Articulo

Applied Energy 135 (2014) 748777
Contents lists available at ScienceDirect
Applied Energy
journal homepage: www.elsevier.com/locate/apenergy
Recent development and application of several high-efciency surface

heat exchangers for energy conversion and utilization
Qiuwang Wang , Min Zeng, Ting Ma, Xueping Du, Jianfeng Yang
Key Laboratory of Thermo-Fluid Science and Engineering, MOE, Xian Jiaotong University, Xian, Shaanxi 710049, PR China
h i g h l i g h t s
Recent research of three kinds of surface heat exchangers are reviewed.
The STHXs with novel combined helical bafes have been reviewed.
The primary surface heat exchangers with compact surfaces have been reviewed.
The direct and indirect air-cooled heat exchangers have been reviewed.
a r t i c l e
i n f o
Article history:
Received 28 January 2014
Received in revised form 30 April 2014
Accepted 3 May 2014
Available online 23 May 2014
Presented at the 8th Conference on
Sustainable Development of Energy, Water
and Environment Systems, 2013.09.22-27,
Dubrovnik, Croatia (Original Paper Title:
High-Efciency Surface Heat Exchangers for
Energy Conversion and Recovery Systems,
by Qiuwang Wang, Min Zeng, Ting Ma,
Paper No. SDEWES2013.00689).
a b s t r a c t
In the present study, the recent research of three kinds of surface heat exchangers, i.e., shell-and-tube
heat exchangers with helical bafes, air-cooled heat exchangers used in large air-cooled systems, and primary surface heat exchangers are reviewed. They are used in the energy conversion and utilization for
liquid to liquid, gas to gas and liquid to gas heat exchange, respectively. It can be concluded that the helical bafed shell-and-tube heat exchangers (STHXs) should be used to replace the conventional segmental
bafed STHXs in industries, despite there are a lot of research work have to be done, especially on the
novel combined helical bafes. The primary surface gas to gas heat exchangers are developing towards
to the more complex 3D CC primary surfaces, such as the double-wave CC primary surface, offset-bubble
primary surface and 3D anti-phase secondary corrugation. The whole performance for the air-cooled heat
exchangers in the air cooling system and the multi-objectives optimization for air-cooled heat exchangers should be paid more attention, considering the heat transfer, pumper power, space usage and other
economic inuence factors.
2014 Elsevier Ltd. All rights reserved.
Keywords:
Helical bafed STHX
Primary surface gasgas heat exchanger
Air-cooled heat exchanger
Enhanced heat transfer
Energy conversion and utilization
1. Introduction
Surface heat exchangers are devices that transfer the heat from
the hot uid to the cold uid while keeping them from mixing with
each other. Due to the energy crisis, the importance of surface heat
exchangers has increased immensely in the energy conservation,
conversion, recovery, and new energy utilization. The other reason
to the importance of surface heat exchanger is the thermal pollution, air pollution, water pollution and waste disposal, which have
been concerned by more and more countries. The surface heat
Corresponding author. Tel./fax: +86 (29)82665539.

E-mail address: [email protected] (Q. Wang).
http://dx.doi.org/10.1016/j.apenergy.2014.05.004
0306-2619/ 2014 Elsevier Ltd. All rights reserved.
exchangers are widely used in many energy conversion and

utilization industries, such as the process, power, transportation,
air-conditioning and refrigeration, cryogenic, heat recovery, alternate fuels, manufacturing industries.
Shell-and-tube heat exchanger is generally built of a bundle of
round tubes mounted in a cylindrical shell with the tube axis parallel to that of the shell. One uid ows inside the tubes, the other
ows across and along the tubes. More than 40% of the market
share in process and petrochemical industry heat exchangers is
held by the traditional shell-and-tube heat exchanger, for the following reasons: its versatility for handling a wide range of operating conditions with a variety of materials, design experience of
about 100 years, proven design methods, and design practice with
codes and standards [1].
Q. Wang et al. / Applied Energy 135 (2014) 748777
749
Nomenclature
Aw
C
Dh
Eu
h
Hi
Luc
N
Nu
P
Re
vfr
U
amplitude of waviness (mm)

compactness (m2 m3)
hydraulic diameter (mm)
dimensionless pressure drop
heat transfer coefcients (W m2 K1)
internal height (mm)
length of unitary cell (mm)
number of longitudinal tube rows
Nusselt number
pitch (mm)
Reynolds number
air frontal velocity (m s1)
wind speed (m s1)
Greek symbols
a
apex angle of air-cooled radiator ()
h
air inlet angles, inclination angle ()
Dp
pressure drop (Pa)
Abbreviations
ACC
air-cooled condenser
ACHX
air-cooled heat exchanger
ACSC
air-cooled steam condenser
The gasgas heat exchangers are necessary in many waste heat

recovery systems, such as the microturbine recuperated cycle
system, the fossil fuel power system and the fresh air ventilation
system. The primary surface heat exchanger is a kind of highefcient plate heat exchangers without any secondary heat transfer surfaces, which is different from the plate-n heat exchanger.
Because there is no n efciency and complex structures are easy
to be manufactured, the heat transfer performance is very high and
the weight is very light. Due to these advantages, the primary surface heat exchanger is widely used in the gasgas heat transfer
applications. The most popular primary surface sheets include
cross corrugated (CC), cross wavy (CW), cross undulated (CU) and
double notched (DN) surfaces. In the microturbine system, the efciency of microturbine with a regenerative cycle can approach 30%
or higher, while that with a simple cycle is less than 20% [2].
In an air-cooled heat exchanger, or air cooler, heat is transferred
from the process uid to the cooling airstream via extended surfaces or nned tubes. Air-cooled heat exchangers are found in
the electronics industry, vehicles, air conditioning, and refrigeration plants as well as chemical and process plants where uids at
temperatures of approximately 60 C or higher are to be cooled
[3]. The use of air-cooled or dry-cooling systems in industry or in
power plants is often justied where cooling water is not available
or is very expensive. Movement of the cooling air is achieved by
mechanical means, fans, or buoyancy effects, e.g., natural draft
dry-cooling towers. The performance of air-cooled heat exchangers
is determined by the types and shapes of the n surface, the dry
bulb temperature of the air, and so on. In large air-cooled systems,
heat exchanger bundles are often arranged in the form of A-frames,
deltas, or V-arrays to conserve space. The obliquity of the incoming
ow to the bundle front face and the jet formed at the downstream
face may be regarded as two interrelated effects, which causes a
considerable loss, contributes to overall aerodynamic losses, and
will affect the performance of the cooling system.
The objective of this paper is to provide the state-of-the-art
development of the three kinds of surface heat exchangers, e.g.,
shell-and-tube heat exchangers with helical bafes, primary surface heat exchangers and air-cooled heat exchangers used in large
CC
CU
CW
CH-STHX
cross corrugated
corrugated undulated
cross wavy
shell-and-tube heat exchanger with continuous
helical bafes
CH&CH-CSSP-STHX continuouscontinuous
helical
bafed
combined single shell-pass shell-and-tube heat
exchanger
CMSP-STHX combined multiple shell-pass shell-and-tube heat
exchanger with continuous helical bafes
CPSSP-STHX combined parallel single shell-pass shell-and-tube
heat exchanger with continuous helical bafes
CSSP-STHX combined single shell-pass shell-and-tube heat
exchanger with continuous helical bafes
DCH-STHX shell-and-tube heat exchanger with discontinuous
helical bafes
DCH&CH-CSSP-STHX discontinuouscontinuous helical bafed
combined single shell-pass shell-and-tube heat
exchanger
FTHX
nned tube heat exchanger
SG-STHX
shell-and-tube heat exchanger with segmental
bafes
air-cooled systems. What is more, the typical applications now or

in the future will be presented.
2. Shell-and-tube heat exchangers with helical bafes
With the improvement of industrialization, the effective utilization of energy should be increased accordingly. The shell-and-tube
heat exchanger with segmental bafes (SG-STHX) is still one of the
most important heat equipment. However, in past decades, the
drawbacks in principle of the SG-STHX have been increasingly
obvious. One of the new technologies, named the shell-and-tube
heat exchanger with helical bafes, was proposed to improve the
comprehensive performance of the commonly used SG-STHX.
And it has been continuously developed until now, from the STHX
with discontinuous helical bafes (DCH-STHX) in an early stage to
the STHX with continuous helical bafes (CH-STHX). To further
enhance the heat transfer performance and overcome the disadvantages of the CH-STHX, two major categories combined bafes
were invented, which are the combined single shell-pass STHX
with continuous helical bafes (CSSP-STHX) and the combined
multiple shell-pass STHX with continuous helical bafes (CMSPSTHX). The CSSP-STHX includes three subclasses: the discontinuouscontinuous helical bafed combined single shell-pass STHX
(DCH&CH-CSSP-STHX), the continuouscontinuous helical bafed
combined single shell-pass shell-and-tube heat exchanger
(CH&CH-CSSP-STHX), and the combined parallel single shell-pass
STHX with continuous helical bafes (CPSSP-STHX). Different types
of helical bafed STHXs depicted above are shown in Fig. 1. This
part will review the history of the shell-and-tube heat exchanger
with helical bafes according to the aforementioned logical order
as illustrated in Fig. 1.
2.1. Principle of heat transfer enhancement of helixchangers
Due to the principle shortcomings of the conventional segmental bafes which are the zigzag ow pattern resulting in large dead
zones and a relatively high amount of back mixing, during the last
decades, a shell-and-tube heat exchanger with helical bafes
750
Fig. 1. Different types of shell-and-tube heat exchangers with helical bafes.
known as HELIXCHANGERS has been increasingly adopted in

industrial applications to supersede the traditional STHXs with
segmental bafes.
The principle of helixchangers [4] design is fairly easy: circular
or elliptical sector-shaped plates are arranged in a pseudo-helical
bafe system (see DCH-STHX Fig. 1). Each bafe occupies one
quadrant or one third of the heat exchanger shell cross section
area and has an inclination angle to the central axis of the shell,
forming a helix ow path for the working uid. Hydraulic results
from Kral and Stehlik [5] pointed out that helical bafe arrangements have less back mixing producing compared with the segmental bafe arrangement in the heat exchanger. Wang [6]
carried out an experimental investigation on the velocity eld of
the STHXs with helical bafes by applying the Laser Doppler Anemometry (LDA). The results from the experiment make sure that
the helical bafes result in increasing turbulence and local mixing
occurs owing to the induced inhomogeneous component of the
velocity in the shell side of heat exchanger. Lutcha and Nemcansky
[7] demonstrated that large differences in effectiveness of heat
exchanger are result from different ow patterns, two extremes,
those are, perfect mixing ow and perfect plug ow. They manifested that the perfect mixing ow has signicant disadvantages
in thermal performance versus perfect plug ow, because local

driving force for heat transfer, i.e., the temperature difference
between two working uids has substantially reduced by the mixing ow. Consequently, a ow pattern which is close to ideal plug
ow is the design target for a proper bafe arrangement. It has also
been reported that the convective heat transfer across a tube bundle decreased with the increasing velocity angle if the free stream
is homogeneous, while the thermal performance of the shell-and
tube heat exchangers with helical bafes signicantly depended
on the inclination angle. At the range of small helix angles, the
ow eld of working uid resembled that in a homogeneous cross
ow. However, in the range of 2540, the thermal performance in
terms of the Nusselt number increases with larger inclination
angles. In case of beyond 40, the heat transfer performance drops
rapidly.
The principles of heat transfer enhancement on helical bafed
STHXs can be summarized as follows [8]: higher temperature
driving force or larger temperature difference resulting from
approaching plug ow conditions on the shell side; higher surface
heat transfer efciency through improving the velocity gradient of
shell side uid; and less fouling appealed to more uniform ow
distribution.
2.2. Improvements on bafe structure of helixchangers

2.2.1. STHXs with discontinuous helical bafes
In the STHXs with discontinuous helical bafes, the quadrant or
third plates on the shell side can shape a close helical ow pattern,
which has higher conversion of pressure drop to heat transfer
while an optimal inclination angle is selected. In addition, reducing
fouling and ow-induced vibration and prolonging the service life
of working uid in the shell side are several obvious benets from
the helix ow manner which can achieve near plug ow
conditions.
In the STHXs with discontinuous helical bafes, triangle zones
exist in the shell side. Zhang et al. [9] have studied the effects of
leakage on ow performance in shell side of middle overlapped
helical bafed STHXs by using the computational uid dynamic
(CFD) method. The leakages in the triangle spaces were apparent
and seriously affected the helix ow in the shell side. Inserting
block plates in these triangle spaces is one useful method to solve
the leakage problem. Experiments were carried out to study the
impacts of inserting block plates on thermal performance and
resistance performance in overlapped discontinuous helical bafed
STHXs [10]. Zhang et al. [11,12] made several careful experiments
studies on the segmental bafed STHX and middle-overlapped discontinuous helical bafed STHXs with different helix inclination
angles and their results clearly stated that the optimal helix inclination angle was 40, which agreed well with the results of the
previous studies [7]. Another important geometrical parameter
for the helical bafes is the bafe space. Five heat exchangers corresponded to the same optimal inclination angle of 40 with different bafe spaces were designed [13], and the simulated results
exhibited that the performances will be impacted greatly and
when a new helical bafed heat exchanger is to be designed, the
bafe space should be considered properly according to the real
constraint factors. Van Der Ploeg and Master [14] reported that a
double helical bafe arrangement design proposal with a helix
inclination angle up to 45 was adopted for large feed/efuent
exchangers and this leading to converting pressure drop to heat
transfer more effectively and exibly. After that, Lei et al. [15] studied the single-helical bafes, double-helical bafes, and segmental
bafes numerically which adopted the periodic boundary condition. According to their simulation results, it can be concluded that
the STHX with double helical bafes had obviously better effectiveness (10%) of heat exchangers than that of the STHX with single
helical bafes. Barely any research focused on the helix inclination
angles smaller than 20, so Zhang et al. [16] designed an experiment to compare the comprehensive performances of four heat
exchangers whose helix angle ranges from 7 to 25. Xiao et al.
[17] provided a new perspective that the heat transfer performance comparison among different heat exchangers is better conducted under the same required heat transfer capacity. And then
ve different kinds of uids were involved in their research to
reveal the effect of Prandtl number on heat transfer characteristics.
A surprise conclusion was obtained that helix inclination angle 40
is no longer the best selection if the Prandtl number for the shell
side uid increases, especially, when Prandtl number is large
enough, heat exchanger with small bafes inclination angle reveals
to be the optimal choice. Leong et al. [18] were the rst guys
applied the nanouids to the DCH-STHX. He et al. [19] applied
the second-law-based thermodynamics analysis to helical bafed
STHXs. The computed results indicated that both entropy generation number and exergy losses of the STHXs with single-helical
bafes and double-helical bafes were lower than those of STHXs
with segmental bafes. From the second-law-based thermodynamic viewpoint, it means that the heat exchangers with helical
bafes have higher efciency than the heat exchangers with
segmental bafes. Liu et al. [20] performed an experiment and
751
numerical simulations to investigate the ow and heat transfer

characteristics in shell and tube exhaust gas recirculation (EGR)
helical bafed cooler with spirally corrugated tubes (HBCSCTs)
and the results revealed that it had better performance than that
of the EGR segmental bafed cooler with smooth tubes (SBCSTs).
Zhang and co-workers [21,22] experimentally studied the heat
exchangers with discontinuous helical bafes combined with
two-dimensional and three-dimensional nned tubes. After that,
they performed experiments [23] to compare the shell-side condensation heat transfer coefcients of an integrally helical bafed
condenser with petal-shaped n (PF) tubes to those of that with
low n (LF) tubes while the zeotropic refrigerant mixture R407C
employed as the working uid. Chen [24] put forward a novel
structural named circumferential overlap trisection helical bafe
shell-and-tube heat exchanger (cothSTHX). It is illustrated in
Fig. 2, which shows the helical bafes, cylindrical shell, and heat
exchange tube bundle.
The cothSTHX is more suitable for use with the equilateral triangle tube layout and includes anti-shortcut leakage features.
Related experimental and numerical research [24,25] on this type
of heat exchanger has been performed and demonstrated that
the cothSTHX exhibiting excellent performance. And some other
typical numerical simulations utilizing entire heat exchanger
model and porous media model can be referred to Refs. [2629].
The experimental correlations for heat transfer and pressure drop
of the DCH-STHXs with can be seen in Table 1.
According to the available results of experiments and numerical
simulations mentioned earlier, the DCH-STHXs have higher heat
transfer coefcients under the same shell-side pressure drop. However, shortcomings are cannot be avoided, for example, under the
conditions of identical shell-side mass ow rate and the same
bafe pitch, the DCH-STHXs usually have relatively lower heat
transfer coefcient than the traditional SG-STHXs. When the
shell-side mass ow rate is small, the percentage of leakage ow
of the discontinuous helical bafes may be relatively large in the
triangle spaces, which will reduce the shell-side comprehensive
heat transfer performance. Furthermore, when the diameter of
shell increases in large-scale STHXs, the triangle zone area will
become quite large. In order to overcome these problems, continuous helical bafed and combined helical bafed STHXs have been
introduced in the next section.
2.2.2. STHXs with continuous and combined single shell-pass helical
bafes
The STHXs with continuous helical bafes (see Fig. 1, CH-STHX)
were proposed by Wang et al. [30,31], and they developed a one
continuous helical cycle method which could dramatically
Fig. 2. Structure of circumferential overlap trisection helical bafed heat exchanger

