5 (2020)

Download as pdf or txt
Download as pdf or txt
You are on page 1of 11

International Journal of Thermal Sciences 155 (2020) 106417

Contents lists available at ScienceDirect

International Journal of Thermal Sciences


journal homepage: http://www.elsevier.com/locate/ijts

Design optimization of a shell-and-tube heat exchanger with novel


three-zonal baffle by using CFD and taguchi method
Nihat Biçer a, Tahsin Engin a, *, Halit Yaşar a, Ekrem Büyükkaya a, Ahmet Aydın b, Adnan Topuz c
a
Sakarya University, Engineering Faculty, Mechanical Engineering Department, Sakarya, Turkey
b
Sakarya University, Research-Development&Application Center-SARGEM, 54200, Sakarya, Turkey
c
Bülent Ecevit University, Faculty of Engineering, Department of Mechanical Engineering, 67100, Incivez, Zonguldak, Turkey

A R T I C L E I N F O A B S T R A C T

Keywords: In this study, a novel and innovative baffle design was offered in order to considerably reduce shell side pressure
Shell-and-tube heat exchanger loss without compromising thermal performance. Computational fluid dynamics (CFD) was utilized to simulate
Optimization and visualize 3-D turbulent flow field in the shell side so as to investigate various shapes of baffles for preliminary
Three-zonal baffle
baffle design purposes. The simulation results showed that a so-called three-zonal baffle could be superior over
Taguchi method
the several other configurations considered. The set of design parameters was then identified for this shape of
baffle and Taguchi method was employed to determine candidate design configurations for optimum. With the
optimized design of shell-and-tube heat exchanger (STHE) with new baffle configuration, it was found that
thermal performance of the heat exchanger with three-zonal baffles was slightly improved, whereas shell-side
pressure drop was significantly decreased compared to the conventional baffled STHE. The shell side pressure
loss was found to lower by 49%, accompanying an increase in the shell side temperature difference up to 7%. In
addition, CFD analyses of the optimized STHE with three-zonal baffles were performed considering specific
boundary conditions, and the results were validated with the experimental data obtained under the same con­
ditions. The results showed that the differences between CFD analyses and experimental data were maximum
7.3% for heat transfer rate and 7.6% for the pressure drop. It was concluded that the three-zonal baffles improved
the STHE performance in terms of both heat transfer rate and pressure loss points of view.

The tube used in the heat exchanger plays an important role in en­
ergy transfer. For this reason, many research studies have been carried
1. Introduction
out for heat exchanger tube developed by using heat improvement
techniques. The efforts to improve the performance of heat exchangers
Heat exchangers are devices used for transferring thermal energy
are still in progress [4–6]. The development of heat transfer in the heat
between a solid object and a fluid, or between two or more fluids. The
exchanger using different geometric models still maintains the agenda.
fluids may be separated by a solid wall to prevent mixing or they may be
The only thing targeted in all the different geometries used is to change
in direct contact. They are widely used in space heating, refrigeration,
the physical behavior of the fluid flow to increase the heat transfer.
air conditioning, power stations, petrochemical, chemical, and phar­
Although there are different geometric shapes of heat exchangers, the
maceutical industries, natural gas processing and wastewater treatment
STHE has more application areas than the others due to the wide range
[1]. Among these, STHEs are the most commonly used ones. In this
of operating temperature and pressure [4–7]. There are many studies in
system, heat transfer performance depends on many parameters such as
the literature to improve the performance of STHEs. An important part
layout of tubes on the tubesheet, number of baffles, number of tubes and
of these studies focuses on baffle design. For this purpose, many in­
length. It is possible to improve the performance of a heat exchanger by
novations such as new baffle design, new baffle configuration have been
changing baffle geometry. Changing the baffle geometry has significant
carried out. Examples of some new types of baffles that are being studied
effects on the flow characteristics and heat transfer on the shell side. The
to improve the performance of Shell and tube heat exchangers are
traditional STHE with segmental baffles are described by high pressure
trefoil-hole baffle [8–10], helical baffle [11–22], flower baffle [8,16,23,
drop, leakage flow in large amount, stagnant flow zones, becoming dirty
24], staggered baffle [23,25], trapezoidal baffle [26], ladder-type fold
and flow induced vibration at high speeds [2,3].

* Corresponding author. University of Sakarya, Engineering Faculty, Mechanical Engineering Department, 54187, Esentepe Campus Serdivan-Sakarya, Turkey.
E-mail address: engint@sakarya.edu.tr (T. Engin).

https://doi.org/10.1016/j.ijthermalsci.2020.106417
Received 10 September 2019; Received in revised form 24 March 2020; Accepted 8 April 2020
Available online 30 April 2020
1290-0729/© 2020 Elsevier Masson SAS. All rights reserved.
N. Biçer et al. International Journal of Thermal Sciences 155 (2020) 106417

