Energy Conversion and Management 47 (2006) 2034–2043

www.elsevier.com/locate/enconman

Experimental study and predictions of an induced

draft ceramic tile packing cooling tower
Esam Elsarrag *

Department of Mechanical Engineering, Technical Studies Institute, P.O. Box 39219, Abu Dhabi, United Arab Emirates

Received 9 January 2005; received in revised form 13 July 2005; accepted 24 December 2005
Available online 28 February 2006

Abstract

Deterioration of the filling material in traditional cooling towers is of serious concern. In this study, long life burned
clay is used as the filling material. It guards against common cooling tower problems resulting from chemical water treat-
ment and deterioration. The size of the ceramic packing material and outlet conditions predictions by theoretical modeling
require heat and mass transfer correlations. An experimental study to evaluate the heat and mass transfer coefficients is
conducted. The previous correlations found in the literature could not predict the mass transfer coefficient for the tested
tower. A mass transfer coefficient correlation is developed, and new variables are defined. This correlation can predict the
mass transfer coefficient within an error of ±10%. The developed correlation is used along with theoretical modeling to
predict the cooling tower outlet conditions within an error of ±5%.
 2006 Elsevier Ltd. All rights reserved.

Keywords: Cooling tower; Ceramic packing; Correlation

1. Introduction

The cooling tower is a steady flow device that uses a combination of mass and energy transfer to cool water
by exposing it as an extended surface to the atmosphere. The water surface is extended by filling, which pre-
sents a film surface or creates droplets. The air flow may be cross flow or counter flow and caused by mechan-
ical means, convection currents or by natural wind. In mechanical draft towers, air is moved by one or more
mechanically driven fans to provide a constant air flow. The function of the fill is to increase the available
surface in the tower, either by spreading the liquid over a greater surface or by retarding the rate of fall
of the droplet surface through the apparatus. The fill should be strong, light and deterioration resistant.
In this study, long life burned clay bricks were used as the filling material. Its hardness, strength and compo-
sition guard against common cooling tower problems resulting from fire, chemical water treatment and
deterioration.

*
Tel.: +971 507160317; fax: +971 84680127.
E-mail addresses: elsarrag@hotmail.com, elsarrag@emirates.net.ae

0196-8904/$ - see front matter  2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.enconman.2005.12.019
 E. Elsarrag / Energy Conversion and Management 47 (2006) 2034–2043 2035

Nomenclature

a area of heat and mass transfer (m2/m3)

Cp specific heat (kJ/kg K)
Dv diffusion coefficient (m2/s)
deq equivalent diameter for structured packing (m)
E efficiency (%)
h enthalpy (kJ/kg)
hc heat transfer coefficient (kW/m2 K)
K thermal conductivity (W/m K)
Kx mass transfer coefficient (kW/m2 K)
Le Lewis number
m flow rate (kg/s)
m0 superficial flow rate (mass velocity) (kg/m2 s)
M molecular weight (kg/kmol)
q heat flux (kJ/kg)
t temperature (C)
Z tower height (m)

Greeks
l viscosity (Ns/m2)
q density (kg/m3)
x humidity ratio (kg water/kg dry air)

Subscripts
a air
db dry bulb
i inlet or interface
m mean
o outlet
t total
v vapor
wb wet bulb
w water