(cothSTHX) [24].
752
Table 1
Correlations for heat transfer and pressure drop of DCH-STHXs.
Researchers
Bafe style
Fluid
Correlations
Range
50 < Res < 1000
400 6 Res
400 < Res < 1000
Lei et al. [15]
Single discontinuous helical bafe
Oil
Nus = 0.275Re0.55
Pr1/3
s
fs = 20.06Re0.56
s
fs = 11.34Re0.47
s
Zhang et al. [12]
Segmental bafe
Oil
Pr1/3
Nus = 0.706Re0.474
s
fs = 25.1Re0.692
s
70 < Res < 650

70 < Res < 650
Middle-overlapped helical bafe

b = 20
Oil
b = 30
Oil
b = 40
Oil
b = 50
Oil
Nus = 0.275Re0.542
Pr1/3
s
fs = 11.0Re0.751
s
Nus = 0.365Re0.516
Pr1/3
s
fs = 13.5Re0.774
s
Nus = 0.455Re0.488
Pr1/3
s
fs = 34.7Re0.806
s
Nus = 0.326Re0.512
Pr1/3
s
fs = 47.9Re0.849
s
100 < Res < 800

100 < Res < 800
110 < Res < 480
110 < Res < 480
190 < Res < 750
190 < Res < 750
150 < Res < 500
150 < Res < 500
Discontinuous helical bafe

Petal-shaped n tube
Integral low-n tube
Oil
Oil
1/3
Nus = 0.026Re0.8
s Pr
1/3
Nus = 0.029Re0.8
s Pr
unknown
unknown
Discontinuous helical bafe

Low-n tubes
Oil
Rib-shaped n tube
Oil
Nus = 0.03(ResPr)0.715
Eus = 19.3(ResPr)0.058
Nus = 0.017(ResPr)0.823
Eus = 54.6(ResPr)0.132
125,000 < ResPr < 500,000

125,000 < ResPr < 500,000
100,000 < ResPr < 425,000
100,000 < ResPr < 425,000
Zhang et al. [11]
Segmental bafe
Middle overlapped helical bafe
Oil
Oil
Pr1/3
Nus = 0.012Re0.98
s
s
Nus = 0.037Re0.75
Pr1/3
s
s
unknown
unknown
Liu et al. [20]
EGR-SBCSTs
Air
3000 < Ret < 9010
EGR-HBCSCTs
Air
Nut = 0.0003Re0.634
t
ft = 0.739Re0.354
t
0.281
Nut = 0.012Ret
ft = 1.097Re0.185
t
Zhang et al. [23]
PF tube
LF tube
R407C
R407C
ao = 888.9exp(q/8.8)
ao = 739.9exp(q/11)
6 < q < 16 kWm2

6 < q < 16 kWm2
Zhang et al. [16]
Segmental bafe
Water
Nus = 0.485Re0.440
Pr1/3
s
s
fs = 51.44Re0.58
s
65,000 < Res < 180,000
Discontinuous helical bafes

b = 7
Water
15,000 < Res < 40,000
b = 13
Water
b = 25
Water
Nus = 0.448Re0.493
Pr1/3
s
s
fs = 66.64Re0.68
s
0.421 1/3
Nus = 0.493Res
Prs
fs = 132.64Re0.77
s
Nus = 0.492Re0.423
Pr1/3
s
s
fs = 156.97Re0.83
s
Zhang and Fang [22]
Zhang et al. [21]
3000 < Ret < 9010
3000 < Res < 90,000

5000 < Res < 15,000
Fig. 3. Fluid ow eld structure of STHXs with continuous helical bafes.

Fig. 4. Schematic diagram of a car air-conditioning system with a recuperator [33].
simplify the manufacturing difculty and reduce the cost signicantly. The uid ow in the shell side of CH-STHX is an ideal helix
in nature (see Fig. 3). The advantages of this ow pattern includes
all of the good points of discontinuous helical bafes, and even
more, excluding the defect triangle zones existing in the discontinuous helical bafes.
Peng et al. [30] conducted some experiments on the CH-STHXs
with side-in/side-out and middle-in/middle-out congurations. As
expect, the experimental results proved that the CH-STHXs have
better comprehensive performance than that in the SG-STHXs.
And it was also suggested that the side-in/side-out style is the better choice for the CH-STHXs as the shell-side inlet and outlet locations. The CH-STHXs have been considered as the replacement of
the SG-STHXs. There are two examples, one is the building air-conditioning system as an evaporator, and another is the car air-conditioning system (see Fig. 4) as a recuperator. For the former, the
most important index coefcient of performance (COP) of the
CH-STHX was 5.4% exceed than that of the SG-STHX [32]. For the
latter, the most important index heat exchanger effectiveness
increases from 0.21 of the original recuperator to 0.29 of the
753
CH-STHX [33]. Two more examples of successful applications of

helix ow could be found in Refs. [34,35]. One is a successful application of a helical shaped-pipe heat exchanger, and another is a
magnesium based metal hydride reactor incorporating helical coil
heat exchanger.
Usually after doing some experiments or numerical simulations,
authors need to supply correlations between the variables which
readers most care about, such as Colburn factor and friction factor.
Wang and Xie [36,37] employed different optimization algorithms
(genetic algorithm (GA) and articial neural network (ANN)) for
better tting to data, and their results illustrated that the predict
values from these optimization algorithms had a closer t than
those by the regression analysis. For the continuous helical bafes,
the bafe inclination angle is still a decisive role of the shell-side
heat transfer and resistance performance. Lei et al. [38] performed
numerical investigations on the effects of various bafe inclination
angles on the comprehensive performance of CH-STHXs using the
periodic boundary. The results declaimed that the best integrated
performance occurs at approximately the bafe inclination angle
45, which well agreed with the previous studies [7]. Improved
enhancement nned tubes is another way to enhance the heat
transfer performance of STHXs. Zhang and co-workers [3941]
have done a mass of research works on the STHXs with continuous
helical bafes combined with three-dimensional nned tubes. The
experimental and numerical results both indicated that the nned
tubes, especially petal-shape nned tubes, can signicantly
increase the shell-side heat transfer performance relatively to the
smooth tubes, while pressure drop just only slight increased.
Recently, Yang [42] noticed that for the STHXs with helical bafes,
the bypass stream still exists in the gap between the tube bundle
and shell. The bypass ow which has larger velocity magnitude
than that in the SG-STHX moves a helix path along the shaped of
the helical bafes under the effect of centrifugal force, without
rushing cross the tube bundle. Actually the bypass stream is so
important that a complete thermal design need a correction factor
to deal the impacts of the bypass on the shell-side heat transfer
coefcient. They exhibited that the ratio between the gap width
and the maximal free ow area is about 28.4%, and the ratio
between the mass ow rate in the bypass and the total shell-side
mass ow rate is as high as 47.7%. And some other typical numerical simulations utilizing entire heat exchanger model and heat
exchanger unit model with periodic boundaries can be referred
to Refs. [4345]. The experimental correlations for heat transfer
and pressure drop of the STHXs with continuous helical bafes
can be seen in Table 2.
The CH-STHXs are not perfect without any defects, researchers
have been aware of the two mainly troubles in the continuous helical bafes. Firstly, Song [46] reported that the helix ow will get
weak in the central axis zone and the uid almost goes through
in a straight line path resulting in decreasing heat transfer performance, seen in Fig. 5. Secondly, when the helical pitch is large, the
continuous helical bafe surface becomes considerably steep at
portions close to the central axis, especially in the STHXs with large
shell diameters. In such situations, it is so difcult to manufacture
the continuous helical bafes surfaces. A central sleeve tube may
have to be employed to x the helical structure inside the shell
to solve the above problem. On the other hand, heat exchange
tubes cannot be arranged at the place where the central tube is
located, and thus the heat transfer area of the heat exchanger will
be decreased; meanwhile, part of the heat exchanger volume is
occupied by the central tube and cannot be used for heat exchange,
which will result in the decreasing of STHX compactness.
Due to the above reasons, Wang et al. [4750] have invented
three different novel structures with continuous helical bafes
(see Fig. 1) to simply the manufacture and make full use of the
advantages of continuous helical bafes, which are the DCH&CHCSSP-STHX, CH&CH-CSSP-STHX, CPSSP-STHX. The DCH&CHCSSP-STHX is a combination of discontinuous helical bafes and
continuous helical bafes, in which discontinuous helical bafes
are mounted in the central region, while the continuous helical
bafes are equipped in most part of the outer space of the shell
side. The two types of helical bafes work together to forming
the whole continuous helicoids. The results from Chen at al. [43]
revealed that the comprehensive performance heat transfer coefcient per unit pressure drop was nearly 81.7% higher than that of
the SG-STHX. The novel structure CH&CH-CSSP-STHX is just proposed recently by Wang [47], and a photograph of this type of heat
exchangers can be found in Fig. 6 at the end of this section. As to
Table 2
Correlations for heat transfer and pressure drop of CH-STHXs.
Researchers
Bafe style
Fluid
Correlations
Range
Peng et al. [30]
Segmental bafe
Oil
Nus = 0.0889Re0.717
Pr1/3
s
fs = 2.79Re0.269
s
fs = 1.41Re0.167
s
Res 6 752
Continuous helical bafe

Side-in/side-out
Oil
Res 6 1960
Middle-in/middle-out
Oil
Nus = 0.0599Re0.669
Pr1/3
s
fs = 3.76Re0.578
s
fs = 0.316Res0.251
Nus = 0.0451Re0.699
Pr1/3
s
fs = 1.40Re0.437
s
fs = 0.226Re0.206
s
Segmental bafe
Oil
Nus = 0.0743Re0.747
Pr1/3
s
fs = 2.05Re0.22
s
Unknown
Continuous helical bafe

Side-in/side-out
Oil
Unknown
Middle-in/middle-out
Oil
Nus = 0.0533Re0.688
Pr1/3
s
fs = 1.64Re0.45
s
Nus = 0.0323Re0.742
Pr1/3
s
0.33
fs = 0.63Res
Zhang et al. [41]
Continuous helical bafe with petal-shaped nned tubes
Oil
1/3
Nus = 0.026Re0.8
s Pr
Unknown
Zhang et al. [39,40]
Continuous helical bafe with smooth tube

Continuous helical bafe with petal-shaped n tube
Oil
Oil
0.4
Nus = 0.030Re0.8
s Pr
0.4
Nus = 0.062Re0.8
s Pr
0.4
Nus = 0.063Re0.8
s Pr
0.4
Nus = 0.058Re0.8
s Pr
0.4
Nus = 0.059Re0.8
Pr
s
0.4
Nus = 0.060Re0.8
s Pr
10,000 < Res < 20,000

10,000 < Res < 20,000
10,000 < Res < 20,000
10,000 < Res < 20,000
10,000 < Res < 20,000
10,000 < Res < 20,000
Wang et al. [37]
752 < Res
1960 < Res

Res 6 2757
2757 < Res
Unknown
Fig. 6. Photograph of the CH&CH-CSSP-STHXs used in oil rening.
the CPSSP-STHXs, the working uid ows through the outer and
inner shell passes simultaneously, the inlet and outlet of two
passes join together at the both ends. To learn more about this type
of STHXs, Wang at al. [51] compared the CPSSP-STHX with the SGSTHX by CFD method and the results showed that the comprehensive performance, which evaluated by heat transfer coefcient per
pressure drop of the CPSSP-STHX is nearly 13.2% higher than that
of conventional SG-STHX for the same mass ow rate, and for
the same overall pressure drop, the heat transfer rate of the
CPSSP-STHX is nearly 20% higher than that of the SG-STHX.
2.2.3. STHXs with combined multiple shell-pass helical bafes
In the CMSP-STHX, the working uid ows through the outer
and inner shell passes in sequence, therefore, for each individual
shell pass, the cross sectional ow area is reduced dramatically,
and thus the velocity magnitude of the uid could be increased
for the same mass ow rate, and the heat transfer performance
can be anticipated to have a great enhancement. Wang et al. [52]
found out that the average overall pressure drop of the CMSP-STHX
was nearly 13% lower than that of SG-STHX under the same mass
ow rate and overall heat transfer rate, and under the same shellside overall pressure drop, the shell-side overall heat transfer rate
of the CMSP-STHX was nearly 5.6% higher than that of SG-STHX
and the mass ow rate in the CMSP-STHX is higher than that in
the SG-STHX by about 6.6%.
2.3. Shell-side thermal design method for STHXs with helical bafes
2.3.1. Correction factors method for discontinuous helical bafes
It can be found that both heat transfer and ow performance of
DCH-STHXs are different from traditional SG-STHX, which result in
many difculties in designing these heat exchangers. In order to
simply the design process, the classic Bell-Delaware method of
shell-and-tube heat exchanger design [53] is used as a basis for calculating several heat transfer and pressure drop correction factors.
Stehlik et al. [54] carried out a study of correction factors for SGSTHXs as compared to STHXs with one quarter discontinuous helical bafes. The results of the study shown that DCH-STHXs, when
properly designed, offered a signicant improvement in heat transfer while providing a reduced cost in pressure drop. In designing
2.3.2. Maximal velocity design method for continuous helical bafes