Taguchi method in heat exchangers. Gunes et al. [36] applied the


Nomenclature Taguchi optimization method for copper coil heat exchanger design
parameters. They used this method to determine the optimization of
A distance between the baffles (mm) heat transfer with minimum pressure drop. Similarly, Chamoli [37] also
B rotation angle of the baffle (degree) worked on the optimization of flow and geometric parameters in a
C ratio of outer diameter to inner diameter of the baffle rectangular channel roughened with V geometry baffles using Taguchi
cp specific heat of the fluid (kJ/kg-K) for the same purpose. The selected parameters for performance predic­
CFD Computational Fluid Dynamics tion of V down perforated baffle roughened rectangular channel were
D angle of openness to center (degree) relative roughness pitch, relative roughness height, open area ratio and
E ratio of openness to closure Reynolds number. Tang et al. [38] investigated the effects of the vortex
m_ mass flow rate (kg/s) generator fin-tube heat exchanger parameters have been optimized with
Q_ heat transfer rate (kW) the Taguchi method. The levels of each factor were combined to form
K coverage factor sixteen models and analyze the heat transfer and flow friction charac­
SHTE Shell-and-Tube Heat Exchanger teristics of each model. Another similar study was done by Zeng et al.
S/N Signal/Noise ratio They examined the parameters of vortex-generator fin-and-tube heat
S2 variance of observed values exchangers, such as attack angle, length of vortex generator, height of
ΔT temperature difference (K) vortex generator, fin material, fin thickness, fin pitch and tube pitch by
U uncertainty (%) the Taguchi method [39]. Aghaie et al. [40] investigated on optimized
y average of observed values geometry of angled ribs for enhancing the thermohydraulic behavior of
yi performance characteristics of the performance a solar air heater channel by Taguchi method. They were used L16
orthogonal array to optimize the geometry factors accounts for the
Subscript maximum thermal performance of the ribbed channel. Tingting et al.
Ins instrument [41] investigated the influence of helix angle, overlap size, diameter of
rep repeatability tube, central distance of tube and tube layout on the performance of
overlapped helical baffled heat exchangers using Taguchi method.
Sivasakthivel et al. [42] used Taguchi method to carry out the para­
metric optimization for heating or cooling mode operation. Etghani and
baffles [27], clamping baffle [28,29], round rod baffle [28,30] and Baboli [43] investigated heat transfer coefficient and exergy loss. In
louwer baffle [31]. Leoni et al. [32] investigated the effect of baffle their study, four design parameters (pitch coil, tube diameter, hot and
clearances on a shell and tube heat exchanger performance. Their results cold flow rates) were considered and Taguchi method were used to
showed that the pressure drop was about 40% smaller when clearances obtain the optimum levels of the design factors. Zhang et al. [44]
were considered. Mellal et al. [33] investigated a three-dimensional investigated the effect of the structural parameters of three-dimensional
numerical simulation of turbulent fluid flow and heat transfer in the finned tubes on the heat transfer and pressure drop characteristics in the
shell side of a shell and tube heat exchanger. They tested two primordial air cross-flow by the Taguchi method to improve the heat transfer effi­
parameters: baffles spacing of 106.6, 80 and 64 mm and six baffles ciency. Chamoli et al. [45] used Taguchi grey relational analysis method
orientation angles of 45� , 60� , 90� , 120� , 150� and 180� . Their results approach for the multi response optimization of different geometrical
showed that the baffle orientation angle of 180, at 64 mm of baffle and flow parameters on thermo-hydraulic performance of heat
spacing was the best design. Arania and Moradi [34] focused on the fluid exchanger tube with perforated disk insert. They concluded that the
flow and heat transfer of water inside the segmental baffle shell and tube most significant parameter with respect to grey relational analysis was
heat exchanger optimization using combined baffle and longitudinal diameter ratio. Miansari et al. [46] optimized a helically grooved shell
ribbed tube configuration. At maximum mass flow rates (2 kg/s), the and tube heat exchanger by using Taguchi experimental design method.
average value of shell-side heat transfer coefficient of disk baffle shell They investigated the performance of the heat exchanger for different
and tube heat exchanger and combined segmental-disk baffle shell and working conditions, such as different level of cold fluid inlet tempera­
tube heat exchanger were 26.6% and 31.9% higher than disk baffle shell ture, cold fluid flow rate, and groove height. They concluded that the
and tube heat exchanger with longitudinal circular ribbed tube and optimum groove height was 10 mm.
combined segmental-disk baffle shell and tube heat exchanger with In this study, it is aimed to improve the performance of shell-and-
longitudinal circular ribbed tube, respectively. tube heat exchangers by using a new baffle design. For this aim, a
Baffle range and baffle gap are an important parameter in the design three-zonal baffle has been proposed to use in shell and tube heat ex­
of the STHEs. If the baffle spaces are left larger or smaller than the op­ changers. As a result of the investigations, it was found that there were
timum design, they usually result in large vortices of poorly distributed no studies performed using this type of baffle in the literature. Using
flow, dead zones, and higher pressure losses than expected. Determi­ these new three-zonal baffles, it is ensured that the fluid in the shell-side
nation of these parameters by experimental methods also causes a sig­ is more effectively mixed by creating a propeller effect leading to rela­
nificant amount of material waste. In addition, it is very difficult to tively higher convective heat transfer coefficient. Thus, it is intended to
investigate the effect of each parameter on heat conduction, flow direct the fluid towards the shell walls by striking the baffles. The heat
resistance and thermo-hydraulic performance on the shell side. There­ exchanger with three-zonal baffles was then optimized by using the
fore, some optimization methods are used to reduce the number of ex­ Taguchi experimental design method considering the heat transfer rate
periments and their costs. One of them is Taguchi optimization method. and pressure drop. The optimized heat exchanger was then manufac­
Taguchi method is an experimental design method based on parameter, tured and tested. The experimental results were compared to CFD re­
system and tolerance design. This method is widely used in statistical sults, and it was found that a good agreement between measured and
analysis of data collected within the scope of quality assurance systems. simulated results exist. Overall, the results showed that the STHE with
However, it is also a very useful method to determine the optimal new baffle design lead to considerable lower pressure drop in the shell-
combination between different levels of different parameters. Thus, it is side with an improved thermal performance.
possible to reach a much less number of experiments by using the
Taguchi method in cases where too much experimental work is required
2. CFD analyses
to determine the effect of each parameter [35].
In the literature, there are some studies performed by applying
In this study, a new type three-zonal baffle has been developed for