Several studies on cooling tower analysis were developed with different points of view. Webb and Villacres
[1] described three computer algorithms that have been developed to perform rating calculations of three
evaporatively cooled heat exchangers. The heat and mass transfer characteristic equation of one of the heat
exchangers is derived from the manufacturer’s rating data at the design point. Jaber and Webb [2] produced
the effectiveness-NTU (number of transfer unit) design method for counter flow towers using Merkel’s simpli-
fication theory [3]. Braun et al. [4] presented effectiveness models for cooling towers and cooling coils. The
results of the models were compared with those of more detailed numerical solutions to the basic heat and
mass transfer coefficients and experimental data. Osterle [5] pointed out that the Merkel assumptions under-
estimated the required NTU. Soylemez and Unsal [6] presented an interactive computer code for sizing forced
draft, counter flow cooling towers using a closed formula offered by Unsal and Varol [7]. El-Dessouky et al. [8]
modified the analysis done by Jaber and Webb. They presented a solution for the steady state counter flow wet
cooling tower with new definitions of tower effectiveness and number of transfer units. Bernier [9] presented an
analysis of the basic heat and mass transfer processes occurring around a droplet in transient cooling of a
spray counter flow tower. The influence of fill height, water retention time and water–air flow ratio on the
tower performance was represented.
2036 E. Elsarrag / Energy Conversion and Management 47 (2006) 2034–2043

Nimr [10] presented a mathematical model to describe the thermal behavior of packed cooling towers. A
closed form solution was obtained for both the transient and steady temperature distribution in a cooling
tower.
Jameel et al. [11] investigated the heat and mass transfer mechanism and performance characteristics using
a detailed theoretical model in counter flow cooling towers.
Fisenko et al. [12] developed a new mathematical model of a mechanical draft cooling tower performance.
The model represented a boundary value problem for a system of ordinary differential equations describing a
change in the droplets velocity, its radii and temperature, and also a change in the temperature and density of
the water vapor in mist air in a cooling tower. Söylemez [13] presented a thermohydraulic performance opti-
mization analysis yielding a simple algebraic formula for estimating the optimum performance point of coun-
ter current mechanical draft wet cooling towers. The effectiveness-NTU method was used, together with
derivation of the psychrometric properties of moist air based on a numerical approximation method, for ther-
mal performance analysis of counter flow type towers. Kloppers and Kröger [14] investigated the effect of the
Lewis factor on the performance prediction of natural draft and mechanical draft wet cooling towers.
The design of a ceramic tile cooling tower requires heat and mass transfer correlations to estimate the pack-
ing height and to predict the water and air outlet conditions. Several workers have measured the heat and mass
transfer coefficient in cooling towers. Thomas and Houston [15] developed heat and mass transfer correlations
using a tower of 2 m height and 0.3 m2 cross section. They gave the following relations for the heat and mass
transfer coefficients: hc a ¼ 3:0m00:26
w ma
00:72
and k x a ¼ 2:95mw00:26 m00:72
a .
Lowe and Christie [16] measured the heat and mass transfer coefficients using 1.3 m2 experimental column
fitted with a number of different types of packing. They showed that in most cases hc aam01n 0n
w ma . This showed a
close agreement with the results of Thomas and Houston, when n = 0.75.
Jorge and Armando [17] tested a new closed wet cooling tower. They obtained experimental correlations for
the heat and mass transfer coefficients. They concluded that the existing thermal models were found to predict
0:8099
reliably the thermal performance of cooling towers. They found that K x a ¼ 0:1703ðm=mmax Þsair and hc a ¼
0:6584
700:3ðm=mmax Þwater .
Lebrun and Silva [18] generated a correlation between the global heat transfer coefficient UA from exper-
imental analysis as a function of water and air flow rates entering the tower. The best fit of the results gave the
following correlation: UA ¼ 745m00:43w ma
01:03
.
Here, an induced draft counter flow ceramic tile packed cooling tower has been designed and tested for heat
and mass transfer. The packing material used in the column was burned clay bricks. Some investigators have
used theoretical models to predict the performance of cooling towers. However, the theoretical model requires
that the heat and mass transfer coefficients be experimentally determined. In addition, these coefficients can be
used to estimate the size of the packed tower. Other investigators developed correlations to predict the heat
and mass transfer coefficients, but these correlations did not consider the effect of the packing type and the
mass transfer driving force on the mass transfer coefficient.
This paper presents the heat and mass transfer as calculated from the experimental measurements. The
results from the experimental data were used to develop a new mass transfer coefficient including new param-
eters. The developed correlation is used along with Merkel’s theoretical model to predict the ceramic tower air
and water outlet conditions.

