Available online at www.sciencedirect.

com

ScienceDirect
Energy Procedia 75 (2015) 3239 – 3252

The 7th International Conference on Applied Energy – ICAE2015

1-D Mathematical Modeling and CFD Investigation on

Supersonic Steam Ejector in MED-TVC
A.S.Hanafia , G.M.Mostafaa , A.Waheeda* , A.Fathya
a
Department of power mechncial engineering, Cairo university, Cairo, Egypt

Abstract

Thermo Vapour Compressor TVC is the heart or prime mover of Multi-Effect Desalination MED system,
it is also described as a steam ejector. Steam ejector is an economical device that utilizes high pressure
steam to compress low pressure steam without any rotating parts or external power sources. The ejector
performance is highly dependent on its geometry and operating conditions. The aim of the present work
is to design and obtain steam ejector geometry, based on pre-specified operating conditions, and to
investigate the flow behavior inside the ejector by using Computational Fluid Dynamics CFD through
using “ANSYS FLUENT 14.5” software. In first section of this paper, 1-D mathematical model is
carried out to predict the ejector geometry at designed operating conditions. The second part, describes
the flow behavior inside the designed model of the ejector by CFD. CFD is a most reliable tool to reveal
the phenomenon of the mixing process at different parts of the supersonic turbulent flow ejector and to
study the effect of operating conditions on the effective ejector area. Obtaining such effects, using
experimental work, is much costly and difficult. Finally, the results show the effect of the operating
conditions on the entrainment ratio and the critical back pressure.
© 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2015 The Authors. Published by Elsevier Ltd.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Selection
Peer-reviewand/or peer-review of
under responsibility under responsibility
Applied of ICAE
Energy Innovation Institute

Keywords: Thermocompressor; ejector; performance evaluation; entrainment ratio; MED desalination; CFD.

1. Introduction
Thermo Vapour Compressor TVC is simply a steam ejector which is widely used as an essential part in
Multi-Effect Desalination MED system, due to its simple construction, easy installation and maintenance
and low capital and operating costs. TVC is responsible for the energy recovery in the MED unit, through
transferring the energy contained in the high pressure steam to lower pressure vapour, in order to produce
a mixed discharge vapour at intermediate pressure. Steam ejector is also applied in MED system to
maintain vacuum levels and remove non-condensable gases inside the successive effects and low pressure
condenser. Accordingly, increasing TVC efficiency will result higher MED overall performance. Good
understanding of flow behaviour, shock interaction and mixing process inside the ejector is necessary in
designing high performance ejector. Keenan and Neumann developed a classical one-dimensional theory
based on the gas dynamics rules in order to design the ejectors [1]. Then, Emas et al. [2] modified this

* Corresponding author. Tel.: +2-0100-846-2025.

E-mail address: [email protected]

1876-6102 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of Applied Energy Innovation Institute
doi:10.1016/j.egypro.2015.07.690
3240 A.S.Hanafi et al. / Energy Procedia 75 (2015) 3239 – 3252

theory to consider the loss coefficient at different parts of the ejector. However, this theory is used to
predict performance, only when the ejector operates at its design conditions. Huang et al. [3] developed a
1-D model for analyzing ejector performance at critical mode operation, considering hypothetical throat.
El-Dessouky et al. [4] developed a simple empirical model to design and evaluate the performance of
steam jet ejectors, based on a large database, that have been extracted from several ejector manufacturers
and a number of experimental literatures.
Due to the difficulty of predicting flow behaviour inside the ejector, because of highly turbulent and
supersonic speed flow, computational fluid dynamic CFD and numerical simulation became the most
efficient and reliable method to obtain the flow field inside complex geometries with less time and lower
cost than the experimental methods. CFD software shows very good agreement with the experimental
results that has been proved by a number of researchers. Rusly et al. [5, 6] investigated the flow
characteristics of the ejector in a refrigeration system by using the real gas model through the commercial
code, “FLUENT”, and found a good agreement. Sriveera kul et al. [7] used CFD to analyze the flow
phenomenon inside a steam ejector, according to the validation of the static pressure profile along the
wall of the ejector, entrainment ratio, and critical back pressure. CFD is also applied for visualizing the
change in the flow structure and the mixing process inside the steam ejector as influenced by interested
parameters, ejector’s operating conditions and geometries. Pianthong et al. [8] investigated the flow
phenomenon and performance of two typical ejectors used in refrigeration systems. The results are
validated with experimental work and simulation in other various conditions. MyoungKuk et al. [9] used
CFD technique to study the effects of compressibility and turbulence on the flow structure inside steam
ejector in MED system. The study examines the effects of operating pressure and ejector geometry on
the performance of the steam ejector. Ariafar, K. [10] studied the flow behaviour along the axis of a
designed model ejector. The author investigated the pressure distribution profile along the axis. Then,
the author analyzed the ejector performance by choosing different operating conditions at the boundaries.
In the present work, operating conditions of the considered steam ejector are according to that of Abu-
Qir power station, which uses MED-TVC unit to produce distilled water from sea water for the steam
generator make-up. 1-D model is used to obtain ejector geometry. The flow behaviour inside the ejector
is investigated, using CFD code “ANSYS FLUENT 14.5”, by showing the Mach number and pressure
distributions along the axis and analyzing the performance to obtain the optimum operating points by
changing boundary conditions.

