Applied Thermal Engineering
Applied Thermal Engineering
Applied Thermal Engineering
Research Paper
PII: S1359-4311(17)32182-8
DOI: http://dx.doi.org/10.1016/j.applthermaleng.2017.06.115
Reference: ATE 10634
Please cite this article as: S. nal, M.T. Erdin, C. Kutlu, Optimal Thermodynamic Parameters of Two-Phase Ejector
Refrigeration System for Buses, Applied Thermal Engineering (2017), doi: http://dx.doi.org/10.1016/
j.applthermaleng.2017.06.115
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Optimal Thermodynamic Parameters of Two-Phase Ejector
Department of Mechanical Engineering, Osmaniye Korkut Ata University, 80000 Osmaniye, Turkey
Abstract
However, installing an air-conditioning system can add extra load on the engine and result
in extra fuel cost. Therefore, an improvement in the air-conditioning system can lower the
fuel consumption of the buses and reduce the size of the evaporator and the condenser. It is
known that using two-phase ejector as an expansion valve in the air-conditioning system
can improve the system performance. This study offers a model to predict the optimal
thermodynamic parameters for a two-phase ejector refrigeration system for buses using
the optimal thermodynamic parameters of the system. The effect of evaporation and
condensation temperatures on the heat transfer surface area are discussed and graphically
illustrated. Moreover, an experimental study to validate the developed model has been
carried out in a midibus air-conditioning system. The study findings revealed that the heat
transfer surface area can be reduced by about 4% and 55% in the condenser and
evaporator, respectively.
Key Words: Ejector, Refrigeration, Evaporator, Condenser, Heat Transfer Surface Area
*
Corresponding author. Tel.:+90 328 827 1000 (Ext:3454); Fax: +90 328 825 0097;
1
Nomenclature
diameter [m]
height [m]
enthalpy [J kg-1]
length [m]
number [-]
2
heat [W]
temperature [oC]
velocity [m s-1]
width [m]
Greek symbols
efficiency [-]
quality [-]
difference [-]
Subscripts
air
condenser
evaporator
primary evaporator
3
secondary evaporator
fin or fouling
gas
inlet
inside
liquid
mean or mixture
outlet
outside
air pass
refrigerant
total or tube
wall
4
1. Introduction
to provide air-conditioning systems in almost all the buses and midibuses, as passengers of
engine shaft, would generate an extra load on the engine. Therefore, the fuel consumption
of the engine increases with this extra load. The empty weight of the vehicle is another
major issue to be considered. There are obligatory standards regarding the weight of buses.
For instance, the gross vehicle weight of a bus with two axles must not exceed 18 tons, and
the rear axle load must not be greater than 11.5 tons [1]. The passenger and baggage
capacity of the bus increases when the empty vehicle weight of the bus decreases.
Installing a cooling system using a two-phase ejector can be beneficial, as the coefficient
of performance (COP) of the bus air-conditioning system can be increased while the fuel
The focus of most of the research in this field is on reducing throttling losses in the
expansion valve. For this purpose, the ejector is used as an expansion valve. When the
throttling losses are regenerated, the compressor work decreases and the COP of the
system increases. The two-phase ejector was first used by Kornhauser [2] as an expansion
R12 as a refrigerant. The study reported 21% improvement in COP. Thereafter, various
ejector refrigeration systems using different substances have been reported. Takeuchi et al.
