Solar Cooling: Experiences and Lessons Learned

with two Different Systems

Ahmed Hamza H. Ali,
Department of Energy Resources Engineering, Egypt-Japan University of Science and Technology (E-JUST), P.O. Box 179, New
Borg El-Arab City, Alexandria 21934, Egypt
ahmed.hamza@ejust.edu.eg

Abstract The aim of this study is to report the performance
as well as the gained experiences and lessons learned with two
different solar cooling systems. The first cooling system is an
integrated cooling plant having both free cooling system and
solar powered single-effect lithium bromidewater absorption
chiller in operation since August 2002. A floor space of 270 m
2
is
air-conditioned by the plant. The plant includes 35.17 kW cooling
absorption chiller, vacuum tube collectors aperture area of
108m
2
, hot water storage capacity of 6.8 m
3
, cold water storage
capacity of 1.5m
3
and a 134 kW cooling tower. For this solar
driven cooling plant, the performance data show that free cooling
in some cooling months can be up to 70% while it is about 25%
during the 5 years period of the plant operation. Moreover, for
sunny clear sky days with equal incident solar radiation, the daily
solar heat fraction used to drive the chiller ranged from 0.33 to
0.41; collectors field average efficiency ranged from 0.352 to
0.492 and chiller COP varies from 0.37 to 0.81, respectively. The
monthly average value of solar heat fraction varies from 31.1%
up to 100% with five years average value of about 60%. The
monthly average collectors field efficiency value varies from
34.1% up 41.8% and the five-year average value amounts about
28.3%. The second solar driven cooling system is a steam jet
ejector chiller (SJEC) driven by parabolic trough collector. The
chiller has a capacity of one kW cooling. The experiments show
that the particular characteristics of a SJEC with the strong
influence of the condenser cooling water temperature and
evaporator temperature on the system coefficient of
performance. The coefficient of performance reaches high values
in part load and at good re-cooling conditions, so that the mean
efficiency is clearly higher than the nominal efficiency of the
system. A first calculation of profitability leads to specific cold
cost of 0.62 /kWh in Germany and 0.15 /kWh in Egypt.
KeywordsSolar cooling, Absorption chiller, Steam jet
ejector chiller, Free cooling, Lithium bromide-water
Nomenclatures
COP: Coefficient of performance,
h: Steam specific enthalpy, kJ/kg
I: Solar insolation on the collector plane, W/m
I
hor
:
Solar insolation on the horizontal, W/m
K1: Linear heat losses, W/(m. K)
K2: Quadratic heat losses, W/(m. K)
0
m : Suction steam mass flow, kg/h
1
m : Motive steam mass flow, kg/h
P: Pressure, mbar
P
1
: Motive pressure, bar or kPa
P
o
: Suction steam pressure, mbar
P
c
: Condenser pressure, bar or kPa
PTC: Parabolic Trough Collector
Q: Energy rate, kW
SJEC: Steam Jet Ejector Chiller
T
amb
: Ambient air dry bulb temperature, C
T
col, in
: Inlet water temperature to the collector,
o
C
T
col,out
: Outlet water temperature from the collector,
o
C
T
evap
: Condenser temperature,
o
C
T
evap
: Evaporator temperature,
o
C
T
m
: Collector mean temperature, C
q
col
: Collector efficiency,
I. INTRODUCTION
The rapid growth and changes in the energy utilization sector
with its related impact on environmental awareness have led to
more interest in the utilization of renewable energies cooling
techniques in the last years with a focus on solar energy. Solar
energy cooling comprises attractive features as the solar
insolation generally matches the cooling load profile in
buildings during the cooling season, together, with the existing
technology that solar energy can be converted to either
electricity or heat. Pridasawas [1] cited that the first recorded
solar-driven cooling machine was in 1872 by Albel Pifre, in
Paris, for producing a small amount of ice. In addition, he
presented the most recent overview of possible technologies for
solar driven cooling and refrigeration systems. The solar
cooling system is generally comprised of three sub-systems:
the solar energy conversion system, cooling machine, and the
cooling load. Solar-driven cooling system is mainly classified
in to two main groups depending on the solar energy supply
form to the cooling machine: thermal/work driven system and
electricity (Photovoltaic) driven system. The appropriate
cooling machine in each application depends on cooling
demand, form of the required driving power either thermal or
electric (energy source), the required cold temperature levels of
the cooled object, as well as the system operating environment.
For air conditioning applications, which required a service
temperature ranged from 12 to 20C, the cooling capacity is
generally high and energy removed from the chilled space has
a low potential to convert to a useful energy. In this case, the
solar thermal-driven system is more suitable for the air-
conditioning system than the photovoltaic-driven one due to a
lower installation cost of the solar energy conversion system.
Furthermore, the performance of the thermal driven
refrigeration machine is high at high evaporator temperatures
such as the case of air conditioning applications. The solar
thermal-driven air-conditioning system could be the absorption
978-1-4673-6374-7/13/$31.00 2013 IEEE


