Budding ondEnoironmenr, Vol. 31, No. 6, pp.

503-507, 1996
Copyright 0 1996 Elsevier Science Ltd. All rtghts reserved
Pergamon Printed in Great Britain
036%1323/96 $15.00+0.00

PII: SO360-1323(96)00024-g

A Passive Evaporative Cooling System by

Natural Ventilation

ZAHRA GIABAKLOU* (Received 4 July 1995; accepted 4 March 1996)

JOHN A. BALLINGER*

Evaporative cooling is used extensively for cooling in climates with medium to low humidity. In
residential buildings the conventional mechanical systems tend to be both noisy and unsightly
protuberances on buildings, whilst traditional cooling systems incorporated into the building struc-
ture which employ chimneys and the like tend to be designed for the specific building and so may
not have wide application. Here a proposal is presented for a passive evaporative cooling system
which makes use of natural ventilation at the building facade. The system makes use of the
evaporative effect from water falling vertically along guides to produce a reduction in the tem-
perature of the air entering the building. It can also be used as a design element in the building
facade. Such a system provides an inexpensive, energy efficient, environmentally benign (not
requiring ozone-damaging gas as in active systems) and potentially attractive cooling system.
A numerical study is presented to demonstrate the system efficiency and airflow rate through a
building, making use of measured outside wind speed and direction, building geometry and sur-
roundings. The likely effect of the system on the indoor air temperature is discussed; further work
is beina undertaken to explore the integration of such a system into the building fabric. Copyright
0 1996 Elsevier Scienceitd

INTRODUCTION water cascade associated with openings and balconies of

the individual units.
THERE is a growing demand for space cooling in hot
The present study aims to assess the technical feasi-
climates as people seek to raise their standard of living
bility of climate control systems which could be inex-
and improve work performance, resulting in an increas-
pensive and visually attractive cooling systems. In prin-
ing demand for energy during the day, competing with
ciple there are several possible ways of lowering the
other more pressing needs such as refrigeration and the
indoor temperatures by means of the natural conditions
demands of the manufacturing sector. Countries in such
that exist both in arid and semi-arid environments. One
climates are facing serious peak demand problems as a
such solution, shown in Fig. 1, is a system which includes
result of such trends. Pakistan is one such country where,
evaporation of an exposed water film with the help of
today, electricity demand outstrips supply and so is
natural ventilation. Water is allowed to fall vertically
rationed to the residential sector in favour of the manu-
over the elements such as nylon lines or other filaments,
facturing sector. This can be directly attributed to the
thus exposing the maximum surface area of the water to
growth in the use of air-conditioning in the commercial
the passing air flow. The arrangement of the lines is such
and residential sectors. The National Energy Con-
that the water forms a curtain through which the air flows
servation Centre (ENERCON) has estimated that air-
horizontally.
conditioning loads will grow substantially by early in
The water is raised to an upper level trough by a small
the next century [l]. They predicted in 1990 that energy
consumption for commercial and residential air-con-
ditioning will grow from 15 to 30% by 2005. hea&tithoutlets
Traditionally, evaporative cooling systems that have J-
been designed to provide relief from the intense daytime
heat are based on water sprays and the like associated
with natural air movement from chimneys and court-
yards. Such schemes are dependent on building design
and are not easily transported from one application to
another, although the application of water sprays in
chimney like structures has been well researched [2]. The
present paper presents a possible method of cooling
apartments in low-rise multi-storey buildings by a simple

*School of Architecture, The University of New South Wales, ‘t resehvoir

Sydney 2052, Australia. Fig. 1. Schematic diagram showing evaporative cooling system.

