Journal of Applied Fluid Mechanics, Vol. 11, No. 6, pp. 1511-1520, 2018.

Available online at www.jafmonline.net, ISSN 1735-3572, EISSN 1735-3645.

DOI: 10.18869/acadpub.jafm.73.249.28989

Computational Analysis of Integrated Engine Exhaust

Nozzle on a Supersonic Fighter Aircraft
I. Arif1†, J. Masud2 and I. Shah1
1 Department of Aerospace Engineering, College of Aeronautical Engineering, National University of
Sciences and Technology, H-12, Islamabad, Pakistan
2 Department of Mechanical Engineering, Air University, Islamabad, Pakistan

†Corresponding Author Email: [email protected]

(Received March 14, 2018; accepted May 9, 2018)

ABSTRACT

A unique approach of analyzing jet exhaust nozzle integrated to aircraft and propulsion system is presented in
this paper. Engine exhaust nozzle is usually omitted in Wind Tunnel Testing and numerical analysis of
aircraft due to complexities involved in integration of nozzle and presence of high pressure / temperature
inside exhaust nozzle. Also, the flow properties are non-uniform and highly turbulent in the vicinity of nozzle.
Therefore, exhaust nozzle is usually analyzed in isolation and these results often lead to inaccuracies from
actual scenario where nozzle is integrated with aircraft and its propulsion system. This research aims to
integrate engine exhaust nozzle on a supersonic fighter aircraft and analyze its flow characteristics and
variation in performance parameters due to its integration. Engine propulsion characteristics and parameters
such as nozzle inlet temperature and total pressure have been analyzed through an in-house validated engine
analytical model developed by some of the authors of this study. In the first part of paper, exhaust plume
structure has been analyzed to study the flow behaviour (flow turbulence and flow distortion etc) at nozzle
exit. Later, nozzle performance parameters such as Exit Velocity, Nozzle Pressure Ratio (NPR), Engine
Pressure Ratio (EPR), and Engine Temperature Ratio (ETR) have been calculated when exhaust nozzle is
integrated with the aircraft. Finally, the results are compared and validated with analytical calculations to
compare the performance of nozzle when it is in isolation and when it is integrated on aircraft. It is observed
that nozzle flow has no significant effect on aircraft major surfaces such as fuselage, wing upper and lower
surfaces, and nose section. However, there is a prominent effect of exhaust nozzle flow on horizontal
stabilizers, vertical tail and rear fuselage area of the aircraft. An average difference of 18% in NPR, 12% in
EPR, and 9% in ETR is observed between integrated nozzle and isolated nozzle which further signifies the
importance of integrating exhaust nozzle in aircraft analysis. This proposed methodology will allow more
accurate analysis of the effects of exhaust nozzle on the overall performance of aircraft. The methodology can
further be used for proposing design changes in existing nozzle configurations.

Keywords: Aerodynamics; Internal Flows; Nozzle; Plume; Nozzle Pressure Ratio.

NOMENCLATURE

α Angle of Attack P Static pressure (Pa)

AoA Angle of Attack S-A One equation Spalart-Allmaras
CFD Computational Fluid Dynamics turbulence model
Mach number Mach number RANS Reynolds Averaged Navier-Stokes
NPR Nozzle Pressure Ratio y+ Non-dimensional length scale
EPR Engine Pressure Ratio associated with turbulence model
ETR Engine Temperature Ratio

1. INTRODUCTION shape, size and location on aircraft. Exhaust nozzle

integrates propulsion system with aircraft systems
Engine exhaust nozzle is a simple yet sensitive and its design is critical to engine performance
device which allows the hot gasses to flow through (Boyce 2011). Its design includes installed and
it. Nozzle performance largely depends upon its uninstalled performance estimation. Other design
challenges include area variations through
I. Arif et al. / JAFM, Vol. 11, No. 6, pp. 1511-1520, 2018.

