Energy Conversion and Management 108 (2016) 309–321

Contents lists available at ScienceDirect

Energy Conversion and Management

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

Numerical investigation of wet-bulb effectiveness and water

consumption in one-and two-stage indirect evaporative coolers
Shahab Moshari, Ghassem Heidarinejad ⇑, Aida Fathipour
Department of Mechanical Engineering, Tarbiat Modares University, P.O. Box: 14115-143, Tehran, Iran

a r t i c l e i n f o a b s t r a c t

Article history: In this study, three configuration for two-stage indirect/indirect evaporative cooling systems (IEC/IEC)
Received 20 August 2015 were proposed (Type A, Type B and Type C) to determine what configuration produces a better wet-
Accepted 9 November 2015 bulb effectiveness (or better energy-saving). For this purpose, six cities with a variety of hot weather con-
Available online 21 November 2015
ditions with the dry-bulb in range of 31.9–46.66 °C were selected. Results show that under these three
configuration, the wet-bulb effectiveness of Type A, Type B and Type C varies over ranges of 62–68%,
Keywords: 76–81% and 85–91% respectively, whereas the effectiveness of a one stage IEC varies over a range of
Dimensionless water evaporation rate
54–60%. There is a common misconceive belief in the concept of water evaporation rate of an evaporative
Two-stage indirect/indirect evaporative
cooling system
cooling system, which were fueled by many articles; this belief is, if a cooler consumes less water it is an
Wet-bulb effectiveness environmentally friendly cooler for dry areas. A more accurate and practical definition is proposed in this
Regenerative evaporative cooler article named Dimensionless Water Evaporation Rate (DWER). The numerical results showed that Type B
is the optimum configuration, because of a range of 4–24% DWER saving could be obtained by Type B in
comparison with Type C whereas Type B increases the product air up to 32%. As well as IEC, in a counter-
flow regenerative evaporative cooler the DWER decreases as the primary airflow rate increases whereas
water consumption increases. Moreover, using Type B the index of thermal comfort was investigated
which showed that Type B could meet thermal comfort condition in two climatic zones of temperate-
dry and hot-dry.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Also, some disadvantages of evaporative cooling systems are as

Direct Evaporative Cooling system (DEC) and Indirect Evapora-
tive Cooling system (IEC) are two main groups in evaporative cool-
ing systems. In hot and humid or temperate and humid regions
where the DEC and IEC systems do not provide comfort condition,
mechanical vapor compression cooling systems are used. some
advantages of evaporative cooling systems in comparison with
mechanical vapor compression systems are as following:
(a) Reduction in electrical input power and input current.
(b) Using environmentally friendly liquid as the working fluid.
(c) Elimination of ozone-depleting fluid from refrigeration
cycle.
(d) Ability to induce fresh air into the room.
(e) low base price.
⇑ Corresponding author. Tel.: +98 21 82883361; fax: +98 21 88005040.
E-mail addresses: s.moshari@modares.ac.ir (S. Moshari), gheidari@modares.ac.ir,
gheidari@alum.mit.edu (G. Heidarinejad).
following:
(a) Water supply restrictions in an area where water is very
scarce.
(b) Evaporative cooling systems will have little or no cooling
effect in a moist environment.
(c) Due to supplied humid air in the room, the air of room will
get saturated.
(d) Evaporative cooling systems need more frequent
maintenance.
(e) Evaporative cooling systems need ductwork.
The simplest and oldest form of the evaporative cooler is DEC. In
a DEC a fan is used in order to propel outdoor air through a porous
wetted pad. In a DEC the pads are wetted using the gravity-driven
water which are supplied from the water sump using water pump.
In a direct evaporative cooler, sensible heat changes into the latent
heat therefore, the dry bulb temperature of the air decreases. In a
DEC, the heat and mass transferred between air and water,
decreases the dry bulb temperature of the inlet air while increases
the inlet air relative humidity. This process can be considered as an

http://dx.doi.org/10.1016/j.enconman.2015.11.022
0196-8904/Ó 2015 Elsevier Ltd. All rights reserved.
310 S. Moshari et al. / Energy Conversion and Management 108 (2016) 309–321

Nomenclature

cp specific heat capacity ðkJ=kg  CÞ W width of channel ðmÞ


Cw; CF water to air and fluid to air heat capacity rate ratios, WBT wet bulb temperature ð CÞ
respectively WER water evaporation rate ðl=hÞ
DH hydraulic diameter of channel ðmÞ x coordinate along length of channel ðmÞ

DBT dry bulb temperature ð CÞ
 y coordinate along width of channel ðmÞ
kg
DWER dimensionless water evaporation rate kg Primary air

h heat transfer coefficient ðW=m2 CÞ Greek letters
hc convective heat transfer coefficient ðkW=m2  CÞ g wet-bulb effectiveness
0 k
hfg evaporation heat of water at reference temperature k ratio of thermal conductivity kyx
(0 °C) ðkJ=kgÞ q density ðkg=m3 Þ
hv specific enthalpy of water vapor at water film tempera- l dynamic viscosity coefficient ðPa sÞ
ture ðkJ=kgÞ r surface wettability
k thermal conductivity ðW=mKÞ x Humidity ratio of moist air ðkg=kgðaÞÞ
Km convective mass transfer coefficient ðkg=m2 sÞ
L length of channel ðmÞ
Subscripts
Le Lewis factor a secondary air or working air (wet channel)
m mass flow rate ðkg=sÞ asw moist air in equilibrium with water surface
NTU number of heat transfer units
db dry bulb
Nu Nusselt Number f primary air or air (dry channel)
p pressure ðPaÞ i inlet
Q secondary airflow rate ðm3 =hÞ o outlet
Rv a ; Rwa water vapor and liquid water to dry air specific heat
pl wall
capacity ratios, respectively Rmax maximum living room
RH Relative Humidity ð%Þ Rmin minimum living room
Re Reynolds Number v water vapor
sp spacing between palate ðmÞ
vs saturation vapor
t thickness of the wall ðmÞ w water film