503
504 Z. Giahaklou und J. A. Ballinyer

pump from the reservoir in the bottom of the system and velocity occurring within the tube bank
then allowed to flow down the lines and back to the Pr Prandtl number
reservoir. The air passing through the system is cooled Pr, Prandtl number at water temperature
and humidified. If contact between water and air i\ ( ‘./?I constants.
sufficient to bring the water and the exit air into equi-
The Reynolds number ReumsX for the foregoing cor-
librium, the air leaving the system will become saturated
relations is based on the maximum fluid velocity occur-
at a temperature close to the wet bulb temperature of the
ring within the tube bank.
exiting air. Since the water evaporated into the air is
lost from the system. make-up water is supplied to the
reservoir.
This form of cooling is most suitable for arid regions
where
where the relative humidity is low and undesirable. In
addition a case can be made for the psychological benefit /’ mass density (kg, m ‘)
to people in a hot climate of the sound of water falling I 111.1, maximum velocity (m/s)
131. There are many other ways in which water can be n diameter (m)
used as a design element that provides a linkage between l’ viscosity (kg/s. m)
various components of a design. including buildings. I’ kinematic viscosity (m’:s)

For the aligned arrangement. from the mass con-

SATURATING EFFICIENCY servation requirement for an incompressible fluid
In the direct evaporative process. the air stream is
cooled by the evaporation of water into the air. The
I’,,,,= -S! b’.
(4)
8s;- D
addition of water vapour increases the latent heat of the
airvapour mixture. As the overall process is adiabatic where
this increase is offset by a sensible heat reduction and a s, transverse pitch (m)
consequent lowering of the dry-bulb temperature of the I. wind speed (m/s).
air. During the cooling process the wet-bulb temperature
of the air remains constant. The maximum cooling 15 The average heat transfer coefficient in cross how
achieved when the outlet air is saturated, implying equal according to Hilpert [9], is given in the following equa-
dry-bulb and wet-bulb temperatures [4]. tion:
Equation (I) defines the performance factor of the
direct evaporative cooling system [5] l;_Nu/i-
D'
1, -~ 1 ,,
E, = rI-r;. (1) where
I ,

II average convection heat transfer coefficient

where
(W/m2. K)
E, cooling or saturation efficiency I, thermal conductivity (W/m. K).
r, dry-bulb temperature of the inlet air
A simplified commonly used equation for solving prob-
r,, dry-bulb temperature of the outlet air
lems involving simultaneous heat and mass transfer has
T\ thermodynamic wet-bulb temperature of the
been developed with the use of the Lewis relation and
entering air.
gives satisfactory results for most air-conditioning pro-
From this equation it can be seen that a performance cesses [IO].
factor of 1 indicates maximum possible cooling of the ait A mass transfer coefficient is conveniently defined
Row. using the log-mean humidity ratio difference, as the driv-
ing potential
SYSTEM EFFICIENCY

The efficiency of the system is controlled by a number ril, = K,,Au (6)

’
of factors such as the average heat transfer coefficient (I;)
and the mass transfer coefficient (K,,). The convection
heat transfer associated with cross flow over a bank ol where
tubes. is described by Holman [6] and Incropera [7].
Assuming that the filaments carrying the water in the Cl,, rate of mass transfer (g/s)
system are tube bundles, according to Zhukauskas [8]. KI,, mass transfer coefficient (g/s. m’)
the correlation takes the form .4H area of water (m’)
W, inlet moisture (kg water/kg air)
w,, outlet moisture (kg water/kg air)
.
W, moisture at saturation (kg water/kg air)

where The coefficient K,,, is defined as

Nun Nusselt number

K,, = ; (7)
Re D.mlr Reynolds number based on the maximum fluid pm
 Passive Evaporative Cooling System by Natural Ventilation 505

where the humid specific heat, C,, (kJ/kg (dry air). K), A, = A,A,/(A;+ Af)“2, (10)
of the airstream is, by definition
where A, and Ai are the open outlet and inlet areas respec-
c,, = (1 + JY)C,, (8) tively (m’).
where The pressure coefficient over a building surface varies
with the position on the surface, particularly near the
C, specific heat at constant pressure (kJ/kg (dry edges, where it varies considerably with changes in the
air) K). approach wind angle and, to a lesser extent, with changes
By calculating W, from equation (6) the dry-bulb tem- in the geometry of the building. The normalised pressure
perature of the outlet air (To) can be computed, and hence coefficient is given by Swami and Chandra [12], as fol-
the efficiency of the system will be known. lows:

NCP = ln(C0 + Cl t sin(a/2) + C2 sin*(E)

NATURAL VENTILATION + C3 sin’(2. LXG)+ C4 cos(@)
A simple building structure is assumed to simplify the + C5 G’. sin*(z/2)+ C6. cos’(a/2)), (11)
description. A building with dimensions 10m x 10m x
2.8m high and continuous window openings in two where
opposite walls is assumed for modelling purposes. The NCP the normalised CP
area of window in the north wall is 11.2 m* and in the In the natural logarithm
south wall, 2.8m2, with the passive evaporative system c( the wind angle in degrees
fixed in the north window. The system being modelled G the natural log of the side ratio.
assumes water lines of 3 mm diameter at 9 mm centres in
The coefficients used in this equation are [I 21 as fol-
three rows as shown in Fig. 2. A structured procedure
lows:
for calculating the wind-driven natural ventilation rate
is used, as proposed by Vickery [1 11. The assumptions CO = 1.248 Cl = -0.703 C2 = -1.175 c’3 = 0.131
inherent in this calculation procedure are C4 = 0.769 C5 = 0.07 C6 = 0.717.
that there is no stack effect The actual CP is calculated by multiplying the nor-
that there is no pressure drop inside the building. i.e. malised value by the CP at zero incidence, which is 0.6.
there is negligible effect due to partitions
that there is perfect mixing of the room air
that the wind profile can be described by the power COMPUTER SIMULATION
law A modified version of the computer simulation pro-
that the pressure coefficient (CP) data on an average gram CHEETAH [13] is used to predict the thermal
wall is used for low-rise buildings behaviour of the building with a passive evaporative cool-
that the method is valid for window or other wall ing system. The program calculates hourly temperatures,
apertures only, not for roof level apertures. and heating and cooling energy requirements, using
The flow through a cross-ventilated building with one hourly recorded climatic data for periods ranging from
effective inlet and one effective outlet is given in the fol- one day to a full year. A response-factor method is the
lowing equation: basis of the calculation. For this study, the ventilation
model used in CHEETAH has been replaced with the
CQ = Q/(AeVr.,j = Cd(ACP)‘,‘, (9) hourly calculated data from the above equations.
where The following assumptions are made to simplify the
analysis.
CQ the flow coefficient
l The house is modelled as single zone.
Q the flow (m’/s)
the effective window area (m’) l No overhangs are assumed.
A,
Vref the reference velocity (m/s) at the building l No furnishings are included.
height
Cd the discharge coefficient RESULTS
ACP the pressure coefficient difference across the
inlet and outlet Figure 3 shows the simulated indoor air temperature
for a 168-hour period compared with the outdoor air
and temperature for a hot summer week in Wagga Wagga, a
semi-arid location in New South Wales, an eastern state
of Australia. The data illustrate the effect of continuous
ventilation on the indoor air temperature and the hourly
0 0 0 0 0 variation of indoor air temperature when the evaporative
system is installed. It is shown that although the average
0 0 of daily maximum inside temperature (325°C) in con-
tinuous ventilation is lower than average maximum ambi-
0 0 0 ent (35.8”C), it is not comfortable for the room occu-
pants. During the hottest hours of the week when the
Fig. 2. Horizontal section of the system evaporative cooling system is installed, however, the
506 Z. Giabaklou and J. A. Ballinger

5-

0 L_+ + -+~_+ ~I-~+ + +- + + / I I ~t-~+ttr-+-w+-

0 612180 612180 612180 612180 612180 612180 61218

Time (hour)