chillier, the adsorption chillier, and the desiccant cooling
system, the Rankin system or the ejector refrigeration system.
For solar energy conversion system to thermal energy form
solar collectors are used. There are different collector types
each can produce different levels of the working fluid (heat
transfer carrier) outlet temperature. However, the temperature
working fluid outlet temperature level can be matched to
various cooling machine demands. For example, the ejector
refrigeration machine requires a rather high driving pressure
and temperature of the inlet working fluid whereas the
desiccant cooling system manages at a lower temperature level
of the supply working fluid. Recently, there are numerous
prototypes and demonstration solar driven cooling plants have
been erected worldwide. Several techniques and concepts are
currently discussed and there are remain high research and
development necessity. For solar driven cooling systems,
further access to the market is impeded both by high
investment costs that are caused by their complicated
construction and design. Therefore, conventional vapour
compression systems, i.e. electrically driven cooling systems,
are often preferred by consumers. Therefore, for further
development of solar cooling and implementation of these
systems into the market, it is important to evaluate and interpret
the available gained operational experiences and lessons
learned from real working plants. In this paper the operational
experiences gathered by the two solar cooling systems is
presented. The first is an integrated cooling plant having both
free cooling system and solar-powered single-effect lithium
bromide-water absorption chillier in operation since August
2002. While, the second system is a solar driven Steam Jet
Ejector Chillier (SJEC).

II. AN INTEGRATED COMBINED FREE COOLING AND SOLAR
POWERED SINGLE EFFECT LITHIUM BROMIDE-WATER
ABSORPTION COOLING PLANT
Ali et al. [2 and 3] and Ali [4] summarizes the published data
so far for solar operated lithium bromide-water absorption
chillier for space air-conditioning application that are working
in different locations worldwide having a solar heating fraction
ranging from 39 % to 95 % with different COP values.
Moreover, they reported that no information have been
published so far concerning an integrated combined free
cooling and solar powered single effect lithium bromide-water
absorption cooling system as a combination of both cooling
techniques. This because such combined system is always
limited by the prevailing local weather conditions. Therefore,
this section presents a performance assessment of an integrated
cooling plant with combined free cooling and solar-powered
single effect lithium bromide-water (LiBr-H2O) absorption
chillier in operation for five years duration. The plant is a part
of the infrastructure at Fraunhofer Institute UMSICHT in
Oberhausen-Germany. The plant was driven by solar energy
only from 2002 to 2004 under the typical weather conditions in
the Central Europe. While, from 2005 until now, the plant has
been integrated in the institute heating system that can provide
solar heating portion in heating season and utilize the available
hot water of the Institute heating system (gas boiler and micro
gas turbine) as supplementary source in case the solar collector
field supply heat is not enough to drive the chillier at cooling
season. The plant includes a 35.17 KW cooling (10-RT-RT is
Refrigeration Ton) absorption chillier, vacuum tubes collectors
with gross and net areas of 108 m
2
and 72 m
2
, a hot water
storage tank with capacity of 6.8 m
3
, a cold water storage tank
with capacity of 1.5 m
3
and a 134 kW cooling tower. The plant
provides air-conditioning for a floor space of 270 m
2
.
A. Plant Description, Measurements, Data Acquisition and
Processing System
As the demands for energy efficiency and reduction in
carbon footprints continue to bite, the popularity of free
cooling chillers is increasing. However, the most significant
point in this plant is the free cooling, which is essentially
chilled water at no cost. This is an integrated cooling plant
possessing both combined free cooling and solar-operated
absorption chiller provides the cooling demands for the
Fraunhofer Institute (UMSICHT), laboratories, meeting rooms
and three offices during the cooling season. The office rooms
only require cold demand during working hours, whereas the
laboratories cold demand is continuous during 24 hours. The
solar energy collection system is also integrated with the
Institute heating system providing support to the central
heating system during sunny days in heating season. Figure (1)
shows the plant schematic diagram. The major components of
the plant are a roof-mounted vacuum tubes solar collector field,
a 35.17 KW cooling (10-RT) single-effect Lithium Bromide-
Water (LiBr-H2O) absorption chiller, a hot water storage tank,
a cold water storage tank, a cooling tower, a free cooling heat
exchanger, a roof top heat release heat exchanger, pumps, a
control system, a water treatment system, a pressure
maintaining system and some other auxiliary equipment. In
addition, it is comprised of four main flow circuits which are
the solar circuit (with anti-freezing agent), the hot water circuit,
the chilled water circuit, and the cooling water circuit. These
circuits interrelate with the absorption chilliers at the generator,
the evaporator, the absorber and the condenser, respectively.
While both the cooling water and chilled water circuits are
interrelate at the free cooling heat exchanger as shown in
Figure (1). In addition, the solar circuit is connected to the hot
water circuit by a plate type heat exchanger HE1.

