Applied Thermal Engineering 188 (2021) 116616

Contents lists available at ScienceDirect

Applied Thermal Engineering

journal homepage: www.elsevier.com/locate/apthermeng

Numerical investigation of small plate heat exchangers performance having

different surface profiles
Saeed Mohebbi, Farzad Veysi *
Department of Mechanical Engineering, Razi University, Kermanshah, Iran

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

Keywords: In the present study, the numerical investigation of single-phase and steady-state water flow in a small-sized
Plate heat exchanger plate heat exchanger with a chevron type corrugation pattern, is performed in three-dimensional. For vali­
Numerical dating the numerical modeling, an empirical model of a commercial PHE is designed and a good agreement
Geometry
between the results of experimental tests and numerical modeling is obtained. Moreover, the influence of flow
Heat transfer
Friction
regime (150 < Re 〈5 5 0), as well as geometrical parameters of chevron angle (β), corrugation depth (b), and
corrugation pitch (λ) on the performance of the PHE are evaluated. In all corrugation depths and corrugation
pitches, the maximum Nu-number and friction coefficient occur at β = 60◦ and β = 30◦ , respectively. Consid­
ering the simultaneous effect of heat transfer and friction, the JF factor is defined. The reducing trends of heat
transfer and the increasing trends of friction are inferred by decreasing Re-number; however, the JF factor is
increased. Besides, β = 60◦ is the most favorable case from the performance point of view due to the formation of
different flows. In this case, the JF factor increases compared to the other chevron angles. On the other hand, the
increasing behavior of this factor is observed with increasing corrugation depth (b), which is 55% at β = 60◦ and
λ = 8.3 mm compared to the commercial PHE. The decrease in the corrugation pitch (λ) also has a similar effect
on the JF factor and increases it by 40% at β = 60◦ and b = 2.3 mm compared to the commercial PHE. In general,
it is inferred that the superior performance of the small-sized PHEs is associated with the decreased mass flow
()
rate, the chevron angle of 60◦ , and the increased bλ ratio. By increasing the corrugation depth or decrementing

the corrugation pitch, the weight and volume of the PHE showed an enhancement. In small PHEs, however, this
weight and volume gain versus performance improvement can be ignored. It is recommended to reduce the
corrugation pitch rather than increasing its depth as a decline in corrugation pitch does not enhance the volume
of the small PHE.

heat exchangers [4]. The improved heat transfer can be attributed to the
increased area and formation of turbulence in the flow [5].
1. Introduction plate heat exchangers with a chevron type corrugation pattern are
considered as a class of PHEs with their characteristics. Chevron angle is
Heat exchangers are considered as one of the most widely used referred to as the corrugations created at a particular angle on each
equipment in a wide variety of applications in different industries. The plate. By the use of these corrugations, the heat transfer would be
different goals, including increased efficiency, reduced fuel consump­ improved and better performance can be achieved [6-8]. It should be
tion, reduced size, and consequently, the reduced economic cost of in­ noted that, however, the presence of these corrugations, increases the
dustrial equipment necessitate the demand for precise and optimum friction coefficient and ultimately fluid pumping power. Therefore, it is
design of this equipment. A plate heat exchanger (PHE) has gained desired to determine an optimum corrugation geometry that forms a
extensive applications in processing, pharmaceutical, petroleum and balance between the maximum heat transfer and minimum friction in­
petrochemical, HVAC, and other industries [1,2]. PHEs were first stud­ crease. The schematic view of a PHE with a chevron type corrugation
ied by Mckillop and Dunkley [3]. The enhanced heat transfer between pattern is available in [9] and is depicted in Fig. 1.
hot and cold fluids is regarded as the advantage of PHEs so that at the Different parameters such as heat transfer rate and pressure drop
same heat transfer, the Re-number of them is five times lower than tube

* Corresponding author.
E-mail address: [email protected] (F. Veysi).

https://doi.org/10.1016/j.applthermaleng.2021.116616
Received 29 September 2020; Received in revised form 9 December 2020; Accepted 18 January 2021
Available online 25 January 2021
1359-4311/© 2021 Elsevier Ltd. All rights reserved.
S. Mohebbi and F. Veysi Applied Thermal Engineering 188 (2021) 116616

Nomenclature Q Convective heat flux (W/m2)

ṁD
Re Reynolds number, NbL wμ
A Area (m2) T Temperature (K)
b Corrugation depth (m) T Thickness (m)
cp Specific heat capacity of fluid (J/kgK) ui ,uj ,uk Components of velocity vector (m/s)
D Hydraulic diameter (m), 2b φ u’i , u’j Fluctuations of velocity (m/s)
E Total energy (J) V Volume (m3)
f Friction coefficient,f = 1 ρ4uτw 2 W Width (m)
2 ave
H Channel height (m)
I0 Turbulent intensity Greek symbols
j Nu
Colburn factor,RePr 1/3
μ Dynamic viscosity (kg/ms)
( ) μt Eddy viscosity (kg/ms)
j
jref
ρ Density (kg/m3)
JF JF factor, 1/2
β Chevron angle (deg)
(f /fref ) λ Corrugation pitch (m)
Realarea
k Turbulent kinetic energy (m2/s2) φ Surface enlargement factor,projectedarea
kf Thermal conductivity of fluid (W/mK) δij Kronecker delta
keff Effective thermal conductivity (W/mK) ε Turbulent dissipation rate (m2/s3)
L Length (m) τ Shear stress
m Mass (kg) τijeff Deviatoric stress tensor (N/m2)
ṁ Flow rate (kg/s)
N Number of hot or cold channels Subscripts
Nu Nusselt number Ave average
PHE Plate Heat Exchanger B bulk
P Pressure (Pa) L Local
Pr Prandtl number T turbulent
Prt Turbulent prandtl number w wall

H b

L2
x L1

W1

W2

Fig. 1. Geometry parameters of a plate with a chevron type corrugation pattern.