-Outside temperature
-~ Inside temperature with natural ventilation
- - Inside temperature with evaporative cooling system
Fig. 3. The impact of natural ventilation and evaporative cooling system on the indoor air temperature,

inside average maximum air temperature is reduced to would seem that on a theoretical basis the system is
25.9”C, which is 9.9”C less than the incoming ambient feasible and could have wide application in medium den-
air temperature. The average relative humidity inside the sity housing. It can be installed independently on the
building is 73.2% with average air velocity of 0.28mls facade of the building, say across a balcony, to provide
and average temperature of 24.18”C, which is close to the a passive cooling method for multi-storey residential
comfort range. buildings.
At times of high evaporative cooling effect, air change Equations and charts have been presented, which
rates were found to be in the lower range, thus at times of enable the calculation of the saturating efficiency of the
high wind speed the windows could be closed or partially system due to natural ventilation through openings.
closed to optimise cooling effect. In addition, ceiling fans Work is underway to define the application constraints
could be used to increase air movement at times of low and quantify the comfort benefits of such systems in other
air flow and high humidity. climates. Further work is still required to incorporate the
calculation routines into suitable simulation models that
will allow the building designer to make full use of this
CONCLUSION cooling device in building designs.
.4cknowledgementsThis project is part of the doctoral studies
Evaporation from water droplets is one of the most of the principle author, being funded by the Ministry of Culture
potent of the passive cooling processes as it can con- and Higher Education of Iran. The authors would like to thank
siderably lower room air temperatures. The system pro- Dr. A. Delsante of the CSIRO, Highett for his advice and the
posed and analysed in this study provides a potentially customisation of the computer program, CHEETAH, Dr. D. K.
Prasad of Solarch, School of Architecture, University of New
inexpensive, energy efficient, environmentally safe and
South Wales for his contribution, and also Dr. T. Pham, School
attractive cooling method. Whilst it has not been possible of Chemical Engineering. University of New South Wales for
to construct such a system due to financial constraints, it his advice on the evaporative exchange process.

REFERENCES
I The potential of electricrty conservation In the Industrial and commercial sector in Pakistan. ENER-
CON, Pakistan (1986).
2 C. D. Moody, T. L. Thompson and J. F. Peck. The downdraft chimney: passive evaporative cooling
for the South West. Report to the Arizona Solar Energy Commission (1978).
3 C. S. Campbell, Water in Landscape Architecture. Van Nostrand Reinhold, New York (1978).
4 E. H. Mathews, M. Kleingeld and L. J. Grobler, Integrated simulation of buildings and evaporative
cooling systems. Building and Environment 29, 197-~206 (1994).
5. J. R. Watt, Evaporative Air Conditioning, The Industrial Press, New York (1963).
6. J. P. Holman, Hear Transfer, McGraw-Hill, New York (1989).
7. F. P. Incropera and D. p. De Witt, Fundamentals ofHeat and Mass Transfer. John Wiley, New York
(1990).
8. A. Zhukauskas and A. B. Ambrazyavichyus, Heat transfer from tubes in crossflow. International
Journal ofHeat and Mass Transfer 3, 305 (1961).
9. R. Hilpert. WLrmeabgabe von Geheizen Drahten und Rohren. Forsch. Geb. Ingenieurwes. 4, 215
(1933).
10. ASHRAE, ASHRAE Handbook, Fundumenfals, Chapter 5, American Society of Heating, Refri-
gerating and Air-Conditioning Engineers, New York (1985).
 Passive Evaporative Cooling System by Natural Ventilation 507

11. B. J. Vickery, R. E. Baddour and C. A. Karakatsanis, A study of the external wind pressure
distributions and induced internal ventilation flow in low-rise industrial and domestic structures.
Report No. BLWT-SS2, Boundary Layer Wind Tunnel Laboratory, University of Western Ontario
(1983).
12. M. V. Swami and S. Chandra, Correlations for pressure distribution on buildings and calculation of
natural ventilation airflow. ASHRAE Transactions 94(l), 243-266 (1986).
13. P. J. Walsh and A. E. Delsante, Calculation of the thermal behaviour of multi-zone buildings. Energy
and Buildings 5, 231-242 (1983).