1) Solar Collector System
The plant primary energy source is the solar energy, which is
converted into thermal energy by the vacuum tubes solar
collector field. The water is the heat transfer medium in the
collector field circuit that interrelate with the hot water circuit
that includes the hot water storage tank as shown in Figure (1).
The solar collector field is composed of a 108 m
2
apparent area
and a 72 m
2
absorbing area for an array of 432 evacuated tubes
in 9 collector fields which work in the range of 97/105
o
C and a
capacity of 50 kW (for I=1000 W/m
2
), where I is the solar
insolation at the collector plane. The collectors field is
mounted on a slightly tilted roof by 2
o
with the horizontal for
rain derange, while the collectors absorber element is tilted by
30
o
with the roof.
1) Absorption Chiller
The absorption chiller is a Water Fired Chiller (WFC-10 RT)
with a rated capacity of 35.17 KW cooling (10-RT) when it is
operating at a driving hot water temperature of 87
o
C, coolant
water temperature of 29.5
o
C and output chilled at 9
o
C with
coefficient of performance (COP) of 0.7 as reported by the
manufacture. The cooling demand of the plant is needed from



Vacuum-tube collectors field
Outdoor heat
release HE
Free cooling HE
Absorption
chiller
Chilled water
network


Hot water
network

HE1
HE2
P7
B6
P6
P1
B1
P3
P10
P2
P8
Chilled
water
storage
Cooling
tower
Hot water
storage

Fig. 1. A schematic diagram of the solar cooling plant,
where B is storage tank, HE is heat exchanger and P is pump,
respectively.

May to September as will be presented in the results section.
However, as the local weather data in Oberhausen city,
Germany has a wet-bulb temperature in cooling season ranging
from 15 to 22
o
C. Thus, the cooling tower can provide cold
water at a temperature which is less than its rated value
(24/31
o
C). Moreover, the vacuum tubes collector field is also
able to provide driving hot water at a higher temperature than
the rated value. Therefore, the chiller cooling capacity can be
rising up to 58 kW with a nominal COP of 0.75, corresponding
to the manufacture reported characteristics.
2) Cooling Tower
The cooling tower is serving to reject the heat from the chiller
coolant water to the ambient as well as a source of free cooling
in the plant. The cooling tower is located at the rooftop of the
building with a capacity of 134 kW at 24/31
o
C when ambient
air wet-bulb temperature is 21
o
C.
3) Thermal Buffer System
The thermal buffer system is composed of two tanks. The first
is the hot water storage tank, B6, as shown in Figure (1) with a
capacity of 6.8 m
3
. This tank acts as a storing and buffers that
supplies hot water which enters the chiller generator at constant
temperature value. The stored heat in the tank can be from
either the solar system or the Institute heating system. The
second tank is the cold-water storage buffer with a capacity of
1.5 m
3
, B1, as shown in Figure (1).
4) Cooling Load
A floor space of about 270 m
2
is air-conditioned by this plant.
The load network is distributed into a central air-handling unit,
which provides 100 % fresh air to labs, several air cooler units
at offices and convective radiators at the central computers labs
of the institute. The chilled water is supplied to each of these
loads installations via a supply network with a length of about
390m.
5) Measurements, Data Acquisition and Processing System
The volume flow rate at six different locations in the plant is
measured by different water meters. Multi-jet dry dial vane
impeller hot water meters are used in solar circuit and hot water
circuit with a measuring range from 0.2 to 20 m/h, woltmann
water meters are used in the re-cooling circuit and the chilled
water system with a measuring range from 0.45 to 90 m/h, and
multi-jet water meters are used for tap up water and waste
water with a measuring range from 0.05 to 5 m/h, respectively.
The temperatures were measured with PT-1000 sensors in
protection sleeves, with measuring ranges of 0-60 C for
chilled water and re-cooling water, 0-150 C for hot water and
0-200 C for the solar circuit, respectively. The resolution of
the temperature measurement is 0.01 C. The incident total
solar radiation on the plane of the absorber level was measured
by a Tritec Spectrum Irradiation Sensor 300 having a
measuring range of 0-1500 W/m
2
, with a standard 4-20 mA-
signal output. Different pumps were used to circulate the fluids
in each circuit as shown in Figure (1). In addition, flow-
controlling valves integrated with plant controller system were
used to adjust the flow rates.

B. Presentation of Parameters
From the measured data, results are presented based on using a
simple data reduction. The thermal capacity of the equipments
is determined by:
T c m Q
p
A = kW (1)
where m is the mass flow rate in kg/s, c
p
is the specific heat at
constant pressure in kJ/(kg.C) and AT is the temperature
difference in
o
C, respectively. The energy during a certain
period is determined by the integration of the capacity during
this time as follows:
dt Q E
tf
to
=
}
kWh (2)
Where t
0
and t
f
are the initial and final times. The solar
collectors efficiency () is determined by:
absorber
net solar
A I
Q

=
,
q (3)
Where Q solar, net is the net thermal power gained by the
water from the collectors field, I is the solar radiation in the
collector absorber plane and Aabsorber is the total absorbing
area of the collectors field, respectively. Neglecting the chiller
pump power as the used chiller has no solution pump, the
chiller coefficient of performance, COP, is defined as the ratio
of the evaporator cold capacity to the heat input to generator
as follows:
G
E
Q
Q
COP = (4)
III. SOLAR DRIVEN STEAM JET EJECTOR CHILLER
Using water as a refrigerant has the advantage that it is the
most environmentally friendly option and shows the highest
efficiency in refrigeration as cited by Pollerberg et al. [5 and 6].
The literature on solar driven ejector chiller studies with water
as working is limited as cited by Pollerberg et al. [5 and 6].


