.2 C, and
the tube side Reynolds number was approximately 38,000. Frontal
velocities of inlet air ranged from 1 to 4 m/s. The energy un-balance between air side and tube side was within 2%. The water side
resistance (evaluated as 1/hiAi) was less than 10% of the overall
resistance in all cases. The test n-and-tube heat exchangers are
tension wrapped having an L type n collar. Thermal contact
conductance provided by the manufacturers ranged from 11,000
to 16,000 W m2 K 1.
To obtain the heat transfer coefcient and pressure loss characteristics of the test coil from the experimental data, the e-NTU
method is applied to determine the UA product in the analysis.
Detailed reduction can be seen from previous work (e.g. Wang
et al. [27]).
h (W/m2K)
284
60
60
40
40
20
20
ufrontal (m/s)
(a) Fp = 1.6 mm
100
120
Plain h
VG
Louver
Plain
P
VG
Louver
80
100
80
60
P (Pa)
h (W/m2K)
60
40
40
20
20
0
0
ufrontal (m/s)
(b) Fp = 2.0 mm
Fig. 3. Test results for N = 1.
285
C.-C. Wang et al. / International Journal of Heat and Mass Transfer 80 (2015) 281287
Plain h
VG
Louver
Plain
P
VG
Louver
120
200
100
150
60
100
P (Pa)
h (W/m2K)
80
40
50
20
ufrontal (m/s)
(a) Fp = 1.6 mm
Plain h
VG
Louver
Plain
P
VG
Louver
120
200
100
150
60
100
P (Pa)
80
h (W/m2K)
Fp = 2.0 mm at a frontal velocity of 5 m s 1. The results are somewhat expected when the frontal velocity is low. This is because
the entering air ow for the highly interrupted surface may behave
as a duct ow where the main ow simply bypass the louver n
(Webb and Trauger [32]) without sufcient mixing, leading to a
similar heat transfer performance amid plain, louver, and semidimple VG surface. With a rise of frontal velocity, a better mixing
resulting from the louver/semi-dimple VG geometry prevails, and
accordingly a better heat transfer coefcient develops when compared to the plain n geometry. Notice that the semi-dimple VG
geometry shows a comparative performance against louver n
geometry for Fp = 2.0 mm when Vfr < 2 m s 1. For Fp = 1.6 mm, the
louver n outperforms that of semi-dimple VG geometry throughout the test range. The results are associated with the basic
enhanced natures of these two n geometry. The highly interrupted surface like louver n employs boundary layer restating
and mixing mechanisms for heat transfer augmentation while
semi-dimple VG mainly entails swirled motion alongside the n.
The engendered swirled motion is somewhat limited or conned
when the n spacing is reduced, leading to a lower heat transfer
performance. For a larger n pitch like Fp = 2.0 mm and Vfr < 2
m s 1, the generated longitudinally swirled motion is quite effective and shows a comparable performance against the louver n
geometry. As a result, an almost identical heat transfer performance for these two geometries emerges. However, with the rise
of frontal velocity, a much better mixing phenomenon caused by
the louver n geometry takes control and gives rise to a better heat
transfer performance for the louver n geometry as compared to
that of semi-dimple VG. The test results of pressure drop also
reect the heat transfer augmentation mechanisms for the louver
and semi-dimple VG, i.e. mixing, boundary layer restarting, and
swirled motion. From Fig. 3(a) where the swirled ow is less effective due to connement of n pitch, the pressure drops of the louver n geometry is appreciably higher than that of semi-dimple
VG. On the other hand, for Vfr < 2 m s 1 and Fp = 2.0 mm where
semi-dimple may process effective swirled ow, the difference in
pressure drops amid louver and semi-dimple VG is negligible. With
a rise of the frontal velocity, the mixing by louver n becomes
more effective, and the measured pressure drop is moderately
higher than that of the semi-dimple geometry. However, it should
be mentioned that the difference in pressure drop is comparatively
small for Fp = 2.0 mm as compared to that of Fp = 1.6 mm.