convergent divergent principle, thrust vectoring and characteristics and performance is quite different
thrust reversing (Gamble, et al. 2004). Flow when propulsion system and aircraft external flow
ingested into engine through intake or propelled interaction with exhaust plume is taken into account.
through exhaust have a prominent effect on flow Effect of jet exhaust plume on control surfaces of
surrounding the body. Generally, analysis of aircraft and change in pitching and yawing moments
exhaust nozzle is excluded from aircraft numerical are also important in determining the stability
analysis as it involves complexities related to parameters of aircraft (Pandya, et al. 2004).
internal flows and its interaction with external flow
over the aircraft surface. Furthermore, integration of A unique approach of analyzing jet exhaust nozzle
exhaust nozzle with aircraft requires comprehensive integrated to aircraft and propulsion system is
data about the installed propulsion system to presented in this paper. Internal flow through
evaluate boundary conditions for exhaust nozzle at exhaust nozzle interacts with external flow over the
each flight condition which is only available to aircraft and plume structure is observed which is
manufacturers. However, this simplification and different than the plume structure observed
assumption often results in considerable downstream of a nozzle in isolation. Engine
inaccuracies in analysing the aircraft characteristics propulsion characteristics and parameters such as
and nozzle performance as well. For any design nozzle inlet temperature and total pressure were
engineer, it is critical to evaluate nozzle uninstalled analyzed through an in-house developed engine
and installed performance to accurately predict the analytical model developed by some of the authors
aircraft performance. of this study (Arif, et al. 2018).

A significant number of studies on exhaust nozzle Owing to the complexities involved in the
design and its performance in isolation has been integrated solution of internal and external flows,
conducted in the past and available in literature previous numerical studies on the selected aircraft
(Rao 1958, Mikhail, et al. 1980, Stitt 1990, Li and did not consider aircraft propulsion system and
Gutmark 2005, Jassim 2016, Virdi, et al. 2017). exhaust nozzle while carrying out flow simulations
Many researchers studied the flow downstream of around the aircraft (Hassan, et al. 2015). Separate
exhaust nozzle to analyze the jet plume structure analysis of propulsion system have however, been
(Mc Ghee 1970, Robinson and High 1974, conducted (Arif, et al. 2018). In order to integrate
DalBello, et al. 2004). Jet plume characteristics has the propulsion system and exhaust nozzle with
been studied in detail by Korst, et al. and Adamson baseline aircraft for numerical analysis, it was
Jr. (Korst, et al. 1981, Adamson Jr 2012). Similarly, important to ascertain different engine parameters at
computational analysis of exhaust plumes of axi- all flight conditions. Calculation and finalization of
symmetric nozzle was performed to analyze the jet boundary conditions is considered one of the most
plume characteristics(Dash, et al. 1978, DalBello, et critical step in CFD analysis. Input conditions for
al. 2004). The study conducted by Dash et al. intake, exhaust and aircraft were required for each
included shock structure, viscous and shear effects flight condition in order to carry out accurate
and mixing effects of jet (Dash, et al. 1978). Later, numerical analysis. Calculation of exhaust nozzle
a parabolized Navier-Stokes model was developed boundary conditions is quite complex and
to simulate the viscous flow (Dash, et al. 1985). cumbersome as it requires complete details of
The model was found to be quite compatible in propulsion system. For this purpose, mathematical
analyzing the effect of viscous flow turbulent jet modelling for RD-93 engine was carried out to
mixing. Chuech, et al. (Chuech, et al. 1989) ascertain the exhaust nozzle boundary conditions
modified the k-epsilon turbulence model to study (pressure, temperature, pressure ration etc) at
the turbulent mixing phenomena in under expanded different flow conditions.
jet flow in stagnant medium. Later, the k-epsilon
model was further modified to analyze the The analysis in this research is carried out at subsonic
compressibility effects at high speed. This model and supersonic regimes at different flight conditions.
yielded much better results to predict the pressure at The final geometry includes exhaust nozzle and
different points in under expanded nozzle. A two intake duct along with aircraft external surfaces. The
equation k-epsilon model was successfully applied paper is divided into two major parts, the first part
to predict the jet flow of under-expanded nozzle presents the exhaust flow characteristics observed
using Reynolds Averaged Navier Stokes Equations plume structure downstream of exhaust nozzle when
(RANS) (Woodmansee, et al. 1999). This model it is integrated with aircraft. Exhaust plume structure
allowed to ascertain the Mach number upstream of is analyzed to observe the flow behavior (flow
Mach disc and yielded accurate results in agreement turbulence and flow distortion) at nozzle exit. In the
with experimental data. second part, nozzle performance is analyzed when it
Though a significant literature is available on jet is integrated with the aircraft. Nozzle performance
nozzle design, characteristics and performance. parameters such as Exit Velocity, Nozzle Pressure
However, most of the research on exhaust nozzle Ratio, Engine Pressure Ratio, and Engine
flows are done with jet exhaust nozzle in isolation or Temperature Ratio are calculated for comprehensive
their associated base flows (Rao 1958, Mikhail, et al. analysis. Finally, the results are compared and
1980, Abdol-Hamid, et al. 1993, Li, et al. 2009, Butt validated with analytical calculations to compare the
and Arshad 2015). These studies mainly focused on performance of nozzle when it is in isolation and
flow behavior inside and downstream of exhaust when it is integrated on aircraft. The aircraft and
nozzle, whereas, the actual nozzle flow exhaust nozzle geometries are shown in Fig. 1 below.