This due to that the disadvantage of water as a refrigerant in the
vapour phase is has high specific volume which requires a
relatively large-sized ejector. In addition, the water evaporation
temperature cannot be lower than 0 with the necessary of
higher temperature and pressure level of the water outlet from
the solar collector field to the steam drum/generator. However,
when water is used as heat transfer fluid between the solar
energy subsystem and the SJEC subsystem, the unique
advantage is that an open system can be designed without
hydraulic separation between solar collector subsystem, steam
driven ejector and chilled water subsystem. However, this
section presents a study of solar cold production by steam-jet
ejector technology which is driven by a parabolic-trough
collector field. Moreover, is to experimentally investigate the
performance of steam jet ejector chiller under load condition as
well as the effects of other operational parameters through
outdoor experimental measurements.
A. Apparatus and Procedures
A schematic diagram of a solar driven jet ejector chillier is
shown in Figure 2, while screenshot from the process diagram
of present solar driven steam jet ejector chillier in operation is
shown in Figure 3. during outdoor measurements with an SJEC
connected to parabolic trough collector (PTC) modules. From
figures (2 and 3), the experimental apparatus consists mainly of
the following: parabolic trough collector (PTC) field which
covers an area of 10.5 m, steam drum/generator which delivers
the motive steam for the refrigeration process, steam jet
compressor located above the evaporator with a nozzle
diameter of either 1.5 or 1.7 mm, a plate heat exchanger which
is used as condenser and convective cooler which simulates the
cold load with appropriate measuring instrumentations. The
SJEC has a cooling capacity of 1 kW and it is installed in a
movable frame in order to facilitate further investigations in
different locations.
1) Solar Collector System
The outdoor experimental measurements on a steam jet ejector
chillier driven by parabolic trough collector modules are
performed with parabolic trough collector (PTC) modules. The
tracking system turns the PTC around its horizontal axis. The
PCT was orientated in southerly direction, so that its rotary axis
is aligned from east to west. The receiver/absorber is a steel
tube coated with a selective blackened nickel surface and
surrounded by tough Pyrex glass..









Fig. 2. Screenshot from the process diagram of an experiment
run of steam jet ejector chiller
The radiative properties of the collector are reported by the
manufacturer as follows: selective blackened nickel surface,
absorptance ranged from 0.96 to 0.98, emittance (at 80
o
C)
ranging from 0.15 to 0.25, envelope material borosilicate glass
with transmittance ranging from 0.95 to 0.965. Water is the
heat transfer fluid and it is circulated in the solar collector
subsystem and the steam drum/generator. The collector cycle is
designed as a pressurized water cycle, which means that the
water will not boil in the absorber pipes of the collector. The
steam drum has a volume of 75 l and serves as steam generator
as well as heat energy buffer to ensure constant steam supply
conditions for the SJEC. The maximum pressure of the solar
cycle is 900 kPa (abs).
2) The steam jet ejector
The steam jet ejector compressor is the main device of the
SJEC, with a motive high pressure steam; it sucks the vapor
refrigerant from the evaporator and pumps it into the
condenser, thus reducing the pressure in the evaporator
corresponding to the evaporator temperature. This pressure
reduction causes water evaporation in the evaporator followed
by its temperature drop. This leads to the production of cold
water that circulates to the convective cooler The steam jet
ejector compressor is designed for an absolute motive steam
pressure P1 of 300 kPa, absolute suction steam pressure Po of
10 kPa and absolute condensation pressure Pc of 50 kPa. The
evaporator tank has a volume of 150 l and is equipped with a
sight glass. After being mixed in the steam ejector compressor
the motive and suction steams are condensed in the condenser
and feed back into the steam drum as well as into the
evaporator respectively. In these experiments, the tap water is
used as cooling medium in the condenser. A water-ring pump
is used to remove the leaked-in air from the system at the
beginning of experiments.
B. Measurements, Data Acquisition and Processing System
A flow-controlling valve controls the cooling load. The volume
flow rate was measured by rotameters and electromagnetic
flow meters having different measuring ranges. The
temperatures were measured with PT-100 with a measuring
range of 0-175 C with a resolution of 0.1 C. The incident total
solar radiation was measured at the collector level by a
pyranometer having a measuring range of 0-1100 W/m2 with
an error of 0.1% due to its temperature variation. The
uncertainties were calculated through the data reduction of the
experimental measurements and are estimated by determining
the COP as ranging from 13.5% to 15.6%, respectively.
C. Presentation of Parameters
From the measured data, experimental results are presented
based on using a simple data reduction; they are described in
the following. For the solar collector efficiency () which is
known as the Hottel-Whillier-Bliss an equation is presented in
[7] as follows:
I
T T
k
I
T T
k
amb m amb m
2
0
) (
2
) (
1

=q q
(5)
Equation (5) describes the efficiency as a function of the
solar radiation in the collector plane and the temperature
difference between the collector mean temperature Tm and
ambient air temperature T
amb
.
W
Steam drum/

Generator
Evaporator
Condenser
Pump
Throttle
valve
Solar Collector
Qevap= Cold load
Q
c Ejector
QSolar
W



Neglecting the work input to the pumps, the thermal COP
of the ejector refrigeration system is defined as the ratio of cold
capacity
0
Q
to motive heat
1
Q
according to Eq. (6).
( )
( )
0 1 1
0 0
1
0
h h m
h h m
Q
Q
COP
o
' ' '
' ' '
~ =