T temperature ð CÞ wb wet bulb

U overall heat transfer coefficient ðkW=m2 CÞ
V velocity ðm=sÞ

adiabatic cooling process, so, the enthalpy of the inlet air remains
constant. Therefore, in a DEC, the wet bulb temperature of the inlet
and outlet air has the same value. By adiabatic saturation process
in a DEC, minimum temperature of the product air will be limited
by the wet-bulb temperature of the ambient air. Parameters that
can affect the wet bulb effectiveness of a DEC are: air velocity
through the pad, pad thickness, inlet air temperature and wetted
surface area per unit volume of the material.
When the increase in the air moisture content is not desired, the
indirect evaporative cooling system shall be used. In an indirect
evaporative cooling system there are two air streams, primary
and secondary air. In the wet channel, secondary air is directly
cooled by evaporation of the water. In fact, water film absorbs sen-
sible heat of the secondary air and converts it into the latent heat
by water evaporation, therefore the secondary air will be humidi-
fied. In the dry channel, due to the heat conduction through the
separating wall between wet and dry channels, primary air stream
will be cooled without any increase in the humidity ratio of the pri-
mary air. The product air (primary air) leaves the dry channel at a
lower wet bulb temperature.
Indirect evaporative cooling systems can be classified into the
two main groups: above-wet bulb cooling and sub-wet bulb cool-
ing. Using pre-cooling the secondary air (working air, before it
enters the wet channel) indirect evaporative cooling systems can
be used for sub-wet bulb cooling. This sub-wet bulb cooler is called
as Regenerative Evaporative Cooler (REC). In a regenerative evapo-
rative cooler, a portion of inlet air stream is extracted for product
air at the end of the dry channel (turning point) and the remaining
air stream is being diverted into the wet channel. Fig. 1(a) and (b)
shows the schematic of working principle for an indirect evapora-
tive cooling system and a regenerative evaporative cooling system
respectively. Fig. 1(c) shows the element that was used in mathe-
matical model of an indirect evaporative cooling system. This ele-
ment consists of the dry channel, the plate wall, the water layer
and the wet channel. In addition, the origin of the used x-y-
coordinate and exchanger dimensions are shown in the Fig. 1(c).
An early study on characteristics of an IEC was done by Pescod
[1]. With respect to low thermal conductivity of the plastic plates,
Pescod [1] showed that in an exchanger the heat transfer resis-
tance of thin plastic plates would be less than the thermal resis-
tance between the air and plate in dry channel. His model
showed discrepancy with experimental data in numerical study
of efficiency for an IEC. Maclaine-cross and Banks [2] based on
analogy to dry surface heat exchangers proposed a simplified
model for heat and mass transfer process in the exchanger. They
assumed stationary water layer, whereas water continuously
replenished with water at the same temperature. Chen et al. [3]
proposed a heat and mass transfer model for calculations of ther-
mal and hydraulic performance of an indirect evaporative cooler.
They have presented a universal model, which could be used in
simulation of tube- and plate-types of indirect evaporative cooling
systems. Their numerical results were presented for water evapo-
ration rate, air pressure drops, cooling capacity, COP, effectiveness,
and power demand of the unit. In order to calculate the perfor-
mance of an indirect evaporative cooler, a user-friendly model
were proposed by Alonso et al. [4]. Their model was universal
and could be used in order to analyze different indirect evaporative
coolers and energy analysis as well as for system or product opti-
mization. Joudi and Mehdi [5] evaluated two arrangements of an
indirect/direct evaporative cooling system. They considered four
 S. Moshari et al. / Energy Conversion and Management 108 (2016) 309–321
Fig. 1. Schematic of heat and mass exchanger, (a) counter-flow IEC, (b) counter-flow REC, (c) studied geometry.
operating modes for variable ambient temperature and cooling
load conditions during the summer season. They showed that the
coefficient of performance of the system tends to be very high
because the system consumes only fan and water pumping power.
Ren and Yang [6] developed an analytical solution for simultaneous
heat and mass transfer in a counter-flow IEC, Using a set of coupled
four Ordinary Differential Equations (ODEs), they described sec-
ondary airflow temperature, mass balance equation of the mois-
ture in the wet channel, water layer temperature and the
primary airflow temperature. Their model was one of the most
complete models for heat and mass transfer in an indirect evapora-
tive cooling system until that time.
Using heat and mass transfer principles in an indirect exchan-
ger, Heidarinejad and Bozorgmehr [7] studied a two-stage indi-
rect/direct evaporative air cooler. They showed that a two-stage
indirect/direct cooling system could provide comfort conditions
for some cities whereas they used only water as the cooling refrig-
erant. Heidarinejad et al. [8] experimentally studied the perfor-
mance of an indirect evaporative cooling stage followed by a
direct evaporative cooling stage in various climatic conditions.
They showed that the wet-bulb effectiveness of a stand-alone IEC
is in range of 55–61% whereas is in range of 108–111% for a two-
stage IEC/DEC. They showed that more than 60% power saving
can be obtained using this two-stage IEC/DEC in comparison to a
mechanical vapor compression system while its increase in water
consumption was around 55% over a DEC system. Farmahini
et al. [9] studied a two-stage cooling system consists of a nocturnal
radiative unit, a cooling coil and an indirect evaporative cooler. In
their study, in summer during the night, using a nocturnal radia-
tive cooling unit, chilled water was stored in a storage tank. During
the next day, chilled water was used for the cooling stage followed
by an indirect evaporative cooler. They proposed three models in
order to supply the secondary airflow for indirect evaporative
cooler. They proposed that this energy-efficient system can be used
as an alternative to the mechanical vapor compression systems.
311
Heidarinejad et al. [10] studied the performance analysis of a
ground-assisted hybrid evaporative cooling system. In their simu-
lation a Ground Coupled Circuit (GCC) followed by a direct evapo-
rative cooler, provides the necessary pre-cooling effects for sub
wet-bulb cooling of inlet air. They showed that the combination
of GCC and DEC system is capable in order to provide comfort con-
ditions whereas DEC alone did not. They proposed that this novel
hybrid system is an environmentally clean and energy efficient
system, which can be used as an alternative to the mechanical
vapor compression systems. Khalajzadeh et al. [11] investigated
thermal analysis of a novel integrated system consists of a ground
heat exchanger followed by an indirect evaporative cooler in sum-
mer conditions of Tehran. In their simulation, ground heat exchan-
ger was used in order to pre-cool the inlet air of indirect
evaporative cooler. Using three-dimensional computational fluid
dynamics (CFD), they investigated a four vertical ground heat
exchanger in series configuration. They showed that this combina-
tion of ground-coupled circuit and the indirect evaporative cooler
could easily provide comfort conditions as a clean and efficient
air conditioner. Farmahini and Heidarinejad [12] studied a multi-
step system of nocturnal radiative cooling followed by a two-
stage evaporative cooling system. In their simulation, during the
night, the water was circulated from a storage tank into the two
radiative panels. During the next day, in order to pre-cool the inlet
air of the next stage (indirect/direct) the stored cold water of the
storage tank were used as coolant for cooling coil unit. They
showed that energy-saving of this multi-step system was around
75–79% compared to mechanical vapor compression systems.
Woods and Kozubal [13] presented modeling and experimental
results on a two-stage air conditioner consists of a liquid desiccant
dehumidifier followed by an indirect evaporative cooler. They
reported several experimental data for their new liquid desiccant
air conditioner over a range of inlet temperatures, inlet humidities
and air flow rates. Anisimov et al. [14] numerically studied five dif-
ferent exchangers utilizing the Maisotsenko cycle (M-Cycle) for
312 S. Moshari et al. / Energy Conversion and Management 108 (2016) 309–321