[3] indicated that an increase in COP of 45% to 65% can be obtained for a vehicle
5
refrigeration system. Ersoy and Bilir [4] investigated the standard ejector cycle using
R134a and observed COP improvement in the range of 614% at matched capacity. They
also noted that although the drop in the refrigerant pressure in the ejector system
evaporator was almost negligible, the refrigerant pressure was reported to be as high as 133
kPa in the conventional system. Wang et al. [5] investigated the effect of the mixing
pressure on the performance of the ejector and the hybrid system. An optimum value of the
ejector mixing pressure has been obtained. Lin et al. [6,7] examined the optimum
geometric parameters of the ejector, which is used in variable cooling loads conditions,
using both experimental and numerical approaches. Sarkar [8] analyzed the optimum
Furthermore, ejector refrigeration systems with two evaporators have been studied. Vapor
compression cycle with ejector to improve the refrigeration cycle performance was
developed and patented by Oshitani et al. [9-11]. Lawrence and Elbel [12] investigated the
first and second law analysis and performance characteristics of the two evaporator ejector
cycle with R134a and R1234yf refrigerants. Lawrence and Elbel [13] also performed an
experimental investigation of the two evaporator ejector system with refrigerants R134a
and R1234yf. They found that R134a and R1234yf achieved COP improvements of 5%
and 6%, respectively. In addition, Boumaraf et al. [14] investigated the two evaporator
ejector refrigeration cycle and found that R134a achieved a slightly higher COP than
R1234yf. nal and Ylmaz [15] performed a thermodynamic analysis of the system with
two-phase ejector and two evaporators with R134a refrigerant in bus refrigeration system.
They reported that the COP can be increased by 15% using the two-phase ejector as an
expansion valve in the bus air conditioning system. nal [16] also experimentally
6
investigated the performance of the bus refrigeration system and revealed that the COP of
the vapor-compression system can be increased by 8% when the two-phase ejector is used
as an expansion device.
In order to find the optimal thermodynamic parameters of the system, heat exchanger
analyses also must be performed. Many studies have been conducted on the design of heat
exchanger of the refrigeration systems. Yu and Chan [17] explained how the COP of
chillers can be improved by a new condenser design using evaporative pre-coolers and
variable-speed fans. The study findings revealed that the optimum set-point condensing
temperature is a function of the chiller load and the wet-bulb temperature of the outdoor
air. Granryd [18] analytically expressed the optimum flow rates in the evaporators and
condensers of heat pumping systems. The obtained expressions can be used for design
purposes as well as for checking the suitable flow velocities in existing plants. Sanaye and
Dehghandokht [19] performed thermal modeling of parallel flow condenser. The modeling
results were validated using the experimental data. Shen and Lu [20] established a
design parameters on the performance of evaporator. Castro et al. [21] characterized the
models of an automotive air-conditioning evaporator and condenser were developed for the
prediction of their performance under operating conditions. Hyun and Pit [22] carried out a
thermal fluid flow analysis for the multi-flow channel condenser of automotive air-
7
When designing a two-phase ejector refrigeration system for buses, it is critical to
minimize the size of the condenser and evaporator of the system. Thus, both the total
weight and cost of the system can be decreased. To the best of the authors knowledge,
there are no studies in the open literature to determine the optimal thermodynamic
parameters for the two-phase ejector refrigeration system. Therefore, the purpose of the
present study is to examine the optimal thermodynamic parameters and determine the
used for buses to be compared with the experimental results obtained for the 14 kW
cooling capacity of minibuses air-conditioning system. Similar models have been used for
both systems, in order to reveal the significance of reducing the size and weight of the
ejector refrigeration system. Thus, the comparison between conventional and ejector
refrigeration system can be applied properly. EES (Engineering Equation Solver) is used to
solve the governing equations. The outside design condition, in terms of temperature and
2. Theoretical Analysis
components of the system are a compressor, a condenser, a liquid tank, two expansion
valves (LH & RH), and two evaporators (LH & RH).
8
Figure 1. Conventional refrigeration system for buses
Fig.2 is a schematic diagram of the two-phase ejector refrigeration system used for buses,
while Fig.3 is a pressure-enthalpy (P-h) diagram of the system. The main components of
this system are a compressor, a condenser, a liquid tank, two expansion valves, an ejector,
and four evaporators. The refrigerant that is collected in liquid tank is divided into the
ejector and expansion valve. The refrigerant, which leaves the expansion valve, first enters
then it enters the ejector as the secondary flow. The primary flow passes through the
ejector under high pressure and expands through a converging-diverging nozzle. The
refrigerant pressure in the nozzle is reduced to the evaporation pressure. The primary flow
has a supersonic velocity at the nozzle outlet, and then it passes through the mixing
section. Meantime, the secondary flow is fed into the mixing section. Next, the mixture
passes into the diffuser and its kinetic energy is converted to the pressure in the diffuser
section. After the mixing process, the refrigerant enters the primary evaporator (marked as
9
Figure 2. Two-phase ejector refrigeration system for buses
10
c) The throttling process in expansion valve is isenthalpic.
f) The process in the mixing section takes place at constant pressure and constant cross-
sectional area.
g) For condenser and evaporators, fin and tube heat exchangers are used in all calculations.