have been studied experimentally in PHEs [10-13]. In addition to the studies to achieve the best results. For example, the utilization of the k-ε
experimental studies, numerical simulations, including finite volume model would lead to a lower estimation for pressure drop, while the
analysis was used to investigate the effect of different geometrical pa­ usage of the laminar flow model results in good agreement with the
rameters [14]. Kanaris et al. [15] simulated the thermal–hydraulic experimental results [18]. Gherasim et al. [19] numerically studied the
behavior of a PHE using CFX-ANSYS software through the use of k-ω and water flow in a PHE with trapezoidal corrugations (b = 2.5 mm, λ = 9
SST models. A good agreement was obtained by comparing the thermal mm, β = 60◦ ). A good compromise between simulation and experi­
profiles of the numerical study with the images of the infrared camera, mental results was obtained for Re <400 with the consideration of the
which, highlighted the capability of CFD codes in predicting the flow laminar flow model. At Re >400 and consideration of the k-ε model and
characteristics and heat transfer. Liu et al. [16] and Tsai et al. [17] the non-equilibrium wall functions (NEWF), the highest consistency
numerically modeled the water flow inside a PHE consisting of two with the experimental results was achieved. Ciofalo et al. [20] compared
plates under the conditions of Re <430, β = 45◦ , and b = 2 mm. The various numerical models inside the corrugated PHE. The compared
obtained results showed a 20% reduction in pressure drop compared to models were: The standard k-ε turbulence model, low Re-number tur­
the experimental results. Different models have been used in numerical bulence model, direct numerical simulation (DNS), and large eddy

2
S. Mohebbi and F. Veysi Applied Thermal Engineering 188 (2021) 116616

simulation (LES). The superiority of the LES model in terms of the best Table 1
agreement with the experimental results was concluded. Blomerius et al. Geometrical parameters of the small PHE
[21] showed the characteristics of periodic flows and the Re-number with a chevron type corrugation pattern.
dependent frequency through numerical analysis. According to the ob­ Parameter Value
tained results, a longitudinal component with oscillatory motion was βo 60
considered as one of the flow components inside the PHE. Pelletier et al. b(mm) 2.3
[22] used Re-Averaged Navier-Stokes (RANS) model under unsteady
λ(mm) 8.3
conditions and the more precise results were obtained for the constant
D(mm) 3.932
flux boundary conditions compared to the constant temperature
Δx(mm) 0.4
boundary condition. Jain et al. [23] found the distribution of tempera­
φ 1.17
ture and velocity using the k-ε (Realizable) method in a steady-state and
t(mm) 0.3
they confirmed it by the experimental results. However, Han et al. [24]
N 7
reported that the k-ε (RNG) method provided acceptable results. In some
L1 (mm) 154
studies, the meshing of the model was performed using GAMBIT soft­
L2 (mm) 193
ware and ANSYS/Fluent software was used for analyzing the flow.
W 1 (mm) 42
Among these studies, it can be mentioned to work of Lee and Lee [25], in
W 2 (mm) 83
which the modeling of the PHE with a chevron type corrugation pattern
and unsteady flow was conducted using an LES model. Lee and Lee [25]
concluded that the angles that gave the best result for laminar and 3. According to the aforementioned outcomes of previous works, they
turbulent flow were β = 30 and β = 60 , respectively and also bλ =
◦ ◦
have some discrepancies in terms of the effect of heat exchanger
2was the best choice in both flows. One year later, they conducted a plate geometry, and hence, further research is desirable.
numerical investigation on the heat transfer and pressure drop of PHE 4. A few studies have assessed the effect of all geometric parameters of
with the aim of geometrical optimization. They inferred that the best heat exchanger plates. In this study, however, the effect of three
performance of the PHE would be occurred at β = 66.5◦ and bλ = 2.73, geometrical parameters and the flow regime were simultaneously
regardless of the Re-number value [26]. evaluated.
The experimental results and correlations from previous studies in 5. In industry, especially the HVAC industry where small PHEs are
the field of heat transfer and friction coefficients can be only used in a widely used, the profiles of heat exchanger plates are made in such a
specific range of Re-numbers and Pr-numbers. One of the first works way that they are not at their highest efficiency. Therefore, in this
which introduced a quasi-experimental-semi-mathematical correlation research, plates with better profiles are introduced which can offer
to calculate heat transfer and friction coefficients was the study by higher performance compared to the conventional commercial
Martin [27], which offered unlimited Re-number and Pr-numbers. But plates.
the deviation of these correlations from the experimental data in some
cases ran up to 50% and even more [28]. Arsenyeva et al. [28] showed 2. Experimental apparatus
that for specified pressure drop, temperature program, and heat load,
the geometrical parameters of the plate and its corrugations can be To validate the accuracy of the modeling and ensure its results, an
found, which can result in a PHE with minimal heat transfer area. Dovic experimental apparatus was designed and constructed. The desired
et al. [29] reported the characteristics of the flow in a plate heat experimental apparatus includes a PHE with a chevron type corrugation
exchanger with a chevron type corrugation pattern through visualiza­ pattern, a closed-loop path for hot water, and an open-loop path for cold
tion tests of channels with β = 28◦ and β = 61◦ . A mathematical model water. Table 1 shows the geometric properties of the studied PHE. In the
was then developed to drive correlations for the prediction of friction hot water path, a 4.5 lit bucket containing two 4 kW heaters, a valve to
coefficient and Nu-number for flow in channels of arbitrary geometry. change the flow rate, a heat exchanger, and a rotameter to measure the
Thermal and hydraulic characteristics were evaluated using analytical flow rate were embedded. Besides, In the cold water path, a valve for
solutions for the entrance and fully developed regions of a sinusoidal changing the flow rate, a rotameter for measuring the flow rate, and a
duct adapted to the basic single cell. manometer for measuring the pressure changes of flow from the inlet to
small-sized PHEs with a chevron type corrugation pattern have been outlet ports were inserted. A data logger was also used to measure the
widely used in diverse industries. Therefore, a close investigation of hot and cold fluid temperatures at the inlet and outlet ports of the PHE.
their thermal–hydraulic performance is of great significance. Based on The above experimental apparatus was designed to calculate the Nu-
previous studies, the chevron angle, corrugation depth, corrugation number and friction coefficient for hot side flow, details of which are
pitch, and flow regime are among the most important parameters available in [9]. The schematic view of the experimental apparatus is
affecting the performance of a PHE. The majority of previous works shown in Fig. 2.
addressed the effect of some of these parameters. Therefore, there is a
lack of a comprehensive study investigating the simultaneous effect of 3. Uncertainty
all these parameters. The authors’ motivation to study small PHEs, as
well as the novelty of this work, are summarized below: To make sure of the validity of the measurements conducted in this
study and hence validate the results, uncertainty calculations were
1. The results of previous studies have been presented for a wide range applied. If G = G(X1 , X2 , ⋯, XN ), and S shows Xj error, total error can be
of sizes and geometric characteristics. Since the size of the PHE has a computed by Equation (1). Moffat [30] presented the details of this
great influence on its thermal–hydraulic parameters, so their results method. The maximum errors in the measurements of temperature
for small PHEs have many errors and their findings within the size (TLMTD ), flow rate, and differential pressure, are ± 4.3%, ±4.2%, and ±
range of a small PHE need to be modified. Furthermore, the present 6.1% respectively. Based on this precision and the propagation of error
study was specifically conducted for small PHEs and the results of analysis, the maximum uncertainties of investigated parameters are
this study can be used with high accuracy for this size of PHEs. listed in Table 2.
2. Regarding the limited number of studies on the field of small-sized [ ]1/2
N ( )
PHEs, the present research is aimed to investigate the effect of ∑ ∂G 2
SG = Sj (1)
plate geometry on their thermal–hydraulic performance. j=1
∂xj