Test results for N = 2 and N = 4 for the three n patterns are
shown in Figs. 4 and 5. As shown in the gures, a further increase
of the number of tube row (N = 2 or N = 4), the airow within the
plain n-and-tube heat exchanger may become periodic developed, and results in the vortex-controlled regime. Consequently,
the effect of n pitch on heat transfer performance is much less
profound for N = 2 and it virtually vanishes for N = 4. The previous
studies [25,26] for plain n geometry also showed that a higher
velocity with a larger number of tube row may result in the occurrence of vortex along the ns, therefore the effect of n pitch on
heat transfer coefcient would be negligible. For the highly interrupted surface like louver or semi-dimple VG n geometry, the
results are analogous to those for N = 1. With a smaller n pitch
(Fp = 1.6 mm), the louver n outperforms that of the semi-dimple
VG geometry in the order of 210%, yet the difference is moderately increased with the rise of frontal velocity. However, the difference is less pronounced as that of N = 1. This is because of the
presence of staggered tube row arrangement that brings about a
better ow mixing mechanism. This can be made clear in
Fig. 5(a) for N = 4 and Fp = 1.6 mm, the difference between louver
and semi-dimple VG is further reduced due to the presence of multiple staggered tube rows. Notice that for N = 1, all the test samples
are regarded as inline arrangement. For a larger n pitch of 2.0 mm
as depicted in Fig. 4(b) (N = 2) and Fig. 5(b) (N = 4), one can see that
40
50
20
ufrontal (m/s)
(b) Fp = 2.0 mm
Fig. 4. Test results for N = 2.
the heat transfer coefcients for the semi-dimple VG and the louver n are comparable. This is because additional mixing augmentation caused by the staggered tube row is imposed upon the
swirled motion. The combined effects result in a similar (or only
marginally lower) heat transfer coefcient. From Figs. 35, one
can also examine the effect of the number of tube row on the heat
transfer coefcient. For highly interrupted surfaces at a same n
pitch, like louver n, the heat transfer coefcient is almost independent of tube row. The results are analogous to previous ndings
(Wang et al. [33,34]). For the present semi-dimple VG surface, the
286
C.-C. Wang et al. / International Journal of Heat and Mass Transfer 80 (2015) 281287
400
Plain h
VG
Louver
Plain
P
VG
Louver
120
100
300
200
60
P (Pa)
h (W/m2K)
80
40
100
20
ufrontal (m/s)
(a) Fp = 1.6 mm
Plain h
VG
Louver
Plain
P
VG
Louver
120
400
100
Conict of interest
300
None declared.
200
60
P (Pa)
80
h (W/m2K)
40
100
Acknowledgements
The authors would like to express gratitude for the nancial
support from Ministry of Science and Technology and Energy
R&D foundation funding from the Bureau of Energy of the Ministry
of Economic, Taiwan. Grants from the Ministry of Science and
Technology, Taiwan under contract 103-3113-E-009-002 is also
highly appreciated.
20
References
0
ufrontal (m/s)
(b) Fp = 2.0 mm
Fig. 5. Test results for N = 4.
[1] C.C. Wang, Technology review a survey of recent patents of n-and-tube heat
exchangers, J. Enhanced Heat Transfer 7 (2000) 333345.
[2] C.C. Wang, A survey of recent patents of n-and-tube heat exchangers from
20012009, Int. J. Air-Conditioning Refrig. 18 (2010) 113.
[3] A.M. Jacobi, R.K. Shah, Heat transfer surfaces enhancement through the use of
longitudinal vortices: a review of recent progress, Exp. Therm. Fluid Sci. 11
(1995) 295309.
[4] ESDU 93024, Engineering Science Data Unit, Vortex Generators for Control of
Shock-Induced Separation Part 1: Introduction and Aerodynamics, 1993.
[5] F.J. Edwards, G.J.R. Alker, The improvement of forced convection surface heat
transfer using surfaces protrusions in the form of (A) cubes and (B) vortex
Generators, Proc. 5th Int. Heat Transfer Conf. 2 (1974) 244248.
[6] S. Tiggelbeck, N.K. Mitra, M. Fiebig, Comparison of wing-type vortex generators
for heat transfer enhancement in channel ows, ASME J. Heat Transfer 166
(1994) 880885.