Details of the thermodynamic analysis of the system that based on these assumptions are
given by nal and Ylmaz [15]. This theoretical study is based on one-dimensional
homogeneous equilibrium model with fixed isentropic efficiencies of nozzle and diffuser.
Schematic view of the two-phase ejector is shown in Fig.4. Refrigerant coming from the
condenser enters the ejector as the primary flow. The refrigerant coming from the
evaporator is referred to as the secondary flow. The secondary flow enters the ejector by
two lines that are connected to the evaporators located at the left and right side of the bus
air-conditioner. The throat diameter is denoted as for the nozzle. The converging and
11
diverging lengths between this point and the nozzle inlet and outlet are shown as and
, respectively. Additionally, the lengths of mixing section and diffuser are expressed as
and , respectively. In the conventional system, the refrigerant passes into the
compressor at point (8), while it enters the compressor at point (1) when using the two-
phase ejector system. Thus, for the same cooling capacity, the energy consumed by the
compressor in the two-phase ejector cooling system is reduced and therefore the COP of
Fin-and-tube type heat exchangers are currently used as the condenser and evaporator in
the buses and midibuses, as shown in Fig.5. The overall heat transfer surface areas of
12
The aim of selecting this type of heat exchanger for the condenser and evaporators is to
easily determine and compare the impact of the different condenser and evaporator sizes
used in the currently existing system and the ejector air-conditioning system.
The heat rejected from the condenser can be obtained by Eq. (1), as follows:
(1)
Eq. (2) can be used to determine the overall heat transfer surface area of the condenser.
(2)
Hence, the logarithmic mean temperature of the condenser can be calculated using Eq.(3).
(3)
In addition, the overall heat transfer coefficient of the condenser is calculated by Eq.(4), as
follows:
(4)
13
where the total outside heat transfer surface area and the mean surface area of the tube wall
(5)
(6)
Areas of bare tubes and fins can be calculated using Eqs. (7) and (8) given below,
respectively.
(7)
(8)
In Eq.(4), and refer to the wall thickness and thermal conductivity of the condenser
tube, respectively. Additionally, the fouling resistance for the inside of the tube can be
assumed as for R134a with copper tubes, and the fin efficiency
can be used as [23]. The total number of tubes and fins can be obtained by
(9)
(10)
In the condenser, the refrigerant flows through the tubes while air flows over the finned
tubes. The forced convection heat transfer coefficient for the air-side depends upon the
type of fins, fin spacing, fin thickness, tube diameters, etc. Air-side heat transfer coefficient
(11)
(12)
14
where is the air velocity between fins, is hydraulic diameter, and is kinematic
viscosity of air passed through the condenser. The hydraulic diameter is calculated by Eq.
(13).
(13)
(14)
flow direction). Therefore, the wetted perimeter, Pc can be expressed by Eq. (15), as
follow:
(15)
According to definition of the Nusselt number, the outside heat transfer coefficient is
(16)
In order to calculate the heat transfer coefficient between the refrigerant and the inside
surface of the condenser tubes ( ), Nusselt number expressed by Eq.(17) can be used
[25].
(17)
where is two-phase flow Reynolds number, which is calculated by Eq. (18) [25].
(18)
where, , and are the Reynolds number, density of liquid, and gas phase of the
refrigerant, respectively. Thus, the inside heat transfer coefficient is calculated using Eq.
(19).
15
(19)
Ejector refrigeration system for buses has two evaporators that work at different pressures,
as shown in Fig.2 and Fig.3. The cooling load of the evaporators can be obtained by Eq.