In industries, the tube bundle sometimes needs to be resigned
or replaced, because of failure or improving its heat transfer efciency. How to design new tube bundle and bafe arrangements
to make the CH-STHX has better performance than the SG-STHX
in the same shell structure is very important for designers. From
the correlations mentioned above, it can be concluded that the
maximal velocity in the minimum ow area, which determines
the Reynolds number, can affect the heat transfer performance
greatly. In order to enhance heat transfer performance of the CHSTHX, one effective method is to increase the maximal velocity
in the shell side for the same mass ow rate. The maximal velocity
ratio, R, is dened when the CH-STHX and the SG-STHX have the
same heat transfer coefcient in the shell side: R = uCH,max/uSG,max,
in which uCH,max and uSG,max are the maximal velocities in the minimum ow area for helical bafed heat exchangers and segmental
bafed heat exchangers. As shown in Fig. 7, the maximal velocity
rates decrease with the increase of the heat transfer coefcient in
the shell side. When the maximal velocity rates are above the line
with circle points(R > 2.4), the heat transfer coefcient of CH-STHX
will be higher than that of SG-STHX. On the other hand, the
corresponding friction factor rates between the CH-STHX and the
SG-STHX keep almost constant fo,CH/fo,SG = 0.28 with the increase
of heat transfer coefcient in the shell side [57].
The schematic diagram of the design process for maximal velocity ratio design method for STHXs with continuous helical bafes is
shown in Fig. 8.
There are two core steps in this method: rst, design a SG-STHX
by Bell-Delaware method which could meet the requirements of
heat loads and allowable pressure drop (see the left part of
Fig. 8); second, design a CH-STHX by the maximal velocity ratio
RCH,SG
Fig. 5. Shell-pass ow path line of continuous helical bafed heat exchangers [46].
and optimizing of helixchangers, pitch angle, bafes arrangement,

and the space between two bafes are important parameters.
Zhang et al. [55] have replaced the curve-type factors in the
literature by mathematical expressions for the convenience of
engineering design. Based on Ref. [55], Azar et al. [56] developed
a similar calculation procedure and established a computer program for convenience to designing a helixchanger. A rapid design
algorithm and computer code for discontinuous helical bafed heat
exchangers was also proposed by Jafari Nasr and Shafeghat [28].
But the detailed design procedure and the determination of geometry parameters of both sides in DCH-STHX were not supplied. The
details of correction factors design and rating method proposed by
Stehlik [54] and developed by Zhang [55] and Azar [56] can be
found in the references.
RCH,SG=2.4
fo,CH/fo,SG
754
fo,CH/fo,SG=0.28
0.4
0.2
0.0
0.5
1.0
1.5
2.0
2.5
3.0
ho ,kW/(m K)
Fig. 7. Maximal velocity ratio versus shell-side heat transfer coefcient [57].
755
solid particles). As a result of serious fouling of the solid particles,

the original SG-STHXs have to be washed and cleaned every half
year, at most not exceeding one year. After using the CH-STHXs,
continuously running twelve months, the inlet and outlet temperatures and shell-side pressure drop of the original SG-STHXs and
CH-STHXs are listed in Table 4. It can be clearly obtained that the
increase of outlet temperature of CH-STHXs is considerably smaller
than that of the SG-STHXs. Meanwhile the shell-side pressure drop
of the CH-STHXs is almost 30% lower than that of the SG-STHXs.
Finally, it should be emphasized that the CH-STHXs are well antifouling and their service life are obviously prolonged.
3. Primary surfaces heat exchangers for gas-to-gas heat
exchange
Fig. 8. Schematic diagram of the design process for maximal velocity ratio design
method.
curve and its mathematical expression (see the right part of Fig. 8).
The detailed procedures of the method are depicted in Fig. 8.
2.4. Industrial applications of helical bafed STHXs
More and more helical bafed shell-and-tube heat exchangers
including both DCH-STHXs and CH-STHXs are now in service globally in numerous applications. STHXs with helical bafes have
obvious advantages in cost saving in pumping equipment and
especially in reducing fouling during operation. Two examples
have been provided to testify the excellent features of helical bafed STHXs.
Crude oil stabilization is a vital part of the petroleum extraction.
After the crude oil is exploited, it will exchange heat with the stabilized crude oil to achieve the proper temperature for preparing to
be pumped into the fractionating tower. The schematic diagram of
ow paths of stabilizing crude oil and stabilized crude oil are
shown in Fig. 9.
The original heat exchangers are the DCH-STHXs, and then the
CH-STHXs are adopted to replace the DCH-STHXs. These heat
exchangers have been running continuously one entire year. The
data collected from the operating eld is listed in Table 3. It can
be concluded from Table 3 that the overall heat transfer rate of
the CH-STHXs is nearly 25.5% higher than that of the DCH-STHXs,
and the overall heat transfer coefcient of the CH-STHXs is about
80.9% higher than that of the DCH-STHXs, however, the heat transfer area of the CH-STHXs is 5.6% lower than that of the DCH-STHXs.
Four CH-STHXs are employed in the ammonia hydrate processes as the intercooler and aftercooler for the twelve 2MCL705
centrifugal compressor units. The tube-side working uid is water,
and the shell-side working uid is hybrid gas (gaseous CO2 with
In the microturbine system, the thermal efciency is only about

20% or less if no recuperator is used for heat recovery. However, it
could be increased to about 30% if a recuperator with 87% effectiveness is used. Because the cost of the recuperator is about 2530% of
the total cost, the recuperator should have high performance and
minimum cost [58]. Typically, the detailed requirements of microturbine recuperator include high heat transfer effectiveness (>90%),
low pressure loss (<5%), primary surface geometry (no secondary
surface inefciency), high surface compactness, superior thermalhydraulic characteristics, and so on [59].
In the fossil fuel power system, an air preheater is an important
component to recover the low grade heat from the combustion
product gases. In the typical fossil fuel power system, the air preheater is used to cool the ue gases leaving the nal water-heating
stage from about 300 to 100 C, and support warm air at about
250 C for the furnace [60]. It is typically comprised of many closely packed heat transfer elements. The main requirements for
an air heater are the high heat transfer rates, low pressure losses
and low sensitivity to fouling. The heat transfer elements with high
thermal hydraulic performance could substantially reduce the fuel
costs [61].
The indoor air quality and thermal comfort of air ventilation
have attracted more and more attention in recent years. The fresh
air ventilation system is such an efcient and healthy technology
that adopts the air-to-air heat exchanger to provide fresh air as
well as saving exhaust energy. To replace the traditional metal heat
transfer plate, Zhang et al. [62] further used the hydrophilic polymer membranes to manufacture the air-air heat exchanger, which
could transfer both the heat and moisture simultaneously. The efciency of the heat and mass transfer of the heat exchanger is
important for the market application.
The primary surface heat exchanger is a kind of high-efcient
plate heat exchangers without any secondary heat transfer surfaces, which is different from the plate-n heat exchanger. Because
there is no n efciency and complex structures are easy to manufacture, the heat transfer performance is very high and the weight
is very light. Due to these advantages, the primary surface heat
exchanger is widely used in the gasgas heat transfer applications.
In this section, we will give a review of the primary surfaces development used for gas-to-gas waste heat recovery systems in recent
years.
3.1. Traditional primary surfaces
Fig. 9. Schematic diagram of ow paths of stabilizing and stabilized crude oil.
The traditional so-called primary surfaces include the cross corrugated (CC) [63], the cross wavy (CW) [63], the cross undulated
(CU) [64] and the double notched (DN) surfaces [65]. The CC surface is also called chevron surface or herring-bone surface. It is
formed by two symmetrical plates which placed with a phase
angle. The cross section of CC plate is corrugated while the main
756
Table 3
comparisons of DCH-STHXS and CH-STHXs for crude oil stabilization.*
Item
Temperature (C)
Stabilizing crude oil

Stabilized crude oil
Inlet
Outlet
Inlet
Outlet
Heat transfer rate (kW)

Logarithmic mean temperature difference (C)
Heat transfer area (m2)
Overall heat transfer coefcient (kW m2 K1)
*
DCH-STHXs
CH-STHXs
Deviation
46
145
200
68
16,046
35
9018
58.2
42
190
207
67
20,146
21
8504
105.3
/
/
/
/
+25.5%
/
5.6%
80.9%
The data is provided by Qinhuangdao Dongyan Energy-Saving Technology Corporation of China (http://www.dyjn.com/).
Table 4
Comparisons of SG-STHXs and CH-STHXs for ammonia hydrate processes.*
Item
Shell-side working uid (gaseous CO2 with solid particles)
Shell-side pressure drop (kPa)
Inlet temperature (C)
Outlet temperature (C)
SG-STHXs
Initial operation
Twelve months later
145
145
43.6
46.7
6.6
8.4
CH-STHXs
Initial operation
Twelve months later
145
145
41.2
42.5
4.7
5.1
The data is provided by Qinhuangdao Dongyan Energy-Saving Technology Corporation of China (http://www.dyjn.com/).
direction is straight, as shown in Fig. 10(a). The main geometrical

parameters are the channel height, corrugation pitch, plate thickness and chevron angle. The structures of CU plate is similar to that
of CC plate, but the dimensional sizes of upper and lower plates are
different, as shown in Fig. 10(b). Compared to the CC plate, an additional waviness is introduced in the ow direction of CW plate. The
CW plates bend as a sinusoidal wave in the main direction so that
there is 180 out of phase between upper and lower plates waviness, as shown in Fig. 10(c). The CU and CW primary surfaces are
the variants of CC primary surface, which is suitable for the situation when there is a large difference in the volume ow rate
between hot and cold gases. For example, in the recuperated process of microturbine system, the non-compressed gas-side ow
may lead to higher pressure drop than the compressed air-side
ow under the balanced mass ow rate on both sides. The higher
pressure drop in the gas-side ow can be compensated for by
arranging a bigger channel in the gas side. The DN primary surface
is comprised of notches, undulations and at sections, as shown in
Fig. 10(d). Here, the longitudinally straight notches are arranged in
an equidistantly spaced apart and parallel arrangement. The
inclined undulations are designed on both sides of notches and
there is a at surface between the adjacent inclined undulations.
The hydraulic diameter of primary surface channels is usually
bigger than 4 mm because the plate heat exchangers are more used
to liquid-to-liquid heat transfer. Khan et al. [66] summarized many
reported experimental parameters of CC primary surface heat
exchangers and it was shown that the hydraulic diameter was usually bigger than 4 mm and the Prandtl number was usually bigger
than 2.0. Metwally and Manglik [67] studied the laminar developed forced convection of different viscous uids (Pr = 5, 35, and
150) in the sinusoidal CC channel. It was found that the temperature distribution of low Pr uids was relatively more stratied,
while that of high Pr uids was more uniform. The Nusselt number
of the uid with Pr = 35 was 99% higher than that of the uid with
Pr = 5, and 43% smaller than of the uid with Pr = 150 at Re = 700
and c = 0.5 (channel corrugation aspect ratios). It can be seen that
there is a signicant difference between the liquid-to-liquid and
gas-to-gas heat transfer processes. Due to the convective heat
transfer coefcient of gas-to-gas is much smaller than that of
liquid-to-liquid, the hydraulic diameter must be much smaller to

get high efciency and reduce the physical size of recuperator.
For example, the surface compactness (i.e., the ratio of heat transfer surface area and enclosed volume) of CC and CU primary surfaces with hydraulic diameter of 1.54 mm is about 1299 m2 m3.
The surface compactness of CW primary surfaces using the hydraulic diameter 1.54 mm is about 13431496 m2 m3 [63].
Utriainen and Sundn [68] conducted the numerical study to
assess the thermal and hydraulic performances of CW ducts aimed
for recuperators. Compared to the straight ducts, the Nusselt number of CW duct was improved by over 6 times and the pressure
drop was increased in a similar order of magnitude except CW2Z3. The heat transfer could be signicantly enhanced by the amplitude of waviness, which deected the direction of the main ow
and opened up the passages to adjacent ducts.
The trapezoidal cross wavy (TCW) channel is a special CW channel and the only difference between them is the cross sectional
shape. The cross section of TCW is trapezoidal with sharp corners
whereas that of CW is rounded. The Nusselt number of the TCW
duct could be improved by over 4 times than the straight ducts
accompanied with a similar increased pressure drop [69]. The
small changes of geometric parameters would result in signicant
changes of thermal and hydraulic performances of the TCW. The
Nusselt number and product f Re of CW and TCW were functions
of the Reynolds number in the laminar ow, while these of straight
ducts were constants.
Stasiek [61] investigated the full-eld temperature and heat
transfer coefcient distributions of the CU channels by the truecolor image processing of liquid crystal images. Six specied
shapes of CU channels were measured to study the effect of angle,
shape factors (height and pitch) and Reynolds number on the heat
transfer and pressure drop. It was observed that the average Nusselt number increased by about 8090%, while the friction coefcient increased by almost 34 times when the angle varied from
20 to 70. The Nusselt number would increase by almost 50%
when the ratio of undulated height to corrugation height increased
from 0.24 to 0.43. The large undulation depth could improve the
heat transfer performance for both the corrugated and undulated
plates, while the aspect ratio of P/H varied from 3.87 to 5.8 had
757
(a) Cross corrugated surface
(b) Corrugated undulated surface
(c) Cross wavy (CW) surface
(d) Double notched (DN) surface [65]

Fig. 10. Traditional primary surfaces.
little effect on the heat transfer performance under a constant corrugation height.
Zhang and Che [65] conducted a single-blow experiment and a
numerical simulation to test the overall and local ow and heat
transfer performances of DN primary surface channels. It was
found that there were wavy channel and pipe ows, but no swirling ow in the DN channels. The Nusselt number side was high at
the luff, but was low at the lee side of the undulations and notches.
The thermal-hydraulic performance of DN surface was between

that of the CC and parallel surfaces.
Stasiek et al. [60] conducted an experiment to study the effect
of corrugation angle, geometry and Reynolds number on the ow
and heat transfer performances of CC channel under the condition
of 2.22 6 P/Hi 6 4.00, 30 6 h 6 79 and 500 6 Re 6 5000. The local
Nusselt number distribution of CC surface was measured by applying thermochromic liquid crystals and true color image processing.
758
It was found that the averaged Nusselt number increased approximately as Re2/3, while the equivalent friction coefcient decreased
roughly as Re1/2.
Zhang and Che [70] studied the effect of corrugation prole on
the heat transfer and pressure drop performances of CC primary
surface. Six kinds of corrugation proles, i.e. sinusoidal, isosceles
triangular, trapezoidal, rectangular and elliptic corrugations, are
compared by using a low Reynolds number ke turbulence model.
The Nusselt number and friction factor of trapezoidal channel were
14 times higher than those of elliptic channel. The trapezoidal
channel showed the best heat transfer performance, while its pressure drop was also the largest.
Wang et al. [71] investigated the Nusselt number and friction
factor of CC primary surface channels with different included
angles by experiment. Compared to the CC channel with 60, the
Nusselt number of that with 120 was 4986% higher, while the
friction factor was 5.56 times higher.
Lin et al. [72] conducted an experiment to test the heat transfer
performance of CC duct under an approximately constant temperature boundary condition. Then the dimensionless heat transfer
correlations were derived by using the Buckingham Pi theorem.
It revealed that the local Nusselt number was determined primarily by Re, R/Dh (Radius of curvature), x/Dh (Coordinate) and b
(Corrugated angle), whereas the averaged Nusselt number was
determined by Re, R/Dh and b.
Zhang [73] compared the CC primary surface channel with triangular cross section with that with sinusoidal cross section by
numerical method. It was found that the Nusselt numbers of the
sinusoidal cross sections was bigger than that of triangular cross
sections under uniform temperature condition and smaller than
that under uniform heat ux condition, but the friction factors of
triangular cross sections were much higher. The momentum transfer was mainly produced by the laminar ows at low Reynolds
number, while it was induced by the secondary and swirl ows
at high Reynolds number.
Utriainen and Sundn [74] gave the heat transfer and pressure
drop correlations for three CC surfaces, four CW surfaces and three
CU surfaces, as shown in Tables 5 and 6. Utriainen and Sundn [63]
also compared the comprehensive performances of the CC, CW, CU
and plate-n heat transfer surfaces. The result indicated that the
performances of CC and CW were the best due to their small volume and weight. The CC was recommended due to the simple
manufacture as well as the small passage dimensions. The CC plate
is one of the most popular surfaces in the available primary surface
plates because of its relatively high thermal performance, low
pressure drop, simple structure and high rigidity [70].
3.2. Novel primary surfaces
Unlike the traditional CC primary surface which is single-wave,
Kim et al. [75] proposed a newly double-wave CC primary surface,
as shown in Fig. 11(a). Here the vertical corrugations are
Table 6
Heat transfer and friction factor correlations in the form [Nu, f Re] = C1 + C2 Re
[74].
Surface
CC-45
CC-60
CC-75
CW2-z3
CW2-z5
CW3-z3
CW3-z8
UCS-30
UP2-30
US-50
Nu
f Re
C1
C2
C1
C2
2.9241
3.8988
6.1107
2.4101
1.0441
0.5256
1.8194
6.7538
3.07377
2.85
7.66E03
1.02E02
1.48E02
2.32E02
1.57E02
2.31E02
3.88E03
1.16E03
2.93E03
5.13E03
21.3186
42.4535
85.6395
61.8672
35.1707
51.7276
26.172
37.0463
16.2233
17.7878
2.95E02
5.93E02
0.1362
0.4427
0.1168
0.2524
3.13E02
1.39E02
9.53E03
2.04E02
perpendicularly added to the traditional single-wave CC primary