[7] S. Tiggelbeck, N.K. Mitra, M. Fiebig, Experimental investigation of heat transfer
enhancement and ow losses in a channel with double rows of longitudinal
vortex generators, Int. J. Heat Mass Transfer 36 (1993) 23272337.
[8] G. Biswas, N.K. Mitra, M. Fiebig, Heat transfer enhancement in n-tube heat
exchangers by winglet type vortex generators, Int. J. Heat Mass Transfer 37
(1994) 283291.
[9] M. Fiebig, Vortices generators and heat transfer, Trans. IChemE 76 (1998) 108
123.
[10] M. Fiebig, A. Valencia, N.K. Mitra, Wing-type vortex generators for n-andtube heat exchangers, Exp. Therm. Fluid 7 (1993) 287295.
C.-C. Wang et al. / International Journal of Heat and Mass Transfer 80 (2015) 281287
[11] Y.L. He, H. Han, W.Q. Tao, Y.W. Zhang, Numerical study of heat-transfer
enhancement by punched winglet-type vortex generator arrays in n-andtube heat exchangers, Int. J. Heat Mass Transfer 55 (2012) 54495458.
[12] J.F. Fan, Y.L. He, W.Q. Tao, Application of combined enhanced techniques for
design of highly efcient air heat transfer surface, Heat Transfer Eng. 33 (1)
(2012) 5262.
[13] J.M. Wu, W.Q. Tao, Impact of delta winglet vortex generators on the
performance of a novel n-tube surfaces with two rows of tubes in different
diameters, Energy Convers. Manage. 52 (2011) 28952901.
[14] J. Li, S. Wang, J. Chen, Y.G. Lei, Numerical study on a slit n-and-tube heat
exchanger with longitudinal vortex generators, Int. J. Heat Mass Transfer 54
(2011) 17431751.
[15] P. Chu, Y.L. He, W.Q. Tao, Three-dimensional numerical study of ow and heat
transfer enhancement using vortex generators in n-and-tube heat
exchangers, J. Heat Transfer 131 (2009) 091903-1091903-9.
[16] J.Y. Jang, L.F. Hsu, J.S. Leu, Optimization of the span angle and location of vortex
generators in a plate-n and tube heat exchanger, Int. J. Heat Mass Transfer 67
(2013) 432444.
[17] W.L. Hu, K.W. Song, L.M. Chang, L.B. Wang, Secondary ow intensity
determines Nusselt number on the n surfaces of circle tube bank n heat
exchanger, Int. J. Heat Mass Transfer 62 (2013) 620631. Y.G., S.L.
[18] M.J. Li, W.J. Zhou, J.F. Zhang, J.F. Fan, Y.L. He, W.Q. Tao, Heat transfer and
pressure performance of a plain n with radiantly arranged winglets around
each tube in n-and-tube heat transfer surface, Int. J. Heat Mass Transfer 70
(2014) 734744.
[19] K.W. Song, L.B. Wang, The effectiveness of secondary ow produced by vortex
generators mounted on both surfaces of the n to enhance heat transfer in a
at tube bank n heat exchanger, J. Heat Transfer 135 (2013) 041902-1
041902-11.
[20] W. Hu, M. Su, L. Wang, Q. Zhang, L. Chang, S. Liu, L.B. Wang, The optimum n
spacing of circular tube bank n heat exchanger with vortex generators, Heat
Mass Transfer 49 (2013) 2711285.
[21] H. Huisseune, C. TJoen, P.D. Jaeger, B. Ameel, S.D. Schampheleire, M.D. Paepe,
Performance enhancement of a louvered n heat exchanger by using delta
winglet vortex generators, Int. J. heat Mass Transfer 56 (2013) 475487.
[22] H. Huisseune, C. TJoen, P.D. Jaeger, B. Ameel, J. Demuynck, M.D. Paepe,
Numerical study of ow deection and horseshoe vortices in a louvered n
round tube heat exchanger, J. Heat Transfer 134 (2012) 091801-1091801-11.
287