(20), as follows:
(20)
The cooling loads of the primary and secondary evaporators are expressed by Eqs. (21) and
(22), respectively.
(21)
(22)
In order to determine the overall heat transfer surface areas of the evaporators, Eqs. (23)
(23)
(24)
16
Hence, the logarithmic mean temperatures of the primary and secondary evaporators can
(25)
(26)
Additionally, the overall heat transfer coefficients of the primary and secondary
(27)
(28)
The expressions of heat transfer surface areas for fin-and-tube type evaporators are similar
to the previously given equations for the air-cooled condensers. Therefore, the total outside
heat transfer surface areas and the mean surface areas of the evaporator tube walls are
(29)
(30)
(31)
(32)
17
Surface areas of bare tubes and fins of primary and secondary evaporators can be
calculated by Eqs.(33)(36).
(33)
(34)
(35)
(36)
In addition, total number of tubes and fins of the primary and secondary evaporators can be
obtained by Eqs.(37)(40).
(37)
(38)
(39)
(40)
The expression of the overall heat transfer coefficient is similar to the fin-and-tube
condenser as long as no phase change takes place. In this study, it is assumed that the fluid
that passes the external side of evaporators does not undergo any phase change. Thus, the
air-side heat transfer coefficient of the finned tube of the primary and secondary
(41)
(42)
(44)
(45)
(46)
Air pass areas and can be calculated by Eqs. (47) and (48).
(47)
(48)
Wetted perimeter of the primary and secondary evaporators can be calculated as follows:
(49)
(50)
Outside heat transfer coefficients of the primary and secondary evaporators are calculated
(51)
(52)
The heat transfer includes both the convective and nucleate boiling contribution to the
evaporation under forced-flow conditions. The boiling heat transfer coefficient is primarily
19
a function of vapor quality, refrigerant velocity, heat flux, flow channel geometry, two-
phase flow pattern, and refrigerant properties. In order to calculate the heat transfer
coefficient between the refrigerant and inside surface of the evaporator tubes, Gungor and
Winterton [26] correlation can be used. According to this correlation, the inside heat
transfer coefficients for the primary and secondary evaporators are obtained by Eqs.(53)
and (54).
(53)
(54)
where and are the liquid phase heat transfer coefficients that can calculated by
(55)
(56)
The nucleate pool boiling coefficients and are obtained by Eqs.(57) and (58)
[28].
(57)
(58)
This correlation expresses the heat transfer coefficient in . The heat flux must be
entered as input in . M and indicate the molecular weight and the reduced
pressure, respectively. The reduced pressure is defined as the ratio of evaporator pressure
to the refrigerant pressure at the critical point. In the enhancement factors and are
(60)
Also, the suppression factors and can be obtained by Eqs. (61) and (62) given
below [26].
(61)
(62)
where is the quality, and and indicate the density of the saturated liquid and
saturated vapor refrigerant, respectively. Also, and are the boiling number and
Reynolds number of liquid phase, respectively. They can be calculated as follows [26]:
(63)
(64)
(65)
(66)
where is the mass flow rate per cross sectional area and is inside diameter of the tube.
and indicate the latent heat and liquid phase dynamic viscosity of refrigerant,
respectively.
21
3. Results and Discussions
In this study, the optimal thermodynamic parameters are examined for two-phase ejector
refrigeration system for buses to determine the minimum heat transfer surface area of the
between 40-60 based on the ambient temperature. There are two separate evaporators in
the two-phase ejector refrigeration system for buses. The temperature range for the primary
evaporator is considered as 10-20 , while that for the secondary evaporator is 2-8 . It is
found that the superheating and sub-cooling temperatures of the refrigerant do not result in
a significant change in the heat transfer surface areas of the evaporators and condenser.