2. Principal of operation
The main function of the ejector is to suck vapours or gases from low pressure places to high
pressure ones, by converting the kinetic energy into pressure energy. The ejector consists of three main
parts; nozzle, mixing chamber and diffuser. The present ejector as well as its nozzle has the converging
diverging form. The high pressure steam that enters the nozzle is referred to as “the motive steam” or
“primary fluid”. The recovered low pressure steam is known as “the entrained vapour” or “secondary
fluid”, and the steam that exits the ejector from the diffuser is named as “the discharge steam”. Fig.(1)
shows a schematic of the ejector with pressure-velocity profile along its axis. Motive steam enters the
nozzle at point "P", with high pressure and low velocity. The motive steam velocity in the nozzle
increases until reaches sonic value at the nozzle throat. Beyond the throat, the steam velocity increases to
supersonic until reaches the nozzle exit at point "2", while the pressure decreases to lower than entrained
vapour pressure at point "e". The mixing process begins in the suction chamber. At point "3" in the
converging section of the ejector, mixing between the two streams is completed. Pressure recovery of the
mixed stream begins at the later section of the converging ejector and continues until the outlet of the
ejector. After the mixed stream leaves the converging part of the ejector, it reaches the constant area
throat of the ejector, where sudden rise in the pressure occurs and flow becomes subsonic again due to
normal shock wave that formed at point "4". As the fluid moves further down the diverging part of ejector,
the velocity drops further and pressure is more recovered. Finally, the fluid reaches the design outlet
pressure at the exit of the subsonic diffuser.
 A.S.Hanafi et al. / Energy Procedia 75 (2015) 3239 – 3252 3241

Fig. 1. Variation in stream pressure and velocity as function of location along the ejector.

3. Mathematical model
Sun and Eames [11] pointed out that there are two basic approaches for ejector analysis, based on mixing
of motive steam and entrained vapour, namely constant pressure and constant area. Keenan [1] concluded
that, the constant pressure ejector shows better performance than the constant area ejector. Consequently,
the constant pressure model is commonly applied in the ejector mathematical modeling. Mundy and
Bagster [12] developed constant pressure mixing model by assuming that the motive steam doesn’t mix
with the entrained vapour after the primary nozzle exit immediately. Primary fluid creates a convergent
duct for entrained vapour, subsequently the entrained vapour is accelerated until reached its sonic speed at
the hypothetical throat area (effective area), as seen from Fig. (2), after which the mixing of two streams
starts at uniform pressure [3].
In the present work, constant pressure model is applied, as indicated in Fig.(3), based on the following
assumptions:
x Steam follows ideal gas behavior with constant specific heat.
x Steady, 1-D conditions and adiabatic flow.
x Inner wall of the ejector is adiabatic.
x Velocities of motive steam, entrained vapour and discharge flow are negligible.
x Friction losses are introduced by applying isentropic efficiency to primary nozzle, diffuser and
mixing chamber.
x Hypothetical throat occurs at constant area section of the ejector.
x Two streams mixing occurs at uniform pressure Ppy = Psy = Pm at section y-y, before shock wave.