2. Experimental setup

The tested cooling tower is an induced draft counter flow type. The schematic diagram of the tower is
shown in Fig. 1. The tower cross sectional area is 0.64 m2, the total height of the tower is 2 m and the filling
portion is 0.8 m. Burned clay bricks were used as the packing material with a maximum number of 12 layers
each consisting of 18 bricks. The brick dimensions are 235 · 120 · 64 mm, and its void fraction is 0.4. Water
distributors were used to distribute the water uniformly over the packing. An axial fan was fixed on the top of
the tower to extract the air from the bottom of the tower. Two pumps were used to circulate the hot water over
the packing in a counter manner to the air flow.
Before each experiment, the mains water was heated by solar collectors and stored in the tank. The water
was allowed to re-circulate through the solar collectors to obtain the desired water temperature. The flow
 E. Elsarrag / Energy Conversion and Management 47 (2006) 2034–2043 2037

Fig. 1. Schematic diagram for the tested ceramic cooling tower.

Table 1
Measuring devices specifications
Device Type Accuracy Operative range Fluid
Sling psychrometer Hand 0.1 C 0–60 C Air (DB & WB)
Thermometer Ordinary 0.1 C 0–100 C Water and air
Flowmeter Rotameter 0.06 l/s 0–0.4 l/s Water
Anemometer Vane 5% of measured air flow rate 0–5 m/s Air

regulating valve was adjusted and the water and air flow rates were measured. During the experiment, the hot
water was distributed uniformly over the filling material. The axial fan extracts the air from the bottom of the
tower in a counter manner to the water flow. The temperatures measurements were taken after allowing
enough time for steady state readings. The inlet air and water temperatures were measured before and during
the experiment. Three sets of experiments were performed for different heights of packing (0.77, 0.7 and
0.63 m). For each height, the experiments were performed for different water to air flow ratios, (1.13, 0.96,
0.8 and 0.7). The inlet water temperature was varied for each liquid to air flow rate ratio (42.5, 40, 38 and
35 C).
The measurements during the experiment are shown in Fig. 1. The specifications of the different measuring
devices are shown in Table 1. The rotameter was calibrated using the weighing cylinder method, ASME
Standard [19]. The uncertainties of the experimental parameters are shown as error bars in the discussion
figures.

3. Cooling tower theory

When air flow passes a wetted surface there is a transfer of sensible and latent heat. If there is a difference in
temperature between the air and the wetted surface, heat will be transferred. If there is a difference in the par-
tial pressure of water vapor in the air and that of the water, there will be a mass transfer. This transfer of mass
causes a thermal energy transfer because if some water evaporates from the water layer, the latent heat of this
vaporized water will be supplied to the air. The concept of enthalpy potential is a very useful one in quanti-
fying the transfer of heat (sensible and latent) in those processes and components where there is a direct con-
tact between the air and water.
2038 E. Elsarrag / Energy Conversion and Management 47 (2006) 2034–2043

The expression for transfer of the total heat dqt through a differential area dA is expressed by Stoecker [20]:
hc dA
dqt ¼ ðhi  ha Þ ð1Þ
Cp m
The name of enthalpy potential originates from the above equation because the potential for the transfer of
the sum of the sensible and latent heats is the difference between the enthalpy of the saturated air at the wetted
surface temperature hi and the enthalpy of the air stream ha.
The rate of heat removed from the water is equal to the rate gained by the air, so the following expression
can be written:
dqt ¼ ma dha ¼ 4:19mw dt ð2Þ