3
S. Mohebbi and F. Veysi Applied Thermal Engineering 188 (2021) 116616

T T
Flow meter Flow meter Cold water

Control valve
Hot water
tank P PHE
Control valve

Hot water Pump

T T

T: Thermocouple
P: Pressure

Fig. 2. Schematic of the experimental apparatus and, the picture of data logger and thermocouples [9].

4.1. Boundary conditions

Table 2
Maximum uncertainties of investigated parameters. After plotting the PHE geometry and before the numerical simulation
Maximum uncertainties (%) Parameter by ANSYS Fluent 18 software, the boundary conditions of the problem
±4∙3 Re-number should be determined in the software. The inlet boundary conditions
±10.5 Nu-number were of the velocity-inlet type. The outlet boundary condition was of a
±7.3 Friction coefficient pressure-outlet type, and zero relative pressure was used for it. Further­
more, the non-slip adiabatic boundary condition was used for the side
walls. Also for the inlet and outlet ports, the periodic condition was
utilized.

4.2. Governing equations

After examining the numerical solutions of the available viscosity

models, the k-ε turbulence (RNG) model was selected in standard wall
conditions considering the closest results to the experimental results. In
this model, k and ε are defined as Equations (3) and (4):
2
k = (I0 ui )2 (3)
3

Cμ 0.75 k1.5
ε= (4)
Fig. 3. The channel and simulated plates. (0.07DH )
Semi-Implicit Method for Pressure-Linked Equations (SIMPLE) al­
4. Numerical modeling gorithm was used to discretize the governing equations. The pressure,
momentum, and energy equations were second-order upwind; while
The geometry of the studied heat exchanger should be precisely turbulence equations were first-order upwind. Under-relaxation co­
drawn before the numerical analysis. In 3D PHE geometry, two plates efficients of 0.3, 0.7, and 1 were also used to control the solution of
with a chevron angle that can create a channel were plotted in Solid­ pressure, momentum, and energy equations, respectively. The conver­
Works 2016 software. The exact geometry of the corrugation was gence criteria for continuity equations, momentum, energy, turbulence
extracted by cutting the commercial PHE. The obtained relation for the kinetic energy, and turbulence dissipation rates were considered to be
corrugation profile is defined in Equation (2). 10-6. The equations were solved under single-phase, steady, three-
( ) dimensional, and incompressible conditions. Besides, the natural con­
y = 1.15sin
2π(x − 2.075)
+ 1.15 (2) vection and radiative heat transfer were neglected and constant physical
8.3 properties were assumed for the fluids. The continuity, momentum, and
After extracting the required relation for the corrugation, a channel energy equations are presented in Equations (5), (6), and (7),
of the heat exchanger was simulated as shown in Fig. 3. It should be respectively.
noted that the plate was made of 316L stainless steel with a thickness of ∂
0.3 mm. (ρui ) = 0 (5)
∂xi
[ ( )]
∂ ( ) ∂p ∂ ∂ui ∂uj 2 ∂uk ∂ ( )
ρui uj = − + μ + − δ + − ρu’i u’j (6)
∂xj ∂xi ∂xj ∂xj ∂xi 3 ij ∂xk ∂xj
Table 3 [ ]
Investigating the mesh independency. ∂ ∂ ∂T ( )
(u (ρE + p) ) = k + ui τij eff (7)
Number of nodes Nu f ∂xi i ∂xj eff ∂xj
1 439,468 33.50 0.159 The effective heat transfer coefficient, keff , is defined as:
2 809,261 35.64 0.173
3 1,662,342 35.7 0.175

4
S. Mohebbi and F. Veysi Applied Thermal Engineering 188 (2021) 116616

0.2
4τw
36
f =1 (10)
2
ρuave 2
0.19 34

32 4.4. Validating the numerical model

0.18
30
After the creation of the mesh on the PHE geometry and running the
0.17
numerical simulation, the obtained results were compared with the
f

28

Nu
experimental results. To understand the type of flow regime in terms of
26
0.16 laminar or turbulent, both laminar and turbulent models have been
f 24 implemented. In the case of laminar flow, different results are obtained
0.15 from the experiment, but in the case of turbulent flow, the results are
Nu 22
better coordinated. Among the various models of turbulent flow, model
0.14 20 k − ε, and among k − ε models, the RNG model shows the best agreement
0 750000 1500000 2250000 3000000 3750000 with the test results. The basis for selecting solution methods is to
Number of nodes compare their results with the test results. To ensure the numerical
problem-solving method, all numerical solution methods have been
Fig. 4. Comparison of Nu-number and friction coefficient in different
tested and the results are given in Tables 4 and 5. Also, the behavior of
mesh sizes.
the fluid near the wall has been selected as the standard wall function.
The comparison of the k − ε (RNG) model with the experimental re­
sults is provided Figs. 6, 7, and 8. The comparison was performed at β =
60◦ , b = 2.3mm, and λ = 8.3mm, which were identical to the geomet­
rical properties of the tested PHE. Two cases were considered for this
purpose. In the first case, the constant flow rate of 7 Lpm was considered
for the cold side to flow along with variable hot side flow and in the
other case, the constant flow rate of 7 Lpm was considered for the hot
side to flow along with variable cold side flow. The maximum discrep­
ancies between the experiment and simulation for the temperatures of
the cold and hot fluids at the outlet port corresponded to 12% and 10%
respectively. Besides, the pressure difference between the inlet and
outlet ports of the cold and hot side flows was less than 12% in all
simulation cases, which is acceptable and validates the numerical
procedure.

Fig. 5. Meshing three channels of PHE.