3.1. Governing equations

Mass balance equation :
mm m p  ms (1)
Where "‫ "ۦ‬is mass flow rate and subscripts m, p and s, defined mixed, primary and secondary mass flow
rates, respectively.
Entrainment ratio:
ms
Z (2)
mp
3242 A.S.Hanafi et al. / Energy Procedia 75 (2015) 3239 – 3252

Fig. 2. Effective area occurring in the ejector throat.

Fig. 3. Schematic diagram of ejector performance.

For given primary flow pressure Pp, temperature Tp and mass flow rate ীp , at chocking conditions, with
following gas dynamic relations, the cross section area for the primary throat nozzle Ath , is obtained as:
J 1
A th P p J § 2 · J 1
m p ¨ ¸ Kn (3)
Tp R ©J 1¹

Where K n is the nozzle isentropic efficiency. Using isentropic gas dynamic relations between Mach
number at nozzle exit M p and exit primary nozzle pressure p p and exit cross section area A p , it can
1 1 1

be found that :
J 1
ª º
2K n « § P p · J
»
M ? ¨ ¸¸ 
p1
J  1 « ¨© P p 1 ¹
»
(4)
¬« ¼»
2
J 1
§ A p1 · 1 ª 2 § J 1 2 · º J 1
¨ ¸ 2 «J 1 ¨1  M p1 ¸»
© A th ¹ M P1 ¬ © 2 ¹¼ (5)
Applying the previous assumptions, chocking of the entrained flow occurs at section y-y and Msy=1, for a
given stagnation secondary pressure Ps , as follows :
J 1
ª º
2 « § Ps · J
»
M ¨ ¸¸  1»
sy
J  1 « ¨© P s y
«¬ ¹ »¼ (6)

Which results from the following assumption:

Pm Pp y Psy (7)
Mach number Mpy for primary flow at section y-y can be obtained from the relation:
 A.S.Hanafi et al. / Energy Procedia 75 (2015) 3239 – 3252 3243

J
§ J  1 2 · J 1
¨1  M p1 ¸
P py © 2 ¹
J
P P1 (8)
§ J  1 2 ·J 1
¨1  M py ¸
© 2 ¹
Using isentropic gas dynamic relations to calculate cross section area of primary flow Apy at section y-y,
where coefficient ‫׎‬p is included to consider primary flow losses from section 1-1 to y-y, as:
J 1
ª º
2 J  1 2 ? J  
A
« ‡ p / M py  ¨§ §
¨1  M 2
py
· ·
¸ ¸
»
py « © J  1 © 2 ¹ ¹ »
« J 1 »
A « » (9)
p1
2 J  1
 §¨ · · 2 ? J  
§
« 1 / M p1 ¨1  M 2
p1 ¸ ¸ »
¬« © J  1 © 2 ¹ ¹ ¼»
For given secondary flow pressure Ps , temperature Ts and mass flow rate ীs at chocking condition, and
following gas dynamic relations, the cross section area for primary throat nozzle Asy , is obtained as:
J 1
A s y Ps J § 2 · J 1
ms ¨ ¸ Kn (10)
Ts R © J 1¹
The constant cross sectional area of the ejector A3 at section y-y is calculated by summing up the primary
flow area Apy and secondary flow area Asy , as:
A3 A p y  A sy (11)
Appling gas dynamics relations, the Mach M and the critical Mach M* numbers at any point is related by:
2

M * M J  1
2 (12)
M J  1  2
Where critical Mach number M* is the ratio between the local fluid velocity to the velocity of sound at the
critical conditions.
The mixing process is modelled by one dimensional continuity, momentum and energy equations.
These equations are combined to define the critical Mach number of the mixture M*m in terms of the
critical Mach number for the primary and entrained fluids at section y-y, as follows:
* *
*
M Py ZM sy ? Ts T p
M m (13)
1  Z  1  Z T s / Tp 
From equations (12) and (13), the Mach number of the mixing flow at section m-m can be obtained. The
Mach number M3 after shock wave at section s-s, is given by:
§ J 1 · 2
1 ¨ ¸ M m
M © 2 ¹
3
§ J 1 ·
(14)
J M m2  ¨ ¸
© 2 ¹
From Mach after shock wave M3 and pressure Pm and Mach Mm before shock wave, the pressure after
shock wave is expressed by:
2
P3 1 J M m
2 (15)
Pm 1 J M 3

The steam pressure at the exit of the diffuser follows isentropic gas dynamic relations according to:
J
Pc § K ? J  M ·J 1
¨1 
d
3
2
¸ (16)
P3 © 2 ¹

Where K d is the diffuser isentropic efficiency.