indirect evaporative cooling system. In order to find thermal calcu- (d) Temperate (moderate) and dry climate (DBT < 40 C
lations of the indirect evaporative cooling system they used mod- and WBT < 23  C)
ified e-NTU method. They showed high efficiency gains, which 
(e) Temperate (moderate) and humid climate (DBT < 40 C
were sensitive to various inlet air conditions. 
and WBT P 23 C)
Cui et al. [15] presented an analytical model for indirect evapo-
rative heat exchangers based on a modified Log Mean Temperature These six cities are included all of five mentioned climate con-
Difference (LMTD) method. Their analysis was carried out to ditions ((a) Yazd, (b) Bam, (c) Abadan, (d) Tehran and Mashhad,
demonstrate a practical method to provide an accurate result with (e) Rasht). In this study, the optimum configuration for two-stage
a short computational time. Buker et al. [16] experimentally ana- IEC/IEC is proposed which can be used for air conditioning applica-
lyzed a new concept on solar thermal energy driven liquid desic- tion in some cities as an alternative to the mechanical vapor com-
cant based dew point cooling system, which used several green pression systems. As other novelty of this study, dimensionless
technologies. They showed that their proposed tri-generation sys- water evaporation rate is clearly presented in this article as a more
tem was capable to provide about 3 kW of heating, 5.2 kW of cool- accurate and practical definition for investigation of water con-
ing power and 10.3 MW h/year power generation. Pandelidis and sumption of evaporative cooling systems. None of existing studies
Anisimov [17] presented numerical simulation of heat and mass conducted up to now have not presented comparative study of
transfer in eight different types of the M-Cycle heat and mass dimensionless water evaporation rate for two-stage IEC/IEC sys-
exchangers. Their numerical simulation reveals many unique fea- tems. Some of the above articles may studied water consumption
tures of the heat and mass exchangers, enabling an accurate pre- of evaporative cooling system, but it should be noted that this
diction of HMXs performance. Furthermore, they showed that the study presents an environmentally friendly definition for water
effectiveness of the heat and mass exchangers (HMXs) depends evaporation rate of evaporative cooling systems in dry climatic
strongly on the inlet air parameters. Alklaibi [18] experimentally conditions.
studied the performance of internal two-stage evaporative cooler
in comparison with direct evaporative cooler. In addition, he theo- 2. Numerical analysis
retically studied the performance of internal two-stage evaporative
cooler in comparison with both direct evaporative cooler and Two dimensional equations of heat and mass transfer in wet
external two-stage evaporative coolers. He showed that the effi- and dry channels of IEC and REC are reported by Moshari and
ciency of the internal evaporative cooler was less sensitive to air Heidarinejad [21] as following:
 

speed than direct evaporative cooler. His results also showed that @T a 1 r
¼ NTU 1 þ Rv a ðxasw  xa Þ ðT w  T a Þ ð1Þ
the supply air of the internal evaporative cooler had higher humid- ^
@y 1 þ xa Rv a Le
ity content in comparison with direct evaporative cooler. Younis
et al. [19] developed predictive mathematical models of the condi- @ xa r
tioned space in order to study the performance of the integrated ¼ NTUðxasw  xa Þ ð2Þ
^
@y Le
system. Using displacement ventilation (DV) system combined
with a novel evaporative cooled ceiling (ECC) they showed that 0
!
@T w R1 1 hfg
better efficiency was achieved at higher supply flow rates and ¼ NTU ðT pl  T w Þ  ðT W  T a Þ  þ T w ðRv a  Rwa Þ
^
@y Cw Cw C pa
lower supply air relative humidity. In addition, they showed that

DV/ECC achieved 36.2% energy saving compared to typical DV/ 1 r
 ðxasw  xa Þ ð3Þ
chilled ceiling system. Heidarinejad and Moshari [20] proposed a C w Le
new mathematical modeling of a cross-flow indirect evaporative
cooling system with consideration of wall Longitudinal Heat Con- @ 2 T pl @ 2 T pl
duction (LHC) and the effect of gravity-driven water temperature þk ¼ RR1 ðT w  T pl Þ  RR2 ðT pl  T f Þ ð4Þ
@ ^x2 @y^2
variation along the exchanger plate. They numerically investigated
a two-stage system of indirect/direct evaporative cooling system. @T f R2
Their results showed that with the same parameters of inlet air ¼ NTUðT pl  T f Þ ð5Þ
@ ^x CF
and exchanger a two-stage indirect/direct evaporative cooling sys-
tem in comparing to a one-stage IEC has around 50% higher wet- In above equations parameters are defined as:
bulb effectiveness. ky x y
As shown above, the interest in simulations of two-stage evap- L  W ¼ A; k ¼ ; ^x ¼ ; y ^ ¼ ; dx ¼ dy;
kx L W
orative cooling systems and energy-saving systems is constantly mf cpf mw cpw U1
growing. With respect to aforementioned researches, there has CF ¼ ; Cw ¼ ; R1 ¼ ;
ma cpa ma cpa hc
been no focus on different configuration for two-stage indirect/
indirect evaporative cooling systems. A purposed key of this study U2 AU 1 AU 2
R2 ¼ ; RR1 ¼ ; RR2 ¼ ;
is to present three configuration for two-stage indirect/indirect hc tkx tkx
evaporative cooling system (IEC/IEC) whereas two configuration hc A c pv cpw
NTU ¼ ; Rv a ¼ ; Rwa ¼ ;
(Type A and C) of these three configuration are presented for the ma C pa cpa cpa
first time in this article. In order to investigate the potential of  t  1  t  1
1 1
using two-stage IEC/IEC for air conditioning applications, condi- U1 ¼ þ 2 ; U2 ¼ þ 2 ;
hw kpl hf kpl
tions of six cities of Iran have been selected. Based on summer
qVDH l
design conditions, Iran is a multi-climate country, which, can be Re ¼ ; DH ¼ 2sp ; Pr ¼
divided to the five different regions as following [8]: l qa

Due to the small hydraulic diameter and low velocity in dry and wet
(a) Hot and dry climate (DBT P 40 C and WBT < 23  C) channels (Re < 2000), air streams can be treated as laminar flow

(b) Hot and semi-humid climate (DBT P 40 C and 23  C 6 [13–15], in additions, the air streams are assumed to be fully devel-

WBT 6 27 C) oped [15]. Nusselt number correlation was derived by Ranz and

(c) Hot and humid climate (DBT P 40 C and 27  C < WBT) Marshall [22,23] as following:
 S. Moshari et al. / Energy Conversion and Management 108 (2016) 309–321 313

Nu k Table 2
h¼ ; Nu ¼ 2:0 þ 0:6Re1=2 Pr 1=3 ð6Þ Specifications of cross-flow IEC.
DH
Parameters Unit Values
Also, the value of hw obtained as following [24]:
WLH cm  cm  cm 50  50  40
Nuw kw Plates spacing mm 7
hw ¼ ; Nuw ¼ 0:023Re0:8
w Pr
0:3
ð7Þ Plates thickness mm 0.3
DH
Height of projections mm 7
The convective mass transfer coefficient can be found from the Spacing between projections mm 20
Lewis factor as following: Wall thermal conductivity (ky = kx) W=ðm KÞ 0.1

hc
Km ¼ ð8Þ
LeC pa
The Lewis factor Le for air water mixture, proposed by Ren and
2
Yang [6] is equal to 0:873 at standard atmospheric conditions. Rel-
ative humidity and humidity ratio can be found as following [25]:

pv s  0:01 X8
Ln ¼ ð374:136  TÞ F i ð0:65  0:01TÞi1
22087:87 T þ 273:15 i¼1

F 1 ¼ 741:9242; F 2 ¼ 29:7210;F 3 ¼ 11:55286; F 4 ¼ 0:8685635

F 5 ¼ 0:1094098; F 6 ¼ 0:439993;F 7 ¼ 0:2520658; F 8 ¼ 0:05218684
   
pv s pasw
RH ¼ ; xasw ¼ 0:622 ð9Þ
pasw p  pasw
The applied boundary conditions for cross-flow IEC are:
^ ¼ 0Þ ¼ T a;i ; xa ðy
T a ðy ^ ¼ 0Þ ¼ xa;i ; T w ðy
^ ¼ 1Þ ¼ T w;av erage ðy^ ¼ 0Þ ð10Þ Fig. 2. Primary air temperature profile, comparing to experimental data [26].
   