surface. The ow mixing is more obvious because both small and
large scale eddy are produced and the wave motion of air ow
reacts jointly between the double-wave plates, as indicated in
Fig. 11(b). The heat transfer coefcients will thus be increased by
the eddy motion in the double-waved inection points. Moreover,
the heat transfer areas are also increased due to the additional
wave. The heat transfer performance of double-wave primary surface heat exchanger is enhanced by approximately 50%, with 30%
additional pressure drop compared to single-wave primary surface
heat exchanger.
In order to increase the heat transfer performance by vortex
motion, Liu et al. [76] proposed smaller sine-wave corrugations
in the cross section of traditional CC primary surfaces, as shown
in Fig. 12(a). The additional corrugations can increase the heat
transfer area. Moreover, it can strengthen the uid disturbance
and mixing near the walls. It can be seen from Fig. 12(b), obvious
spiral secondary ow is formed along the primary surface, especially in the peak area. The spiral secondary ow can take more
uid from the mainstream to the peak area so that more heat
can be transferred. Under the condition of P/H = 2.2 and b = 75,
the heat transfer performance of CC primary surface with smaller
sine-wave corrugation can increase 10% compared to that without
corrugation.
Inspired from the ow disturbance of offset-strip plate-n
channels, Luo and Ma et al. [77,78] imported the offset-bubble into
the primary surface CC channels, as shown in Fig. 13(a). As can be
seen in Fig. 13(b), the uid ow in the offset-bubble channel bends
transversely in the front of offset bubble, which is similar to that in
the offset-strip plate-n channel. The heat transfer performance
was improved due to the additional transverse disturbance. Further, all the heat transfer surfaces are primary surfaces and there
is no n efciency. The averaged area goodness factor could be
increased by 41% and 71% compared to the offset strip plate nned
channel and CC3.1-60 (pitch-over-height ratio of 3.1 and inclined
angle of 60) channel.
Table 5
Geometrical parameter of typical primary surfaces [74].
Type
Surface
P /mm
Hi/mm
Luc/mm
Aw/mm
C/m2 m3
h/
Dh/mm
CC
CC
CC
CW
CW
CW
CW
CU
CU
CU
CC-45
CC-60
CC-75
CW2-z3
CW2-z5
CW3-z3
CW3-z8
UCS-30
UP2-30
US-50
3.48
3.48
3.48
1.38
1.38
1.38
1.38
2.24, 3.17
2.78, 2.15
2.74, 2.66
0.87
0.87
0.87
2.28
2.28
2.28
2.28
1.30, 0.79
1.61, 0.45
1.59, 0.44
4.54
3.48
2.85
2.98
4.96
2.98
7.94
6.33
4.3
3.47
0.99
0.99
0.79
0.79
1299
1299
1299
1717
1422
1496
1343
1299
1299
1299
45
60
75
30
30
50
1.54
1.54
1.54
1.54
1.54
1.54
1.54
1.54
1.54
1.54
(a) Double-wave surface
(b) Eddy motion of air flow between double-wave plates

Fig. 11. Double-wave CC primary surface [75].
(a) CC surface with corrugations
(b) Velocity vectors of CC surface with corrugations
Fig. 12. Double-wave CC primary surface [76].
(b) Velocity field of offset-bubble channel with P=12 mm
(a) Offset-bubble surface
Fig. 13. Offset-bubble primary surface [78].
(a) 3-D CC primary surface
(b) Velocity field of 3-D CC primary surface at Re2h=2000
Fig. 14. 3-D CC primary surface with inclined angle of 45 [79].
759
760
Blomerius and Mitra [79] developed a new 3-D CC primary surface by superposition of an additional wave perpendicular to the
waves in the traditional CC primary surface. Fig. 14 shows the
3-D CC primary surface with corrugation of 45 and the corresponding ow eld. It can be seen that extensive longitudinal vortices are generated by the 3-D wavy, which distribute at the lower
and upper walls as well as in the core of channel. It is different
from the traditional CC primary surface, in which the uid mainly
follows the furrows of the upper and lower plates. The comparison
evaluated by VG1 criteria of Webb indicated that, under the Reynolds number being 1800, the required heat transfer area of 3-D
CC primary surface with corrugation of 45 is about 5% smaller
than that of traditional CC primary surface with corrugation of
45, and about 10% smaller than that of 3-D and traditional CC
primary surfaces with corrugation of 90.
The ow in the middle plane of CC channel is a spiral mixing pattern induced by the interactive effect of the upper and lower plates,
but near the trough is mainly along the corrugation direction [80].
Therefore, the heat transfer performance in the trough is lower than
that in the peaks. In order to enhance the ow mixing and heat
transfer in the trough, Doo et al. [81,82] proposed three modied
primary surfaces with different secondary corrugations, i.e., inphase secondary corrugation, full-wave rectied trough corrugation
and anti-phase secondary corrugation, as shown in Fig. 15(a)(c).
Unlike the straight corrugations in the traditional CC primary surface, there are secondary corrugations on both the peak and trough
of the CC primary corrugation. The secondary corrugations on the
peak and trough are arranged in phase and anti-phase for the
(a) In-phase secondary corrugation
(c) Anti-phase secondary corrugation
primary surfaces with in-phase secondary corrugation and antiphase secondary corrugation. For the primary surfaces with fullwave rectied trough corrugation, the secondary corrugation on
the peak and trough are dissimilar. According to the comparison,
it is found that the anti-phase secondary corrugation has the greatest potential to improve the compactness of heat exchanger matrix.
For the primary surfaces with big anti-phase secondary corrugation,
the ow separation, recirculation and impingement are produced
and the uid mainly ows along the stream-wise direction and
the two ows become more parallel to each other, as shown in
Fig. 15(d), while the uid mainly ows along the trough of each corrugation plate in the traditional CC channel. Compared to the conventional CC channel, the pressure drops of the primary surface
channels with anti-phase could be reduced by approximate 15%
without big changes for the heat transfer capacity.
In many advanced systems such as the intercooled-cycle gas
turbine engine, the operating pressure and pressure drop requirements on the hot and cold sides are different. For these cases, Kim
et al. [83] proposed a novel primary surface channel with asymmetric proles, as shown in Fig. 16(a). The pressure drop on each
side can be controlled by adjusting the level of asymmetry of this
prole. It can be seen from Fig. 16(b), the ow separation and
impingement in the high pressure side are stronger than those in
the low pressure side due to smaller cross-sectional area. The ow
difference is more obvious as the generation of an asymmetric prole increases. The asymmetric prole could balance the pressure
drops between the high pressure and low pressure sides with a
slight loss of effectiveness.
(b) Full-wave rectified trough corrugation
(d) Streamlines of anti-phase secondary corrugation ( h= 0.8)
Fig. 15. Modied primary surfaces with different secondary corrugations [81,82].
761
(a) Primary surface channels with asymmetric profiles
(b) Streamlines of high pressure side
(c) Streamlines of low pressure side
Fig. 16. Primary surfaces with asymmetric proles [83].
3.3. Potential effects

3.3.1. Longitudinal heat conduction
Because the thickness of primary surface plate is about 0.1 mm,
the transverse heat conduction in such a thin plate could be
ignored. However, the heat transfer coefcients on the opposite
surfaces of primary surface plate are much non-uniform due to
the complex ow separation, recirculation and impingement. Big
longitudinal heat conduction may exist between the high heat
transfer regions in the opposite surfaces of primary surface plates.
Shah and Sekulic [1] pointed out that the effect of longitudinal heat
conduction was signicant and should be considered to achieve a
higher heat transfer performance when the heat transfer effectiveness was above about 85%. Doo et al. [84] conducted a CFD simulation to study the longitudinal heat conduction effect on the
traditional CC primary surface channels. It was found that the maximum performance degradation was up to 8.4% in the numerical
ranges. As the plate thickness increases, both the small-scale longitudinal heat conduction and thermal performance increase. As the
corrugation pitch increases, the small-scale longitudinal heat conduction increases, but the local heat transfer coefcient uniformity
reduces. Therefore, there is an optimal corrugation pitch for the
thermal performance of primary surfaces.
3.3.2. Numerical model and method

In the primary surface model, there are many singular contact
points between the upper and lower plates. Ciofalo et al. [80] used
the thin surface contact to replace the line contact between the
upper and lower plate when the width surface was arbitrarily
chosen to be equal to the plate thickness, as shown in Fig. 17. In
fact, the surface contact usually has a signicant inuence on the
grid quality and computational accuracy. Bu et al. [85] found that
Fig. 17. CC computational domain (inset showing replacement of line contact by

surface contact at corners) [80].
effects of contact treatments on the pressure drops were remarkable for different structured packings, especially when the porosity
was relatively low, while such effects on the Nusselt numbers were
relatively small. Hence, much attention should be paid on the contact points between the upper and lower primary surface plates.
Du et al. [64] modied the traditional computational model of
CW duct by adopting multiple periodic boundary conditions in
both the main ow and transverse directions, as shown in
Fig. 18. It was more realistic and suitable for large amplitude of
waviness when the ratio of amplitude of waviness to channel pitch
was greater than 0.25. It was found that the Nusselt number of
model with linear-variation wall temperature boundary condition
increases by about 15% with the same pressure drop compared to
that with the constant wall temperature boundary condition.
762
Fig. 18. Division of multi-periodic boundary conditions for CC model with a gap
[64].
Utriainen and Sundn [68,69] said that the laminar ow prevailed

when the Reynolds number was less than 2000.
Zhang [86] investigated the complex 3D turbulent ow and
heat transfer of air inside a CC duct with triangular corrugation.
It showed that compared to the experimental data, the low
Reynolds ke turbulent model gave the best prediction for
2000 6 Re 6 6000, while the standard ke turbulent model gave
the best prediction for 6000 6 Re 6 20,000. The heat transfer of
the duct was 4060% higher than the parallel at plate channel
with twice friction pressure drop. However, the above studies were
analyzed under uniform temperature boundary conditions. Zhang
and Chen [62,73] conducted an experiment and numerical simulations to disclose the uid ow and heat transfer behaviors of CC
duct with triangular corrugation under uniform heat ux boundary
condition. The Nusselt numbers were different under uniform temperature and uniform heat ux boundary conditions.
Ciofalo et al. [80] tested the applicability of different models,
i.e., laminar model, standard ke turbulence model with wall functions, low-Reynolds number ke model, direct simulations and
large eddy simulations on the predictions of ow and heat transfer
performances of CC channel under the condition of 2 6 P/Hi 6 4,
30 6 h 6 150 and 1000 6 Re 6 10,000. The results indicated that
the low-Reynolds number ke model and LES gave the best predictions for friction factors and average or local heat transfer coefcients on the whole range of geometrical parameters. However,
the standard ke turbulence model with wall functions was completely inapplicable to Reynolds number lower than 2000, while
the laminar ow was only effective for Reynolds number lower
than 3000 and moderate angle. It is difcult to establish the mesh
for primary surface channels due to the complex structure and singularities at the corners.
To nd a suitable model for CC channels, Zhang and Che et al.
[87] numerically investigated the uid ow and heat transfer characteristics of CC ducts by using eight turbulence models, i.e., low
Reynolds number ke, standard ke, realizable ke model, renormalization group ke, Reynolds stress model, standard kx model,
shear-stress transport kx and large eddy simulations. The low
Reynolds number ke and shear-stress transport kx performed
the best predictions compared with the available experimental
data when the Reynolds number ranged from 1000 to 10,000.
Although some models provided the similar averaged heat transfer
and pressure drop performances, but very different local velocity
and temperature distributions. They concluded that it was not
enough to only compare the overall heat transfer and pressure
drop data between simulation and experiment for comprehensive
model evaluations.
3.3.3. Total heat transfer coefcient separation

Although the primary application of primary surface heat
exchangers is for gas to gas heat exchange, the experiment is usually conducted in an air-water loop to obtain a constant wall temperature or constant wall heat ux conditions for the primary
surface channels [60]. The reason for this is that the heat transfer
coefcients on both sides are on the same order of magnitudes
and it is difcult to separate the one-side heat transfer coefcient
from the total heat transfer coefcient. Wang et al. [71] conducted
an experiment in the wind tunnel to test the heat transfer and
pressure drop characteristics of a CC primary surface heat exchanger with different corrugation angles. The results indicated that the
critical Reynolds number from laminar ow to turbulent ow is
about 948 for CC2-60 and 1284 for CC2-120. The Genetic Algorithm
was used to separate the single-side heat transfer from the total
heat transfer coefcient without wall temperatures and establish
the heat transfer correlations for hot and cold sides. The wall temperatures predicted by established heat transfer correlations were
agreed well with the experimental data. It gave a possibility for
total heat transfer coefcient separation where the Wilson-plot
technique could not be used, because the method required that
the thermal resistant of one side should be kept constant. Ngo
and Tsuzuki et al. [88,89] assumed that the Nusselt number correlations were the polynomials of Reynolds number and Prandtl
number and then adopted the least-squares method to obtained
one-side heat transfer coefcient from the total heat transfer coefcient in the printed circuit microchannels.
3.4. Industrial applications of primary surface heat exchangers
The primary surfaces have been used for many industry applications. One of the typical applications is the completely laser
welded primary surface recuperator developed by the Rekuperator
Svenska AB (RSAB), which is owned by VOLVO Technology Transfer
Corporation and Avesta Polarit, as shown in Fig. 19(a) [90]. The CW
primary surface plates are selected as the main heat transfer element. One pair of the stainless steel CC plates is laser welded
around the perimeter to from an air cell. As shown in Fig. 19(b),
the air cell is comprised of the cross ow inlet section, counter ow
section and cross ow outlet section. The inlet and outlet sections
are designed to be triangular to allow a uniform ow distribution.
Then 390 air cells are stacked and laser welded together to form
the recuperator core. The recuperator is 295 420 2.22 mm,
whose compactness is up to 1600 m2 m3. At last, the manifolds
for air inlet and outlet as well as piping system are welded to the
core so that a complete recuperator is accomplished.
The recuperator was installed and tested in a 100 kW micro-gas
turbine power plant for combined heat and electricity generation
by a customer. The design parameters are shown in Table 7. It
was found that the effectiveness and total relative pressure losses
are better than the design requirements. As described in Fig. 20,
the effectiveness under different mass ow rates is higher than
89%, while the relative pressure loss is lower than 3.5%. The prototype is shown in Fig. 21(a) [90]. Some other typical primary surface
recuperators are described in Fig. 21(b)(d) [9193], which are
developed by the Solar Turbines Company, Liang and Zhang et al.
from Xian Jiaotong University. All the heat transfer cores are manufactured by the CW primary surfaces.
4. Heat exchangers for liquid-to-gas heat exchange in
air-cooling system
The nned tube heat exchanger (FTHX) is a type of compact
heat exchangers and the most commonly utilized type in a variety
of industries, such as chemical engineering, heating and ventilation
763
(a) Recuperator structure
(b) Typical air passage geometry
Fig. 19. Primary surface recuperator developed by RSAB [90].
Table 7
Design parameters of primary surface recuperator developed by RSAB [90].
Operating conditions
Design requirements
Mass ow rate
Inlet air temperature
Inlet gas temperature
Inlet air pressure
Inlet gas pressure
0.78 kg/s
210 C
650 C
4.5 kPa
1.1 kPa
Fig. 20. Heat transfer and pressure drop performances of recuperator developed by
RSAB under different mass ow rates [90].
compressor intercoolers, HVAC & R, automobiles, air cooling industries, residential air-conditioning, and refrigerating appliances.
FTHXs are preferred on the condition of high pressure uid at
one side or high heat transfer coefcient of another uid. Especially
in the air cooling towers and air separators, they play an important
role. So the extended surface is employed effectively either on the
internal or external or on both sides to enhance the heat transfer
performance of the gas side in such compact heat exchangers. Fins
Effectiveness
Total relative pressure losses
Durability
Min. 89%
Max. 4.5%
60,000 hours
3000 start-up cycles
are basically external surfaces on the tube, which can increase the
surface area of the bare tube and be used to destroy the boundary
layer and change the ow pattern to meet the demands of heat
transfer enhancement.
The method of attaching the ns over the tube is very important, since even a slightest air gap between the tubes and ns will
degrade the whole purpose of ns over the tube. The nned tube of
GEA Heat Exchangers is shown clearly in Fig. 22.
If a gas-to-liquid heat exchanger is applied in engineering elds,
the liquid with high heat transfer coefcient is placed on the tube
side, while the gas with low heat transfer coefcient is on the shell
side. The main thermal resistance is usually on the gas side in
practical applications, and therefore special surfaces should be
employed to effectively improve the gas side heat transfer performance of FTHXs. It is an effective way to reduce the air-side
thermal resistance occupying up to 90% of the overall thermal
resistance.
In order to reduce the thermal resistance of the gas-side effectively and study the effect of n types on the hydraulic thermal
performance of FTHXs with different type of ns, different new
types of ns were developed, such as the plain n, wavy n, the
louvered n, herringbone wavy n, crimped spiral n, the slit n
and the vortex generator (VG) n [94100]. Vijaisri and her coworkers [101] proposed a n named rip saw n and applied it to
a high temperature ceramic plate-n heat exchanger. They also
compared the hydraulic and thermal performances of the rip saw
n with those of other types of ns. Experimental, numerical,
and optimal studies for these different types of ns have been
extensively performed.
The air-cooled heat exchanger (ACHX) is a typical type of
FTHXs. An ACHX is a heat transfer device for rejecting the thermal
energy from a hot uid directly to the ambient environment. The
main advantage of an ACHX is that it does not need water, which
means that the plant requiring large cooling capacities does not
need to be located close to a supply of large quantity of cooling
764
(a) Primary surface recuperator for 100 kW microturbine developed by RSAB [90]
(b) Annular primary surface recuperator for 30 kW and 60 kW microturbine developed