Thus, the superheating temperatures of the refrigerant at the primary and secondary
Figure 8. The effect of condensation temperature on total heat transfer surface areas
22
The effect of the condenser temperature on the total heat transfer surface areas of
condenser and evaporators is investigated and the results are shown in Fig.8. The heat
transfer surface area decreases and the secondary evaporator heat transfer surface area
increases. The surface areas of primary evaporator and secondary evaporator are equal
Figure 9. The effect of primary evaporation temperature on total heat transfer surface areas
Fig.9 shows the effect of the primary evaporation temperature on the total heat transfer
surface areas of the condenser and evaporators. It can be seen that the primary evaporation
temperature slightly affects the condenser surface area. Based on the given parameters, the
calculated total heat transfer surface area of the condenser is approximately 45 . On the
other hand, the primary evaporation temperature has a significant effect on the primary and
secondary evaporator surface areas. As the primary evaporation temperature increases, the
secondary evaporator heat transfer surface area decreases; however, the primary evaporator
23
heat transfer surface area increases with respect to the conservation of total cooling
capacity. Fig.9 clearly indicates that when the primary evaporation temperature is about
13 , the primary and secondary evaporators have nearly the same heat transfer surface
areas. This finding enhances the applicability of the developed system as the standard sized
The effect of the secondary evaporation temperature on the total heat transfer surface areas
of the condenser and evaporators is shown in Fig.10. Similar to the primary evaporation
temperature, the secondary evaporation temperature slightly affects the condenser surface
area. While the secondary evaporation temperature increases, the primary evaporator
surface area decreases. Unlike the primary evaporation temperature, the secondary
evaporation temperature has a significant effect on the secondary evaporator heat transfer
surface area. It is found that the surface areas of the primary and secondary evaporators are
Figure 10. The effect of secondary evaporation temperature on total heat transfer surface
areas
24
In order to reduce the weight of the condenser and evaporators, the heat transfer surface
areas of these components should be minimized. There are many parameters to design the
the design of the condenser and evaporators. The condensation and secondary evaporation
temperatures can be predicted based on the outside and inside air temperatures,
determined to minimize the heat transfer surface areas according to the condensation and
Figure 11. Variations in total heat transfer surface areas of the condenser and evaporators
Fig.11 illustrates that the total heat transfer surface area of the condenser and evaporators
vary with the primary evaporation temperature. In the calculations, the midibus cooling
25
load is assumed as 14 kW and the inside air temperature and relative humidity to achieve
comfort conditions are considered as 25 and 50% , respectively. While the outside air
In order to determine the effect of the primary evaporation temperature on the total heat
transfer surface area of the condenser and evaporators, the condensation temperature is
varied between 40-50 and the secondary evaporation temperature is assumed as 2-8 .
The obtained theoretical results based on these parameters are shown in Fig.11 regarding
the total heat transfer surface area of the condenser, primary evaporator, and secondary
evaporator. As shown in Fig.11, there is an optimal point for each operating conditions.
This optimal point rises as the secondary evaporation temperature increases. When the
condensation temperature drops, the total heat transfer surface area increases because of
the increment in condenser surface area. Hence, this figure indicates that it is necessary to
select the appropriate design criteria. The primary evaporation temperatures reported at the
minimum total heat transfer surface area of the condenser and evaporators are listed in
Table 1.
Table 1. Primary evaporation temperatures at the minimum total heat transfer surface area
26
In this study using the given data in Table 1, an empirical correlation is derived for the
optimal evaporation temperature of the refrigerant in the primary evaporator as shown Eq.
(67).
(67)
As seen in Table 2, Eq. (67) can be used to calculate the optimal primary evaporation
temperature with an error margin of 1%. Hence, this empirical equation can be used to
determine the optimal value of the primary evaporation temperature, which is one of the
most important parameters in a two-phase ejector cooling system. Therefore, both the size
and the weight of the condenser and the evaporators of the bus air conditioning system can
, the heat transfer surface area of the condenser and evaporator for a
midibus air conditioner, which has 14 kW cooling capacity, can be reduced by 4% and
results, the total heat transfer surface area for the condenser to achieve 14 kW cooling
27
evaporator. However, the total heat transfer surface areas of the condenser, primary and
secondary evaporators required for the two-phase ejector refrigeration system are ,
significantly reduced. Details of the weight and cost analysis were provided in [16].