@T pl  @T pl  @T pl  @T pl 
¼ ¼ ¼ ¼ 0; T f ð^x ¼ 0Þ ¼ T f ;i
@ x^ ^x¼0 @ ^x ^x¼1 @ y ^ y^¼0 @ y^ y^¼1
flow IEC, the inlet primary air temperature, inlet secondary air
Boundary conditions for REC with cross-flow configuration are: temperature and inlet Relative Humidity (RH) of the secondary
air are 35 °C, 25.5 °C and 41.5% respectively. More inlet data for
^ ¼ 0Þ ¼ T f ;av erage ð^x ¼ 1Þ; xa ðy
T a ðy ^ ¼ 0Þ ¼ xa;i ;
experimental setup are shown in Table 1.
^ ¼ 1Þ ¼ T w;av erage ðy
T w ðy ^ ¼ 0Þ ð11Þ In order to check the independency of the mesh size, four grid
   
@T pl  @T pl  @T pl  @T pl  sizes of 20  20, 50  50, 100  100 and 200  200, are investi-
¼ ¼ ¼ ¼ 0;T f ð^x ¼ 0Þ ¼ T f ;i
@ ^x ^x¼0 @ ^x ^x¼1 @ y ^ y^¼0 @ y^ y^¼1 gated. The grid simulation was carried out using a 100  100 grid
points. The geometrical specifications of the cross-flow IEC are
For numerical simulation of counter-flow REC with the afore- shown in Table 2.
mentioned model, Eq. (5) and the boundary condition of the sec- Fig. 2 shows the primary air temperature profile at the outlet of
ondary air will be changed as following: the dry channel of the present work in comparison with the exper-
imental data of Bozorgmehr [26]. With respect to Fig. 2, the max-
@T f R2
¼ NTUðT pl  T f Þ ð12Þ imum discrepancy between the calculated outlet primary air
^
@y CF temperatures and experimental data is around the 3.5% (i.e. the
highest differences in outlet temperatures is 0.89 °C) of primary
T a ðinletÞ ¼ T f ;av erage ðend of the dry channelÞ ð13Þ air temperature at ð^ ^ ¼ 0:5Þ. The primary air enters the
x ¼ 1; y
The finite difference method was used for discretization of the exchanger with the temperature of 35 °C at the boundary ^ x¼0
governing equations of heat and mass transfer. Further details and exits the exchanger with the average temperature 24.94 °C
about the discretization of the governing equations can be found at the boundary ^ x ¼ 1. The secondary air enters the exchanger with
in [20]. 25.5 °C and exits the exchanger with the average temperature
23.46 °C. In this simulation, the maximum decrease of the temper-
ature is 10.06 °C and for the primary air.
3. Validation of the model accuracy

3.1. Validation of cross-flow IEC

Table 3
The numerical results have been verified using the experimen- Input data for counter-flow REC and specifications of the exchanger [27].
tal data of cross-flow IEC [26]. In numerical simulation of cross-
Input data/parameters Unit Specification/values
Inlet air temperature (Tf,i) °C 30, 35, 40 and 45
Table 1 Inlet air Humidity g/kg 6.9, 11.2, 20 and 26.4
Input data for cross-flow IEC. Intake air velocity m/s 2.4
Wall material Cotton sheet coated with polyurethane
Input data Unit Values
Wall thickness mm 0.5
Inlet primary air temperature (Tf,i) °C 35 Channel length mm 1200
Inlet secondary air temperature (Ta,i) °C 25.5 Channel width mm 80
Inlet Relative Humidity of the secondary air (RHa,i) % 41.5 Channel gap mm 5
Mass flow rate of primary air (mf) kg/s 0.512 Working air to intake kg/kg 0.33
Mass flow rate of secondary air (ma) kg/s 0.253 air ratio
Mass flow rate of water (mw) kg/s 0.267 Water supplied g/h 60
314 S. Moshari et al. / Energy Conversion and Management 108 (2016) 309–321
Fig. 3. The counter-flow dew point evaporative cooling system [27].
3.2. Validation of counter-flow REC
The numerical results of the REC with counter-flow configura-
tion have been verified using the experimental data of Riangvi-
laikul and Kumar [27]. In this study, the inlet air temperature
varying from 30 to 45 °C and inlet humidity ratio ranging from
6.9 to 26.4 g/kg. The input data and parameters, which were used
for REC, are listed in the Table 3.
Fig. 3 Shows the schematic diagram of the REC with counter-
flow configuration, which was reported by Riangvilaikul and
Kumar [27]. In this dew point evaporative cooling system, intake
air enters the exchanger at the top of the exchanger (point No 1)
and passes vertically through, the dry channel and then circulates
to the wet channel with upward direction. Then a portion of the air
stream is extracted at the turning point (point No 2) in order to
provide the product air. Intake working air enters the wet channel
at the bottom of the exchanger (point No 3) while the water is
sprayed onto the top of the wet channel and exits the exchanger
at point No 4.
Fig. 4 shows the outlet air temperature in the dry channel of
the present work in comparison with the experimental data
of Riangvilaikul and Kumar [27]. In reported experimental data of
Fig. 4. Validation of counter-flow REC with the experimental data of Riangvilaikul
and Kumar [27].
Riangvilaikul and Kumar [27], the accuracy of the thermocouples
and hot-wire anemometer were ±0.2 °C and ±0.5 °C respectively.
As shown in Fig. 4, the average outlet air temperature in the dry
channel of the present work has a good agreement with the exper-
imental data of Riangvilaikul and Kumar [27]. Fig. 4 shows that the
maximum discrepancies between the numerical results of these
simulations and experimental data are around 7% of inlet air tem-
perature for the humidity ratio of 6.9 g/kg (at inlet air temperature
equal to 45 °C). The numerical results indicated that discrepancies
between the numerical results of these simulations and experimen-
tal data were considerably influenced by increases in inlet air tem-
perature. This means that in numerical simulation of a counter-flow
REC, the model is more sensitive to change in inlet air temperature
and less sensitive to change in inlet air humidity ratio.
As shown in Fig. 4, the lower outlet air temperature was
achieved for the humidity ratio of 6.9 g/kg. The outlet air temper-
ature increases by increases in the inlet humidity ratio. Moreover,
at a constant humidity ratio of the inlet air, the outlet air temper-
ature linearly increases by increases in the inlet air temperature. In
addition, by increases in the inlet air temperature, the differences
between the maximum and minimum outlet air temperature was
found to be around 13 °C. In order to obtained outlet air tempera-
tures below the 24 °C as an air conditioner using this system, these
simulations showed that the inlet humidity ratio should be below
11.2 g/kg.
4. Results and discussions
4.1. Numerical results of two-stage cooling IEC/IEC
If a first indirect evaporative cooling system is added to a sec-
ond indirect evaporative cooling system, a two-stage IEC/IEC is
gained. In this section, three configuration of two-stage IEC/IEC
are presented. The outlet primary air of the first stage (IEC 1) can
be used in order to supply the secondary air (Type A), primary
air (Type B) and both of the primary and secondary air (Type C)
of the second stage (IEC 2). The above mentioned configurations
are shown schematically in Fig. 5(a)–(c).
4.1.1. Numerical results of two-stage IEC/IEC (Type A)
Table 4 shows the numerical results of (Type A) IEC/IEC evapo-
rative air cooler for six cities of Iran. As it seen form Table 4 the
maximum decrease in dry bulb temperature (DB) in the second
stage (DB2-DB3) and two-stage IEC/IEC (DB1-DB3) are around
1.55 and 12.82 °C for Yazd and Abadan respectively.
Fig. 6 shows the psychrometric chart for (Type A) IEC/IEC evap-
orative air cooler with respect to the ASHRAE evaporative comfort
zone [29]. For all cities, points indicated by 2 and 3 indices repre-
sent air condition after first stage of IEC and second stage of IEC
(IEC/IEC) respectively. As shown in Fig. 6, using (Type A) IEC/IEC
evaporative air cooler, only Mashhad can meet the thermal com-
fort condition. Using of (Type A) IEC/IEC can be recommended for
some cities in temperate and dry climate regions (DBT < 40 °C
and WBT < 23 °C).
The mathematical expression of the wet-bulb effectiveness
(EFF) is as following:
T db;i  T db;o
gwb ¼ ð14Þ
T db;i  T wb;i
Fig. 7 showed that the wet-bulb effectiveness of (Type A) IEC/IEC
evaporative air cooler for six cities are in the range of 62–68%
(EFF 1-3). As shown in Fig. 7, the wet-bulb effectiveness of the sec-
ond stage IEC/IEC (EFF 1-3) is around 8% more than the first stage
IEC/IEC (EFF 1-2). This show the inefficiency of this two-stage con-
figuration (Type A) for thermal comfort condition applications.
 S. Moshari et al. / Energy Conversion and Management 108 (2016) 309–321 315