by Solar Turbines [91]
(c) Primary surface recuperator for 100 kW microturbine developed by Wang et al. [92]
(d) Annular primary surface recuperator for 30 kW microturbine developed by Zhang et al. [93]
Fig. 21. Typical prototypes of primary surface recuperator (see [9093]).
water. The ACHX may be as small as an automobile radiator or

large enough to reject the heat of turbine exhaust steam condensation in a cooling system.
There are two basic types of air cooling systems, namely the
direct cooling system and the indirect cooling system, in the power
plant applications, which are shown in Figs. 23 and 24, respec-
tively. As is well known, the nned tubes in ACHXs are positioned

in the form of an A-frame or delta to reduce the required land area
[102].
ACHXs are classied according to their shape and draft created.
An ACHX can be installed in an A-frame form or a horizontal form,
which is shown in Fig. 25. The horizontal conguration is easier to
765
(b) Extruded tube type
(a) Groovy Fin tube type
(c) L Fin tube type

Fig. 22. Finned tube in detail [94].
4.1. Research development of air-cooled heat exchanger
Fig. 23. Schematic of the direct air-cooling condensing system.
Fig. 24. Schematic of the indirect air-cooling system.
be maintained than the A-frame one, which is commonly

employed for condensation. The draft in ACHXs can be either natural or mechanical generated by electric fans. Induced and forced
mechanical drafts are the two main types often used at present
[103].
The ACHX is universally employed where a large uid ow must
be cooled, or the water is expensive. The capital investment for an
ACHX is representatively three times that of an equivalent traditional STHX. The investment cost for ACHXs is very high but the
operating and maintenance expenses are lower. So much attention
and foresight are paid to the design of the ACHX [103].
This part will give a review about the investigations and applications of ACHXs, including direct ACHXs and indirect ACHXs.
Pirompugd et al. [99] proposed a new method, namely the

nite circular n method (FCFM), to analyze the heat and mass
transfer characteristics of FTHXs having plain n conguration
under fully and partially wet surface conditions. It was shown that
the sensible heat transfer and mass transfer performance are
insensitive to changes of n pitch. The inuence of inlet relative
humidity on the sensible heat transfer performance is small, and
is almost negligible when the number of tube rows is above four.
Wang et al. investigated the air-side performance of herringbone wavy FTHXs with larger diameter tube [96] and that under
the dehumidifying condition [97]. Their results show that the inuence of the n pitch on the air-side performance is comparatively
small for N = 1 or N = 2, while a remarkable decrease of heat transfer performance is found when the tube-row number is increased
and normally higher thermal-ow performance is associated with
that of the larger n pitch [96]. The pressure drops in wet condition
is much higher than that in dry condition, while the difference of
pressure drop between dry and wet condition is less profound
for the wavy n conguration as compared to that of plain n
geometry [97].
Pongsoi et al. experimentally studied the effect of n pitches
[98] and number of tube rows [104] on the air-side performance
of crimped spiral FTHXs with a multipass parallel and counter
cross-ow conguration. The schematic diagram of the heat
exchangers algorithm for multipass parallel cross ow, multipass
counter cross ow and multipass parallel and counter cross-ow
(N = 2, 3, 4 and 5) was shown in Ref. [104]. Their results showed
that the n pitch casts have small inuence on the heat transfer
performance, while a rise of the resistance factor was detected
when the n pitch is increased to 6.2 mm [98]. It was found that
either the number of tube rows or the n materials has no important inuence on the heat transfer characteristics at high Reynolds
number. Besides, the experimental correlation for the air-side
thermal-ow performances of this type of heat exchangers at high
Reynolds number (400013,000) was developed for engineering
applications [104].
Furthermore, the authors group [105112] also studied FTHXs
with different types of ns (plain n, offset-strip n, slit n,
crimped spiral n, mixed n, and VG n) in recent years, using
766
(a) A-frame configuration with forced draft
(b) Horizontal with forced draft
(c) Horizontal with induced draft

Fig. 25. Types of ACHXs [103].
experimental, numerical and articial neural network (ANN),

genetic algorithm (GA) and Taguchi methods. Refs. [105107]
showed the experimental and numerical investigations for the
inuences of the n patterns and geometry parameters on the
air-side performance of FTHXs, and presented the multiple correlations of heat transfer and friction characteristics. The heat transfer
enhancement in narrow rectangular channel with LVGs was studied experimentally in Ref. [108], and optimized by Taguchi method
in [109]. The performances of laminar and turbulent uid ow and
heat transfer of FTHXs with large tube-diameter and large tuberow was predicted through ANN approach [110]. Finally, GA was
applied in the thermal and optimal design of compact heat
exchangers [111,112].
4.1.1. Direct air-cooled heat exchanger
An air-cooled condensers (ACC) is shown in Fig. 26. The turbine
exhaust steam is admitted to ACHXs and condenses there. Heat
transfer to the cooling air is by convection only. Therefore, there
is no water demand for the cooling system. Cooling air is moved
by fans (forced circulation). The ACC is designed modularly [113].
ACCs have been widely employed to reject waste heat in power
industry to save water resources. The of performance ACC is so sensitive to ambient wind that almost all the air-cooled power plants
in China are less efcient compared with the design conditions.
Ambient wind plays a signicant part in the thermo-hydraulic
characteristics of ACCs, but the wind effect mainly imposes on
the upwind and bilateral condenser cells and will decay immediately. So it is benecial for the design and operation optimization
of ACCs in an air-cooling system to study the wind effect and its
decay performances [114]. Yang et al. [115] investigated the
Fig. 26. Air-cooled condenser [113].
wind-break wall employed as an effective way to weaken the inlet

ow distortions and hot plume recirculation of ACCs in a power
plant. To study the effects of the windbreak wall congurations
on the thermo-hydraulic characteristics of ACCs, the layout of the
ACCs and the corresponding main buildings in a typical
2 600 MW direct dry cooling power plant is schematically shown
in Fig. 27, which is consisted of the boiler and turbine houses,
steam duct, wind break wall and chimney.
Both the inuences of winds across and parallel to the longitudinal axis on the fan performance and recirculation in a forced
draught ACHX bank were numerically studied [116]. It was concluded that the cross wind mainly leads to the reduction in air volume ow rate, while the wind along the longitudinal axis leads to
an increase in plume recirculation along the sides of the ACHX
bank. The effect of both wind directions on the heat transfer of
an ACHX bank consisting of 6 bays, are important, which must
be considered when designing an ACHX.
Based on a representative 2 600 MW direct dry air cooling
power plant, the physical and mathematical models for the air-side
ow and heat transfer in the ACCs at different ambient wind
speeds and directions were built [114] by introducing the radiator
model to the nned tube bundles [117]. The schematic view of an
A-frame condenser cell is shown in Fig. 28(a), and the ACC platform
and condenser cell serial number is shown in Fig. 28(b). From Ref.
[114], the simulation results showed that the cooling capability of
the air at high wind speeds was worsened due to the high inlet air
temperature caused by the reversed ows happened in the upwind
condenser cells. The ambient wind effect decays rapidly along the
wind direction due to the composited behaviors combined with
the hot plume discharge and ambient wind. At the wind direction
of 90, it is the ow rate decrease of the upwind condenser cells
that primarily causes the decline of the thermo-hydraulic characteristics of ACCs, not the hot plume recirculation. The hot plume
recirculation ow plays an adverse role in deteriorating the comprehensive performances of ACCs at the wind direction of 0. Bothe
of the total volume ow rate and heat rejection of the two ACCs
decrease with the increase of the wind speed under different wind
directions, which are shown in Fig. 29, respectively. From Ref.
[117], it was shown that the comprehensive thermal performances
of the ACCs, condenser cells and nned tube bundles at the downstream are generally superior to those at the upwind. It is of benet
to investigate the space characteristics of the heat transfer performance for the upwind fan regulations and the A-frame condenser
cell geometrical optimization of the ACCs in power plants.
A numerical model including two identical ACC cells was built
and the different effects on the performance of the cells were
767
Fig. 27. Schematic of air-cooled condensers and main buildings in a typical 2 600 MW direct cooling power plant [115].
(a) A-frame condenser cell
(b) ACC platform and condenser cell serial number

Fig. 28. Schematic view of A-frame condenser cell, ACC platform and condenser cell serial number [114].
shown and analyzed through the numerical simulation method

[118]. The grid distributions around two air-cooled steam condensers (ACSCs) cells are shown in Fig. 30 (a), and the temperature plot
at Y = 0 m under windless condition is shown in Fig. 30 (b). It is
shown that the symmetric inuence disappears until the wind
angle changes. Both the ow and heat transfer performances of
the leeward cell are improved a lot with the increase of the wind
angle, while small difference appears for the windward cell. The
related results can supply the basic supports for improving the
comprehensive performance of the ACHXs.
The air-side thermal-hydraulic characteristics for the ACCs in a
representative 2 600 MW direct dry cooling power plant were
investigated numerically with three different wind-break walls
[115]. According to the simulation results of the ow and temperature elds at various ambient wind speeds and directions, the
volume ow rate and heat rejection of the ACCs are obtained and
768
(a) Volume flow rate
(b) Heat rejection
Fig. 29. Total volume ow rate and heat rejection of ACCs versus wind speed and direction.
(a) Grid distribution around two ACSC cells
(b) Temperature plot at Y=0 m under windless

conditions
Fig. 30. Numerical mode of two ACSC cells and temperature distribution [118].
compared. It was shown that the thermal-hydraulic performances

of the ACCs are improved due to the extensions of the inner and
outer walkways and elevation of the wind-break wall, especially
at the wind directions varying between 0 and 90. And the
improvement resulting from the width increase of the inner or
outer walkway is better than that caused by the elevated windbreak wall.
The performance of direct ACCs in power plant is inuenced signicantly by the air-side thermal-hydraulic performances of the
wavy nned at tube heat exchangers. So experimental studies

were performed for the air-side thermal-ow performances with
and without delta winglet pairs punched on the surface of the
wavy n [119]. The schematic of the wavy nned at tube used
in ACC is shown in Fig. 31. The different temperature elds of the
heated wavy n surface were acquired by the infrared thermography technology. Both of the results from the experiments and
numerical simulations indicated the high potential of heat transfer
enhancement to direct ACCs by LVGs.
Fig. 31. Schematic of the wavy nned at tube used in ACC [119].
Fig. 32. Air cooler and air-cooled condenser system [120].
A brief summary of the product categories in Alfa Lavals air heat

exchangers (AHXs) range and how each type of system works were
given by Nicholson [120]. There are air coolers, ACCs, air-cooled gas
coolers, air-cooled liquid coolers and air heaters in those categories.
The nned coils are much essential to these AHXs, and designed to
maximize the heat exchange between a uid and the air. An air
cooler and ACC system is shown in Fig. 32. Air coolers and ACCs
are the signicant elements in the air cooling and air conditioning
systems [120]. The liquid circulated in the coils is often brine or a
refrigerant, such as ammonia or carbon dioxide. Air-cooled gas
coolers become more and more popular and they run at much
higher pressures than that in ACCs. The AHX may be referred to
as a radiator applied in the cooling of process liquids [120].
The ow and heat transfer exhibit complex non-linear
characteristics for the ACC of power generating unit because
numerous factors, including the meteorological and the geographic
conditions, as well as the congurations of wavy nned tube
769
bundles, can affect its performances. In order to quickly and accurately predict the air side velocity and temperature elds of ACC,
the reduced order models (ROMs) were established based proper
orthogonal decomposition (POD) method [121]. The physical
model was an ACC cell in the cooling system of a 2 330 MW
power plant, which was shown in Fig. 33. As shown in Fig. 33(a),
the typical structure of ACC of power generating unit is A-frame
conguration. The ACC cell is simplied to a two-dimension model
as shown in Fig. 33(b), and the computational domain was set to be
472 138 m (X Y), which was shown in Fig. 33(c). It was found
that it was more appropriate for the POD solution of extrapolated
cases and this research may provide a rapid and reliable approach
for the optimization of real-time operation of air-cooled power
generating units.
In order to remove the strong wind effect, an effective method is
proposed to enhance the thermal performance of the ACSC by setting up deecting plates under the air-cooled platform to form a
(a) Structure of ACC (from SPX Co.)
(b) Two-dimension model of ACC
(c) Computational domain

Fig. 33. Physical model of ACC. [121].
770
(a) Schematic of the ACSC
(b)Velocity vector plot in the middle plate of column 3
Fig. 34. Numerical model of ACSC and velocity distribution [122].
homogeneous air mass ow rate in the axial fans through leading

enough cooling air to the fans in the upwind region [122]. The
schematic of the ACSC unit is shown in Fig. 34(a). A heat exchanger
model was introduced to study the ow and heat transfer characteristics in the improved ACSC with deector, by introducing a porous medium model and a fan model simultaneously. The ow
velocity vector plot in the middle plate of column 3 of the aircooled units is shown in Fig. 34(b), from which it can be found that
the ambient air can be smoothly guided by the deecting plates to
the fans near the windward side, the rst and second rows, and
only a much smaller vortex exists there. It can be obviously seen
that the improved ACSC with deector shows an important
enhancement in both the cooling air mass ow rate and the heat
rejection rate, comparing with the conventional ACSC. The effect
of the plate inclination is also investigated, and the inclination
angle of 45 is found to be the optimum value for the arrangement
of the deector [122].
To evaluate the feasibility of the thermal plant project, the overall ow and temperature distribution of the air in the whole power
plant, especially around the ACHX were numerically investigated.
A CFD model was built to obtain the heat transfer and pressure
drop for the air ow through an ACHX inside the whole thermal
power plant, which was shown in Fig. 35(a) [123]. The heat
exchangers adopted in ACHX are the elliptical tube-n tanks, a
new type of convective heat exchangers, which is shown in
Fig. 35(b). It concluded that the CFD model was very useful for
the prediction of the ow and temperature eld around ACHX
inside the whole thermal power plant or in a large space. The standard ke turbulence model is used to simulate the temperature and
ow eld inside the ACHX to get the heat transfer coefcient, h,
and the pressure drop, Dp, and the corresponding correlations
are as follows:
h 25:345v 0:283
fr
Dp 13:6v 1:36
fr
The wave-nned at tube heat exchangers are widely used in