5. Results and discussion

C p μt After validating the simulation results and comparing them with the
keff = kf + (8)
Prt experimental results, the variation of the Nu-number and friction coef­
ficient by changing the flow regime or geometry of the PHE can be
4.3. Mesh independency investigated. It should be noted that in all different simulation cases, the
Nu-number and the friction coefficient of the hot side flow were calcu­
To ensure the appropriate meshing on the PHE geometry, several lated and the cold side flow rate was considered the constant value of 7
meshes with various sizes should be created and the Nu-number and Lpm. Moreover, the hot and cold side inlet temperatures were constant at
friction coefficient of each must be calculated. Table 3 and Fig. 4. show 67.3 ◦ C and 20.2 ◦ C, respectively. In this study, in contrast to other
the values of Nu-number and friction coefficient at different mesh sizes. studies, in which the constant value was considered for the channel
The difference between Nu-number and friction coefficient, between height despite the changes in the corrugation depth [31], the channel
cases 1 and 2, and cases 2 and 3 are 6.4% and 8.8%, and 0.17% and height (b + Δx) was changed for each case. In the commercial design of a
1.16%, respectively. Therefore, the mesh of case 2 was chosen. The PHE with a chevron type corrugation pattern, due to the overlap of the
geometry of the meshed PHE is shown in Fig. 5. It is worth mentioning plates and the inverse chevron angle of each plate relative to the adja­
that in all the numerical solutions, the corresponding average values for cent plate, the channel height is equal to the corrugation height plus a
skewness and aspect ratio were less than 0.25 and 2, respectively. The fixed distance (b + Δx) which is varied regarding the changes in the
equations (9) and (10) are proposed for the calculation of the Nu- corrugation depth. The velocity vectors and temperature contours in the
number and friction coefficient using numerical results. mid-plan of a channel in PHE at Re = 400 and different chevron angles
∫ are represented in Figs. 9 and 10, respectively. In general, two main
1 ql D
Nu = Nul dA, Nul = (9) directions can be attributed to the flow within the PHE channels: lon­
A (Tw − Tb )kf
gitudinal component and furrow component [29]. As can be seen in
Figs. 9 and 10, by passing the flow through the PHE channels at the

Table 4
Comparison of the results of different numerical solution models with the experimental results.
k − ε(RNG) k − ε(standard) Spalart- Allmaras Laminar Experiment

24.1 24 27.7 20.3 25.7 Th,o (◦ C)

25.3 25.1 30.1 23.4 26.2 Tc,o (◦ C)
3.29 3.11 2.73 3.95 3.45 ΔP(kPa)
1.16 4.21 1.06 5.07 0 Flow rate difference in inlet and outlet*10-10

5
S. Mohebbi and F. Veysi Applied Thermal Engineering 188 (2021) 116616

Table 5
Comparison of the results of different numerical solution models with the experimental results.
Reynolds k − ω(Realisable) k − ω(RNG) k − ω(standard) k − ε(Realisable)
stress

21.3 27.9 27.5 27.6 23.5 Th,o (◦ C)

24.2 28.3 28.1 28 24.9 Tc,o (◦ C)
2.85 3.75 3.67 3.86 3.15 ΔP(kPa)
5.24 6.21 4.09 2.14 2.00 Flow rate difference in inlet and outlet*10-10

angles other than 0◦ and 90◦ , low-velocity vortices are formed in a

60 cold flow rate=7 lpm longitudinal component and the main flow path is splatted into two
hot flow rate=6 lpm +12% separate paths. The location of these vortices relative to the main flow in
50 the channel is dependent on the chevron angle. The lower values of the
chevron angle are attributed to the closest location to the mainstream;
Numerical(oC)

40 -12% while the greater values are the representor of farther locations. At the
angles of 0◦ and 90◦ , these low-velocity vortices are not formed, which
30 affects the flow parameters significantly. In the other analyzed Re-
number, such fluid flows are present.
20
The variations of Nu-number, friction coefficient, Colburn factor,
10
and JF in terms of chevron angle variations in different flow rates are
depicted in Figs. 11, 12, and 13. The corrugation depth (b) and corru­
0 gation pitch (λ) were considered to be the respective values of 2.3 mm,
0 10 20 30 40 50 60 and 8.3 mm, which are equivalent to the commercial PHE. The JF factor
Experimental(oC) incorporates the fluid pumping power and heat transfer coefficient
simultaneously. Therefore, it is a good factor for evaluating the perfor­
Fig. 6. Comparison of the outlet temperature of cold side flow in experiment mance of PHEs. In the definition of the JF factor, the reference mode
and simulation. means the commercial PHE with β = 60◦ , b = 2.3 mm, λ = 8.3 mm, and
ṁ = 6Lpm. This factor increases with increasing heat transfer and
decreasing the fluid pumping power and improves PHE performance. As
60
cold flow rate=7 lpm shown in Figs. 11 and 12, the maximum Nu-number and the coefficient
hot flow rate=6 lpm of friction occur at β = 60◦ and β = 30◦ , respectively. As it is known, the
50 +10%
highest value of JF factor is associated with β = 60◦ , Also β = 0◦ and β =
90◦ have the lowest JF factor. The plate of a heat exchanger with β =
40
0◦ and β = 90◦ as well as the hot and cold side flows are shown in
Numerical(oC)

-10%
Fig. 14a &14b. The reason for the increase in JF at β = 60◦ is associated
30
with an increase in the Nu-number. The plates of a heat exchanger with
20
β = 60◦ are shown in Fig. 14c.
One of the reasons for the maximum Nu-number at β = 60◦ is that
10
the heat flux on the channel wall had the highest value at this chevron
angle. For example, Table 6 presents the heat flux on the channel wall at
0 ṁ = 4Lpm, b = 2.075 mm and λ = 8.3 mm. In other flow rates and other
0 10 20 30 40 50 60 geometric characteristics, the variation trend of heat flux was the same.
Experimental(oC) Also, the reason for the larger heat flux at β = 60◦ is the maximum
average flow velocity inside the PHE hot channel. In fact, the average
Fig. 7. Comparison of the outlet temperature of the hot side flow in the flow velocity increased from β = 0◦ to β = 60◦ and decreased from β =
experiment and simulation.
60◦ to β = 90◦ . This means that the corrugation on the plate at β = 60◦
maximized the average flow velocity. The average velocity of hot flow is
listed in Table 6. In other flow rates and other geometric characteristics,
5
cold flow rate=7 lpm the variation trend of the average velocity is the same.
4.5
hot flow rate=6 lpm The reason for the maximum friction coefficient at β = 30◦ can be
+12%
4 found in Equation (10) which shows the ratio of the wall shear stress and
3.5 velocity. Regarding the high velocity at β = 60◦ , the highest wall shear
Numerical(kPa)