Fig. 5. Schematic diagram of two-stage IEC/IEC with various configuration, (a) Type A, (b) Type B, (c) Type C.

4.1.2. Numerical results of two-stage IEC/IEC (Type B) two-stage IEC/IEC (DB1-DB3) are around 4.41 and 15.33 °C for
Table 5 shows the numerical results of (Type B) IEC/IEC evapo- Yazd.
rative air cooler for six cities of Iran. As shown in Table 5 the max- Fig. 8 shows the psychrometric chart for (Type B) IEC/IEC evap-
imum decrease in temperature of the second stage (DB2-DB3) and orative air cooler. For all cities, points indicated by 2 and 3 indices
316 S. Moshari et al. / Energy Conversion and Management 108 (2016) 309–321

Table 4
Numerical results of a two-stage IEC/IEC (Type A) for six cities of Iran.

ID Cities Outdoor design condition IEC 1 IEC 2 Temperature decrease

DBT 1 WBT RH Humidity ratio DBT 2 WBT RH DBT 3 WBT RH DB2-DB3 DB1-DB3
 
(°C) (°C) (%) kg (°C) (°C) (%) (°C) (°C) (%) (°C) (°C)
kgðdry airÞ
A Abadan 46.66 27.72 25.5 0.0196 35.25 25.28 46.7 33.84 24.96 50.5 1.41 12.82
B Bam 42.4 24.23 24 0.0147 31.92 21.64 42.5 30.51 20.78 51.1 1.41 11.89
M Mashhad 37.22 20.1 22.3 0.0101 27.94 17.38 37 26.5 16.94 40.2 1.44 10.72
R Rasht 31.9 25.72 62.6 0.0218 28.35 24.92 76.7 27.84 24.75 78.7 0.51 4.06
T Tehran 38.5 22.65 27.4 0.0136 29.59 20.31 45 28.32 19.96 48.4 1.27 10.18
Y Yazd 41.3 21.83 19.3 0.0111 30.38 18.83 35.2 28.83 18.38 38.5 1.55 12.47

Fig. 6. Psychrometric chart for two-stage IEC/IEC (Type A).

Fig. 8. Psychrometric chart for two-stage IEC/IEC (Type B).

Fig. 7. Wet-bulb effectiveness of two-stage IEC/IEC (Type A).
represent air condition after first stage of IEC and second stage of
IEC (IEC/IEC) respectively. As shown in Fig. 8, using (Type B) IEC/
IEC evaporative air cooler, Tehran, Yazd and Mashhad can meet
the thermal comfort condition. Type B IEC/IEC can be recom-
mended in two climatic regions: temperate and dry (DBT < 40 °C
and WBT < 23 °C) and hot and dry (DBT P 40  C and WBT <
23  C) climate regions.
As shown in Fig. 9 the wet-bulb effectiveness of (Type B) IEC/IEC
evaporative air cooler for six cities are in the range of 76–81% (EFF
1-3). As shown in Fig. 9, the wet-bulb effectiveness of the second
stage IEC/IEC (EFF 1-3) is around 22% more than the first stage
IEC/IEC (EFF 1-2). This shows the importance of the two-stage
(Type B) IEC/IEC evaporative cooler for energy-saving instead of
the mechanical vapor compression system in two climate zones
of temperate-dry and hot-dry climate.

Table 5
Numerical results of a two-stage IEC/IEC (Type B) for six cities of Iran.

ID Cities Outdoor design condition IEC 1 IEC 2 Temperature decrease

DBT 1 WBT RH Humidity ratio DBT 2 WBT RH DBT 3 WBT RH DB2-DB3 DB1-DB3
 
(°C) (°C) (%) kg (°C) (°C) (%) (°C) (°C) (%) (°C) (°C)
kgðdry airÞ
A Abadan 46.66 27.72 25.5 0.0196 35.25 25.28 46.7 31.5 24.32 59.1 3.75 15.16
B Bam 42.4 24.23 24 0.0147 31.92 21.64 42.5 27.83 20.55 53.7 4.09 14.57
M Mashhad 37.22 20.1 22.3 0.0101 27.94 17.38 37 24.06 16.1 46.5 3.88 13.16
R Rasht 31.9 25.72 62.6 0.0218 28.35 24.92 76.7 26.93 24.59 83.4 1.42 4.97
T Tehran 38.5 22.65 27.4 0.0136 29.59 20.31 45 25.99 19.3 55.5 3.6 12.51
Y Yazd 41.3 21.83 19.3 0.0111 30.38 18.83 35.2 25.97 17.51 45.5 4.41 15.33
 S. Moshari et al. / Energy Conversion and Management 108 (2016) 309–321 317

Fig. 9. Wet-bulb effectiveness of two-stage IEC/IEC (Type B).