2
N. Biçer et al. International Journal of Thermal Sciences 155 (2020) 106417

the STHEs. CFD analyzes of the heat exchanger with three-zonal baffle
were then performed and the obtained results were compared to the
results of the heat exchanger with conventional baffles. CFD analyses
were performed for two different mass flow rates, one low and one high.
In the first case, the mass flow rates of hot water passing through the
tubes, and mass flow rates of cold water passing through the shell were
taken as 0.5 kg/s and 0.25 kg/s, respectively. On the other hand, in the
second case, the mass flow rates of hot and cold water were taken as
2.88 kg/s and 2.19 kg/s, respectively. The inlet temperatures of hot and
cold water were taken as 77 � C and 15 � C, respectively. Table 1 shows
the properties of the fluids flowing through the heat exchanger.
The flow geometry was modeled with the separate ANSYS Design
Fig. 1. The heat exchangers with conventional and three-zonal baffles.
Modeler for the conventional and three-zonal baffle model. In these
models, two separate control volumes were modeled to analyze shell-
pressure drop and increase heat transfer are determined and the
and-tube side flows. The heat exchangers with three-zonal and con­
orthogonal arrays developed by Taguchi are used while blocking the
ventional baffles are shown in Fig. 1. On the other hand, geometric di­
parameters. At the same time, the noise ratio (S/N - Signal/Noise) an­
mensions of the conventional and three-zonal baffles are given in Fig. 2.
alyses are performed. There are three different convenient functions
In this study, a small STHE for CFD simulations was modeled. A
known as Taguchi loss function and also expressed as a function of noise
commercial CFD package, namely ANSYS Fluent was used for numerical
ratio (S/N, Signal/Noise). These are cases where the performance
computations. Since the flow in the tube-side is well-established both
characteristic is called the least favorable result;
experimentally and theoretically, the current study focused on the tube-
, !
side flow. Table 2 gives the heat exchanger specifications. 1X n

The standard k-ε turbulence model was used in the CFD analyses. S N ¼ 10 log yi 2 (1)
n i¼1
Tetrahedral elements were used and the number of elements was taken
as 7,045,950. Fig. 3. Shows element number independency of the nu­ In the case of the highest value is the best:
merical solution based on pressure drop and temperature difference. As , !
can be seen from the figure, when the number of elements is increased 1X n
1
S N ¼ 10 log (2)
by more than 7,045,950, there is no change in the values obtained from n i¼1 yi 2
the analyses. The overall view of the mesh structure is shown in Fig. 4. In
order to model the tube surface fouling resistance, thermal conductivity When the nominal value is the best:
was taken as 3.36 W/m-K at the interface. The simulations were per­ � � 2�
y
formed on a DELL T5600 Workstation (Intel® Xeon®, 3.30 GHz, 2 S N ¼ 10 log 2 (3)
S
processors, 16 cores, 128 GB RAM). The solution time was about 18 h for
each solution. P P
where y ¼ 1n ni¼1 yi , S2 ¼ n 1 1 ni¼1 ðyi yÞ2 , yi is the performance char­
acteristic of performance, n is number of test in trial, y is average of
3. Taguchi optimization
observed values, and S2 is the variance of observation value.
In this study, the three-zonal baffle was optimized to provide the
In this study, Taguchi Experimental Design method was used to
maximum heat transfer rate and the minimum pressure drop. While
optimize the STHE with new design three-zonal baffles. Detailed infor­
designing the geometry of the three-zonal baffle, five factors, and four
mation about Taguchi method is available elsewhere in the literature
levels of these factors were considered. Table 3 gives these factors and
and therefore we will not provide here any mathematical background of
corresponding levels. Cross-sectional view of the three-zonal baffle is
this method. Therefore, this section is more concerned with the adap­
shown in Fig. 5.
tation of the Taguchi method to the study.
When these factors and their levels were taken into consideration, it
Experimental optimization of the STHEs by Taguchi method consists
was decided that the most appropriate orthogonal array was L16
of three main steps: System design, parameter design and tolerance
sequence. The order of the experiments according to this orthogonal
design. In the system design step, it is aimed to determine the parameter
array is shown in Table 4.
values affecting the performance characteristics by designing the STHE.
During this step, layout of the tubes, number of baffles, number of tubes,
4. Experimental studies
length, the distance between baffles, and cross section of the baffle are
defined. In the parameter design step, the best level of parameters are
4.1. Experimental setup and test procedure
defined to optimize the heat exchanger. At this stage, factors that reduce
The experimental studies were carried out on a STHE which has the
geometric dimensions obtained from the optimization. The tube-side
Table 1
flow was supported by a frequency converter pump with a closed
Properties of the fluids in the heat exchanger.
loop. The hot water tank was heated by electrical heaters in order to
Parameter Hot Water Cold Water keep the temperature constant. A continuous stirrer helped to maintain
(Tube side) (Shell side)
the liquid temperature at a constant value for a given test flow rate. On
Mean fluid temperature 338 346 294 298 the other hand, the cold water was controlled with the frequency con­
(K)
verter pump and the heated water was evacuated out in a tank. Flow and
Mass flow rates (kg/s) 0.50 2.88 0.20 2.19
Density (kg/m3) 980.4 977.52 998.0 997.0
temperature control were done at the heat exchanger inlet and outlet
Specific heat (kJ/kg.K) 4.187 4.191 4.182 4.180 points. The experimental setup is shown in Fig. 6. During the experi­
Thermal conductivity 0.659 0.665 0.598 0.607 ments, flow rate and temperature control were done with the control
(W/m.K) 0.444 � 0.399 � 1.004 � 0.894 � panel and necessary controls were provided.
Kinematic viscosity 10 6 10 6 10 6 10 6
The basic elements used in the experimental setup are hot and cold
(m2/s)
Prandtl Number 2.75 2.30 7.01 6.14 water tanks, STHE and control panel. The system also includes globe

3
N. Biçer et al. International Journal of Thermal Sciences 155 (2020) 106417

Fig. 2. Geometric dimensions of the conventional and three-zonal baffles.

When the heater capacities were 15 kW, the analyses were performed for
Table 2
low flow rates to ensure the stability of the temperature. Table 5 presents
Specifications of the heat exchanger.
the conditions for hot and cold fluids entering the heat exchanger. Heat
Shell diameter 161.5 mm transfer rate was calculated by using the following equation;
Tube outlet diameter 17.2 mm
Tube layout and distance between the tubes Triangle, 22 mm Q_ ¼ mc
_ p ΔT (4)
Number of tubes 37
Heat exchanger length 1356.5 mm where, Q_ is heat transfer rate, m_ is mass flow rate, cp is specific heat of
Central baffle distance 193.5 mm
the fluid, and ΔT is temperature difference.
Number of baffle 6
Tube-to-baffle clearance 0.4
Shell-to-baffle clearance 0 4.2. Uncertainty analysis
Shell-to-bundle clearance 12.3
The measuring process of parameters such as mass flow rate and
temperature always have some errors and these lead to an uncertainty in
experimental data. The thermocouples and flow meter outputs are used
to calculate experimental heat transfer rate. In this study, experimental
uncertainties were calculated by Holman [47] method. The following
equations were used to calculate the uncertainty. The uncertainty is
consists of two parts: the uncertainty of instruments (uQ;_ Ins ) and the un­
certainty of repeatability (uQ;_ Rep ).