4. Heat and mass transfer coefficients calculation

The heat transfer coefficient can be calculated by equating Eqs. (1) and (2) and rearranging:
Z out
hc A dt
¼ 4:19 ð3Þ
C p m mw in hi  h a
However, A = aV and V = SZ, so Eq. (3) can be written as:
Z out
hc aZ dt
¼ 4:19 ð4Þ
C p m m0w in h i  ha
The relation between the heat and mass transfer coefficients is expressed by Reynold’s analogy [21]:
hc 2
¼ Le3 ð5Þ
K xCp m
It is found that in most cases of air–water contact, the Lewis number Le can be considered to be unity as a
good approximation [22]:
hc
¼ Kx ð6Þ
Cp m
By substituting Eq. (6) in Eq. (4), the mass transfer coefficient can be expressed as:
Z out
K x aZ dt
0
¼ 4:19 ð7Þ
mw in hi  ha
The integration of Eq. (7) is solved numerically by dividing the packed height into small segments starting
from the bottom to the top of the tower.

5. Results and discussion

The mass transfer coefficient found from the experimental data and the existing correlations were depicted
graphically, along with the design variables. The heat and mass transfer coefficients are related by Reynold’s
analogy, so the factors that influence the mass transfer coefficient also affect the heat transfer coefficient.
As shown in Fig. 2, the mass transfer coefficient increased with the increase of the water to air flow rate
ratio. However, it can be observed that there is some degree of difficulty in the mass transfer when a high water
to air flow rate was employed. The increase of the mass transfer coefficient with the water to air flow rate ratio
gave good agreement with the previous correlations. It can be observed that the Thomas and Houston [15]
values are closer than those of Leburn and Silva [18], but there is a big deviation between the experimental
values and those obtained by their correlations.
The mass transfer coefficient increased when the inlet water temperature increased from 35 to 38 C as
shown in Fig. 3, but it decreased when the water temperature was raised above 38 C (from 38 to 42.5 C).
This is mainly because the driving force increases with the increase of the inlet water temperature and a greater
 E. Elsarrag / Energy Conversion and Management 47 (2006) 2034–2043 2039

Fig. 2. Influence of liquid to air flow rate on mass transfer.

Fig. 3. Influence of inlet water temperature on mass transfer.

heat and mass transfer occurs, but a higher outlet water temperature was obtained by continued increasing of
the inlet water temperature. This is reflected as a decrease in the mass transfer coefficient.
The mass transfer coefficient decreased with the increase of the inlet air enthalpy as shown in Fig. 4. This is
due to the decrease in the driving force, which is reflected as a decrease in the mass transfer coefficient.

Fig. 4. Influence of inlet air enthalpy on mass transfer.

2040 E. Elsarrag / Energy Conversion and Management 47 (2006) 2034–2043

In fact, the correlations developed by Thomas and Houston [15], Lowe and Christie [16] and Leburn and
Silva [18] are functions of the air and water flow rate only. They did not consider the effect of the water tem-
perature and the weather conditions on the mass transfer coefficient in addition to the packing type. The heat
and mass transfer coefficients calculated from the previous correlations will remain constant with the variation
in the air and water temperatures. Therefore, a big deviation was found between the calculated heat and mass
transfer coefficient values and those obtained by these correlations.
From the discussion, it can be found that the factors that have a greater effect on the heat and mass transfer
coefficients are: the water to air flow rate ratio; the inlet water temperature and the inlet air enthalpy.

6. Correlation development

Developing a correlation for the mass transfer coefficient is necessary to predict the packing height and the
outlet conditions from the cooling tower. The factors found to have the greatest effect on the gas phase mass
transfer coefficient are: the water and air flow rates; the inlet air and water conditions; the packing volume and
equivalent diameter and the diffusion coefficient of water in air.
K x a ¼ f ða; d eq ; qv ; Dv ; m0w ; m0a ; ha ; hi Þ ð8Þ

The Buckingham Pi method was employed to obtain the pertinent dimensionless groups [23]:
!
d 2eq
p1 ¼ K x a
Dv qv
G

m0w
p2 ¼
m0a
ð9Þ
ha
p3 ¼
hi
!  
d 2eq ha m0w
Kxa ¼f ;
Dv qv hi m0a

The mass transfer correlation obtained from the dimensional analysis is given by:
!  a  0 b
d 2eq ha mw
K xa ¼c 1 ð10Þ
Dv qv hi m0a