3 -12% stress also occurred at this chevron angle. At β = 30◦ , the shear stress on
the wall and the average velocity of the flow inside the channel are such
2.5
that the friction coefficient reached its highest value compared to other
2
chevron angles. Table 6 shows the shear stress values on the wall of the
1.5 hot flow channel.
1 Moreover, decreasing the fluid flow rate would increase the JF fac­
0.5 tor. This can be justified by the fact that the reduction in fluid flow rate is
0 followed by a decrease in Re-number and Nu-numbers but an increase in
Nu
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 j (j = RePr 1/3 ) Because of the reduction in Nu-number is much less than

Experimental(kPa) the reduction in Re-number. Also, the reduction in fluid flow rate can
increase the friction coefficient; however, the growth of j is more
Fig. 8. Comparison of the pressure difference between the inlet and outlet of
the hot and cold side flow in the experiment and simulation.
dominant compared to the friction enhancement, which increases the JF

6
S. Mohebbi and F. Veysi Applied Thermal Engineering 188 (2021) 116616

Fig. 9. Velocity vectors of mid-plan for one of the channels of PHE at different chevron angles and Re = 400.

main flow
direction

Fig. 10. Temperature contour of mid-plan for one of the channels of PHE at different chevron angles and Re = 400.

factor. According to Fig. 9, it can be seen that the existence of low- industries that is usually ranged between 1 m/s and 2 m/s. Moreover,
velocity vortices formed in chevron angles other than 0◦ and 90◦ , in­ the depth of the corrugations was also selected regarding the commer­
creases the heat transfer rate and friction, which is resulted in the cial PHE. At β = 60◦ and b = 2.91 mm, the JF is maximal and has a 72%
consequent increase in JF. These vortices are induced due to the pres­ increase in comparison with b = 1.66 mm. In fact, at all chevron angles,
ence of the chevron angle. At β = 0◦ and β = 90◦ , no vortices are formed both the Nu-number and friction coefficient increase as the corrugation
and the fluid layers move parallel to each other, which results in a depth increases too. However, the heat transfer growth is more domi­
reduction of heat transfer and friction. The heat transfer reduction is nant compared to the friction enhancement, which increases the JF
more dominant compared to the friction decrease, which decreases the factor. The increase in corrugation depth is followed by the formation of
JF factor. vortices of fluid in the furrow component pattern by passing the fluid
The variations in Nu-number, friction coefficient, Colburn factor, over the plate corrugation. This increases heat transfer and friction.
and JF in terms of chevron angle variations at different corrugation Fig. 18 shows the velocity vectors inside one of the corrugations of the
depths are depicted in Figs. 15, 16, and 17. The corrugation pitch was heat exchanger plate with β = 60◦ andṁ = 6 Lpm at b = 2.91 mm and b
considered to be constant and equal to the commercial PHE. Addition­ = 1.66 mm, respectively. It is evident that at high corrugation depth,
ally, the hot fluid flow rate was also constant at 6 Lpm. The reason for vortices are formed, which changes the furrow movement of fluid par­
choosing this flow rate is related to the flow velocity in small PHEs in ticles at these points and is followed by growth in heat transfer and the

7
S. Mohebbi and F. Veysi Applied Thermal Engineering 188 (2021) 116616

40 and JF in terms of chevron angle variations over different pitches are

b=2.3 mm
35 shown in Figs. 19, 20, and 21. The constant depth equal to the com­
=8.3 mm
mercial specimen for the corrugation, as well as the constant hot side
30
flow rate, were considered. Moreover, the corrugations pitches were also
25 selected regarding the commercial specimen. At β = 60◦ and λ = 6.57
mm, the JF factor is maximal and has a 43% increase compared to λ =
Nu

20
15 11.5 mm. At β = 60◦ , with decreasing λ, the Nu-number and friction
coefficient increase. However, the heat transfer growth is more domi­
10
nant compared to the friction enhancement, which results in the sub­
5 sequent increase in the JF factor. In fact, decreasing the pitch increases
0 the number of corrugations in a given width, which implies the
0 15 30 45 60 75 90 increased effective heat transfer area. Also, decreasing the pitch in­
(degree) creases the number of distortions and this increases the friction. In the
absence of the chevron angle, no significant effect on the JF factor would
hot flow rate=6 Lpm hot flow rate=4 Lpm
be occurred by the variation of the corrugation pitch. Moreover, in the
hot flow rate=2 Lpm case β = 60◦ and λ = 6.57 mm, by increasing the corrugation depth to
Fig. 11. Variations of Nu-number in terms of chevron angle in different
2.91 mm, the JF factor reaches the highest value of 1.6 that is 14%
flow rates. higher compared to b = 2.3 mm. Unfortunately, the performed numer­
ical or experimental investigations available in the open literature re­
ported the results for large-scaled PHE that does not resemble the
consequent increase in friction. Nevertheless, at lower corrugation
depth, the fluid particles move in a certain direction with the parallel
layers.
2.5
Another reason for the rise of the Nu-number with increasing the b=2.3 mm
corrugation depth is that with enhancing the corrugation depth, the =8.3 mm
contact surface between fluid and the channel wall will be incremented; 2
consequently, the surface enlargement factor (φ) incremented, giving
rise to new flow patterns. This can positively affect the heat transfer
1.5
process. Noteworthy, the average velocity of the fluid decreased with
increasing the corrugation depth, this will consequently decline the heat
JF

flux, but the presence of new flow patterns influenced the increase of Nu- 1
number resulting in its overall enhancement.
Another reason for increasing the friction coefficient by increment­
0.5
ing the corrugation depth is the decrease of the parameter (Db ), which is
ṁD
present in the definition of Re-number (NbL wμ
). Since the other parame­
0
ters in the definition of Re-number are constant, a decrement in (Db) can
0 15 30 45 60 75 90
decrease the Re-number, leading to an increase in the friction coeffi­
cient. Table 7 shows the changes in all the mentioned parameters for (degree)
ṁ = 4Lpm, β = 60◦ , and λ = 8.3mm. The mentioned parameters hot flow rate=6 Lpm hot flow rate=4 Lpm
exhibited similar trends for other chevron angles, corrugation pitches,
hot flow rate=2 Lpm
and flow regimes.
The variations of Nu-number, friction coefficient, Colburn factor, Fig. 13. Variations of JF in terms of chevron angle in different flow rates.

0.4 0.3
b=2.3 mm
=8.3 mm
0.35
0.25
0.3
j
0.2
0.25 f
0.2 0.15
j
f

0.15
0.1
0.1
0.05
0.05

0 0
0 15 30 45 60 75 90
(degree)

hot flow rate=6 Lpm hot flow rate=4 Lpm hot flow rate=2 Lpm

Fig. 12. Variations of fricton coefficient and Colburn factor in terms of chevron angle in different flow rates.