U Q_ ¼ K � uQ_ (5)
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi�ffiffiffiffi
(6)
2
uQ¼
_ ðuQ;Ins _ Þ2 þ uQ;Rep _

sffi�
ffiffiffiffiffiffiffiffiffiffiffiffiffiffi�
ffiffiffiffiffiffiffiffiffiffiffi�
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi�
ffiffiffiffiffi
2 2
∂Q_ ∂Q_
uQ;_ Ins ¼ um_ þ uΔT (7)
∂m_ ∂ΔT
sffi�
ffiffiffiffiffiffiffiffiffiffiffiffiffiffi�
ffiffiffiffiffiffiffiffiffiffiffi�
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi�
ffiffiffiffiffi
2 2
∂Q_ ∂Q_
uQ;_ Rep ¼ um_ þ uΔT (8)
∂m_ ∂ΔT
Fig. 3. Element number independency of the numerical solution.

where uQ;_ Ins is the uncertainty of instruments, uQ;_ Rep is the uncertainty of
valves, manometers to measure the pressure differences of the fluids
repeatability, u is the contribution of the uncertainty in the results from
entering and exiting the heat exchanger, and PT100 thermocouples for
parameters m_ and ΔT, K is the coverage factor and was considered as 2 in
measuring the temperature of the hot and cold fluids, and two pumps to
this case. Table 6 shows the measurement ranges, measurement accu­
circulate two streams. Before getting the experimental data, the valve in
racy of the measuring devices used in the experimental setup, and un­
the tube from which the water came from was opened and the system
certainty levels of the calculated parameters based on experimental
was expected to be filled completely. Then the cold water outlet valve
data.
was opened and the control panel provided hot and cold pumping at the
desired flow rate. After a certain period of time, the system became
stable and the necessary measurement results were taken.
Experiments were performed for seven different input conditions.

4
N. Biçer et al. International Journal of Thermal Sciences 155 (2020) 106417

Fig. 4. General view of the mesh structure.

Fig. 8 shows the pressure distributions on heat exchangers having


Table 3 conventional and three-zonal baffles. The figures show that the three-
The design factors and their levels.
zonal baffles give a much more uniform pressure distribution
Level compared to the conventional baffle. It is seen from the figure that when
Parameters 1 2 3 4 three-zonal baffles are used, pressures decreases dramatically compared
A: Distance between baffles (mm) 250 316 400 420
to conventional baffles.
B: Rotation angle of the baffle (degree) 0 60 120 180 Fig. 9 shows velocity streamlines in heat exchangers with conven­
C: The ratio of outer diameter to inner diameter 1.15 1.11 1.08 1.04 tional and three-zonal baffle. As can be seen from the figures, in the heat
(Do/Di) exchanger having conventional baffles, recirculation zones are formed
D: Angle of openness to center (degree) 100 103 105 106
at the rear of the baffles. In the conventional baffle, the fluid is forced to
E: The ratio of openness to closure (So/Sc) 2.52 2.85 3.36 4.11
change direction at an angle of about 90� in front of each baffle on the
shell side. The purpose of this is to forward the flow streams on the shell
5. Results and discussion side perpendicular onto the tube bundle in order to maximize the forced
convective heat transfer coefficient. This increases the heat transfer
In this study, CFD analyzes were performed in two different mass coefficient by a certain amount, but also leads to large pressure drop,
flow rates. One of these mass flow rates was determined as low mass causes the formation of recirculation zones at the junction of the baffle
flow rate (0.5 kg/s), and the other as high mass flow rate (2.88 kg/s). In
Figs. 7–9, temperature, pressure, and velocity distributions obtained Table 4
from CFD analyses for mass flow rate of 0.5 kg/h were given. On the The order of experiments according to the orthogonal array.
other hand, the numerical results obtained from the CFD analyses for
Experiment number A B C D E
both mass flow rates were presented and compared In Table 7.
1 1 1 1 1 1
2 1 2 2 2 2
5.1. Results of the CFD analyses 3 1 3 3 3 3
4 1 4 4 4 4
Fig. 7 shows the temperature distributions on tube surfaces in heat 5 2 1 2 3 4
exchangers having conventional and three-zonal baffles. As can be seen 6 2 2 1 4 3
7 2 3 4 1 2
from the figures, when the three-zonal baffles are used, a much more
8 2 4 3 2 1
uniform temperature distribution is obtained on the tube surfaces 9 3 1 3 4 2
compared to the conventional baffle. In the figure, the temperatures at 10 3 2 4 3 1
two same selected points on tube surfaces were also shown for both heat 11 3 3 1 2 4
exchangers. The tube surface temperatures at these two points were 334 12 3 4 2 1 3
13 4 1 4 2 3

C in the conventional heat exchanger, and 328 � C in the heat exchanger 14 4 2 3 1 4
with three-zonal baffle. This indicate that a better heat transfer from the 15 4 3 2 4 1
hot fluid to the cold fluid is achieved in the heat exchanger with the use 16 4 4 1 3 2
of a three-zonal baffle.

Fig. 5. Cross-sectional view of the three-zonal baffle.

5
N. Biçer et al. International Journal of Thermal Sciences 155 (2020) 106417

Fig. 6. Experimental setup.