(6)
In this equation
0
m is the suction steam mass flow rate,
1
m
the motive steam mass flow rate and h is the enthalpy of the
state point ( h' ' saturated vapor state and h' saturated liquid
state)
IV. RESULTS AND DISCUSSION
A. COMBINED FREE COOLING AND SOLAR
POWERED SINGLE EFFECT LITHIUM BROMIDE-
WATER ABSORPTION COOLING PLANT
There have been enormous amounts of recorded measured
data from the plant since its installation in August 2002.
However, samples from these measurements are used to extract
the results in this contribution, which are in some cases series
of results or represent the average values of five years duration
for the plant operation.
1) Free Cooling Potential
The obtained 5 years cold energy percentage by free cooling to
the total produced cold energy are shown in Figure 5. In
monthly bases, the obtained free cold energy in mainly
depending on ambient air wet-bulb temperature and required
chilled water temperature for the load. The free cooling
percentage in some cooling months can be up to 70 %. It varies
from one month to the other and from year to year. It can be
seen from the figure that in both years the free cooling value is
very low for July. The data presented in Figure 3 is for the
duration from August 2002 to November 2007. During this
period it is found that the total produced cold energy from the
chiller is 31365 kWh while that obtained by free cooling is
10299 kWh. The assessment of the free cooling potential
clearly shows that it offers about 25% of the total cooling
demand during the 5 years period of the plant operation. These
results show an extensive potential for cooling demand by free
cooling means in certain locations based on the local climate
condition. In order to increase this potential, cold water by free
cooling can be produced at nighttime. Moreover, increasing the
cold storage capacity and use of the complete chilled water net
to store the nighttime free cooling cold water has to be taken
into account. It can be concluded that true environmentally
efficient systems must delve into total system design.
1) Solar Energy Heat Fraction and Collector Field
Performance
The fraction of the total driving heat load, which is covered by
solar energy, is referred as solar heat fraction. The total solar
energy supplied to the chillier and as well as the total external
energy and their percentage for the duration from August 2002
to November 2007 are shown in Figure 4. For the duration
from August 2002 to November 2007 the total solar energy
supplied to the chiller is 53914 kWh and the total external
energy (gas energy) supplied to the chiller is 35249 kWh and
their percentage are about 60% and 40%, respectively. The




Fig.3. Obtained cold energy percentage by free cooling to the total
cold load energy during from August 2002 to November 2007.





Fig. 4. Total solar energy supplied to the chiller and as well as
the total external energy and their percentage for the duration from
August 2002 to November 2007
results show that the monthly average value of solar heat
fraction varies from 31.1 % up to 100 %, with a five-year
average value of 60 %. The main factors affecting the solar
heat fraction are the meteorological conditions and the time of
day when the plant is operated as both influence the level of
incident solar insolation. In addition, the monthly average
value of the collectors field efficiency varies from 34.1 % up
to 41.8 %, with a five-year average value of 28.3 % (not
presented). These monthly and yearly average values of the
collector efficiency can be enhanced by adjusting the tilting
angle of the collectors absorber. In addition, cleaning the
collectors tubes outer surface from the stick fouling once a
year could be a further method to enhance the solar system
efficiency.
2) Experiences and Lessons Learned
- The results show that the collector filed has an
instantaneous mean efficiency value of about 0.63. This
result is attributed to the fact that the absorber in the
vacuum tube collectors field has tilt angle of 30o with
roof/horizontal while the system installed in the city at
51o28\ N latitude, therefore this tilt angle is not the optimal
orientation one. Moreover, from the visual inspection of the
collectors glass tube, a thin stick fouling material
combination of residual combustion gasses and other
materials over the tube surfaces is found. These fouling
materials are not removable with heavy rains, is also act on
the decreases of collector field performance. Thus,
periodical cleaning is required to keep the collectors field
working efficiently.
- The time span to reach a first capacity output from the
chiller can be about 15 min. This time became longer,
especially if the chiller has not been in operation for a
longer time, e.g. within the transition time in spring and
autumn or after wintertime. In these cases the driving hot
water temperature (water entering the chillers generator)

Solar energy supplied to
the chiller is 53914 kWh


Gas energy supplied to the
chiller is 35249 kWh

60%
40%



should be close to 100 C to overcome the crystallization
occurred during shutdown period.
- Also, the temperature of water entering the chillers
generator should be continuously observed in order to
ensure delivery of sufficient thermal energy to drive the
internal thermosyphon pump effect within the generator and
to avoid crystallization.
- In case the hot water supplied from the hot water storage
tank drops below 78C the backup heater is activated. In
case of water entering temperature dropping below 75C,
the chiller is shutdown.