The term ð1  hhai Þ represents the driving force (hi  ha), which measures the degree of difficulty of the mass
transfer.
The experimental data of this study were employed along with a curve fitting computer program to obtain
the constants of Eq. (10).
!  0:27  0 0:69
d 2eq ha mw
K xa ¼ 4:62 1  ð11Þ
Dv qv hi m0a

The heat transfer coefficient can be predicted using Reynold’s analogy (Eq. (7))
hc a
¼ K xa
Cp m

The deviation between the predicted values and the experimental data is shown in Fig. 5.
The mass transfer correlation can be predicted within an error of ±10%. This correlation can be used to
estimate the height of the burned clay filling portion by rearranging Eq. (7).
 E. Elsarrag / Energy Conversion and Management 47 (2006) 2034–2043 2041

Fig. 5. Relation between experimental and predicted mass transfer coefficient.

7. Modeling and predictions

Fig. 6 shows a differential volume of a counter flow cooling tower. For the finite difference model, the tower
fill height Z is divided into small segments, dZ, and the mass and energy balances are solved for each segment
from the bottom to the top of the tower. The governing equations that describe the changes in air humidity, air
temperature and water temperature across a segment are given below. A detailed derivation of these equations
is given by Treybal [21].

7.1. Change in air temperature across the segment

The change in air temperature across the segment is obtained by applying an energy balance on the segment
interface, control volume (I) in Fig. 6(b).
hc adZðti  ta Þ ¼ m0a C p m dta
dta hc a ð12Þ
¼ 0 ðti  ta Þ.
dZ ma

ma + dma
mw
ma,o mw,i ta + dta
tw
ωa + dω a

n
n-1
Water
Air
Z ti dZ
3 wi
2 II I
1
III
INTERFACE
ma,i mw,o
mw - dmw ma
tw - dtw ta
ωa
(a) (b)

Fig. 6. Packed Tower: (a) tower Overview, (b) differential segment.

2042 E. Elsarrag / Energy Conversion and Management 47 (2006) 2034–2043

7.2. Change in air humidity across the segment

The change in air humidity is obtained from the mass balance on the segment, control volume (III) in
Fig. 6(b)
K x adZðxi  xa Þ ¼ m0a dxa
dxa K x a ð13Þ
¼ 0 ðxi  xa Þ
dZ ma
Kxa and hca can be calculated by using the correlation (Eq. (11)) and Reynold’s analogy (Eq. (6))
!  0:27  0 0:69
d 2eq ha mw
Kxa ¼ 4:62 1 
Dv qv hi m0a
hc a ¼ K x a  C p m

7.3. Change in water temperature across the segment

By applying an energy balance across the segment, control volume (III) in Fig. 6(b)
ma dha ¼ mw C p w dtw
 
dtw ma dta dxa ð14Þ
¼ ðC p a þ xa C p v Þ þ ðC p v ta þ kÞ
dZ mw C p w dZ dZ

Fig. 7. Relation between experimental and predicted tower outlet temperatures.

 E. Elsarrag / Energy Conversion and Management 47 (2006) 2034–2043 2043

A computer program was written to perform the finite difference analysis with the fill height Z divided into
1000 segments to solve Eqs. (12)–(14), and the mass and energy balances are solved for each segment.
Fig. 7 shows the predicted outlet air temperature, air humidity and water temperature. The outlet air and
water temperatures were predicted with a maximum error of ±5%.