1512
I. Arif et al. / JAFM, Vol. 11, No. 6, pp. 1511-1520, 2018.

ρ
 u u u u 
u v w 
 t x y z 
(4)
p u u u 
2 2 2

 μ   F
 x 
x
x y z
2 2 2

Conservation of Energy:

ρC p  T

t
u
T

x
v
T

y
w
T

z
 

(5)
Φ
 k T    k T    k T 
Fig. 1. Aircraft without exhaust nozzle (Top) and x  x  y  y  z  z 
with exhaust nozzle (Bottom)
where,
2. MODELLING AND ρ is the fluid density; µ is the kinematic viscosity;
COMPUTATIONAL SETUP u,v,w are the component of velocity in Cartesian
coordinates; p is the pressure term; Fx, Fy, Fz are the
During the course of research, number of steps were body force terms; T is temperature in Kelvins; and k
followed which commenced from exhaust nozzle is the heat transfer coefficient.
modelling and integrating it with aircraft and intake
2.1 Overview of Engine Modelling
geometries. Aircraft, intake and exhaust were taken
as separate entities so that changes can be made in The propulsion system of the aircraft under study is
any of the part without affecting other geometry. a dual-duct, dual-rotator turbofan engine, which has
Exhaust geometry was modelled with the help of an afterburner shared by inner/outer channel and a
dimensions provided by Manufacturer’s engine fully adjustable supersonic nozzle. Due to military
manual. The geometries were meshed and were confidentiality, comprehensive engine performance
later appended to Numerical Solver Software data of fighter aircraft engines at different
Fluent® (ANSYS 2013). The meshing strategy operating conditions is not available with its users.
adopted for the research was based on unstructured Hence, it is impossible to ascertain certain flow
scheme for all geometries including aircraft, intake parameters such as pressure and temperature at
and exhaust. different engine stations or propose any design
modification in existing engine design. In order to
Numerical simulations were carried out at zero side
determine the flow properties at certain engine
slip angle which allowed the simulation with
components specially exhaust nozzle for further
symmetry plane. For numerical analysis in this
numerical analysis of aircraft with integrated
study, Reynolds-averaged Navier-Stokes (RANS)
propulsion system, an in-house analytical model
set of equations are used to account for time
was developed and validated at Afterburner
dependent behavior of flow. The governing
configuration (Arif, et al. 2018). A comprehensive
conservation equations are (Hirsch 2007):
scheme was developed for modelling and
Conservation of Mass: simulation engine in accordance with Aircraft
Engine Design literature (Mattingly 2002) and
ρ   ρu    ρv    ρw  software packages ONX® and PERF®. The model
- = + + (1) is verified by thrust matching technique which
t x y z further helped in ascertaining different engine
parameters at off design conditions as well. This
Conservation of Momentum:
analytical model also forms an important part of an
already conducted and published study by the same
ρ
 u u u u 
u v w  authors for characterization of exhaust effects on
 t x y z  aerodynamic behavior of a supersonic aircraft
(2) (Masud, et al. 2017).
p u u u 
2 2 2