Table 7 compares the temperature difference and pressure drops that


Table 5
occur in heat exchangers with conventional and three-zonal baffles
Heat exchanger input parameters.
under different operating conditions. As seen from the table, at the
Tube-Side Shell-Side conditions of 0.5 kg/s hot water mass flow rate and 0.2 kg/s cold water
Mass flow rate Inlet temperature Mass flow rate Inlet temperature mass flow rate, the temperature difference for conventional and three-
(kg/s) (K) (kg/s) (K) zonal baffles were obtained as 44.3 K, and 47.3 K, respectively. On the
0.3 323 0.3 295 other hand, for the mass flow rates of 2.19 kg/s and 2.88 kg/s, a tem­
0.4 323 0.4 295 perature difference of 16.3 K in the conventional heat exchanger, and
0.5 323 0.5 295 17.4 K in the heat exchanger with three-zonal baffle were obtained.
0.6 323 0.6 295
These results show that the temperature difference in the heat exchanger
0.7 323 0.7 295
1.0 323 0.8 295 increases with decreasing mass flow rates and the maximum increase up
2.1 323 1.0 295 to 6.75% with the use of three-zonal baffles.
It is seen from the table that the three-zonal baffle heat exchanger
gives much better results in terms of pressure drop compared to con­
Table 6 ventional heat exchanger. According to the CFD analyses results, the
Measuring ranges and measurement accuracy of the devices used in the exper­ pressure drop in the heat exchanger was 79.9 Pa in the case of con­
iments and uncertainty levels of the calculated parameters. ventional baffle and 51.8 Pa in the case of three-zonal baffle at the lower
Equipment Range Accuracy Uncertainty (%)
mass flow rate condition. As the mass flow rate increases, the decrease in
the pressure drop in the heat exchange with three-zonal baffle becomes
PT100 0/100 � C �1 � C
much more pronounced. At the higher mass flow rate conditions, the
Manometer 0/100 mbar 2 mbar
0/2.5 bar 0.02 bar pressure drop in the heat exchanger was obtained as 9431.2 Pa in the
Flow meter 0/50 l/s 0.01 l/s case of conventional baffle and 4836.7 Pa in the case of three-zonal
Heat transfer rate – – 1.5 baffle. The maximum reduction in pressure drop has increased up to
49%. These results show that the thermal performance of the heat
exchanger can be improved as well as the pressure drop can be
and the shell, and consequently worsens the heat transfer in these zones.
dramatically reduced with the use of the newly designed three-zonal
The recirculation zones are also susceptible to fouling. These recircu­
baffles. In Table 7, the pressure drop values calculated by using the
lation zones reduce the heat transfer from the hot fluid to the cold fluid
Kern method for the conventional heat exchanger were also presented to
on the one hand, while increasing the fouling resistance in these areas.
validate the CFD analyses results. The results show that there is a dif­
Increased fouling resistance reduces the service life of the heat
ference of up to 16% between the pressure drop values calculated by the
exchanger, increases the operating and maintenance costs of the heat
Kern method and the CFD analyses results.
exchanger. In addition, the heat transfer in these recirculation zones
Similar results were obtained in an experimental study conducted by
decreases depending on time and the efficiency of the heat exchanger is
Yang et al. [14]. Their experimental results showed that the use of
reduced. In the case of three-zonal baffle, these recirculation zones are
unilateral ladder type helical baffle resulted in a 15.3–47.1% reduction
almost never formed. With the new baffle design, these recirculation
in pressure drop and an increase in shell side heat transfer coefficient of
zones are substantially eliminated and the average flow velocity on the
9.3–25.5% compared to segmental baffle use. Dandgaval et al. [19],
tube bundle is maintained substantially. As a result, the effective heat
numerically investigated the effects of helical and two different
transfer surface area is increased, and an effective mixture on the shell
segmental baffles usage on the shell and tube heat exchanger perfor­
side is obtained. This also improves the heat transfer mechanism.
mance. They stated that the use of helical baffle prevented the formation

6
N. Biçer et al. International Journal of Thermal Sciences 155 (2020) 106417

Fig. 7. Temperature distributions in the heat exchanger at a mass flow rate of 0.5 kg/s.

of dead zones behind the baffle, thereby pressure drop was decreased, are shown in Fig. 11. As can be seen in the figure, level 4 of factor A,
and thermal performance was improved. Chen et al. [21] investigated level 2 of factor B, level 4 of factor C, level 2 of factor D and level 4 of
the effects of different helical baffle designs on pressure drop and factor E should be considered to minimize the pressure drop
thermal performance of the heat exchanger. Their numerical results (A4B2C4D2E4). Since there is no defined experiment with these levels of
indicated that the helical baffle reduced the pressure drop, and factors among the 16 designed experiments, CFD analyses considering
improved thermal performance compared to the segmental baffle. the optimized factor values was carried out and the pressure drop was
However, the main improvement here is not to increase the rate of heat obtained as 26.5 Pa.
transfer, but to lower the pressure drop in the shell side without wors­ The confirmation tests were performed and the results were pre­
ening the heat transfer. This result was obtained by providing a more sented in Table 9 and Table 10. In Table 9, predicted and actual cold
intense and homogeneous mixing on the shell side and eliminating the water outlet temperature are presented. As can be seen from the table,
recirculation zones and stagnation points. Both CFD results and exper­ there is very good agreement between predicted and actual values. The
imental measurements confirm this conclusion. results show that improvement in S/N ratio for cold water outlet tem­
perature is 0.013 dB. The cold water outlet temperature was also
5.2. Taguchi optimization results approximately 0.15% increased. On the other hand, Table 10 shows
predicted and actual pressure drop values. For pressure drop, the
Table 8 shows the CFD results of the cold water outlet temperature improvement in S/N ratio was 4.3 dB, and the decrease in pressure drop
and pressure drop values in the heat exchanger, which was carried out was approximately 39.27%. The results obtained from confirmation
by taking the experimental conditions determined according to Taguchi tests confirmed the validity of the Taguchi approach used in the opti­
optimization method. The results of the analyses show that the mization of design parameters.
maximum outlet temperature is obtained under test conditions 1 and the Table 11 shows ANOVA variance analyses results. As can be seen
minimum pressure drop value is obtained under test conditions 13. from the table, the most important design factor in terms of temperature
Optimum conditions were determined by using Taguchi method for is A with at least 99.99% confidence level. Factors C and D are less
cold water outlet temperature. As the generated design has not been effective. On the other hand, factors B and E are significant but not
included in the main experimental layout, the process was re-iterated effective. In terms of pressure drop, the most important design factors
until the required criteria are satisfied. Fig. 10 shows the Taguchi ana­ are A, C and E with a confidence level of 95%. However, factors B and D
lyses results performed with Minitab statistical software for cold water are less effective.
outlet temperature. After confirmation test carried out at the 99% con­
fidence level, the optimum design factor combination of the baffle were 5.3. Validation of CFD results
obtained by taking into account level 1 of factor A, level 3 of factor B,
level 1 of factor C, level 1 of factor D, and level 1 of factor E In the study, the CFD analyses results were validated with the
(A1B3C1D1E1). Since this experiment was not included among the experimental data. For this purpose, the heat exchanger with conven­
defined 16 experiments, the outlet temperature was estimated as 331.5 tional and three-zonal baffles were tested for seven non-consecutive
K considering optimized factors. different mass flow rates and the results were used to validate the CFD
The second objective was to minimize the pressure drop in the heat analyses results performed under the same operating conditions. In
exchanger. The results of Taguchi analyses performed for this purpose Figs. 12 and 13, this validation is performed in terms of pressure drop