4. Validation
An experimental setup was constructed for the tested two-phase ejector refrigeration
system for buses (shown in Fig.2) to validate the obtained theoretical results. An air
conditioner currently used in midibuses was turned into a two-phase ejector air
conditioning system. Fig.12 is a schematic diagram that illustrates the experimental setup.
Figure 12. Schematic illustration of experimental setup for ejector bus refrigeration system
28
A midibus without the driveline was used in the experimental study as shown in Fig.13.
The experimental setup has the same dimensions as the currently used midibuses of 7.04 m
in length and 2.22 m in width, and a seating capacity of 28 passengers with one employee.
The maximum cooling load was calculated as 13.8 kW according to the previously
discussed design criteria. Thus, an air-conditioner with the two-phase ejector installed as
an expansion device with 14 kW cooling capacity was used in the experimental setup. For
the purpose of this experiment, the experimental setup was exposed to the sun for a long
time until the inside temperature of the bus reached about 60oC before testing.
29
Temperature and pressure data of the experimental setup were measured and collected
type thermocouples were used to measure pressure and temperature, respectively. Air
velocity and relative humidity at the evaporator inlet and outlet were measured using the
transmitters of hotwire and hygrometer. The measured values were recorded periodically
using a data logger. The specifications of the measuring devices used in the experiments
The compressor of the conditioning system used in the experimental setup shown in Fig.13
is driven by a three-phase electric motor. The A/C compressor was set to 2900 rpm, just as
the case scenario when a midibus is travelling at 90 kmh-1. The power consumption of the
compressor was also measured by a power analyzer. The refrigerant pressure and
valve, and ejector inlet and outlet. Additionally, the refrigerant flow rate was measured in
two separate positions, at the outlet of the liquid tank and at the inlet of the ejector. The air
temperature and relative humidity as well as the air velocity were measured at the inlet and
outlet of the evaporator. The data obtained from the measurements was used to determine
30
It is very important to determine the cooling load of the inter-city buses to appropriately
design and select each of the system components. Hence, the radiant time series (RTS)
method was used to determine the cooling load of the midibus [31]. The air conditioning
system to be used in the experimental setup was chosen in case the midibus has full
passenger capacity (28 person). However, since the experiments are not performed with
full passenger capacity, cooling load calculations have to be done according to the case of
having 1 person in the midibus. Therefore, cooling load calculation was made for two
different situations. In the first case, it was made according to the full passenger capacity
of the midibus. It is also assumed that the midibus is running at a constant speed of 90
kmh-1 in this case. In the second case, cooling load calculation was made according to the
situation of having 1 person in the midibus, that is, the situation in which the experimental
work was done. Due to the structure of the experimental setup, the midibus is stable during
the experimental study. Therefore, the midibus speed is taken as 0 kmh-1 in this case. The
values to be used in the cooling load calculation and the results obtained according to these
Table 4. The model input and output data for the cooling load calculation
31
In the calculations, the external temperature and relative humidity were taken as 35 oC and
45%, respectively. Also, according to the comfort conditions for the summer season given
by ASHRAE [32], the internal temperature and relative humidity were taken as 25oC and
50%, respectively. Sensible and latent heat gain for people given by ASHRE [31] were
taken as 75W and 55W per one person, respectively. In addition, 10W cooling loads per
one person were added because of their clothes. Also, fresh air requirement for ventilation
of the bus was assumed as 15 m3h-1 per one person according to the ASHRAE [33]. The
values of the surface area used in the calculation of the cooling load and the experimental
setup are the same. The variations in the calculated cooling loads during the day are shown
in Fig.14. The maximum cooling loads obtained for the midibus with 28 people and one
The system has two separate evaporator units on the left and right hand side of the
experimental setup. Furthermore, each evaporator has two separate inlet and outlet
32
working at different evaporation pressures. The temperatures and relative humidity values
were measured at the primary evaporator inlet and secondary evaporator outlet that is
located at left and right hand side of the experimental setup. The detected variation in the
mean temperatures and relative humidity values of the air at the inlet and outlet of the
evaporator units are shown in Fig.15. It is reported that the system reached the steady state
The two-phase ejector air conditioning unit has a total of four air blowers for the
evaporators and each one has 1020 m3 h-1 flow rate. Therefore, the total air flow rate of the
air conditioning system is 4080 m3h-1. This flow rate is valid without any counter pressure.