3244 A.S.Hanafi et al. / Energy Procedia 75 (2015) 3239 – 3252

Using the above mentioned 1-D model, performance analysis can be carried out to determine the ejector
cross section areas Ath , Ap1, A3 and entrainment ratio Ȧ for specific boundary conditions (Pp, Tp, Ps and Ts)
and desired back pressure Pc. The performance analyses procedure follows the flow chart shown in
Fig.(4).

Fig. 4. Simulation flow chart in the ejector performance analysis.

3.2. CFD model

Fig.(5) shows the dimensions, which are calculated for the designed ejector, based on the mathematical
model provided in thermodynamic model. “ICEM 14.5” is used to create grid elements for the designed
ejector. 2-D axisymmetric solver is chosen in order to consider 3-D model effect for simulation. 2-D
axisymmetric model uses less computational time and low memory compared with 3-D model.
Moreover, it has been proven that it gives nearly the same results of 3-D model [8]. Several meshes are
created to confirm that the results are grid independent. The final structured mesh consists of 248,621
quadrilateral elements, unstructured mesh, size 1x1 mm. Two pressure inlets are selected for primary and
entrained flows entering the ejector and one pressure outlet is selected for the discharged flow leaving the
ejector.
The current model primary flow is superheated steam at total pressure 6.7 bar and total temperature
180୲C. Entrained vapour set based on the saturation condition at temperature 45୲ C; for discharge vapour
is superheated steam at pressure 0.3 bar and temperature 115୲ C.

Fig. 5. Schematic view of the designed ejector.

 A.S.Hanafi et al. / Energy Procedia 75 (2015) 3239 – 3252 3245

4. Numerical Solution Procedure

The governing equations are solved numerically by using the “ANSYS FLUENT 14.5” software,
which is a commercial CFD package that uses the control volume based methods to convert these
equations into algebraic equations. The nonlinear governing equations are solved using the coupled
implicit solver, and the standard wall function is applied on the near wall treatment. The working fluid is
water vapour, which is assumed to be an ideal gas, with its properties selected in the FLUENT database.
It has been proved before, by other researchers, [9, 10, 13, 14] that, using ideal gas assumption is valid at
low operating pressures.
The realizable k-İ turbulence model is selected to express the turbulence characteristics in this
simulation. This model has been used by previous investigators [8-10, 13, 14] and has proven better
agreement with experimental data and more accuracy in predicating the spreading rate of both planar and
round jets. The governing equations are re-written in their compact Cartesian form as:
wU w
 U ui  0 (17)
wt w xi
w w wP w W ij
wt
 U ul  
wx j
 U u iu j  
w xi

wx j
(18)

w w § & § wT · &
wt
U E  
wx j
[u i
¨¨
U E  P )] ’ .⡻
⡻ ¨ D e ff ¸  ’ .⡻
wxi ¹
ªu
¬ j W  º¼
ij (19)
© ©
And
§ wui wu j · 2 wuk
W ij P e ff ¨  ¸¸  P e ff G ij (20)
¨w x wxi 3 wxk
© j ¹

The modelled transport equations for k and H in the realizable k-H model are:
w w w ª§ P t · wk º
U K   U K ui  «¨ P  ¸ »  G k  U H  YM (21)
wt w xi w x j ¬« © V k ¹ w x j ¼»
w UH   w UH ui  w ª§ P t · wH º H2 H
«¨ P  ¸ »  U C1S H  U C 2  C 1H C 3H G b
wt w xi w x j ¬« © V H ¹ w x j ¼» K  QH K (22)
Where
ª K º
C1 m ax « 0 .4 3 ?K
? SK H (23)
¬ K  5 »¼
In these equations, C1 and C2 are constants. ık and ıছ are the turbulent Prandtl numbers for k and İ,
respectively. The model constants are C1=1.44, C2=1.9, ık=1.0, ıİ=1.2, the eddy viscosity is computed
from:
K2
Pt UCP (24)
H