7
N. Biçer et al. International Journal of Thermal Sciences 155 (2020) 106417

Fig. 8. Pressure distributions in the heat exchanger at a mass flow rate of 0.5 kg/s.

and heat transfer rate, respectively. The results show a good agreement range of 25.3–257.1 Pa. There was a difference up to 7.6% between the
between the CFD analyses results and the experimental values. experimental values and the CFD results.
In Fig. 12, the tube-side experimental pressure drops are compared The heat transfer rates of the heat exchangers calculated using the
with the calculated pressure drops based on CFD analyses for the three- experimental and CFD data are compared in Fig. 13. The horizontal axis
zonal baffle. As can be seen from the figure, the pressure drop in the in the figure shows seven non-consecutive mass flow rate values for
tubes increase with the increase of the mass flow rate as expected. The which experiments and analyzes were performed. As shown in the
experimental pressure drops were calculated in the range of 23.5–240 figure, the heat transfer rate increases as the mass flow rate increases for
Pa, and the pressure drops calculated based on CFD analyses were in the both baffles. A difference of up to 7.3% occurred between the

8
N. Biçer et al. International Journal of Thermal Sciences 155 (2020) 106417

Fig. 9. Velocity streamlines in the heat exchanger at a mass flow rate of 0.5 kg/s.

Table 7
Temperature difference and pressure drops in heat exchangers with conven­
tional and three-zonal baffles.
Baffle type Mass Mass Temperature Pressure Pressure
flow flow difference drop (Pa) drop (Pa)
rate of rate of between inlet (CFD) (Kern)
cold hot and outlet (K)
water water (CFD)
(kg/s) (kg/s)

Three-zonal 2.19 2.88 17.4 4836.7 –


Conventional 2.19 2.88 16.3 9431.2 10693.4
Three-zonal 0.2 0.5 47.3 51.8 –
Conventional 0.2 0.5 44.3 79.9 93.4

Table 8
Pressure and temperature values obtained by CFD analyses.
Experiment number A B C D E Temperature (K) Pressure (Pa)
Fig. 10. Taguchi analyses results performed with Minitab statistical software
1 1 1 1 1 1 331.3 43.14 for cold water outlet temperature.
2 1 2 2 2 2 328.7 32.43
3 1 3 3 3 3 330.1 32.18
4 1 4 4 4 4 328.1 31.77 experimental values and the CFD results.
5 2 1 2 3 4 325.3 38.21
6 2 2 1 4 3 325.7 37.96 6. Conclusions
7 2 3 4 1 2 325.8 33.05
8 2 4 3 2 1 325.8 37.81
9 3 1 3 4 2 326.1 32.81
In this study, a new three-zonal baffle was designed for use in the
10 3 2 4 3 1 326.1 32.61 STHEs. After that, CFD analyzes of the heat exchangers with conven­
11 3 3 1 2 4 326.7 32.54 tional and three-zonal baffles were performed and the results of the
12 3 4 2 1 3 327.4 32.63 analyses were compared. The shell-and-tube heat exchanger with the
13 4 1 4 2 3 327.7 29.96
newly developed three-zonal baffle was then optimized using the
14 4 2 3 1 4 328.6 30.09
15 4 3 2 4 1 328.4 35.31 Taguchi method. In the final stage of the study, the results of CFD an­
16 4 4 1 3 2 329.7 34.01 alyses using optimized baffles were validated with the experimental
results obtained under the same working conditions. The conclusions
obtained in this study are as follows:

9
N. Biçer et al. International Journal of Thermal Sciences 155 (2020) 106417

Fig. 12. Pressure difference on the tube side of the heat exchanger.
Fig. 11. Taguchi analyses results performed with Minitab statistical software
for pressure drop.

Table 9
Confirmation test results for cold water outlet temperature.
Initial Optimum parameters
parameters
Prediction Experiment

Level A1B1C1D1E1 A1B3C1D1E1 A1B3C1D1E1


Temperature 331.3 331.5 331.8
S/N ratio (dB) 50.404 50.409 50.417

Table 10
Confirmation test results for pressure drop.
Initial Optimum parameters
parameters
Prediction Experiment

Level A1B1C1D1E1 A4B2C4D2E4 A4B2C4D2E4


Pressure drop 43.14 26.50 26.20
S/N ratio (dB) 32.70 28.50 28.40
Fig. 13. Tube side heat transfer rate of the heat exchanger.

configuration. Thus, it will be possible to reduce operating costs by


Table 11
using a pump with lower input power.
The variance analysis (ANOVA).
� The use of a three-zonal baffle has resulted in significant increases in
Factors Sum of Degree of Variance, Ffactor temperature differences in the heat exchanger compared to con­
Squares, SS Freedom v V
ventional baffles. The maximum increase was obtained as 6.75%
Temperature A 0.0271 3 0.0090 16.28b depending on the operating conditions of the heat exchanger. This
B 0.0004 3 0.0001 0.25
means that the thermal capacity of the heat exchanger will increase
C 0.0029 3 0.0010 1.74
D 0.0025 3 0.0008 1.52 with the use of a three-zonal baffle, and that the same thermal per­
E 0.0008 3 0.0003 0.39 formance can be achieved with a more compact heat exchanger.
Total 0.0338 15 0.0023 � The flow induced vibration in the shell-side of the heat exchanger
ep 0.0067 12.00 0.0006 could be considerably reduced with the use of new three-zonal
Pressure A 3.1728 3 1.0576 9.91a
drop B 1.0438 3 0.3479 3.26
baffles.
C 3.3662 3 1.1221 10.52a
D 0.3201 3 0.1067 1.00 Declaration of competing interest
E 2.9147 3 0.9716 9.10a
Total 10.82 15 1.06
ep 0.32 3.00 0.11
The authors declare that they have no known competing financial
a
interests or personal relationships that could have appeared to influence
At Least 95% confidence. the work reported in this paper.
b
At Least 99.99% confidence.