B. SOLAR DRIVEN STEAM JET EJECTOR CHILLER
1) Solar Energy Subsystem Performance
As any solar-driven refrigeration system, the system
performance is mainly a function of solar energy collector
subsystem efficiency, , and the refrigeration system
coefficient of performance COP. For the solar energy
subsystem, which consist of the solar collector, the steam
drum, and pump, the solar radiation energy is absorbed and
converted into heat that is transferred to the water which flows
into the steam drum which is serving as heat energy buffer plus
motive steam generator. Outdoor measurements on the solar
collector subsystem only are carried out to investigate its
performance before it is being used to drive SJEC. The
measured data for this subsystem operation at a water mass
flow rate of 0.25 kg/s through the system and the
experimentally determined collector efficiency from different
measurements are shown in Fig 5. The collector was working
in a closed loop with a water pressure of 8 bar. The results
show that the temperature difference between water inlet and
outlet at any time is small and both temperatures rise as the
time of day advances, which leads to higher collector mean
temperature, and consequently to a lower collector efficiency
as can be seen in Fig. 5. However, the constants in Eq. (5) from
the measurements shown in Fig. 5 are as follows:
o
=0.6,
k1=0.1 and K2=0.0075.
2) Operational Characteristics of a Solar Driven SJEC
The results of experiments to clarify the operational
characteristics of a solar driven steam jet ejector chillier system
are shown in Figs. (6 and 7). Figure (6) shows the results for an
almost clear sky free from clouds while the results for a cloudy
summer day are shown in Fig. (7). The two experiments were
conducted with the water mass flow rate in the collector being
0.1111 kg/s in both cases, and motive steam mass flow rates
being 3.8 kg/hr, and suction steam mass flow rates
being1.5kg/hr, respectively. The figures show the measured
meteorological data of the horizontal solar insolation and
ambient air dry bulb temperatures in Figs. (6-a, 7-a), while
Figs. (6-b, 7-b) show the measured inlet and outlet water
temperature from the collector and the motive steam pressure
variation from the cooling process start-up up to the end of the
experiment. Figures (6-b, 7-b) show the variation of the suction
steam pressure and evaporator temperature from the cooling
process start-up up to the end of the experiment. The cooling
process was started up by opening the valve before the ejector
when the temperatures in the steam drum reaches or exceeds
140 C. In both cases the start-up time was around 13:00 hr.
After that time the drop in the temperature values of the inlet
and outlet water from the collector is due to the water
temperature in the steam drum/generator is energy level of the
three collected water from the evaporator, contender and outlet
from the collector which is the inlet to the collector. As can be
seen from the figure, as the refrigeration process start runs, it is
followed by decreases in the evaporator temperature following
the decrease in the pressure as shown in Figs. (6-c, 7-c). As the
figures for the cases presented show, the minimum water
temperature reached in the evaporator is 7 C. However, after
the suction steam reaches the lowest value, the motive pressure
is decreasing as it is the average pressure of the whole system
in addition to the drop in motive steam temperature as its
function of outlet temperature from the solar collector
subsystem. As can be seen from the figures, the water was
delivered to the steam drum/generator at almost the same
temperature as the outlet from the solar subsystem until the
time of motive steam flows to the ejector. Figures (6) and (7)
show that the increase of the motive steam pressure at the valve
opening is the time required for system self-adjustment to reach
the lower condenser pressure. Also, in both experiments the
metrological conditions at the beginning of the refrigeration
process the solar radiation values reached their peaks with little
increase in the ambient air temperature, the minimum
evaporator temperature is reached 2 hours after the ejector

Fig. 5. PCT solar collector field efficiency
8 9 10 11 12 13 14 15 16 17 18
Time, hr.
5
10
15
20
25
30
T
e
m
p
e
r
a
t
u
r
e
,

o
C
.
10
20
30
40
50
P
r
e
s
s
u
e
r
,

m
b
a
r
(c)
T
evap
P
suction steam
20
40
60
80
100
120
140
160
T
e
m
p
e
r
a
t
u
r
e
,

o
C
.
1.2
1.6
2
2.4
2.8
3.2
3.6
4
P
r
e
s
s
u
e
r
,

b
a
r
(b)
T
col,in
T
col,out
P
motive steam
100
200
300
400
500
600
700
800
900
1000
S
o
l
a
r

I
n
s
o
l
a
t
i
o
n
,

W
/
m
2
10
15
20
25
30
35
40
T
e
m
p
e
r
a
t
u
r
e
,

C
o
. (a)
17 Aug., 2005
I
hor.
T
amb.

Fig.6. Operation characteristic of solar driven steam jet ejector
chiller on clear sky day.


8 9 10 11 12 13 14 15 16 17
Time, hr.
5
10
15
20
25
30
T
e
m
p
e
r
a
t
u
r
e
,

o
C
.
10
20
30
40
50
60
P
r
e
s
s
u
e
r
,

m
b
a
r
(c)
T
evap
P
suction steam
20
40
60
80
100
120
140
T
e
m
p
e
r
a
t
u
r
e
,

o
C
.
0.4
0.8
1.2
1.6
2
P
r
e
s
s
u
e
r
,

b
a
r
(b)
T
col,in
T
col,out
P
motive steam
100
200
300
400
500
600
700
800
900
1000
S
o
l
a
r

I
n
s
o
l
a
t
i
o
n
,

W
/
m
2
10
15
20
25
30
35
40
T
e
m
p
e
r
a
t
u
r
e
,

C
o
. (a) 3 Aug., 2005
I
hor.
T
amb.