8. Conclusions

Several field tests using different variables were performed for an induced draft cooling tower using burned
clay bricks as the fill material. It was found that the factors affecting the heat and mass transfer coefficients are:
the water to air flow rate ratio; the inlet water temperature and the inlet air enthalpy. The previous correla-
tions found in the literature could not predict the mass transfer coefficient for the tested tower. A mass transfer
coefficient correlation was developed, and new variables were defined. This correlation can predict the mass
transfer coefficient within a maximum error of ±10%. The developed correlation was used along with theoret-
ical modeling to predict the cooling tower outlet conditions. The model with the correlation showed good pre-
dictions of the outlet water and air temperatures conditions within an error of ±5%.

References

[1] Webb RL, Villacres A. Algorithms for performance simulation of cooling towers, evaporative condensers, and fluid coolers.
ASHRAE Trans 1984;90(2):416–58.
[2] Jaber H, Webb RL. Design of cooling towers by the effectiveness-NTU method. ASME J Heat Transf 1989;111(4):837–43.
[3] Merkel F. Verdunstungshuhlung. Zeitschrift des Vereines Deutscher Ingenieure (VDI) 1925;70:123–8.
[4] Braun JE, Klein SA, Mitchell JW. Effectiveness models for cooling towers and cooling coils. ASHRAE Trans 1989;95(2):164–74.
[5] Osterle F. On the analysis of counter-flow cooling towers. Int J Heat Mass Tran 1991;34(4/5):1313–6.
[6] Söylemez MS, Ünsal M. Computational analysis for sizing of cooling towers. In: Proceedings of the third national refrigeration and
air conditioning congress, Adana, Turkey.1994. p.117–26.
[7] Ünsal M, Varol A. An analysis for sizing of cooling towers. In: Proceedings of the first national refrigeration and air conditioning
congress, Adana, Turkey.1990. p. 52–6.
[8] El-Dessouky HTA, Al-Haddad A, Al-Juwayhel F. A modified analysis of counter flow cooling towers. ASME J Heat Transf 1997;
119(3):617–26.
[9] Bernier MA. Cooling tower performance: Theory and experiments. ASHRAE Trans 1999;100(2):114–21.
[10] Nimr MA. Modeling the dynamic thermal behavior of cooling towers containing packing materials. Heat Transf Eng 1999;20(1):
91–6.
[11] Jameel UK, Yaqub M, Zubair SM. Performance characteristics of counter flow wet cooling towers. Energ Convers Manage 2003;
44:2073–91.
[12] Fisenko SP, Brin AA, Petruchik AI. Evaporative cooling of water in a mechanical draft cooling tower. Int J Heat Mass Tran 2004;
47:165–77.
[13] Söylemez MS. On the optimum performance of forced draft counter flow cooling towers. Energ Convers Manage 2004;45:2335–41.
[14] Kloppers JC, Kröger DG. The Lewis factor and its influence on the performance prediction of wet-cooling towers. Int J Therm Sci
2005;44:879–84.
[15] Thomas WJ, Houston P. Simultaneous heat and mass transfer in cooling towers. Brit Chem Eng 1959;160:217.
[16] Lowe HJ, Christie DG. Heat transfer and pressure drop data on cooling tower packings, and model studies of the resistance of natural
draft towers to air flow. Inst Mech Eng (Steam Group) Symposium on Heat Transfer 1962;113:933.
[17] Jorge F, Armando CO. Thermal behavior of closed wet cooling towers for use with chilled ceilings. Appl Therm Eng 2000;20:1225–36.
[18] Lebrun J, Silva CA. Cooling tower—model and experimental validation. ASHRAE Trans 2002:751–9.
[19] ASME. Measurement of liquid flow in closed conduits by weighing method. American Society of Mechanical Engineers; 1988.
[20] Stoecker WF, Jones JW. Refrigeration and air conditioning. Singapore: McGraw-Hill; 1985.
[21] Treybal RE. Mass transfer operations. New York: McGraw-Hill; 1981.
[22] Kern DQ. Process heat transfer. New York: McGraw-Hill; 1997.
[23] Panton RL. Incompressible flow. New York: John Wiley & Sons; 1984.