8
S. Mohebbi and F. Veysi Applied Thermal Engineering 188 (2021) 116616

It can be concluded that the performance of the small PHE improved

by increasing the corrugation depth and decreasing the corrugation
pitch. The corrugation depth and corrugation pitch can be defined in the
()
form of a bλ ratio whose increment will lead to an improvement in the
PHE performance. The variations of JF in terms of chevron angle in
()
different bλ ratios are shown in Fig. 22.

As mentioned, the small PHE performance improves with an increase

()
of bλ ratio. The increase in the performance of the PHE is accompanied

by an increase in its weight. Table 9 lists the changes in mass and de­
()
viation from the commercial PHE of fifteen plates for different bλ ra­
tios (the commercial PHE and the simulated PHE have fifteen plates with
L2 = 193mm, W2 = 83mm). In Table 9, the JF factor was also calculated
at ṁ = 6Lpm and β = 60◦ . For example, when bλ = 0.3, the JF factor and
the mass of the plates reached 1.35 and 0.52 kg, respectively. In fact, a
2% increase in weight improved the PHE performance by 35% as
compared to commercial PHE. So it seems reasonable to use the plate
with this new profile. It should be noted that changing the chevron angle
did not affect the mass of the PHE.
Table 10 shows the occupied volume of fifteen plates of a small PHE.
The reduction of the corrugation pitch at a constant length and width of
PHE (L2 = constant andW2 = constant) did not affect the volume of the
PHE and only increased its weight. Therefore, it is recommended to
decrease the corrugation pitch of existing commercial small PHEs rather
than increasing its corrugation depth.
By increasing the ratio of (bλ ) in small PHEs, A significant increase
occurred in the performance in terms of weight and volume gain. Upon
further enhancement of this ratio, the volume and weight will increase
more and more. Regarding the compactness of these heat exchangers,
the increase in mass and volume should be limited to some extent. Also,
an increase in the mass and volume involves a rise in the production
materials. Therefore, the economical investigation of this type of heat
exchangers can be addressed in another study.

Fig. 14. Two plates of heat exchanger plates with a:β = 0◦ , b: β = 90◦ and 60
hot flow rate=6 Lpm
c:β = 60◦ . =8.3 mm
50

studied PHE in this paper, and hence no comparison can be made with 40
any other numerical studies.
The incremental behavior of the Nu-number with decreasing the
Nu

30
corrugation pitch can be assigned to the rise of the heat flux and hence
20
the increase in Nu-number upon declining the corrugation pitch
decreased, despite the decrease in average velocity, due to the formation 10
of new flow patterns. Also, with decreasing corrugation pitch, the sur­
face enlargement factor increased, giving rise to a decrement in hy­ 0
draulic diameter (D = 2b 0 15 30 45 60 75 90
φ ). As the hydraulic diameter decreased, the Re-
(degree)
number also declined which enhanced the friction coefficient. Table 8
shows the variations in the mentioned parameters for ṁ = 6Lpm, β = b=1.66 mm b=2.08 mm b=2.49 mm b=2.91 mm
60◦ and b = 2.3mm. In other chevron angles, corrugation depths, and
Fig. 15. Variations of Nu-number in terms of chevron angle in different
flow regimes, the mentioned parameters exhibited similar trends.
corrugation depths.

Table 6
Variations of heat flux, average velocity, and wall shear stress of hot flow with ṁ = 4Lpm, b = 2.075 mm and λ = 8.3 mm.
Chevron 0 15 30 45 60 75 90
angle (degree)

Heat flux (W/m2) 41,783 53,850 57,026 60,152 60,529 51,236 42,633
Velocity (m/s) 0.06 0.063 0.069 0.12 0.15 0.067 0.07
Wall shear stress (N/m2) 0.09 0.14 0.19 0.51 0.7 0.15 0.11

9
S. Mohebbi and F. Veysi Applied Thermal Engineering 188 (2021) 116616

f 0.2
0.3 hot flow rate=6 lpm
0.18
=8.3 mm
0.25 0.16
0.14
0.2 0.12
f 0.1

j
0.15
j 0.08
0.1 0.06
0.04
0.05
0.02
0 0
0 15 30 45 60 75 90
(degree)
b=1.66 mm b=2.08 mm b=2.49 mm b=2.91 mm

Fig. 16. Variations of fricton coefficient and Colburn factor in terms of chevron angle in different corrugation depths.

1.8 Table 7
hot flow rate=6 Lpm Variations in parameters affecting Nu-number and friction coefficient in
1.6 =8.3 mm
different corrugation depth with ṁ = 4Lpm, β = 60◦ and.λ = 8.3mm
1.4
b (mm) 1.66 2.075 2.49 2.905
1.2 heat flux (W/m ) 2
62,511 60,529 58,616 56,467
uave (m/s) 0.18 0.15 0.16 0.14
1
1.095 1.145 1.19 1.255
JF

φ
0.8 D 1.83 1.75 1.68 1.59
b
0.6 Re 389 368 330 312
0.4

0.2
60
0 hot flow rate=6 Lpm
0 15 30 45 60 75 90 b=2.3 mm
50
(degree)
b=1.66 mm b=2.08 mm b=2.49 mm b=2.91 mm 40

Fig. 17. Variations of JF in terms of chevron angle in different corruga­

30
Nu

tion depths.

20

10

0
0 15 30 45 60 75 90
(degree)
=6.57 mm =7.67 mm =9.2 mm =11.5 mm

Fig. 19. Variations of Nu-number in terms of chevron angle in different

corrugation pitches.

investigated numerically. After comparing the results of the numerical

model with the experimental results and verifying the results of the
numerical model, it was be deduced that in small-sized PHE, in addition
to the two main streams, i.e. longitudinal component and furrow
Fig. 18. Velocity vectors in one corrugation with β = 60◦ and.ṁ = 6Lpm
component of the stream, in some parts of the PHE channels, also two
kinds of vortices compared to the main streams were formed. These
6. Conclusions flows were highly influential on the thermal–hydraulic parameters. The
former vortex formed along with the longitudinal component between
In the present study, the effects of PHE geometry (chevron angle, two plates due to the existence of a chevron angle and the latter one was
corrugation depth, and pitch) and flow regime on the heat transfer and formed along with the furrow component and induced by the corruga­
friction in single-phase and steady-state conditions of water flow were tion depth. The heat transfer and friction were increased by both flows.