References
� The three-zonal baffle does not partially block the flow in the shell-
side. Thus, there are no stagnant zones behind the baffles. This re­ [1] S. Kakac, H. Liu, A. Pramuanjaroenkij, Heat Exchangers Selection, Rating and
duces the fouling and ensures long-term operating periods. Thermal Design, third ed., CRC Press, Washington D.C., 2012.
[2] M. Petrik, G.L. Szepesi, Shell side CFD analysis of a model shell-and-tube heat,
� In the case of using three-zonal baffles, very low pressure drops in the Chemıcal Engıneerıng Transactıons 70 (2018) 313–318.
shell-side occur compared to conventional baffles. With the use of [3] C. Yua, T. Chengb, J. Chenb, Z. Renb, M. Zeng, Investigation on thermal-hydraulic
three-zonal baffles, to the pressure loss in the shell-side has lowered performance of parallel-flow shell and tube heat exchanger with a new type of anti-
vibration baffle and wire coil using RSM method, Int. J. Therm. Sci. 138 (2019)
by 49% compared to conventional segmentally baffled
351–366.

10
N. Biçer et al. International Journal of Thermal Sciences 155 (2020) 106417

[4] B.I. Master, K.S. Chunangad, A.J. Boxma, D. Kral, P. Stehlík, Most frequently used [26] X. Gu, Y. Luo, X. Xiong, K. Wang, Y. Wang, Numerical and experimental
heat exchangers from pioneering research to worldwide applications, Heat Tran. investigation of the heat exchanger with trapezoidal baffle, Int. J. Heat Mass Tran.
Eng. 27 (2006) 4–11. 127 (2018) 598–606.
[5] L. Liebenberg, J.P. Meyer, In-tube passive heat transfer enhancement in the process [27] J. Xiao, S. Wang, S. Ye, J. Wang, J. Wen, J. Tu, Experimental investigation on pre-
industry, Appl. Therm. Eng. 27 (2007) 2713–2726. heating technology of coal water slurry with different concentration in shell-and-
[6] S. Yang, Y. Chen, J. Wu, H. Gu, Influence of baffle configurations on flow and heat tube heat exchangers with ladder-type fold baffles, Int. J. Heat Mass Tran. 132
transfer characteristics of unilateral type helical baffle heat exchangers, Appl. (2019) 1116–1125.
Therm. Eng. 133 (2018) 739–748. [28] C. Yu, Z. Ren, M. Zeng, Numerical investigation of shell-side performance for shell
[7] S.K. Singh, D. Stephan, in: https://www.process-worldwide.com/, 2014. and tube heat exchangers with two different clamping type anti-vibration baffles,
[8] J. Mahendran, Experimental analysis of shell and tube heat exchanger using flower Appl. Therm. Eng. 133 (2018) 125–136.
baffle plate configuration, Materials Today:Proceedings 21 (2020) 419–424. [29] C. Yua, J. Chenb, M. Zengb, B. Gao, Numerical study on turbulent heat transfer
[9] K. Wang, C. Bai, Y. Wang, M. Liu, Flow dead zone analysis and structure performance of a new parallel flow shell and tube heat exchanger with sinusoidal
optimization for the trefoil-baffle heat Exchanger, Int. J. Therm. Sci. 140 (2019) wavy tapes using RSM analysis, Appl. Therm. Eng. 150 (2019) 875–887.
127–134. [30] X. Wang, Y. Liang, Y. Sun, Z. Liu, W. Liu, Experimental and numerical investigation
[10] L. Ma, K. Wang, M. Liu, D. Wang, T. Liu, Y. Wang, Z. Liu, Numerical study on on shell-side performance of a double shell-pass rod baffle heat exchanger, Int. J.
performances of shell-side in trefoil-hole and quatrefoil-hole baffle heat Heat Mass Tran. 132 (2019) 631–642.
exchangers, Appl. Therm. Eng. 123 (2017) 1444–1455. [31] Y. Lei, Y. Li, S. Jing, C. Song, Y. Lyu, F. Wang, Design and performance analysis of
[11] C. Donga, X.F. Zhoub, R. Dongc, Y.Q. Zhenga, Y.P. Chend, G.L. Hua, Y.S. Xua, Z. the novel shell-and-tube heat exchangers with louver baffles, Appl. Therm. Eng.
G. Zhanga, W.W. Guo, An analysis of performance on trisection helical baffles heat 125 (2017) 870–879.
exchangers with diverse inclination angles and baffle structures, Chem. Eng. Res. [32] G.B. Leoni, T.S. Klein, R.A. Medronho, Assessment with computational fluid
Des. 121 (2017) 421–430. dynamics of the effects of baffle clearances on the shell side flow in a shell and tube
[12] X. Cao, T. Du, Z. Liu, H. Zhai, Z. Duan, Experimental and numerical investigation heat exchanger, Appl. Therm. Eng. 112 (2017) 497–506.
on heat transfer and fluid flow performance of sextant helical baffle heat [33] M. Mellal, R. Benzeguir, D. Sahel, H. Ameur, Hydro-thermal shell-side performance
exchangers, Int. J. Heat Mass Tran. 142 (2019) 118437. evaluation of a shell and tube heat exchanger under different baffle arrangement
[13] J. Chen, X. Lu, Q. Wang, M. Zeng, Experimental investigation on thermal-hydraulic and orientation, Int. J. Therm. Sci. 121 (2017) 138–149.
performance of a novel shell-and-tube heat exchanger with unilateral ladder type [34] A.A.A. Arania, R. Moradi, Shell and tube heat exchanger optimization using new
helical baffles, Appl. Therm. Eng. 161 (2019) 114099. baffle and tube configuration, Appl. Therm. Eng. 157 (2019) 113736.
[14] S. Yang, Y. Chen, J. Wu, H. Gu, Influence of baffle configurations on flow and heat [35] C.S. Pandey, H. Sharma, M. Dadhich, S.K. Jain, Numerical analysis and