5. Results and Discussion

5.1. Steady state
In this section, the steady state obtained results will be discussed in more details. Fig.(6) Shows the
Mach number contours and the static pressure distributions along the ejector length at steady state
operating conditions. It may be observed that, the motive steam leaves the primary nozzle at supersonic
velocity with pressure higher than the mixing pressure. Shock train can be observed through the
fluctuation of static pressure and Mach number along centre line of the ejector before mixing. Due to the
high difference in the speed between the motive steam and surrounding entrained vapour, a separation
layer between them is created which increases the velocity of entrained vapour until mixing occurs. The
3246 A.S.Hanafi et al. / Energy Procedia 75 (2015) 3239 – 3252

complete mixing between the two streams is obtained at the constant throat area, where the shock train
diminishes and the oblique shock wave is taken place. This may be attributed due to the significant
increase in the static pressure, after which the flow velocity becomes subsonic and the pressure starts to
increase across the diffuser section.

(a) Mach number contours

(B) Static pressure distribution along ejector

Fig. 6. Mach and static pressure distribution inside the ejector under Pp=6.7 bar

5.2. Effect of Operating Conditions

5.2.1 Effect of the back pressure
The effect of changing the ejector back pressure on the entrainment ratio at constant primary and
secondary pressures is shown in Fig.(7). It may be pointed out here that, the flow inside the ejector is
divided, according to the back pressure, into three regions; double choking, single choking and un-choke
(reversed flow). In the double choke region, the primary and secondary flows are choked simultaneously
and the entrainment ratio remains constant, until the back pressure is increased and reached the critical
back pressure. At the critical back pressure, a single choke region is started, where the secondary flow
becomes un-choked, while the primary flow is still choked. In this region, the entrainment ratio decreases
sharply until reaches zero value, at a pressure called “break down pressure”. Then any further increase
in the back pressure will lead to the appearance of reversed flow at the suction inlet in the ejector.
 A.S.Hanafi et al. / Energy Procedia 75 (2015) 3239 – 3252 3247

Fig. 7. Effect of changing back pressure on entrainment ratio at Pp=6.7 bar, Ps=0.094bar.

Fig.(8) shows the Mach number contours at different back pressures. As can be seen from this figure, any
increase of back pressure causes the position of the oblique shock wave to move backward (towards the
primary nozzle) until the back pressure reaches 0.3 bar (critical back pressure). Further increase in the
back pressure results the oblique shock wave to disappear (single choke region). After back pressure 0.33
bar (break down pressure) reversed flow starts, as shown in Fig.(9). Once the critical back pressure is
exceeded, the oblique shock wave is moved backward towards the primary nozzle, causing a decrease in
the axial velocity of the mixed flow. More increase of the back pressure results the oblique shock wave
to reach the primary nozzle, causing disturbance in the primary jet core until the primary flow couldn’t
expand. In turn this will force the primary flow to move back to the entrance of the entrained vapour [14].

5.2.2. Effect of the primary pressure

In order to increase the ejector efficiency and decrease energy cost, minimum primary pressure and
maximum entrainment ratio should be achieved. Fig.(10) shows Mach number contours at different
primary pressures. It can be observed from this figure that, increasing the primary pressure shifts the
oblique shock wave forward to the diffuser section and increases the shock intensity. On the other hand,
as the primary pressure is decreased below the designed (optimum) point, the oblique shock wave moves
backward towards the nozzle exit with lower intensity until vanishes. Further decrease in primary
pressure causes a reversal flow.
Fig.(11) indicates the effect of changing the primary pressure on the entrainment ratio. It may be
noticed that, there is an optimum point of the primary pressure which gives maximum value of the
entrainment ratio. Below this pressure, the entrainment ratio is decreased significantly until reached
reversed flow. On the other hand, after the optimum point, the increase of the primary pressure has no
significance effect on the entrainment ratio and the ejector will waste an extra energy gained from
increasing primary steam pressure. The relation between the entrainment ratio variation with the primary
pressure is attributed to the size of the jet core and effective area (hypothetical throat) [10].
3248 A.S.Hanafi et al. / Energy Procedia 75 (2015) 3239 – 3252

0.24

0.28

0.29

0.30

0.31

0.32

0.33

Fig. 8. Contours of Mach number at various back pressures in bar at Pp=6.7 bar, Ps=0.094 bar.

Fig. 9. Velocity contour showing primary flow reversed to the entrance of secondary flow.