During the testing, the air velocity values passing through the evaporator fins were
measured at the different locations. Thus, the total air flow rate was determined about 2600
m3 h-1. The instantaneous cooling load of the experimental setup was calculated considering
the measured values of the temperatures and relative humidity at the evaporator inlet and
33
outlet, and the air flow rates of the evaporators. The variation in the cooling load in relation
Moreover, it is reported that the system reached steady state conditions after 70 minutes.
According to the experimental results, the instantaneous cooling load sharply increased and
reached the maximum value. It can be seen that the experimental setup air conditioning
system has reached a cooling capacity of approximately 15 kW in the first two minutes.
The cooling load obtained in the experimental study, which conducted with one person,
was about 8.6 kW after the system reached steady state. Accordingly, it is clear that the
obtained theoretical and experimental results regarding the cooling load of two-phase air
The variation in the temperatures of the outside, inside, condenser, and evaporators in
relation to time during the experiment are given in Fig.17. When the change in temperature
values over time is examined, it is found that the system reaches the steady state conditions
34
after about 70 minutes. Experimental apparatus designed and manufactured according to
optimal results from the model. So, when the experimentally obtained temperature values
of the refrigerant in the condenser and the secondary evaporator are applied to equation
(67), it is found that the measured temperature values of the refrigerant in the primary
evaporator are in good agreement with the temperatures obtained from the empirical
equation.
According to the theoretical results given in Table 1, the evaporation temperature of the
found that the evaporation temperature of the refrigerant in the primary evaporator is
35
after reaching the steady state conditions. Hence, the method developed in this study can
be used to determine the optimal thermodynamic parameters to get the minimum total heat
transfer surface area of the condenser and evaporator of the two-phase ejector refrigeration
5. Conclusions
This study examines the optimal thermodynamic parameters and determines the
appropriate size of condenser and evaporator of ejector refrigeration systems used for
buses. Theoretical and experimental studies have been performed on the two-phase ejector
examine the differences between the air-conditioner with two-phase ejector and the
conventional system, finned and tube heat exchangers are used in calculations. The effects
of different parameters on the heat transfer surface areas are discussed. The following
1) Both the primary and secondary evaporation temperatures have significant effect on
the heat transfer surface area of first and second evaporator. However, they slightly
2) The heat transfer surface area of the condenser is inversely proportional to the
evaporator heat transfer surface area decreases and the secondary evaporator heat
3) According to the theoretical results, there are optimal points for the heat transfer
36
4) The condensation and both evaporation temperatures affect the optimal points. Thus,
5) Based on the calculations, the total heat transfer surface areas of condenser and
evaporators for the midibus air-conditioner with two-phase ejector can be reduced by 4
6) The two-phase ejector refrigeration system not only improves the system performance
but also reduce the total weight of the bus. Hence, fuel saving can be achieved.
7) The experiments are performed for an air-conditioner with constant speed compressor.
Further experiments can be performed with variable speed compressor at real road
conditions.
Acknowledgements
Global and SAFKAR companies. The authors would like to thank to TBTAK, TEMSA
37
References
[1]. Council Directive 96/53/EC, No: L235/59-75, (25 July 1996), Official Journal of the
European Communities.
[3]. Takeuchi H., Nishijima H., Ikemoto T., (2004), World's First High Efficiency
Refrigeration Cycle with Two-Phase Ejector: Ejector Cycle, SAE World Congress,
[4]. Ersoy H.K., Sag N.B., (2014), Preliminary experimental results on the R134a
[5]. Wang F., Li D.Y., Zhou Y., (2016), Analysis for the ejector used as expansion valve
[6]. Lin C., Li Y., Cai W., Yan J., Hu Y., (2013), Experimental investigation of the
210.