transfer characteristics of unilateral type helical baffle heat exchangers, Appl. optimization of corrugated heat exchanger with Taguchi method, Int. J. Res. Appl.
Therm. Eng. 133 (2018) 739–748. Sci. Eng. Technol. 6 (2018) 874–879.
[15] S. Shinde, U. Chavan, Numerical and experimental analysis on shell side thermo- [36] S. Gunes, E. Manay, E. Senyigit, V. Ozceyhan, A Taguchi approach for optimization
hydraulic performance of shell and tube heat exchanger with continuous helical of design parameters in a tube with coiled wire inserts, Appl. Therm. Eng. 31
FRP baffles, Thermal Science and Engineering Progress 5 (2018) 158–171. (2011) 2568–2577.
[16] L. He, P. Li, Numerical investigation on double tube-pass shell-and-tube heat [37] S. Chamoli, A Taguchi approach for optimization of flow and geometrical
exchangers with different baffle configurations, Appl. Therm. Eng. 143 (2018) parameters in a rectangular channel roughened with V down perforated baffles,
561–569. Case Stud., Therm. Eng. 5 (2015) 59–69.
[17] T. Dua, Q. Chen, W. Dua, L. Cheng, Performance of continuous helical baffled heat [38] L.H. Tang, S.C. Tan, P.Z. Gao, M. Zeng, Parameters optimization of fin-and tube
exchanger with varying elliptical tube layouts, Int. J. Heat Mass Tran. 133 (2019) heat exchanger with a novel vortex generator fin by Taguchi method, Heat Tran.
1165–1175. Eng. 37 (2016) 369–381.
[18] S. Yang, Y. Chen, J. Wu, H. Gu, Performance simulation on unilateral ladder type [39] M. Zeng, L.H. Tang, M. Lin, Q.W. Wang, Optimization of heat exchangers with
helical baffle heat exchanger in half cylindrical space, Energy Convers. Manag. 150 vortex-generator fin by Taguchi method, Appl. Therm. Eng. 30 (2010) 1775–1783.
(2017) 134–147. [40] A.Z. Aghaie, A.B. Rahimi, A. Akbarzadeh, A general optimized geometry of angled
[19] B.O. Dandgaval, P.D.R. Godase, T. Dey, Study of shell and tube heat exchanger ribs for enhancing the thermo-hydraulic behavior of a solar air heater channel – a
with the effect of types of baffles, Procedia Manufacturing 20 (2018) 195–200. Taguchi approach, Renew. Energy 83 (2015) 47–54.
[20] G. Yang, G. Ding, J. Chen, W. Yang, S. Hu, Experimental study on shell side heat [41] T. Du, W. Du, K. Che, L. Cheng, Parametric optimization of overlapped helical
transfer characteristics of two-phase propane flow condensation for vertical baffled heat exchangers by Taguchi method, Appl. Therm. Eng. 85 (2015)
helically baffled shell-and-tube exchanger, Int. J. Refrig. 107 (2019) 135–144. 334–339.
[21] Y. Chen, H. Tang, J. Wu, H. Gu, S. Yang, Performance comparison of heat [42] T. Sivasakthivel, K. Murugesan, H.R. Thoma, Optimization of operating parameters
exchangers using sextant/trisection helical baffles and segmental ones, Chin. J. of ground source heat pump system for space heating and cooling by Taguchi
Chem. Eng. 12 (2019) 2892–2899. method and utility concept, Appl. Energy 116 (2014) 76–85.
[22] M. Miansari, A. Jafarzadeh, H. Arasteh, D. Toghraie, Thermal performance of a [43] M.M. Etghani, S.A.H. Baboli, Numerical investigation and optimization of heat
helical shell and tube heat exchanger without fin, with circular fins, and with V- transfer and exergy loss in shell and helical tube heat exchanger, Appl. Therm. Eng.
shaped circular fins applying on the coil, J. Therm. Anal. Calorim. (2020). In press, 121 (2017) 294–301.
corrected proof, Available online 07 February 2020. [44] J.N. Zhang, M. Cheng, Y.D. Ding, Q. Fu, Z.Y. Chen, Influence of geometric
[23] E.M.S. El-Saida, M.M.A. Al-Sood, Shell and tube heat exchanger with new parameters on the gas-side heat transfer and pressure drop characteristics of three-
segmental baffles configurations: a comparative experimental investigation, Appl. dimensional finned tube, Int. J. Heat Mass Tran. 133 (2019) 192–202.
Therm. Eng. 150 (2019) 803–810. [45] S. Chamoli, P. Yu, A. Kumar, Multi-response optimization of geometric and flow
[24] J. Chen, P. Zhao, Q. Wang, M. Zeng, Experimental investigation of shell-side parameters in a heat exchanger tube with perforated disk inserts by Taguchi grey
performance and optimal design of shell-and-tube heat exchanger with different relational analysis, Appl. Therm. Eng. 103 (2016) 1339–1350.
flower baffles, Heat Tran. Eng. (2020). In press, corrected proof, Available online [46] M. Miansari, M.A. Valipour, H. Arasteh, D. Toghraie, Energy and exergy analysis
04 February 2020. and optimization of helically grooved shell and tube heat exchangers by using
[25] X. Wang, N. Zheng, Z. Liu, W. Liu, Numerical analysis and optimization study on Taguchi experimental design, J. Therm. Anal. Calorim. 139 (2020) 3151–3164.
shell-side performances of a shell and tube heat exchanger with staggered baffles, [47] J.P. Holman, Experimental Methods for Engineers, eigth ed., The McGraw-Hill,
Int. J. Heat Mass Tran. 124 (2018) 247–259. New York, NY, 2011.

11

You might also like