ACCs of a power plant. The nned tube is easy to be fouling due
to the inclined arrangement of the wavy ns on the A-frame tube
bundles in the ACC, thus the thermo-ow performances of ACCs
are likely to be deteriorated. Yang et al. [124] proposed a new conguration of wave-nned at-tube to prevent the fouling from
gathering on the surface of ns and nned tubes, in which the
n surface rotates to be perpendicular. The schematic views of
the currently employed and proposed ACC cells and nned tubes
are shown in Fig. 36. Then the physical and mathematical models
for the ow and heat transfer of the cooling air across the present
and proposed wave-nned at-tube bundles were setup, and the
thermo-hydraulic performances were compared, respectively. For

the proposed nned tube, the effects of the n pitch, height and
the Reynolds number on the thermal-hydraulic performances were
studied, and the dimensionless correlations of the resistance coefcient and average Nusselt number are obtained. It was shown
that the buoyancy effect should be considered for the correlation
of the resistance coefcient at low Reynolds numbers, but this
buoyancy force could be ignored for the Nusselt number [124].
4.1.2. Indirect air-cooled heat exchanger
As shown in Fig. 24, in the indirect air-cooling system, the turbine exhaust steam ows into the condenser rst, and then the
cooling water is pumped to the air-cooling tower, which is the
main difference between the indirect air-cooling system and direct
air-cooling system. And there are many advantages for indirect aircooling system, such as low initial investment cost, low operating
cost, and simple process. In order to obtain better thermal-hydraulic performances for indirect ACHXs, there are still many problems
need to solve, such as the effect of the atmospheric air, the geometrical parameters, and air inlet angles on the comprehensive heat
transfer performances.
Meyer and co-authors [125130] have done many investigations on forced draught ACHXs experimentally and numerically.
And some conclusions on ACHXs are acquired, such as for a system
comprising of a single bank of ACHXs, and it is concluded that the
numerical models used accurately reect the experimentally
determined differences in inlet ow losses experienced by systems
using either bell-mouth, conical or cylindrical fan inlet sections.
The efciency of the single unit in nned ACHXs with staggered
and in-line bundles of rectangular ducts is numerically investigated, where a hot process uid ows inside nned tubes and
the atmospheric air is circulated outside the nned tubes, over
the extended surface. For the most usual geometrical congurations of ACHXs, with staggered or in-line nned rectangular tubes,
the efciency of the unit cell can be directly deduced the two tables
given in Ref. [131]. Their results can be very useful for designers
and engineers to obtain the optimization of the tube spacing in
heat exchangers to maximize the heat transfer rate.
Kuntysh et al. [132] experimentally investigated the thermalow performances of ve staggered six-row banks of bimetallic
nned tubes in the core sections of ACHXs, which are made up of
tubes with rolled aluminum ns of various heights. The quantitative effect of the relative height h/do on the mean thermal-ow
performance of a bank was evaluated. The experimental data are
described by dimensionless correlations of the form Nu = f(Re)
and Eu = f(Re), where x = h/d0 is the relative height of a n. The
studies of banks IV are generalized for heat transfer and aerodynamic resistance on average, whose results are shown as follows:
771
(a) Sketch of ACHX
(b) Sketch of elliptical tube-fin heat exchanger

Fig. 35. Sketch of ACHX and the elliptical tube-n heat exchanger [123].
Nu 0:0245 0:0824xRe0:810:22x
Eu 2:85 78:3xRe0:1220:3x
A FTHX with hot uids inside the tubes, and cooling air that is
fan-blown (forced draft) or fan-pulled (induced draft) across the
tube bank. An air cooled FTHX ow chart is shown in Fig. 37. Standard fan coil is designed for air ow capacity of 400 cubic feet per
minute (CFM) to 800 CFM for horizontal air discharge and for
vertical air discharge. The tube materials can be chosen copper,
stainless steel, cupronickel, carbon steel, while the n materials
can be chosen aluminum, copper, stainless steel, cold rolled close
annealed sheet steel (CRCA), and hot dip galvanized CRCA, which
is can be designed according to the customer requirement and
the working condition [113].
Besides, the economic protability of ACHX depends on the
capacity of the nned tubes. The capacity results from the transfer
of the maximum heat exchanged in the smallest possible space
while keeping thermodynamic losses on the air side minimum.
Despite the fact that the interrupted surfaces can signicantly

improve the heat transfer performance, the associated penalty of
pressure drop is also tremendous.
A thermodynamic second-law analysis was performed to investigate the effects of different geometry and ow parameters on the
ACHX performance. It was observed that the entropy generation
rate of the external ow of the ACHX reduced when the air-side
Reynolds number decreased. It was also shown that the increasing
of the tube-side Reynolds number resulted in the rise of the
irreversibility of the ACHX [133].
Thermodynamic modeling and optimal design of an ACHX cell
were developed to minimize the two conicting objective functions simultaneously, which is the temperature approaches and
the total annual cost. During the optimal process, a fast and elitist
non-dominated sorting genetic algorithm (NSGA-II) [134] is
employed. At last, a selection procedure of the optimal point was
applied and the nal optimal design point was determined through
some sensitivity analyses varying the objective functions when the
optimal design parameters changed. The result for the distribution
772
(a) Currently used air-cooled condenser cell
(c) New air-cooled condenser cell
(b) Currently used finned tube
(d) New finned tube
Fig. 36. Schematic of the currently used and proposed air-cooled condenser cells and nned tubes [124].
Fig. 37. Air cooled nned-tube heat exchanger ow chart [113].
of Pareto optimal points for the two objective functions through

NSGA-II method is shown in Fig. 38, where A, B, C, D, and E stand
for ve different temperature approaches ACHXs.
Two different types of at-tube heat exchangers have been
investigated theoretically in order to evaluate their performance
potential in indirectly cooled display cabinets [135]; one with serpentine ns and one with continuous plate ns. From the parameter study, it is shown that considerable savings in the required
electric drive power to compressors, pumps and fans can be
obtained in comparison with the traditional cooling coil. The savings may be up to 15%. Besides, the required temperature difference for the at-tube heat exchangers is so small that frost-free
operation is possible, which would result in even larger savings.
An ACHX can be often used in the form of an A-frame in an air
cooling system, such as shown in Figs. 2326, 28, 33, 35 and 36. In
other words, when the ACHX is employed in the air-cooled tower,
the ACHX sometimes is required to be arranged obliquely to reduce
the space occupied. The air inlet angles have important effects on
Fig. 38. The distribution of Pareto optimal points for the two objective functions
using NSGA-II [134].
the thermal-hydraulic performances of the air-cooled FTHX

arranged obliquely. A partial enlarged gure of the apex angle (a)
of a surface type indirect air-cooled radiator and the representative
apex angles are shown in Fig. 39. So the investigations on the
thermal-hydraulic performances under different air inlet angles
are now underway in the authors group to obtain the experimental correlations and comparative results of the comprehensive
performance for different types of FTHXs.
Now three types of nned oval-tube heat exchanges samples
from the indirect air-cooling system were experimentally studied
by the authors group [136,137]. The picture of the main part of
773
Fig. 39. Partial enlarged gure of representative apex angles (a) of a surface type indirect air-cooled radiator.
(a) Experimental system
(b) Finned oval-tube heat exchanger [136]
Fig. 40. Pictures of the experimental system and a nned oval-tube heat exchanger.
the experimental system is shown in Fig. 40(a). There are heat

exchangers with double rows of tubes and two tube-passes [136]
shown in Fig. 40(b), double rows of tubes and one tube-pass
(HE1) and three rows of tubes and one tube-pass (HE2) [137],
respectively. All the materials of the tubes and ns are carbon steel.
The performances of four air inlet angles (h = a/2 = 90, 60, 45 and
30) were tested. The experimental correlations for the three
FTHXs under four different air inlet angles are obtained and the
comprehensive heat transfer performances under different air inlet
angles are compared based on three sets of identical criteria, considering both the heat transfer and pressure drop.
It is concluded that for heat exchangers with different structures positioned obliquely, whether the heat transfer performance
is improved or not is relied on not only their inclined angels, but
also their geometrical structures. The experimental correlations
and comparative results will be very useful for the researches
and engineers to research, design and apply the FTHXs under the
oblique arrangement conditions, especially in the indirect air-cooling tower. And it is also resulted that there must be an optimal air
inlet angle to obtain the best comprehensive heat transfer performance for a given FTHX obliquely employed in an indirect air-cooling tower, considering many factors, such as the heat transfer,
pump power and space saving [136,137].
In order to continue the previous work to acquire more experimental correlations on different types of FTHXs under different air
inlet angles, the experimental investigations of two samples in
indirect air-cooling tower are underway by the authors group,

which are a plain nned oval-tube heat exchanger and wavy nned
at-tube heat exchanger, both with aluminum ns in a cross ow
of air.
4.2. Industrial application of air-cooled heat exchanger
As mentioned above, ACHXs are used extensively in the
petro-chemical, process and power industries. The ACHX has the
advantage of lower comprehensive cost than water-cooled heat
exchanger and scale formation, no corrosion to equipment. So this
part will focus on describing the application of ACHXs employed in
the air cooling system.
4.2.1. Application in air-cooled condensers
As is known to all, the water consumption of the traditional
water-cooled coal-red power generating unit can be as high as
4.96 kg/kW h. In addition to the environmental impact, the rapid
increase of coal-red power generating units in the past decades
speeded the water shortage. In order to overcome the contradiction between the increasing demand of water for the coal-red
power generating unit and water shortage, ACC with obvious water
conservation benet were employed widely in power generating
units in north China from 2002 [138].
Table 8 shows the application data comparison of the indirect
air-cooling system and the ACC system. It can be seen that the heat
774
Table 8
Application data comparison of indirect air cooling-system and ACC system.*
Item
Unit
Tube type
Fin type
Working liquid in tube-side
Inlet temperature in tube-side
Outlet temperature in tube-side
Flow style of air
Heat transfer rate of ACHX unit
Heat transfer area of ACHX unit
Total heat transfer rate
Heat transfer coefcient
Total heat transfer area
Total manufacturing expense
*
C
C
kW
m2
kW
W m2 K1
m2
Yuan (RMB)
ACC system
Oval tube
Rectangular n
Water
65
55
Natural draft
144
2300
420,000
36.8
720,000
130,000,000
Flat tube
Wavy aluminum n
Steam
70
45
Forced draft
140
2500
420,000
30.5
800,000
120,000,000
The data is provided by Beijing Longyuan Cooling Technology Co., Ltd. of China (http://www.bjlylq.com.cn/engver/index.asp).
transfer coefcient the of the indirect ACHX is higher than that of

the ACC due to the smaller heat transfer area of the indirect aircooling system with oval tube under the same total heat transfer
rate. Meanwhile, the total manufacturing expense of the indirect
air-cooling system is 8.3% higher than that of the ACC system.
However, the air is natural draft in the indirect air-cooling system,
which can cut down much operating cost.
4.2.2. Application in indirect air-cooled heat exchangers

HTFS3-AC is the default method for air cooler geometries and
process conditions within the parameter ranges. It is completely
documented in the research report: RS1197. Table 9 shows the
results comparisons of the PFR method in the open literature with
the HTFS3-AC method. The pressure drop predictions of the HTFS3AC method are about 4% averagely lower than those of the general
HTFS3a method, and larger differences may be occurred relating to
the air velocity and temperature; While the heat transfer predictions of the former method are within 2% averagely of the latter
one, but larger differences occurred at the extremes of the air face
Table 9
Results comparisons of PFR method with HTFS3-AC method [139].
Parameters
Units
PFR method
HTFS3-AC
Area, required
Area, actual
Area ratio (Ac tula/Re qd.)
Pressure drop, outside
X-ow outlet Temperature
Pressure drop, tube side
Dp ratio tube side
Total price
Operating cost
Brake power for all fans
m2
m2
5586.5
5893.1
1.05
138
105.23
0.47234
0.94
115.733
17.058
35.538
5622.8
5893.1
1.05
138
106.53
0.47478
0.95
115.733
14,088
29.35
Pa
C
bar
Dollar (US)
Dollar (US)
kW
Table 10
Application data comparison of ACHX with different tubes.*
Item
Unit
Air-cooled heat exchanger
kW
m2
Oval tube
Rectangular
n
Water
144
2300
Flat tube
Wavy
n
Steam
140
2500
kW
W m2 K1
m2
420,000
36.8
720,000
420,000
30.5
800,000
Tube type
Fin type
Working liquid in tube-side
Heat transfer rate of the ACHX unit
Heat transfer area of the ACHX
unit
Total heat transfer rate
Heat transfer coefcient
Total heat transfer area
*
Indirect air-cooling system
The data is provided by Beijing Longyuan Cooling Technology Co., Ltd. of China
(http://www.bjlylq.com.cn/engver/index.asp).
velocity range. From Table 9 it can be seen that operating cost