4.1.3. Numerical results of two-stage IEC/IEC (Type C)

As shown in Table 6 the numerical results of (Type C) IEC/IEC
evaporative air cooler for six cities of Iran are presented. As it seen
form Table 6 the maximum decrease in temperature of the second Fig. 10. Psychrometric chart for two-stage IEC/IEC (Type C).
stage (DB2-DB3) and two-stage IEC/IEC (DB1-DB3) are around 6.09
and 17.14 °C for Yazd and Abadan respectively.
Fig. 10 shows the psychrometric chart for (Type C) IEC/IEC evap-
orative air cooler. For all cities, points indicated by 2 and 3 indices
represent air condition after first stage of IEC and second stage of
IEC (IEC/IEC) respectively. As shown in Fig. 10, using Type C IEC/
IEC evaporative air cooler, Tehran, Yazd, Bam and Mashhad can
meet the comfort condition. Type C IEC/IEC can be recommended
in three climatic regions: temperate and dry (DBT < 40 °C and
WBT < 23 °C), hot and dry (DBT P 40  C and WBT < 23  C) and
hot and semi-humid (DBT P 40 °C and 23 °C 6 WBT 6 27 °C) cli-
mate zones.
As shown in Fig. 11 the wet-bulb effectiveness of (Type C) IEC/
IEC evaporative air cooler for six cities are in the range of 85–91%
(EFF 1-3). As shown in Fig. 11, the wet-bulb effectiveness of the
second stage IEC/IEC (EFF 1-3) is around 30% more than the first
stage IEC/IEC (EFF 1-2). This configuration shows more effective-
ness in comparison to Type A and Type B configuration, but this
configuration has an inherent problem. In order to supply primary Fig. 11. Wet-bulb effectiveness of two-stage IEC/IEC (Type C).
and secondary air of the second stage, using this configuration, pri-
mary air of the first stage is divided into the two parts by the ratio orator medium shall be known. In simplified form, water evapora-
of secondary air flow rate
¼ 7
. Therefore, the primary airflow of the sec- tion rate of IEC is considered by amount of increase in the humidity
primary air flow rate 15
ratio of secondary airflow multiplying by the secondary airflow
ond stage (product air) will decrease (around 32%) in comparison
rate during an evaporation process. Water evaporation rate is cal-
to other configuration. This is a serious limitation for Type C IEC/
culated as following [20]:
IEC configuration.
qa
WER ¼ 1000 Q ðxo  xi Þ ð15Þ
qw
4.2. Calculations of water evaporation rate

In order to calculate Water Evaporation Rate (WER) of a one- 4.2.1. Water evaporation rate of cross-flow IEC system
stage IEC system, some parameters such as airflow rate, dry-bulb The numerical results of the indirect evaporative cooling system
and wet-bulb difference of inlet air and effectiveness of the evap- have been verified using the results of Chen et al. [3]. In these

Table 6
Numerical results of a two-stage IEC/IEC (Type C) for six cities of Iran.

ID Cities Outdoor design condition IEC 1 IEC 2 Temperature decrease

DBT 1 WBT RH Humidity ratio DBT 2 WBT RH DBT 3 WBT RH DB2-DB3 DB1-DB3
 
(°C) (°C) (%) kg (°C) (°C) (%) (°C) (°C) (%) (°C) (°C)
kgðdry airÞ
A Abadan 46.66 27.72 25.5 0.0196 35.25 25.28 46.7 29.52 23.95 64.6 5.73 17.14
B Bam 42.4 24.23 24 0.0147 31.92 21.64 42.5 26.29 20.12 58.8 5.63 16.11
M Mashhad 37.22 20.1 22.3 0.0101 27.94 17.38 37 22.51 15.65 51.1 5.43 14.71
R Rasht 31.9 25.72 62.6 0.0218 28.35 24.92 76.7 26.4 24.47 86 1.95 5.5
T Tehran 38.5 22.65 27.4 0.0136 29.59 20.31 45 24.62 18.9 60.2 4.97 13.88
Y Yazd 41.3 21.83 19.3 0.0111 30.38 18.83 35.2 24.29 16.99 50.3 6.09 17.01
318 S. Moshari et al. / Energy Conversion and Management 108 (2016) 309–321

Table 7 simulations, the outdoor air temperature, primary and secondary

Input data for IEC [3]. airflow rate are as listed in Table 7. The barometric pressure is



Outdoor air Primary airflow m3
Secondary airflow m3 fixed at 101.325 kpa.
s s
DBT (°C) WBT (°C) Table 8 shows the specifications of the exchanger which were
used in these simulations.
42 35 2.3 0.38
32
Fig. 12(a) and (b) shows the water evaporation rate of a one-
29 stage IEC with inlet dry bulb 38 and 42 °C (respectively) comparing
26 to the numerical results of Chen et al. [3]. As shown in Fig. 12
23 (a) and (b) the WER linearly decreases as the outdoor wet bulb
20
increases (when inlet dry bulb is fixed).
38 32 2.3 0.38 This simulation shows that for an indirect evaporative cooler
29
with inlet dry bulb and wet bulb of 38 and 26 °C and primary air-
26 3
23 flow rate of 2.3 ðms Þ ¼ 4873:4 CFM, the calculated water evapora-
20


tion rate is equal to 6.43 ðgsÞ ¼ 23:14 kg
h
. In order to accurately
compare water evaporation rate of various evaporative coolers, it
is better to define water evaporation rate in a dimensionless form.
Dimensionless water evaporation rate is defined as following:

WERðkgÞ
DWER ¼ h
ð16Þ
Table 8 Primary air flow rateðkg
h
Þ
The specifications of the exchanger [3].

Parameters Unit Values 4.2.2. Dimensionless water evaporation rate

Width of primary air passage m 0.0048
Width of secondary air passage m 0.004
Height of primary air passage m 0.48
Length of primary air passage m 0.267
Cross section of secondary air passage mm 0.004  0.267
Length of secondary air passage m 0.535
Number of passage for primary air 101
Number of passage for secondary air 100
Based on Eq. (16), DWER is calculated for (Type B) IEC/IEC for six
cities of Iran with new design operation point of 0.472 and 0.236


m3
s
for primary and secondary airflow rate. The geometrical spec-
ifications of the cross-flow IEC is the same as mentioned in Table 2.

Fig. 12. Water evaporation rate of one-stage IEC comparing to result of Chen et al. Fig. 13. Dimensionless water evaporation rate, (a) IEC and IEC/IEC for Type B, (b)
[3], (a) inlet dry bulb 38 °C, (b) inlet dry bulb 42 °C. Comparison between Type B and Type C.
 S. Moshari et al. / Energy Conversion and Management 108 (2016) 309–321 319

Table 10
Input data for counter-flow REC.

Outdoor air Primary Secondary

3 3
airflow ðms Þ airflow ðms Þ
DBT WBT Humidity ratio


(°C) (°C) g
kgðdry airÞ

34 21.7 13.92 2.3 0.38

1.6
0.9
0.5
38 22.7 13.92 2.3 0.38
1.6
0.9
0.5
42 23.69 13.92 2.3 0.38
1.6
0.9
0.5
Fig. 14. Comparison of DWER of DEC [8], IEC/DEC [8], IEC and IEC/IEC.

Dimensionless water evaporation rate for both two-stage IEC/IEC

and one-stage IEC are shown in Fig. 13(a). As shown in Fig. 13(a),
the DWER of IEC/IEC is in the range of 53–74%, which is more than
one-stage IEC. Using new design operation point of 0.472 and 0.236


m3
s
for primary and secondary airflow rate of the second stage,
the impacts of Type C configuration on DWER are investigated. This
new design operation point indicates that Type C gives the same
amount of product air as Type B. Fig. 13(b) shows the comparison
of DWER between Type B and Type C for six cities of Iran.
As shown in Fig. 13(b), the discrepancy of DWER between Type
C and Type B is in the range of 4–24% of DWER of Type B. As a
result, by decreases in the Relative humidity of the inlet air, the
discrepancy of DWER between Type C and Type B configuration
will decrease. It should be noted that in order to supply the men-
tioned airflow rate for the second stage of the air conditioners, in Fig. 15. Water evaporation rate of counter-flow REC.