Fig.7. Operation characteristic of solar driven steam jet ejector
chiller on cloudy sky day
started operation. It can be seen from both figures that the
cooling rate shown in Fig. (7) is lower than the one in Fig. (6);
this is contrary to what would be expected of the effect of solar
radiation on a cloudy day on the system performance.
Therefore, it can be concluded from this results, that this solar
driven steam jet ejector chiller is working being simple in
design and reliable in operation and it is ecologically friendly
because it uses water as the only working fluid in the whole
system.
C. Effect of Operating Parameters on the SJEC
Performance
Effect of the evaporator and condenser temperatures
The performance is presented in the form of the experimentally
determined COP. In practice, the condenser temperature is
fixed by the environmental conditions, whether is it air cooled
or water cooled. In order to clarify the effect of both the
evaporator and condenser temperatures on the system
performance a set of measured data when the condenser
temperature varied from 24 to 36
o
C simultaneously with three
values of evaporator temperatures, which are 7, 13 and 17
o
C ,
and their effect of on the coefficient of performance are shown
in Fig. (8). The temperature values of 13
o
C and 17
o
C of the
evaporator were chosen with regard to the fact that this cooling
technique can be integrated into a chilled ceiling for air
conditionings in buildings. It can be seen from the figure that
for decreases in the condenser temperature by 12
o
C, in case of
an evaporator temperature of T
evap
= 7
o
C the coefficient of
performance, COP, was increased by 57%, while for Tevap
=13
o
C the COP was increased by 41 % and for Tevap =17
o
C
the COP was increased by 34.4 %, respectively. As can be
seen, the COP varied considerably with both evaporator and
condenser temperatures. These results can be explained as
follows: inside the ejector, in the mixing chamber, the motive
steam and the sucked vapor from the evaporator are mixed.
After mixing, the combined stream will become a transient
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
2.2
2.4
18 20 22 24 26 28 30 32 34 36
T
cond
, C
C
o
e
f
f
i
c
i
e
n
t

o
f

p
e
r
f
o
r
m
a
n
c
e
evaporator temperature 7C
evaporator temperature 13C
evaporator temperature 17C

Fig. 8 Effect of the evaporator and condenser temperatures on
coefficient
supersonic and its velocity must be high enough to increase the
pressure after deceleration in the jet diffuser to be little higher
than the condenser pressure. These results in general indicate
that the COP increases with a decreasing condenser
temperature, which is determined by the source temperature of
the cooling medium. The effect of the evaporator temperature
on the system performance can be seen in Fig. (8). It is shown
that at certain value of the condenser temperature, an increase
in the evaporator temperature leads to increases in the COP.
Estimations for the increases in the COP for the presented
condenser temperature range are shown in Fig. (8) and are as
follows: a change in the evaporator temperature from 7 to 13
o
C, for the presented range of condenser temperatures, leads to
increases in COP values ranging from 14% to 56.7%, while
changes in the evaporator temperature from 7 to 17
o
C, lead to
increases in COP values ranging from 53% to 132%,
respectively. It is also clear from the figure that at a higher
condenser temperature with a moderately high evaporator
temperature, the system COP is about 0.8, which gives reason
to assume that the system will have a reasonable efficiency in
hot arid areas. Furthermore, the COP is higher in case of cold
supply demand at relatively high cold water temperatures than
in case of chilled ceiling cooling applications. On the other
hand, an increase of the evaporating temperature increases the
COP.
A. Economic aspects of a solar SJEC
The solar steam jet ejector chiller was modeled and quasi-
dynamic simulations with an hourly time step were made over
one year for the following locations: Essen in Germany,
Toulouse in France, Geneva in Italy, Safi in Morocco and St.
Katrina in Egypt, respectively. From the metrological data its
clear that that for St. Katrina the solar radiation ratio of direct
to global is nearly 0.7. Under this condition, the Parabolic
Through Collator (PTC) type PT has achieved the highest
annual mean efficiency. The SJEC was simulated in analogy to
the method used in the solar collectors under the same
operational condition described above. The operational
conditions were simulated hourly and the COP value and the
cold capacity were calculated for each time step. In calculation
the cold capacity, it was classified into three load conditions
(50%, 75% and 100% of cold capacity). The concept of
integrated part load values according to ARI Standard