[7]. Lin C., Cai W., Li Y., Yan J., Hu Y., (2013), Numerical investigation of geometry
[8]. Sarkar J., (2017), Performance analyses of novel two-phase ejector enhanced multi-
[9]. Oshitani H., Yamanaka Y., Takeuchi H., Kusano K., Ikegami M., Takano Y.,
Ishizaka N., Sugiura T., (2005), Vapor compression cycle having ejector, U.S. Patent
No. US20050268644A1.
38
[10]. Oshitani H., Takeuchi H., Yamada E., Nishijima H., (2010), Ejector type refrigerant
[11]. Oshitani H., Gocho M., Takano Y., (2008), Ejector-type cool box, SAE International
[12]. Lawrence N., Elbel S., ( 2013), Theoretical and practical comparison of two-phase
ejector refrigeration cycles including First and Second Law analysis, International
[13]. Lawrence N., Elbel S., (2014), Experimental investigation of a two-phase ejector
cycle suitable for use with low-pressure refrigerants R134a and R1234yf,
[14]. Boumaraf L., Haberschill P., Lallemand A., (2014), Investigation of a novel ejector
expansion refrigeration system using the working fluid R134a and its potential
[15]. nal ., Ylmaz T., (2015), Thermodynamic analysis of the two-phase ejector air-
system using analytical and numerical methods, Applied Thermal Engineering, (90)
110-119.
[17]. Yu F.W., Chan K.T., (2006), Improved condenser design and condenser-fan
[18]. Granryd E., (2010), Analytical expressions for optimum flow rates in evaporators
1211-1220.
39
[19]. Sanaye S., Dehghandokht M., (2011), Modeling and multi-objective optimization of
parallel flow condenser using evolutionary algorithm, Applied Energy, (88) 1568-
1577.
[20]. Shen C.C., Lu J.H., (2014), Analysis of the performance of the evaporator of
[21]. Castro F., Tinaut F.V., Rahman Ali A.A., (1993), Automotive evaporator and
[22]. Hyun L.K., Pit W.J., (1997), Thermal flow analysis of HFC-134a MFC condenser for
[23]. Stoecker W.F., Jones J.W., (1982), Refrigeration and Air-conditioning, McGraw-
Hill, NewYork.
[24]. Kays W.M., London A.L., (1964), Compact Heat Exchangers, McGraw-Hill,
NewYork.
[25]. Akers W.W., Dean H.A., Crosser O., (1958), Condensation heat transfer within
[26]. Gungor K.E., Winterton, R.H.S., (1986), A general correlation for flow boiling in
tubes and annuli, International Journal of Heat and Mass Transfer, (29/3) 351-358.
[27]. Dittus F. W., Boelter L.M.K, (1930), Heat Transfer in Automobile Radiators of the
Berkeley.
[28]. Cooper M.G., (1984), Saturation nucleate pool boiling, A simple correlation, 1st
40
[29]. Gungor K.E., (1987), A new correlation for forced convection boiling, Journal of
[32]. ASHRAE Standard 55, Thermal environmental conditions for human occupancy,
[33]. ASHRAE Standard 62, Ventilation for acceptable indoor air quality, American
USA, (2001).
41
Table captions
Table 1. Primary evaporation temperatures at the minimum total heat transfer surface area
Table 4. The model input and output data for the cooling load calculation
42
Figure captions
Figure 8. The effect of condensation temperature on total heat transfer surface areas
Figure 9. The effect of primary evaporation temperature on total heat transfer surface areas
Figure 10. The effect of secondary evaporation temperature on total heat transfer surface
areas
Figure 11. Variations in total heat transfer surface areas of the condenser and evaporators
Figure 12. Schematic illustration of experimental setup for ejector bus refrigeration system
43
Highlights:
Optimal thermodynamic parameters of two-phase ejector refrigeration system.
Empirical correlation for primary evaporator temperature.
Calculation of optimal condenser and evaporator dimensions.
Comparison of calculated values with experimental study.
44