saves approximately 20% using HTFS3-AC method [139].
Table 10 shows the application data comparison of ACHX with
different heat exchange tubes. It can be seen that the total heat
transfer area of the oval tube ACHX is 10% smaller than that of
the at tube ACHX to meet the same heat transfer capacity. And
the same result can be also concluded from Table 9.
5. Conclusions
(1) A general review of the shell-and-tube heat exchangers with
different types of helical bafes, such as discontinuous helical bafes, continuous helical bafes, and several combined
helical bafes, has been presented. Owing to the drawbacks
in principle of the SG-STHXs, industrial processes have
required a new generation of shell-and-tube heat exchangers which have better comprehensive performance and
exclude all disadvantages of the original SG-STHXs. From
the DCH-STHXs in an early stage to the CH-STHXs, then
the various combined helical bafes (DCH&CH-CSSP-STHXs,
CH&CH-CSSP-STHXs, CPSSP-STHXs, and CMSP-STHXs), the
heat transfer performance and comprehensive performance
have been increased continuously. Therefore it can be concluded that the helical bafed STHXs should be used to
replace the conventional segmental bafed STHXs in industries, despite there are a lot of research work have to be
done, especially on the novel combined helical bafes.
(2) A general review of the primary surface heat exchangers
with different types of primary surface plates, such as CC,
CW, CU, DN, and many complex 3D CC primary surface,
has been presented. The CC primary surface shows superior
performance over the CW, CU and DN primary surfaces
because it provides a small volume and weight of the heat
transfer matrix and is easier to be manufactured with small
passage dimensions. In recent years, they are developing
towards to the more complex 3D CC primary surfaces, such
as the double-wave CC primary surface, Offset-bubble primary surface, 3D anti-phase secondary corrugation. The
effect of longitudinal heat conduction in the primary surface
channels is signicant and should be taken into consideration to achieve very high heat transfer effectiveness. The
low Reynolds ke turbulent model is considered as the precise and effectiveness numerical method to simulation the
primary surfaces at Reynolds number higher than 1000.
Under the gas to gas heat transfer conditions in primary surface channels, the total heat transfer coefcient could be
separated combined with GA and least-squares method.
(3) A general review of the direct and indirect air-cooled heat
exchangers has been presented. Ambient wind plays a signicant part in the thermo-hydraulic performances of ACCs.
The extensions of the inner and outer walkways and elevation of the wind-break wall are effective ways to improve
the thermo-hydraulic performances and weaken the inlet
ow distortions and hot plume recirculation for ACCs. The
inuence of the inlet relative humidity on the sensible heat
transfer performance for ACHXs is small, and is almost negligible when the number of tube rows is above four. The
pressure drop under wet condition is much higher than that
under dry condition, while the difference in pressure drop
between wet and day condition is less profound for the wavy
n conguration as compared to that in plain n geometry.
The sensible heat transfer and mass transfer performance
are insensitive to changes of n pitch under fully and partially wet surface conditions for ACHXs. The indirect ACHX
is mostly used in the cooling system because it needs less
operating cost although it needs much capital investments.
The air inlet angle has signicant effect on the comprehensive performance of ACHXs. The whole performance of ACHXs in the air cooling system and the multi-objectives
optimization for ACHXs should be paid more attention, considering the heat transfer, pumper power, space usage and
other economic inuence factors.
Acknowledgements
This work is nancially supported by the China National Funds
for Distinguished Young Scientists (Grant No. 51025623), the International Cooperation and Exchanges Project of NSFC of China
(Grant No. 51120165002), the National Natural Science Foundation
of China (Grant No. 51276139) and the Fundamental Research
Funds for the Central Universities.
References
[1] Shah RK, Sekulic DP. Fundamentals of heat exchanger design. New
Jersey: John Wiley & Sons, Inc.; 2003.
[2] McDonald CF. Recuperator considerations for future higher efciency
microturbines. Appl Therm Eng 2003;23(12):146387.
[3] Kroger D. Air-cooled heat exchangers and cooling towers: thermal-ower
performance evaluation and design. Tulsa, Oklahoma, USA: PennWell
Corporation; 2004.
[4] Master BI, Chunangad KS. Heat exchanger, United States Patent,
US6827138B1, 2004.
[5] Kral D, Stehlik P, Van Der Ploeg HJ, Master BI. Helical bafes in shell-and-tube
heat exchangers, Part I: experimental verication. Heat Trans Eng.
1996;17(1):93101.
[6] Wang SL. Hydrodynamic studies on heat exchangers with helical bafes. Heat
Trans Eng 2002;23(3):439.
[7] Lutcha J, Nemcansky J. Performance improvement of tubular heat exchangers
by helical bafes. Chem Eng Proc 1990;68(3):26370.
[8] Wang QW, Chen GD, Chen QY, Zeng M. Review of improvements on shell-andtube heat exchangers with helical bafes. Heat Trans Eng
2010;31(10):83653.
[9] Zhang DJ, Wang QW, Luo LQ. Numerical simulation of ow performance in
shell side of shell-and-tube heat exchanger with discontinuous helical bafes,
Proceedings of the Second International Symposium on Thermal Science and
Technology, Beijing, China, 2005.
[10] Wang QW, Luo LQ, Zeng M, Wang L, Tao WQ, Huang YP. Shell-side heat
transfer and pressure drop of shell-and-tube heat exchangers with
overlap helical bafes. J Chem Indus Eng (China) 2005;56(4):598601 (in
Chinese).
[11] Zhang JF, Guo SL, Li ZZ, Wang JP, He YL, Tao WQ. Experimental performance
comparison of shell-and-tube oil coolers with overlapped helical bafes and
segmental bafes. Appl Therm Eng 2013;58(12):33643.
[12] Zhang JF, Li B, Huang WJ, Lei YG, He YL, Tao WQ. Experimental performance
comparison of shell-side heat transfer for shell-and-tube heat exchangers
with middle-overlapped helical bafes and segmental bafes. Chem Eng Sci
2009;64(8):164353.
[13] Taher FN, Movassag SZ, Razmi K, Azar RT. Bafe space impact on the
performance of helical bafe shell and tube heat exchangers. Appl Therm Eng
2012;44:1439.
[14] Van Der Ploeg HJ, Master BI, Helical bafes heat exchangers-design option for
reactor feed/efuent heat exchanger, Advances in Industrial Heat Transfer,
IChemE, Birmingham, UK, 1992.
775
[15] Lei YG, He YL, Chu P, Li R. Design and optimization of heat exchangers with
helical bafes. Chem Eng Sci 2008;63:438695.
[16] Zhang L, Xia Y, Jiang B, Xiao X, Yang X. Pilot experimental study on shell and
tube heat exchangers with small-angles helical bafes. Chem Eng Process:
Process Intensication 2013;69:1128.
[17] Xiao X, Zhang L, Li X, Jiang B, Yang X, Xia Y. Numerical investigation of helical
bafes heat exchanger with different Prandtl number uids. Int J Heat Mass
Trans 2013;63:43444.
[18] Leong KY, Saidur R, Khairulmaini M, Michael Z, Kamyar A. Heat transfer and
entropy analysis of three different types of heat exchangers operated with
nanouids. Int Commun Heat Mass Trans 2012;39:83843.
[19] He YL, Lei YG, Tao WQ, Zhang JF, Chu P, Li R. Second-law based
thermodynamic analysis of a novel heat exchanger. Chem Eng Technol
2009;32:8692.
[20] Liu L, Ling X, Peng H. Analysis on ow and heat transfer characteristics of EGR
helical bafed cooler with spiral corrugated tubes. Exp Therm Fluid Sci
2013;44:27584.
[21] Zhang ZG, Wu CS, Fang XM, Gao X, Wang ZY. Experimental Study of shell-side
heat transfer coefcient and pressure drop for an integrally helical bafed
heat exchanger combined with different enhanced tubes. Ind Eng Chem Res
2009;48:40404.
[22] Zhang ZG, Fang XM. Comparison of heat transfer and pressure drop for the
helically bafed heat exchanger combined with three-dimensional and twodimensional nned tubes. Heat Trans Eng 2006;27:1722.
[23] Zhang Z, Li Q, Xu T, Fang X, Gao X. Condensation heat transfer
characteristics of zeotropic refrigerant mixture R407C on single, threerow petal-shaped nned tubes and helically bafed condenser. Appl
Therm Eng 2012;39:639.
[24] Chen YP, Cao RB, Wu JF, Dong G, Sheng YJ. Experimental study on shell side
heat transfer performance of circumferential overlap trisection helical bafe
heat exchangers. In: Proceedings of the ASME international mechanical
engineering congress and exposition, Denver, Colorado, USA; November
2011.
[25] Chen YP, Sheng YJ, Dong C, Wu JF. Numerical simulation on ow eld in
circumferential overlap trisection helical bafe heat exchanger. Appl Therm
Eng 2013;50:103543.
[26] Zhang JF, He YL, Tao WQ. 3D numerical simulation on shell-and-tube heat
exchangers with middle-overlapped helical bafes and continuous bafes
Part I: Numerical model and results of whole heat exchanger with middleoverlapped helical bafes. Int J Heat Mass Trans 2009;52:537180.
[27] Chen GD, Wang QW, Zeng M. Numerical studies on a novel shell-and-tube
heat exchanger with combined helical bafes. In: Proceedings of the seventh
international conference on enhanced, compact and ultra-compact heat
exchangers: from microscale phenomena to industrial applications, Heredia,
Costa Rica; 2009.
[28] Jafari Nasr MR, Shafeghat A. Fluid ow analysis and extension of rapid design
algorithm for helical bafe heat exchangers. Appl Therm Eng
2008;28:132432.
[29] Andrews MJ, Master BI. Three-dimensional modeling of a Helixchanger heat
exchanger using CFD. Heat Trans Eng 2005;26:2231.
[30] Peng B, Wang QW, Zhang C, Xie GN, Luo LQ, Chen QY, et al. An experimental
study of shell-and-tube heat exchangers with continuous helical bafes. J
Heat Transfer Trans ASME 2007;129:142531.
[31] Wang QW, Chen QY, Peng BT, Zeng M, Luo LQ, Wu YN. A shell-and-tube heat
exchanger with continuous helical bafes, China Patent, ZL200510043033.5;
2005.
[32] Wang QW. Investigation on novel combined multi-shell-pass shell-andtube heat exchangers with helical bafes, Workshop on advances in
compact and micro heat exchangers for sustainable development, IIT
Delhi, India; 2008.
[33] Yang JF, Zeng M, Wang QW. A new type of liquid-suction heat exchanger for
an automotive air conditioning system. In: Proceedings of the second
international workshop on heat transfer advances for energy conservation
and pollution control, Xian, China, October, 2013.
[34] Zarrella A, Capozza A, Carli MD. Analysis of short helical and double U-tube
borehole heat exchangers: a simulation-based comparison. Appl Energy
2013;112:35870.
[35] Wu Z, Yang FS, Zhang ZX, Bao ZW. Magnesium based metal hydride reactor
incorporating helical coil heat exchanger: simulation study and optimal
design.
Appl
Energy
2014;130:71222.
http://dx.doi.org/10.1016/
j.apenergy.2013.12.071.
[36] Xie GN, Wang QW, Zeng M, Luo LQ. Heat transfer analysis for shell-and-tube
heat exchangers with experimental data by articial neural networks
approach. Appl Therm Eng 2007;27:1096104.
[37] Wang QW, Xie GN, Peng BT, Zeng M. Experimental study and geneticalgorithm-based correlation on shell-side heat transfer and ow performance
of three different types of shell-and-tube heat exchangers. J Heat Transfer
Trans ASME 2007;129:127785.
[38] Lei YG, He YL, Li R, Gao YF. Effects of bafe inclination angle on ow and heat
transfer of a heat exchanger with helical bafes. Chem Eng Process: Process
Intensication 2008;47:233645.
[39] Zhang ZG, Ma DB, Fang XM, Gao XN. Experimental and numerical heat
transfer in a helically bafed heat exchanger combined with one threedimensional nned tube. Chem Eng Process 2008;47:173843.
776
[40] Zhang ZG, Yu ZS, Fang XM. An experimental heat transfer study for helically
owing outside petal-shaped nned tubes with different geometrical
parameters. Appl Therm Eng 2007;27:26872.
[41] Zhang ZG, Xu T, Fang XM. Experimental study on heat transfer enhancement
of a helically bafed heat exchanger combined with three-dimensional nned
tubes. Appl Therm Eng 2004;24:2293300.
[42] Yang JF, Zeng M, Wang QW. Effects of sealing strips on shell-side ow and
heat transfer performance of a heat exchanger with helical bafes. Appl
Therm Eng 2014. http://dx.doi.org/10.1016/j.applthermaleng.2013.11.064.
[43] Chen GD, Zeng M, Wang Q. Experimental and numerical studies on shell-side
performance of three different shell-and-tube heat exchangers with helical
bafes. J Enhanc Heat Trans 2011;18:44963.
[44] Zhang JF, He YL, Tao WQ. 3D numerical simulation on shell-and-tube heat
exchangers with middle-overlapped helical bafes and continuous bafes
Part II: Simulation results of periodic model and comparison between
continuous and noncontinuous helical bafes. Int J Heat Mass Trans
2009;52:53819.
[45] Huang YQ, Yu XL, Lu GD. Numerical simulation and optimization design of
the EGR cooler in vehicle. J Zhejiang Univ Sci A 2008;9:12706 (in Chinese).
[46] Song S. Numerical studies on performance of double shell-pass heat
exchangers with sleeve tubes and continuous helical bafes. Proc Chinese
Soc Elect Eng 2013;33:827 (in Chinese).
[47] Wang QW, Yang JF, Chen GD, Zeng M. A combined single shell-pass shell-andtube heat exchanger with continuous helical bafes, China Patent,
ZL201210036769.X; 2012.
[48] Wang QW, Chen GD, Zeng M. A combined parallel multiple shell-pass shelland-tube heat exchanger with continuous helical bafes, China Patent,
ZL200910021347.3; 2009.
[49] Wang QW, Zhang DJ, Zeng M, Wu YN. A combined shell-and-tube heat
exchanger with helical bafes, China Patent, ZL200710017395.6; 2007.
[50] Wang QW, Chen QY, Zeng M, Wu YN, Gao Q. A combined multiple shell-pass
shell-and-tube heat exchangers with continuous helical bafes, China Patent,
ZL200710017478.5; 2007.
[51] Wang QW, Chen GD, Zeng M, Chen QY, Peng BT, Zhang DJ, et al. Shell-side
heat transfer enhancement for shell-and-tube heat exchangers with helical
bafes. Chem Eng Trans 2010;21:21722.
[52] Wang QW, Chen QY, Chen GD, Zeng M. Numerical investigation on combined
multiple shell-pass shell-and-tube heat exchanger with continuous helical
bafes. Int J Heat Mass Trans 2009;52:121422.
[53] Kuppat T. Heat exchanger design handbook. New York: Marcel Dekker Inc.;
2000.
[54] Stehlik P, Nemcansky J, Kral D, Swanson LW. Comparison of correction factors
for shell-and-tube heat exchangers with segmental or helical bafes. Heat
Trans Eng 1994;15:5565.
[55] Zhang JF, He YL, Tao WQ. A design and rating method for shell-and-tube heat
exchangers with helical bafes. J Heat Transfer Trans ASME 2010;132.
051802-18.
[56] Azar RT, Khalilarya S, Jafarmadar S. Tube bundle replacement for segmental
and helical shell and tube heat exchangers: experimental test and economic
analysis. Appl Therm Eng 2014;62:62232.
[57] Wang QW, Chen GD, Xu J, Ji YP. Second-law thermodynamic comparison and
maximal velocity ratio design of shell-and-tube heat exchangers with
continuous helical bafes. J Heat Transfer Trans ASME 2010:132.
[58] Shah RK. Compact heat exchangers for microturbines. in micro gas turbines:
2-1-2-18). Educational Notes RTO-EN-AVT-131, Paper 2. Neuilly-sur-Seine,
France: RTO. <http://www.rto.nato.int/abstracts.asp>;2005.
[59] McDonald CF. Low-cost compact primary surface recuperator concept for
microturbines. Appl Therm Eng 2000;20:47197.
[60] Stasiek J, Collins MW, Ciofalo M, Chew PE. Investigation of ow and heat
transfer in corrugated passages.1. Experimental results. Int J Heat Mass Trans
1996;39:14964.
[61] Stasiek JA. Experimental studies of heat transfer and uid ow across
corrugated-undulated heat exchanger surfaces. Int J Heat Mass Trans
1998;41:899914.
[62] Zhang LZ, Chen ZY. Convective heat transfer in cross-corrugated triangular
ducts under uniform heat ux boundary conditions. Int J Heat Mass Trans
2011;54:597605.
[63] Utriainen E, Sunden B. Evaluation of the cross corrugated and some other
candidate heat transfer surfaces for microturbine recuperators. ASME J Eng
Gas Turb Power 2002;124:55060.
[64] Du LX, Ma T, Zeng M, Guo ZX, Wang QW. Numerical investigations on
thermohydraulic performance of cross-wavy channels with multi-periodic
boundary conditions. Numer Heat Trans 2014;65:73249.
[65] Zhang L, Che DF. An experimental and numerical investigation on the
thermal-hydraulic performance of double notched plate. J Heat Transfer
Trans ASME 2012;134:091802.
[66] Khan TS, Khan MS, Chyu MC, Ayub ZH. Experimental investigation of single
phase convective heat transfer coefcient in a corrugated plate heat
exchanger for multiple plate congurations. Appl Therm Eng
2010;30:105865.
[67] Metwally HM, Manglik RM. Enhanced heat transfer due to curvature-induced
lateral vortices in laminar ows in sinusoidal corrugated-plate channels. Int J
Heat Mass Trans 2004;47:228392.
[68] Utriainen E, Sunden B. A numerical investigation of primary surface rounded
cross wavy ducts. Heat Mass Trans 2002;38:53742.
[69] Utriainen E, Sunden B. Numerical analysis of a primary surface trapezoidal