550/560/590-98 is used. A defined cooling start temperature
served as switch-on condition for the SJEC. The mean COP is
evaluated for each load condition and afterwards the annual
mean COP is determined. The results clearly indicated that, the
COP at part load is higher than the COP at full load under
nominal operational condition. This leads to a higher annual
mean COP than under full load condition. Furthermore, the
SJEC reaches good annual mean COP values at locations with
a minimal wet-bulb temperature, especially in St. Katrina in
Egypt. Finally the economic aspects of a solar SJEC for cold
water generation for air-conditioning are calculated based on
the simulation results. At first the heat price is calculated by
using a PTC of the type PT. The heat price is calculated for a
collector area of 100, 500 and 5000 m2 for all five locations.
The calculation shows that PTCs are more economic for large
area collector fields. Furthermore, the location of an
installation, especially the ratio I
dir
=I
global
, has the major
influence on the solar yield and its economy. The costs of the
cold generation by SJEC are calculated for the above five
locations. The cold capacity is assumed to be 200 kW, while
both the COP and the operating time are based on the same
simulation as before. The results clearly indicate that the
specific cold costs range from 0.619 /kWh in Essen,
Germany, to 0.147 /kWh in St. Katrina, Egypt (the specific
cold costs of electrical driven refrigerators with the assumption
above: 0.11 /kWh in Germany and 0.07 /kWh in Egypt).
B. Experiences and Lessons Learned
- A solar driven steam jet ejector chiller equipped with
PTC requires higher pressure for the purpose of
obtaining hot water at moderate temperature level in
order to produce the motive steam. This in turn leads to
decreases in the solar subsystems efficiency with the
advancing time of day. This will have impacts on the
whole system overall efficiency.
- The cooling medium of the condenser is one of the
limiting factors controlling the operation of the solar
steam ejector refrigeration system, and so it is
important to optimize the condenser design in order to
maximize the performance of the system.
- The ambient air temperature affects both the condenser
temperature and the collector efficiency, followed by
motive steam energy. An increase in the ambient
temperature increases the condensing temperature and
decreases the COP.
- The investment costs have the main relevance on the
cold costs. A future standardization of the system,
especially of the SJEC, would lead to a further
reduction of the investment costs and the related cold
costs.
V. CONCLUSIONS
The market potential for solar cooling systems with small-scale
capacity is very large. Therefore recently, there are different
companies developing solar cooling systems/kits for the
product business. In the last 2 year (2008 and 2009), few
companies in the solar business have positioned on the market
as system providers for solar cooling. The small-scale cooling
capacity can reach up to 30kW. It is known that, the basic solar
cooling systems contain solar thermal collectors with
attachment, hot water storage, pump-sets, a chiller, re-cooler
(cooling tower), cold water storage and a control unit. Until
now, almost all of these cooling kits in the market are
developed based on the chillers re-cooling is done by water and
the heat rejected to the atmosphere achieved through a cooling
tower.
- However, in hot arid areas, the water normally is rare,
thus the re-cooling system should be designed based on
dry re-cooling techniques.
- In addition, the major problems facing solar sorption
cooling systems are higher initial capital cost.
- The average value of the specific collector surface for
all unite installed until the year 2006 in Europe is
about 3 m/kW. A value from 3.5 to 4.5 m/kW can be
considered as a reference value for thermally driven
absorption and adsorption chillers. However, these
values are only rough reference values and can never
replace the detailed design and simulation of a system.
- The specific cost of the whole installed solar cooling
system in Europe is so far between 5000 and 8000
EUR/kW. In the year 2008, system prices goes down
to 4500 EUR/kW for the systems installed at maker
countries while it was 7000 EUR/kW for the systems
built in other countries.
ACKNOWLEDGMENT
The author wishes to acknowledge the Alexander von
Humboldt Foundation, Germany, for the fellowship grant
during this work at Fraunhofer Institute for Environmental,
Safety, and Energy Technology UMSICHT, Oberhausen,
Germany.
REFERENCES

[1] Pridasawas, W., 2006, Solar-Driven Refrigeration Systems with Focus
on the Ejector Cycle. Doctoral Thesis, Royal Institute of Technology,
KTH, Stockholm, Sweden.J. Clerk Maxwell, A Treatise on Electricity
and Magnetism, 3rd ed., vol. 2. Oxford: Clarendon, 1892, pp.68-73.
[2] Ali, Ahmed Hamza H., Noeres, Peter and Pollerberg Clemens. 2008,
Performance Assessment of an Integrated Free Cooling and Solar
Powered Single-Effect Lithium Bromide-Water Absorption Chiller.
Solar Energy 82:11, 10211030.
[3] Ahmed Hamza H. Ali, Noeres P., Pollerberg C., and Dtsch C., 2008,
Operational experiences of a solar cooling plant. No. 58, Proceedings of
EuroSun 2008, the 1st Int. Conf. on Solar Heating, Cooling and
Buildings, 7
th
-10
th
, 2008, Lisbon-Portugal.
[4] Ahmed Hamza H. Ali. 2010, Performance and Operational Experiences
of Solar Driven Cooling Plant after Five Years in Operation. Proc Int.
Engineering Conf. on Hot Arid Regions (IECHAR 2010), May 1-2,
2010, Al-Ahsa, KSA. K. Elissa, Title of paper if known, unpublished.
[5] Pollerberg, C., Ali, Ahmed Hamza H. and Dtsch, C., 2008,
Experimental Study on the Performance of a Solar Driven Steam Jet
Ejector Chiller. Energy Conversion and Management. 49:11, 3318
3325.
[6] Pollerberg, C., Ali, Ahmed Hamza H. and Dtsch, C., 2009, Solar
Driven Steam Jet Ejector Chiller Applied Thermal Engineering, 29: 5-
6, 1245-1252.
[7] Duffie, J. A. and Beckman, W. A. (1991). Solar Engineering of Thermal
Process. John Wiley & Sons Inc.Y. Yorozu, M. Hirano, K. Oka, and Y.
Tagawa, Electron spectroscopy studies on magneto-optical media and
plastic substrate interface, IEEE Transl. J. Magn. Japan, vol. 2, pp. 740-
741, August 1987 [Digests 9th Annual Conf. Magnetics Japan, p. 301,
1982].