10
S. Mohebbi and F. Veysi Applied Thermal Engineering 188 (2021) 116616

0.3 0.2
hot flow rate=6 Lpm
b=2.3 mm 0.18
0.25
0.16
0.14
0.2
0.12
0.15 0.1
f

j
j 0.08
0.1
0.06
f
0.04
0.05
0.02
0 0
0 15 30 45 60 75 90
(degree)
=6.57 mm =7.67 mm =9.2 mm =11.5 mm

Fig. 20. Variations of fricton coefficient and Colburn factor in terms of chevron angle in different corrugation pitches.

1.6 1.6
hot flow rate=6 Lpm hot flow rate=6 Lpm
1.4 b=2.3 mm 1.4

1.2 1.2

1
1
0.8
JF

0.8
JF

0.6
0.6
0.4
0.4
0.2
0.2
0
0 0 15 30 45 60 75 90
0 15 30 45 60 75 90 (degree)
(degree) b/ =0.35 b/ =0.3 b/ =0.25 b/ =0.2
()
=6.57 mm =7.67 mm =9.2 mm =11.5 mm b
Fig. 22. Variations of JF in terms of chevron angle in different λ ratios.
Fig. 21. Variations of JF in terms of chevron angle in different corruga­
tion pitches.
Table 9
b
Variations of mass, surface enlargement factor, and JF in different ( ) ratios.
Table 8 λ
Variations in parameters affecting Nu-number and friction coefficient in b/λ 0.2 0.25 0.277 0.3 0.35 0.44
different corrugation pitch with ṁ = 6Lpm, β = 60◦ and.b = 2.3mm.
m (kg) 0.48 0.5 0.51 0.52 0.54 0.6
λ(mm) 6.57 7.67 9.2 11.5 φ 1.095 1.145 1.17 1.19 1.255 1.37

heat flux (W/m2) 84,207 79,065 76,675 64,430 JF 0.86 0.91 1 1.35 1.47 1.6
uave (m/s) 0.14 0.24 0.28 0.31 mass change(%) − 6 − 2 0 +2 +6 +18
φchange(%) − 6 − 2 0 +2 +6 +18
φ 1.255 1.19 1.145 1.095
JF change(%) − 14 − 9 0 35 47 60
D 1.59 1.68 1.75 1.83
b
Re 470 494 547 571 ()
b
increasing the ratio λ resulted in an increase in the JF factor, hence the
In all corrugation depths and corrugation pitches, the maximum Nu- use of such plates in the design of small-sized PHEs provides better
number and friction coefficient occurred at β = 60◦ and β = 30◦ , performance. Also, by increasing the corrugation depth or decrementing
respectively. Also, β = 60◦ was associated with the best performance, so the corrugation pitch, the weight and volume of the PHE showed an
that the JF factor was increased compared to the other chevron angles. enhancement. In small PHEs, however, this weight and volume gain
As the mass flow rate was decreased, the resultant decrease in heat versus performance improvement can be ignored. It is recommended to
transfer and increase in friction occurred, but eventually, the JF factor reduce the corrugation pitch rather than increasing its depth as a decline
was increased. Besides, an increase in corrugation depth (b) at β = 60◦ in corrugation pitch does not enhance the volume of the small PHE.
and λ = 8.3 mm resulted in an increase of 55%, and a decrease in According to the obtained results and their differences with large-scaled
corrugation pitch (λ) at β = 60◦ and b = 2.3 mm, resulted in the increase PHEs with chevron angle, it is proposed to provide correlations to
of 40% in JF factor compare to the commercial PHE. In general, calculate the heat transfer and friction coefficients in small-sized PHEs
in future studies. Besides, the inclusion of geometrical parameters

11
S. Mohebbi and F. Veysi Applied Thermal Engineering 188 (2021) 116616

Table 10
The occupied volume of plates.
b ¼ 2.3 λ = 8.3 λ = 6.57
λ = 6.57 λ = 7.67 λ = 8.3 λ = 9.2 λ = 11.5 b ¼ 1.66 b ¼ 2.08 b ¼ 2.49 b ¼ 2.91 b ¼ 2.91
b/λ 0.35 0.3 0.277 0.25 0.2 0.2 0.25 0.3 0.35 0.44
V (cm3) 6.25 6.25 6.25 6.25 6.25 4.71 5.71 6.71 7.71 7.71
volume change (%) 0 0 0 0 0 − 25 − 9 +7 +23 +23