3
the first stage Type C and Type B need 0.708 and 0.472 ms for pri-
mary airflow rate respectively. Due to the more airflow rate needed
in the first stage of Type C, more electrical power is needed com-
paring to Type B. Based on the obtained results it can be concluded
that, Type B configuration is a more efficient configuration for two-
stage air conditioner in the dry and hot climatic conditions.
In this section, for comparison of dimensionless water evapora-
tion rate of various systems, Tehran, Yazd and Bam are considered.
For this simulation, the experimental data of a stand-alone DEC [8]
and two-stage IEC/DEC [8] are compared with a one-stage IEC and
a two-stage IEC/IEC. As shown in Fig. 14, DWER of a two-stage IEC/
IEC (for Tehran, Yazd and Bam) is around 20–23% less than a two-
stage IEC/DEC, while it can provide thermal comfort condition for
Tehran and Yazd.
The comparisons of wet-bulb effectiveness (EFF) and DWER for
DEC, IEC, IEC/DEC and IEC/IEC (Type B and Type C) for Tehran are
presented in Table 9. It should be noted that all of the mentioned
percentage increase of DWER in Table 9 are calculated based on
Fig. 16. DWER of counter-flow REC.
a stand-alone IEC.
As shown in Table 9, due to energy losses (between wet and dry the wet-bulb effectiveness of direct evaporative cooling system is
channels of IEC in comparison with direct contact of product air much higher (around 37%) than indirect evaporative cooling
and water layer in DEC) in indirect evaporative cooling systems, system. Despite the higher wet-bulb effectiveness of direct

Table 9
Comparison of DWER and EFF for DEC, IEC/DEC, IEC and IEC/IEC.

DEC IEC/DEC IEC IEC/IEC (Type B) IEC/IEC (Type C)

DWER (kg/kg) 0.0059 0.0092 0.0047 0.0075 0.0079
Percentage increase in DWER 25% [8] 95% [8] 0% 59% 67%
Wet-bulb effectiveness (%) 93% [28] 111% [8,20] 56% 79% 87%
320 S. Moshari et al. / Energy Conversion and Management 108 (2016) 309–321
Table 11
Index of thermal comfort of a two-stage IEC/IEC (Type B) for six cities of Iran.
ID Cities Outdoor summer design condition
 
T max ð CÞ T min ð CÞ
A Abadan 46.66 28.88
B Bam 42.4 30.18
M Mashhad 37.22 21.11
R Rasht 31.9 22.73
T Tehran 38.5 20.72
Y Yazd 41.3 26.30
evaporative cooling system in comparison to indirect evaporative
cooling system, it needs 25% more DWER. Therefore, When the
increase in the air moisture content and water consumption are
not desired, the indirect evaporative cooling system should be used.
Also, with respect to Fig. 14 and Table 9 it can be concluded that
DWER of a two-stage IEC/IEC is around 23% (or 36% based on a
stand-alone IEC) less than a two-stage IEC/DEC. It should be noted that