6.00

6.20

6.50

6.70

6.90

7.10

7.30

Fig. 10. Mach number contours at different primary pressures, at Pb=0.29 bar, Ps=0.094 bar.
 A.S.Hanafi et al. / Energy Procedia 75 (2015) 3239 – 3252 3249

Optimum primary
pressure

Fig. 11. Effect of the primary pressure on entrainment ratio, at Pb=0.29 bar, Ps=0.094 bar.

According to Munday and Bagster theory [12], and as described previously, the size of jet core is
directly proportional to the primary steam pressure, while the effective area is inversely proportional to
the size of jet core. It is recommended to have a bigger effective area which will allow a higher entrained
secondary flow. Consequently, increasing the primary pressure (above the optimum point) gives bigger jet
core, and smaller effective area that leads to lower entrainment ratio. Below the optimum point at low
primary pressure, the created effective area is bigger than the critical area A*, needed for choking the
secondary flow, hence the increase of the primary pressure leads to decrease in the effective area until it
reached the critical area A* at the optimum primary pressure which results increasing the entrainment
ratio.

5.2.3. Effect of the primary steam temperature

The effect of increasing the degree of superheating for the primary steam flow is indicated in this
section. It is evident that, raising the primary flow temperature at a constant pressure will increase the
energy content of the flow, which in turn will decrease the primary mass flow rate required for the same
output pressure. Hence, the entrainment ratio and the performance of the ejector are increased as shown in
Fig.(12). Moreover, the discharged flow temperature, at the diffuser exit is increased with increasing the
primary flow temperature, as illustrated in Fig.(13).

Fig. 12. Effect of changing primary flow temperature on entrainment ratio, at Pp=6.7 bar.
3250 A.S.Hanafi et al. / Energy Procedia 75 (2015) 3239 – 3252

Fig. 13. Effect of changing primary flow temperature on exit flow temperature, at Pp=6.7 bar.

6. Mathematical model validation

The present 1-D mathematical model is validated, by comparing its constant area results with those of
current TVC unit operated in MED-TVC of Abu-Qir power station at the same operating conditions. The
results show good agreement with dimensions of Abu-Qir MED-TVC, as listed in Table (1).

Table 1.Comparison between 1-D model results and Abu-Qir ejector geometry.
Ejector Abu-Qir 1-D model Error %
Dth 92.88x10-3 91.83x10-3 - 1.13%
Dp1 331.5x10-3 294.70x10-3 - 11.1%
D3 547.60x10-3 645.60x10-3 + 17.8%

7. Conclusions
The present work presents a 1-D analysis for predicting the ejector geometry at desired operating
conditions and specified entrainment ratio. Constant pressure mixing is assumed to occur inside the
constant area section of the ejector and the entrained flow choking condition is analysed. Flow behaviour
within a designed model of an ejector in MED-TVC desalination unit has been investigated by using a
CFD method. The effect of changing operating conditions on the ejector performance was evaluated. It
has been verified that the CFD is an efficient tool to estimate the entrainment ratio and critical back
pressure of the ejector for different operating conditions. The maximum back pressure, the ejector can
withstand without decreasing the entrainment ratio was determined. Moreover, it is obtained that, there
was an optimum value of the primary steam pressure, which gives maximum entrainment ratio and any
further increase in the primary steam pressure will be considered as waste in energy. It is concluded that,
increasing the degree of superheat for the primary steam temperature will increase the entrainment ratio,
taking into consideration that, the dis-charge steam temperature will also increase which will affect, in
return, the temperature of the first effect in the MED-TVC unit. CFD also helps to reveal the phenomena
inside the ejector in details.

Acknowledgment
The authors acknowledge the support provided by Amr Fathy who has a many contributions in this work.

Nomenclature
Symbol
A area, m2
Ȧ entrainment ratio
A sonic velocity, m/s
 A.S.Hanafi et al. / Energy Procedia 75 (2015) 3239 – 3252 3251

Cp specific heat of gas at constant pressure, kJ/kg.K

Cv specific heat of gas at constant volume, kJ/kg.K
g acceleration of gravity vector
h enthalpy, kJ/kg
ী mass flow rate, kg/s
M Mach number
P Pressure bar.
T temperature K
x-x nozzle position, m
y-y position of the hypothetical throat
ī Cp/CV
R gas constant, kJ/kg.K
K thermal conductivity, turbulent kinetic energy
S modulus of the mean rate-of-strain tensor
T Time
U mean velocity
U fluctuating velocity
V velocity tensor
Ǽ turbulent kinetic energy dissipation rate
Ȃ dynamic viscosity
ȇ fluid density
IJij shear stress tensor
Ȋ kinematic viscosity
ȣt turbulent viscosity
ık turbulent Prandtl numbers for k
ıİ turbulent Prandtl numbers for İ