should be considered. [15] A.G. Kanaris, A.A. Mouza, S.V. Paras, Flow and heat transfer prediction in a
corrugated plate heat exchanger using a CFD code, Chem. Eng. Technol. 29 (8)
(2006) 923–930, https://doi.org/10.1002/(ISSN)1521-412510.1002/ceat.v29:
Declaration of Competing Interest 810.1002/ceat.200600093.
[16] F.-B. Liu, Y.-C. Tsai, An experimental and numerical investigation of fluid flow in a
The authors declared that there is no conflict of interest. cross-corrugated channel, Heat Mass Transf. 46 (5) (2010) 585–593, https://doi.
org/10.1007/s00231-010-0605-7.
[17] Y.-C. Tsai, F.-B. Liu, P.-T. Shen, Investigations of the pressure drop and flow
References distribution in a chevron-type plate heat exchanger, Int. Commun. Heat Mass 36
(6) (2009) 574–578, https://doi.org/10.1016/j.icheatmasstransfer.2009.03.013.
[1] T.A. Elmaaty, A.E. Kabeel, M. Mahgoub, Corrugated plate heat exchanger review, [18] K. Sarraf, S. Launay, L. Tadrist, Complex 3D-flow analysis and corrugation angle
Renew. Sustain. Energy Rev. 70 (2017) 852–860, https://doi.org/10.1016/j. effect in plate heat exchangers, Int. J. Therm. Sci. 94 (2015) 126–138, https://doi.
rser.2016.11.266. org/10.1016/j.ijthermalsci.2015.03.002.
[2] O.P. Arsenyeva, L.L. Tovazhnyansky, P.O. Kapustenko, G.L. Khavin, Optimal [19] I. Gherasim, N. Galanis, C.T. Nguyen, Heat transfer and fluid flow in a plate heat
design of plate-and-frame heat exchangers for efficient heat recovery in process exchanger. Part II: assessment of laminar and two equation turbulent models, Int.
industries, Energy 36 (8) (2011) 4588–4598, https://doi.org/10.1016/j. J. Therm. Sci. 50 (8) (2011) 1499–1511, https://doi.org/10.1016/j.
energy.2011.03.022. ijthermalsci.2011.03.017.
[3] M.M. Abu-Khader, Plate heat exchangers: Recent advances, Renew. Sustain. Energy [20] M. Ciofalo, J. Stasiek, M.W. Collins, Investigation of flow and heat transfer in
Rev. 16 (4) (2012) 1883–1891, https://doi.org/10.1016/j.rser.2012.01.009. corrugated passages – II. Numerical simulations, Int. J. Heat Mass Tran. 39 (1)
[4] R.A. Troupe, J.C. Morgan, J. Prifiti, The plate heater versatile chemical engineering (1996) 165–192, https://doi.org/10.1016/S0017-9310(96)85014-9.
tool, Chem. Eng. Prog. 56 (1960) 124–128. [21] H. Blomerius, C. Hlsken, N.K. Mitra, Numerical investigation of flow field and heat
[5] M. Reppich, Use of high performance plate heat exchangers in chemical and transfer in cross-corrugated ducts, ASME J. Heat Transf. 121 (1999) 314–321.
process industries, Int. J. Therm. Sci. 38 (11) (1999) 999–1008, https://doi.org/ https://doi.org/10.1115/1.2825982.
10.1016/S1290-0729(99)00109-X. [22] O. Pelletier, F. Strmer, A. Carlson, CFD simulation of heat transfer in compact
[6] T.S. Khan, M.S. Khan, M.-C. Chyu, Z.H. Ayub, Experimental investigation of single brazed plate heat exchangers, ASHRAE Trans. 111 (2005) 846–854.
phase convective heat transfer coefficient in a corrugated plate heat exchanger for [23] S. Jain, A. Joshi, P.K. Bansal, A new approach to numerical simulation of small
multiple plate configurations, Appl. Therm. Eng. 30 (8-9) (2010) 1058–1065, sized plate heat exchangers with chevron plates, ASME J. Heat Transfer 129 (2007)
https://doi.org/10.1016/j.applthermaleng.2010.01.021. 291–297, https://doi.org/10.1115/1.2430722.
[7] A. Durmuş, H. Benli, İ. Kurtbaş, H. Gül, Investigation of heat transfer and pressure [24] X.-H. Han, L.-Q. Cui, S.-J. Chen, G.-M. Chen, Q. Wang, A numerical and
drop in plate heat exchangers having different surface profiles, Int. J. Heat Mass experimental study of chevron, corrugated-plate heat exchangers, Int. Commun.
Tran. 52 (5-6) (2009) 1451–1457, https://doi.org/10.1016/j. Heat Mass 37 (8) (2010) 1008–1014, https://doi.org/10.1016/j.
ijheatmasstransfer.2008.07.052. icheatmasstransfer.2010.06.026.
[8] K. Nilpueng, T. Keawkamrop, H.S. Ahn, S. Wongwisesb, Effect of chevron angle and [25] J. Lee, K.S. Lee, Flow characteristics and thermal performance in chevron type
surface roughness on thermal performance of single-phase water flow inside a plate plate heat exchangers, Int. J. Heat Mass Tran. 78 (2014) 699–706, https://doi.org/
heat exchanger, In. Commun. Heat Mass 91 (2018) 201–209, https://doi.org/ 10.1016/j.ijheatmasstransfer.2014.07.033.
10.1016/j.icheatmasstransfer.2017.12.009. [26] J. Lee, K.S. Lee, Friction and Colburn factor correlations and shape optimization of
[9] S. Mohebbi, F. Veysi, An experimental investigation on the heat transfer and chevron-type plate heat exchangers, Appl. Therm. Eng. 89 (2015) 62–69, https://
friction coefficients of a small plate heat exchanger with chevron angle, Heat Mass doi.org/10.1016/j.applthermaleng.2015.05.080.
Transf. 56 (3) (2020) 849–858, https://doi.org/10.1007/s00231-019-02749-0. [27] H. Martin, A theoretical approach to predict the performance of chevron-type plate
[10] W.W. Focke, P.G. Knibbe, Flow visualization in parallel-plate ducts with heat exchangers, Chem. Eng. Process. 35 (4) (1996) 301–310, https://doi.org/
corrugated walls, J. Fluid Mech. 165 (1986) 73–77, https://doi.org/10.1017/ 10.1016/0255-2701(95)04129-X.
S0022112086003002. [28] O. Arsenyeva, P. Kapustenko, L. Tovazhnyanskyy, G. Khavin, The influence of plate
[11] G. Gaiser, V. Kottke, Flow phenomena and local heat and mass transfer in corrugations geometry on plate heat exchanger performance in specified process
corrugated passages, Chem. Eng. Technol. 12 (1) (1989) 400–405, https://doi.org/ conditions, Energy 57 (2013) 201–207, https://doi.org/10.1016/j.
10.1002/(ISSN)1521-412510.1002/ceat.v12:110.1002/ceat.270120157. energy.2012.12.034.
[12] A. Muley, R.M. Manglik, Experimental study of turbulent flow heat transfer and [29] D. Dović, B. Palm, S. Švaić, Generalized correlations for predicting heat transfer
pressure drop in a plate heat exchanger with chevron plates, ASME J. Heat Transf. and pressure drop in plate heat exchanger channels of arbitrary geometry, Int. J.
121 (1999) 110–117, https://doi.org/10.1115/1.2825923. Heat Mass Tran. 52 (19-20) (2009) 4553–4563, https://doi.org/10.1016/j.
[13] B. Kumar, A. Soni, S.N. Singh, Effect of geometrical parameters on the performance ijheatmasstransfer.2009.03.074.
of chevron type plate heat exchanger, Exp. Therm Fluid Sci. 91 (2018) 126–133, [30] R.J. Moffat, Describing the Uncertainties in Experimental Results, Exp. Therm
https://doi.org/10.1016/j.expthermflusci.2017.09.023. Fluid Sci. 1 (1) (1988) 3–17, https://doi.org/10.1016/0894-1777(88)90043-X.
[14] R.Y. Miura, F.C.C. Galeazzo, C.C. Tadini, J.A.W. Gut, The effect of flow [31] S. Harikrishnan, S. Tiwari, Heat transfer characteristics of sinusoidal wavy channel
arrangement on the pressure drop of plate heat exchangers, Chem. Eng. Sci. 63 (22) with secondary Corrugations, Int. J. Therm. Sci. 145 (2019) 105973, https://doi.
(2008) 5386–5393, https://doi.org/10.1016/j.ces.2008.07.029. org/10.1016/j.ijthermalsci.2019.105973.

12