3
this 23% less DWER (equal to 3.55 kg/h for 0:472 ms ¼ 1000 CFM)
for Tehran (Population of Tehran (metropolitan area): 13700000) with
around 4000000 house (If a cooler continuously operates for 12 h
daily), will be resulted in around 5 000 000 cubic meters of water
per one month. This show the importance of two-stage IEC/IEC
for megacities in regions with limited water resources.
4.2.3. Water evaporation rate of counter-flow REC system
In order to study the water evaporation rate of counter-flow
REC, the various outdoor air condition are assumed as listed in
Table 10.
Fig. 15 shows the water evaporation rate of a counter-flow REC
using input data of Table 10.
As the first result of Fig. 15, water evaporation rate increases as
the primary airflow rate increases. As the second result, water
evaporation rate increases with increase of inlet dry bulb temper-
ature (while humidity ratio is constant). This conclusion (first
result) is a rough investigation, using the concept of DWER, a
new understanding of water consumption will be presented.
Fig. 16 shows the DWER for above mentioned counter-flow REC
using input data of Table 10.
With respect to Fig. 16, DWER decreases as the inlet dry bub
temperature of the primary air decreases. As shown in Fig. 16,
the discrepancy of DWER between three inlet air temperatures
(34, 38 and 42 °C) decreases as the primary airflow rate increases.
In addition, by increases in the primary airflow rate, the discrep-
ancy of DWER decreases non-linearly with decreasing slope. The
numerical results indicate that the DWER is more sensitive to
changes in inlet primary airflow rate and less sensitive to changes
in inlet air temperature. Using the concept of DWER resulted that,
in order to minimize the water consumption of a cooler, primary
airflow rate should be at the maximum value of that cooling class
capacity.
4.3. Index of temperature variation in the indoor environment
The most widely used thermal comfort index is the predicted
mean vote (PMV) which indicates the mean thermal sensation vote
on a standard scale for a large group of persons [30]. In order to
find the PMV the following parameters shall be specified: dry
and wet bulb temperature, globe temperature, air velocity, clothing
insulation and human activity (for more detail see [31]). In this
study, the index of temperature variation in the indoor environ-
ment is investigated. In order to measure the thermal comfort of
the indoor condition in two-stage IEC/IEC systems, the index of
thermal comfort (IOC) is investigated in this study. IOC is the index
IEC/IEC
T Rmax ð CÞ T Rmin ð CÞ
Climate Thermal Comfort IOC
Hot and humid 29.52 20 No 0.22
Hot and semi-humid 26.29 20 No 0.16
Temperate and dry 22.51 20 Yes 0.09
Temperate and humid 26.4 20 No 0.15
Temperate and dry 24.62 20 Yes 0.13
Hot and dry 24.29 20 Yes 0.13
of temperature variation in the indoor environment of a building
[32]. It is expressed in terms of the difference between the maxi-
mum and minimum living room temperatures as following [33]:
T Rmax  T Rmin
IOC ¼ ð17Þ
T Rmax þ T Rmin
Smaller ICOs mean that room’s temperature has less fluctuations
and it has stable condition, which is a feature of a comfortable con-
dition [33]. In this simulation, the minimum temperature of the
room is assumed to be 20 °C (thermostat is set at 20 °C), which is
calculated based on the minimum temperature of the outdoor.
Moreover, the maximum temperature of the room is assumed equal
to the outlet air temperature of two-stage IEC/IEC (Type B). Table 11
shows the index of temperature variation for six cities of Iran.
As shown in Table 11, using two-stage IEC/IEC (Type B) the ICO
for Mashhad is less than other cities. This mean that using this sys-
tem, room air temperature has less fluctuations and has stable con-
dition. In addition, for three cities, which meet thermal comfort
condition (ASHRAE evaporative comfort zone), the value of IOC is
less than 0.13 while is in the range of 0.15–0.22 for other cities.
The numerical results indicate that in numerical simulation of a
two-stage IEC/IEC (Type B), the IOC is more sensitive to changes
in inlet wet bulb temperature and less sensitive to changes in inlet
dry bulb temperature. Aforementioned results showed the applica-
bility of two-stage IEC/IEC (Type B) for air conditioning applica-
tions in two climatic regions of temperate-dry and hot-dry climate.
5. Conclusions
In this study, the wet-bulb effectiveness of three-configuration
for two-stage IEC/IEC (Type A, Type B and Type C) were investi-
gated in various climatic conditions of six cities of Iran. These six
cities are consists of five climatic regions, hot-dry, hot-semi-
humid, hot-humid, moderate-dry and moderate- humid. The
numerical results of these simulations were validated against
experimental data. As the first study, the wet-bulb effectiveness
of these three configuration were investigated. Numerical results
show that under these three configuration, the average wet-bulb
effectiveness of Type A, Type B and Type C are around 8%, 21.5%
and 31% higher than that of the one stage IEC respectively. As a
result, the product air of the Type B is 32% higher than that of
the Type C. As the second study, the dimensionless water evapora-
tion rate is presented in this article, in order to accurately compare
water evaporation rate of various evaporative coolers. The study
results of DWER for two-stage evaporative cooling systems indi-
cated that, a two-stage IEC/IEC needs around 20–23% less DWER
comparing to a two-stage IEC/DEC. also, the numerical study of
water evaporation rate for counter-flow REC showed that DWER
decreases as the primary airflow rate increases whereas water
evaporation rate increases. The major conclusion of these two
studies is that Type B is the optimum configuration with a saving
range of 4–24% DWER in comparison with Type C.
As the third study, index of thermal comfort was investigated
for aforementioned six cities using two-stage IEC/IEC (Type B).
 S. Moshari et al. / Energy Conversion and Management 108 (2016) 309–321
The third study showed that for three cities, which meet thermal
comfort condition (ASHRAE evaporative comfort zone), the value
of IOC is less than 0.13. Based on the obtained results, Type B can
be used in order to save DWER and also as an energy efficient sys-
tem (using fan and pump) instead of the conventional mechanical
vapor compression systems (using compressors), while can pro-
vide comfort condition in vast region of Iran.
References
[1] Pescod D. A heat exchanger for energy saving in an air conditioning plant.
Trans ASHRAE 1979;85:238–51.
[2] Maclaine-Cross IL, Banks PJ. A general theory of wet surface heat exchangers
and its application to regenerative evaporative cooling. J Heat Transfer
1983;103(3):579–85.
[3] Chen PL, Qin HM, Huang YJ, Wu HF. A heat and mass transfer model for
thermal and hydraulic calculations of indirect evaporative cooler performance,
ASHRAE Transactions, LBL Report 28201, 1989.
[4] Alonso JF, Martinez FJ, Gomez E. Simulation model of an indirect evaporative
cooler. Energy Build 1998;29:23–7.
[5] Joudi KA, Mehdi SM. Application of indirect evaporative cooling to variable
domestic cooling load. Energy Convers Manage 2000;41(17):1931–51.
[6] Ren C, Yang H. An analytical model for the heat and mass transfer processes in
indirect evaporative cooling with parallel/counter flow configurations. Int J
Heat Mass Transfer 2006;49:617–27.
[7] Heidarinejad G, Bozorgmehr M. Heat and mass transfer modeling of two stage
indirect/direct evaporative air coolers. ASHRAE J Thailand Chapter 2008.
[8] Heidarinejad G, Bozorgmehr M, Delfani S, Esmaeelian J. Experimental
investigation of two-stage indirect/direct evaporative cooling system in
various climatic conditions. Build Environ 2009;44:2073–9.
[9] Farmahini M, Heidarinejad G, Delfani S. A two-stage system of nocturnal
radiative and indirect evaporative cooling for conditions in Tehran. Energy
Build 2010;42:2131–8.
[10] Heidarinejad G, Khalajzadeh V, Delfani S. Performance analysis of a ground-
assisted direct evaporative cooling air conditioner. Build Environ 2010;45
(11):2421–9.
[11] Khalajzadeh V, Farmahini-Farahani M, Heidarinejad G. A novel integrated
system of ground heat exchanger and indirect evaporative cooler. Energy Build
2012;49:604–10.
[12] Farmahini M, Heidarinejad G. Increasing effectiveness of evaporative cooling
by pre-cooling using nocturnally stored water. Appl Therm Eng
2012;38:117–23.
[13] Woods J, Kozubal E. A desiccant-enhanced evaporative air conditioner:
numerical model and experiments. Energy Convers Manage 2013;65:208–20.
[14] Anisimov S, Pandelidis D, Danielewicz J. Numerical analysis of selected
evaporative exchangers with the Maisotsenko cycle. Energy Convers Manage
2014;88:426–41.
321
[15] Cui X, Chua KJ, Islam MR, Yang WM. Fundamental formulation of a modified
LMTD method to study indirect evaporative heat exchangers. Energy Convers
Manage 2014;88:372–81.
[16] Buker MS, Mempouo B, Riffat SB. Experimental investigation of a building
integrated photovoltaic/thermal roof collector combined with a liquid
desiccant enhanced indirect evaporative cooling system. Energy Convers
Manage 2015;101:239–54.
[17] Pandelidis D, Anisimov S. Numerical analysis of the heat and mass transfer
processes in selected M-Cycle heat exchangers for the dew point evaporative
cooling. Energy Convers Manage 2015;90:62–83.
[18] Alklaibi AM. Experimental and theoretical investigation of internal two-stage
evaporative cooler. Energy Convers Manage 2015;95:140–8.
[19] Younis M, Ghali K, Ghaddar N. Performance evaluation of the displacement
ventilation combined with evaporative cooled ceiling for a typical office in
Beirut. Energy Convers Manage 2015;105:655–64.
[20] Heidarinejad G, Moshari S. Novel modeling of an indirect evaporative cooling
system with cross-flow configuration. Energy Build 2015;92:351–62.
[21] Moshari S, Heidarinejad G. Numerical study of regenerative evaporative
coolers for sub-wet bulb cooling with cross-and counter-flow configuration.
Appl Therm Eng 2015;89:669–83.
[22] Ranz WE, Marshall WR. Evaporation from drops. 1. Chem Eng Prog
1952;48:141–6.
[23] Ranz WE, Marshall WR. Evaporation from drops. 2. Chem Eng Prog
1952;48:173–80.
[24] Dittus PW, Boelter LMK. Heat transfer in automobile radiators of the tubular
type. Heat Mass Transf 1985;12:3–22.
[25] Heidarinejad G, Pasdarshahri H. The effects of operational conditions of the
desiccant wheel on the performance of desiccant cooling cycles. Energy Build
2010;42(12):2416–23.
[26] Bozorgmehr M. Mathematical modeling of one dimensional heat and mass
transfer in evaporative cooling and performance evaluation of indirect/direct
cooling systems, Tarbiat Modares University. PhD. thesis, (2009). (In Persian).
[27] Riangvilaikul B, Kumar S. An experimental study of a novel dew point
evaporative cooling system. Energy Build 2010;42:637–44.
[28] Moshari S. Modeling heat and mass transfer in indirect evaporative cooling
systems with considering the effect of two-dimensional wall heat conduction,
Tarbiat Modares University. PhD. thesis, 2015. (In Persian).
[29] ASHRAE handbook – fundamentals. American Society of Heating, Refrigerating
and Air-Conditioning Engineers, 2005).
[30] Atthajariyakul S, Leephakpreeda T. Neural computing thermal comfort index
for HVAC systems. Energy Convers Manage 2015;46(15):2553–65.
[31] Fathollahzadeh MH, Heidarinejad G, Pasdarshahri H. Prediction of thermal
comfort, IAQ, and energy consumption in a dense occupancy environment
with the under floor air distribution system. Build Environ 2015;90:96–104.
[32] Holman JP. Heat transfer. N.Y: McGraw-Hill; 2010.
[33] Owrak M, Aminy M, Jamal-Abad MT, Dehghan M. Experiments and
simulations on the thermal performance of a sunspace attached to a room
including heat-storing porous bed and water tanks. Build Environ
2015;92:142–51.