Subscripts
P primary flow
S secondary flow, entrained flow
B back flow, discharged flow
M mixed flow
Th nozzle throat
p1 nozzle exit
Py primary flow at the location of choking for the
entrained flow
Sy entrained flow at the location of choking for the
entrained flow
1 nozzle exit
2 entrance of the constant-area section
3 exit of the constant-area section

References
1. Keenan, J. and E. Neumann, A simple air ejector. ASME Journal of Applied Mechanics, 1942.
64: p. 75-82.
2. Eames, I., S. Aphornratana, and H. Haider, A theoretical and experimental study of a small-scale
steam jet refrigerator. International Journal of Refrigeration, 1995. 18(6): p. 378-386.
3252 A.S.Hanafi et al. / Energy Procedia 75 (2015) 3239 – 3252

3. Huang, B.J., et al., A 1-D analysis of ejector performance. International Journal of Refrigeration,
1999. 22(5): p. 354-364.
4. El-Dessouky, H., et al., Evaluation of steam jet ejectors. Chemical Engineering and Processing:
Process Intensification, 2002. 41(6): p. 551-561.
5. Rusly, E., et al. Ejector CFD modeling with real gas model. in Mechanical engineering network
of Thailand the 16th conference. 2002.
6. Rusly, E., et al., CFD analysis of ejector in a combined ejector cooling system. International
Journal of Refrigeration, 2005. 28(7): p. 1092-1101.
7. Sriveerakul, T., S. Aphornratana, and K. Chunnanond, Performance prediction of steam ejector
using computational fluid dynamics: Part 2. Flow structure of a steam ejector influenced by
operating pressures and geometries. International Journal of Thermal Sciences, 2007. 46(8): p.
823-833.
8. Pianthong, K., et al., Investigation and improvement of ejector refrigeration system using
computational fluid dynamics technique. Energy Conversion and Management, 2007. 48(9): p.
2556-2564.
9. Ji, M., et al., CFD investigation on the flow structure inside thermo vapor compressor. Energy,
2010. 35(6): p. 2694-2702.
10. Ariafar, K., Performance evaluation of a model thermocompressor using computational fluid
dynamics. International Journal of Mechanics, (1): p. 35-42.
11. Sun, D.-W. and I.W. Eames, Recent developments in the design theories and application of
ejectors: a review. Journal of the Institute of Energy, 1995. 68(475): p. 65-79.
12. Munday, J.T. and D.F. Bagster, A New Ejector Theory Applied to Steam Jet Refrigeration.
Industrial & Engineering Chemistry Process Design and Development, 1977. 16(4): p. 442-449.
13. Cai, L. and M. He, A Numerical Study on the Supersonic Steam Ejector Use in Steam Turbine
System. Mathematical Problems in Engineering, 2013. 2013: p. 1-9.
14. Ruangtrakoon, N., et al., CFD simulation on the effect of primary nozzle geometries for a steam
ejector in refrigeration cycle. International Journal of Thermal Sciences, 2013. 63: p. 133-145.


Biography
Ahmed Waheed is a current research assistant in Cairo University. He works on two research topics; the
first involves one dimensional mathematical modeling and CFD investigation for steam jet ejectors and
the second involves a Thermo-economic analysis of combined cycle MED-TVC desalination system.
Waheed has been graduated from Mechanical power engineering department, Cairo University in 2009
and finished his master courses of Fluid Mechanics in 2011 of GPA 3.3. He currently works for Shotec
Company as a senior technical sales engineer. He has received many trainings in the engineering field
such as PLC S7 basic and S7SYS1 “Siemens”, Training for Bornemann Multiphase Pumps , Training for
NETZSCH pumps in oilfield industry¬, NETZSCH Tornado rotary lobe pumps Familiarization
Training, OSHA Occupational Safety & Health Administrator, Design And maintenance of Air
Conditioning and ventilation, and Design of Automatic Firefighting system.