cross wavy duct. Int J Numer Method H 2000;10:63448.
[70] Zhang L, Che DF. Inuence of corrugation prole on the thermalhydraulic
performance of cross-corrugated plates. Numer Heat Trans 2011;59:26796.
[71] Wang QW, Zhang DJ, Xie GN. Experimental study and Genetic-AlgorithmBased correlation on pressure drop and heat transfer performances of a crosscorrugated primary surface heat exchanger. J Heat Transfer Trans ASME
2009;131:061802.
[72] Lin JH, Huang CY, Su CC. Dimensional analysis for the heat transfer
characteristics in the corrugated channels of plate heat exchangers. Int
Commun Heat Mass Trans 2007;34:30412.
[73] Zhang LZ. Numerical study of periodically fully developed ow and heat
transfer in cross-corrugated triangular channels in transitional ow regime.
Numer Heat Trans 2005;48:387405.
[74] Utriainen E, Sundn, B. A comparison of some heat transfer surfaces for small
gas turbine recuperators. Proceeding of ASME Turbo Expo 2001, 47 June,
2001, Louisiana, USA, ASME 2001-GT-0474.
[75] Kim M, Baik YJ, Park SR, Ra HS, Lim H. Experimental study on corrugated
cross-ow air-cooled plate heat exchangers. Exp Therm Fluid Sci
2010;34:126572.
[76] Liu XD, Liu HP, Zhou L. Numerical simulation of heat transfer in primary
surface with corrugations recuperators. World Acad Sci Eng Technol
2010;41:6216.
[77] Luo T, Ma T, Zeng M, Wang QW. Stress study on different designs of ceramic
high temperature heat exchangers. Chem Eng Trans 2012;29:66772.
[78] Ma T, Zeng M, Luo T, Chen YT, Wang QW. Numerical study on thermohydraulic performance of an offset bubble primary surface channels. Appl
Therm Eng 2013;61:4452.
[79] Blomerius H, Mitra NK. Numerical investigation of convective heat transfer
and pressure drop in wavy ducts. Numer Heat Trans 2000;37:3754.
[80] Ciofalo M, Stasiek J, Collins MW. Investigation of ow and heat transfer in
corrugated passages.2. Numerical simulations. Int J Heat Mass Trans
1996;39:16592.
[81] Doo JH, Ha MY, Min JK, Stieger R, Rolt A, Son C. An investigation of crosscorrugated heat exchanger primary surfaces for advanced intercooled-cycle
aero engines (Part-I: Novel geometry of primary surface). Int J Heat Mass
Trans 2012;55:525667.
[82] Doo JH, Ha MY, Min JK, Stieger R, Rolt A, Son C. An investigation of crosscorrugated heat exchanger primary surfaces for advanced intercooled-cycle
aero engines (Part-II: Design optimization of primary surface). Int J Heat Mass
Trans 2013;61:13848.
[83] Kim M, Ha MY, Min JK, Stieger R, Rolt A, Son C. Numerical study on the crosscorrugated primary surface heat exchanger having asymmetric crosssectional proles for advanced intercooled-cycle aero engines. Int J Heat
Mass Trans 2013;66:13953.
[84] Doo JH, Ha MY, Min JK, Stieger R, Rolt A, Son C. Theoretical prediction of
longitudinal heat conduction effect in cross-corrugated heat exchanger. Int J
Heat Mass Trans 2012;55:412938.
[85] Bu SS, Yang J, Zhou M, Li SY, Wang QW, Guo ZX. On Contact point
modications for forced convective heat transfer analysis in structured
packed bed of spheres. Nucl Eng Des 2014;270:2133.
[86] Zhang LZ. Turbulent three-dimensional air ow and heat transfer in a crosscorrugated triangular duct. J Heat Trans Trans ASME 2005;127:11518.
[87] Zhang L, Che DF. Turbulence models for uid ow and heat transfer between
cross-corrugated plates. Numer Heat Trans 2011;60:41040.
[88] Ngo TL, Kato Y, Nikitin K, Ishizuka T. Heat transfer and pressure drop
correlations of microchannel heat exchangers with S-shaped and zigzag ns
for carbon dioxide cycles. Exp Therm Fluid Sci 2007;32:56070.
[89] Tsuzuki N, Utamura M, Ngo TL. Nusselt number correlations for a
microchannel heat exchanger hot water supplier with S-shaped ns. Appl
Therm Eng 2009;29:3299308.
[90] Lagerstrom G, Xie M. High performance & cost effective recuperator for
micro-gas turbines. In: Proceedings of ASME turbo expo 2002, 36 June 2002
Amsterdam, The Netherlands: GT-2002-30402.
[91] Treece B, Vessa P, Mckeirnan R. Microturbine recuperator manufacturing and
operating experience. In: Proceedings of ASME turbo expo 2002, June 36,
2002, Amsterdam, The Netherlands: GT-2002-30404.
[92] Wang QW, Liang HX, Luo LQ, Wang JW, Huang ZP, Feng ZP, et al. Experimental
investigation on heat transfer and pressure drop in a microturbine
recuperator with cross-wavy primary surface channels. In: Proceedings of
ASME turbo expo 2005, June 69, 2005, Reno-Tahoe, Nevada, USA: GT200568255.
[93] Zhang DJ. Performance studies and conguration optimization of primary
surface heat exchangers. Doctor Dissertation, Xian Jiaotong University;
2008.
[94] GEA air n coolers, nned tube types. <www.gea-heatexchangers.com/
products/air-cooled-heat-exchangers/air-n-coolers/nned-tubes/> 25/4/25.
[95] Wang CC, Hwang YM, Lin YT. Empirical correlations for heat transfer and ow
friction characteristics of herringbone wavy n-and-tube heat exchangers. Int
J Refrig 2002;25:67380.
[96] Wang CC, Liaw JS, Yang BC. Airside performance of herringbone wavy nand-tube heat exchangers data with larger diameter tube. Int J Heat Mass
Trans 2011;54:10249.
[97] Wang CC, Liaw JS. Air-side performance of herringbone wavy n-and-tube
heat exchangers under dehumidifying condition data with larger diameter
tube. Int J Heat Mass Trans 2012;55:305460.

[98] Pongsoi P, Pikulkajorn S, Wang CC, Wongwises S. Effect of n pitches on the
air-side performance of crimped spiral n-and-tube heat exchangers with a
multipass parallel and counter cross-ow conguration. Int J Heat Mass Trans
2011;54:223440.
[99] Pirompugd W, Wang CC, Wongwises S. Finite circular n method for heat and
mass transfer characteristics for plain n-and-tube heat exchangers under
fully and partially wet surface conditions. Int J Heat Mass Trans
2007;50:55265.
[100] Liu MS, Wang CC, Liaw JS, Chang YJ. Fin-tube heat exchanger with vortex
generator. 2002, US6349761B1.
[101] Nagarajan V, Chen YT, Wang QW, Ma T. Hydraulic and thermal performances
of a novel conguration of high temperature ceramic plate-n heat
exchanger. Appl Energy 2014;113:589602.
[102] Krger DG. Air-cooled heat exchangers and cooling towers, vol. I. Tulsa,
Oklahoma: PennWell Corporation; 2004.
[103] Karimi IA, Rangaiah GP, Shastri SS, Yap C. Enhance air-cooled heat exchanger
performance. Hydrocarb Process. 2001, <www.hydrocarbonprocessing.com/
Article/2600224/Enhance-air-cooled-heat-exchanger-performance.html> 25/
4/2014.
[104] Pongsoi P, Pikulkajorn S, Wang CC, Wongwises S. Effect of number of tube
rows on the air-side performance of crimped spiral n-and-tube heat
exchanger with a multipass parallel and counter cross-ow conguration.
Int J Heat Mass Trans 2012;55:140311.
[105] Xie GN, Wang QW, Sunden B. Parametric study and multiple correlations on
air-side heat transfer and friction characteristics of n-and-tube heat
exchangers with large number of large-diameter tube rows. Appl Therm
Eng 2009;29:116.
[106] Tang LH, Zeng M, Wang QW. Experimental and numerical investigation on
air-side performance of n-and-tube heat exchangers with various n
patterns. Exp Therm Fluid Sci 2009;33:81827.
[107] Tang LH, Zeng M, Xie GN, Wang QW. Fin pattern effects on air-side heat
transfer and friction characteristics of n-and-tube heat exchangers with
large number of large-diameter tube rows. Heat Trans Eng 2009;30:17180.
[108] Wang QW, Chen QY, Wang L, Zeng M, Huang YP, Xiao ZJ. Experimental study
of heat transfer enhancement in narrow rectangular channel with
longitudinal vortex generators. Nucl Eng Des 2007;237:68693.
[109] Zeng M, Tang LH, Lin M, Wang QW. Optimization of heat exchangers with
vortex-generator n by Taguchi method. Appl Therm Eng 2010;30:177583.
[110] Xie GN, Sunden B, Wang QW, Tang LH. Performance predictions of laminar
and turbulent heat transfer and uid ow of heat exchangers having large
tube-diameter and large tube-row by articial neural networks. Int J Heat
Mass Trans 2009;52:248497.
[111] Xie GN, Wang QW, Sunden B. Application of a genetic algorithm for thermal
design of n-and-tube heat exchangers. Heat Trans Eng 2008;29:597607.
[112] Xie GN, Sunden B, Wang QW. Optimization of compact heat exchangers by a
genetic algorithm. Appl Therm Eng 2008;28:895906.
[113] Website: <sipep.energy.siemens.com/manual/backcoolingsystem.htm> 25/4/
2014.
[114] Yang LJ, Du XZ, Yang YP. Wind effect on the thermo-ow performances and
its decay characteristics for air-cooled condensers in a power plant. Int J
Therm Sci 2012;53:17587.
[115] Yang LJ, Du XZ, Yang YP. Inuences of wind-break wall congurations upon
ow and heat transfer characteristics of air-cooled condensers in a power
plant. Int J Therm Sci 2011;50:205061.
[116] Duvenhage K, Krger D. The inuence of wind on the performance of forced
draught air-cooled heat exchangers. J Wind Eng Ind Aerod 1996;62:25977.
[117] Yang LJ, Du XZ, Yang YP. Space characteristics of the thermal performance for
air-cooled condensers at ambient winds. Int J Heat Mass Trans
2011;54:310919.
777
[118] He W, Dai Y, Li M, Ma Q. Inuence mechanism on ow and heat transfer

characteristics for air-cooled steam condenser cells. Heat Mass Trans
2012;48:166374.
[119] Du XZ, Feng LL, Yang YP, Yang LJ. Experimental study on heat transfer
enhancement of wavy nned at tube with longitudinal vortex generators.
Appl Therm Eng 2013;50:5562.
[120] Nicholson M. Air heat exchanger categories. 2013, <airheatexchanger.co.uk/
2013/04/04/air-heat-exchanger-categories/> 25/4/2014.
[121] Du XZ, Hu HM, Shen YQ, Yang LJ, Yang YP. Reduced order analysis of ow and
heat transfer for air-cooled condenser of power generating unit. Appl Therm
Eng 2013;51:38392.
[122] Gao XF, Zhang CW, Wei JJ, Yu B. Performance prediction of an improved aircooled steam condenser with deector under strong wind. Appl Therm Eng
2010;30:26639.
[123] Wang QW, Zhang DJ, Zeng M, Lin M, Tang LH. CFD simulation on a thermal
power plant with air-cooled heat exchanger system in north China. Eng
Comp 2008;25:34265.
[124] Yang L, Tan H, Du X, Yang Y. Thermal-ow characteristics of the new wavenned at tube bundles in air-cooled condensers. Int J Therm Sci
2012;53:16674.
[125] Duvenhage K, Vermeulen JA, Meyer CJ, Krger DG. Flow distortions at the fan
inlet of forced-draught air-cooled heat exchangers. Appl Therm Eng
1996;16:74152.
[126] Meyer CJ, Krger DG. Numerical investigation of the effect of fan performance
on forced draught air-cooled heat exchanger plenum chamber aerodynamic
behaviour. Appl Therm Eng 2004;24:35971.
[127] Meyer CJ, Krger DG. Air-cooled heat exchanger inlet ow losses. Appl Therm
Eng 2001;21:77186.
[128] Meyer CJ, Krger DG. Plenum chamber ow losses in forced draught aircooled heat exchangers. Appl Therm Eng 1998;18:87593.
[129] Meyer CJ. Numerical investigation of the effect of inlet ow distortions on
forced draught air-cooled heat exchanger performance. Appl Therm Eng
2005;25:163449.
[130] Hotchkiss PJ, Meyer CJ, von Backstrm TW. Numerical investigation into the
effect of cross-ow on the performance of axial ow fans in forced draught
air-cooled heat exchangers. Appl Therm Eng 2006;26:2008.
[131] Buzzoni L, DallOlio R, Spiga M. Efciency of the unit cell in rectangular nned
tube arrangements. Appl Therm Eng 1999;19:114756.
[132] Kuntysh VB, Sykhotskii AB, Piir AE. Heat transfer and resistance of staggered
banks of air-cooled heat exchangers formed from tubes with rolled
aluminum ns of various height. Chem Petrol Eng 2011;46:71321.
[133] Salimpour M, Bahrami Z. Thermodynamic analysis and optimization of aircooled heat exchangers. Heat Mass Trans 2011;47:3544.
[134] Alinia Kashani AH, Maddahi A, Hajabdollahi H. Thermal-economic
optimization of an air-cooled heat exchanger unit. Appl Therm Eng
2013;54:4355.
[135] Stignor CH, Sundn B, Fahln P. Energy-efcient at-tube heat exchangers for
indirectly cooled display cabinets. Int J Refrig 2009;32:146071.
[136] Du XP, Zeng M, Wang QW, Dong Z. Experimental investigation of heat
transfer and resistance characteristics of a nned oval-tube heat exchanger
with different air inlet angles. Heat Trans Eng 2014;35:70310.
[137] Du XP, Zeng M, Dong ZY, Wang QW. Experimental study of the effect of air
inlet angle on the air-side performance for cross-ow nned oval-tube heat
exchangers. Exp Therm Fluid Sci 2014;52:14655.
[138] Ge ZH, Du XZ, Yang LJ, Yang YP, Li YL, Jin YS. Performance monitoring of direct
air-cooled power generating unit with infrared thermography. Appl Therm
Eng 2011;31:41824.
[139] New Airside Methods for Heat Transfer and Pressure Drop. <www.htfs.org/
newsletters/sep08.htm> 25/4/2014.

Articulo

Uploaded by

Copyright:

Available Formats

Articulo

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Articulo

Uploaded by

Copyright:

Available Formats

Applied Energy 135 (2014) 748777

Contents lists available at ScienceDirect

Recent development and application of several high-efciency surface

Corresponding author. Tel./fax: +86 (29)82665539.

exchangers are widely used in many energy conversion and

Q. Wang et al. / Applied Energy 135 (2014) 748777

amplitude of waviness (mm)

The gasgas heat exchangers are necessary in many waste heat

air-cooled systems. What is more, the typical applications now or

Q. Wang et al. / Applied Energy 135 (2014) 748777

Fig. 1. Different types of shell-and-tube heat exchangers with helical bafes.

known as HELIXCHANGERS has been increasingly adopted in

in thermal performance versus perfect plug ow, because local

Q. Wang et al. / Applied Energy 135 (2014) 748777

2.2. Improvements on bafe structure of helixchangers

numerical simulations to investigate the ow and heat transfer

Fig. 2. Structure of circumferential overlap trisection helical bafed heat exchanger

Q. Wang et al. / Applied Energy 135 (2014) 748777

Lei et al. [15]

Single discontinuous helical bafe

Zhang et al. [12]

70 < Res < 650

Middle-overlapped helical bafe

100 < Res < 800

Discontinuous helical bafe

Discontinuous helical bafe

125,000 < ResPr < 500,000

Zhang et al. [11]

Liu et al. [20]

3000 < Ret < 9010

Zhang et al. [23]

6 < q < 16 kWm2

Zhang et al. [16]

65,000 < Res < 180,000

Discontinuous helical bafes

15,000 < Res < 40,000

Zhang and Fang [22]

Zhang et al. [21]

3000 < Ret < 9010

3000 < Res < 90,000

Fig. 3. Fluid ow eld structure of STHXs with continuous helical bafes.

Q. Wang et al. / Applied Energy 135 (2014) 748777

CH-STHX [33]. Two more examples of successful applications of

Peng et al. [30]

Continuous helical bafe

Continuous helical bafe

Zhang et al. [41]

Continuous helical bafe with petal-shaped nned tubes

Zhang et al. [39,40]

Continuous helical bafe with smooth tube

10,000 < Res < 20,000

Wang et al. [37]

752 < Res

1960 < Res

Q. Wang et al. / Applied Energy 135 (2014) 748777

Fig. 6. Photograph of the CH&CH-CSSP-STHXs used in oil rening.

2.3.2. Maximal velocity design method for continuous helical bafes

and optimizing of helixchangers, pitch angle, bafes arrangement,

Q. Wang et al. / Applied Energy 135 (2014) 748777

6 < q < 16 kWm2