 μ   F 2.2 Grid Density

 x 
x
x y z
2 2 2

Computational grid size and type plays an important

part in computational analysis. The grid generation
ρ
 v v v v  must be fine enough to resolve all flow gradients.
u v w 
 t x y z  However, this aspects needs to be balanced with
(3) available computational power which puts limit on
p  v v v 
2 2 2
grid size. Aircraft and exhaust geometries were
 μ   F meshed separately using unstructured meshing
 x 
y
y y z
2 2 2

scheme. Tetrahedral mesh generated for aircraft and

domain cannot capture the near wall effects for a
reasonable mesh size. For this purpose, a five layer

1513
I. Arif et al. / JAFM, Vol. 11, No. 6, pp. 1511-1520, 2018.

prism was applied on aircraft boundary to 2.4 Solution Strategy

accurately capture boundary layer and near wall
effects. A very dense mesh was created in exhaust For numerical analysis, Reynolds-Averaged Navier-
plume area to analyze the flow behavior Stokes (RANS) set of equations are used to account
downstream of exhaust. Turbulent y+ values were for time dependent behavior of flow. Double
kept at optimum level for different flight speeds. precision solver was used for better accuracy and
Prime importance was given to mesh consistency fluid was taken as air with ideal gas properties.
with previous research (Masud, et al. 2017) to Density based solver was selected with explicit
validate and compare the results. algorithm. Due to sensitivity and importance of
temperature effects in the research, energy equation
2.3 Grid Independence option was also used. 2nd order upwind scheme was
Size and type of grid has significant impact on the selected in flow discretization and 1st order upwind
accuracy of results. A highly dense and fine mesh scheme was used to cater for turbulent viscosity.
can achieve high accuracy but it is usually Suitable relaxation parameters were applied and a
computationally expensive in terms of time and courant number of 1 was selected (Masud, et al.
resources. Hence, it is important to maintain a 2017).
balance between the generated size of mesh and
accuracy of results achieved. In order to select an
2.5 Boundary Conditions
optimum mesh size, a grid independence study All surfaces of aircraft were selected as no slip
was carried out before the final selection of mesh. walls. Input conditions for exhaust and aircraft were
Three different meshes were generated based on calculated separately for this research. The
the number of cells and size function. Nozzle evaluation of boundary conditions at exhaust nozzle
Pressure Ratio at Mach number. 0.6 and AoA 00 was a complex task as it required complete analysis
for each mesh was evaluated for comparative of propulsion system. For this purpose, a complete
analysis. Details of mesh are presented in analytical model of engine was developed and
following Table 1 and results are shown in Fig. 2 validated with manufacturer’s provided data by
below. some of the authors of this study (Arif, et al. 2018).
Table 1 Grid Independence Analysis For all Mach number (subsonic and supersonic),
Grid Cells static pressures values were taken in accordance
with atmospheric sea level conditions. The case was
Grid 1 8.9 million analyzed at subsonic and supersonic Mach number
Grid 2 11.2 million with varying angles of attack and mass flow rates.
Design mass flow rate is set as 44 kg/s, 50 kg/s and
Grid 3 13. 3million
105 kg/s for Mach numbers 0.6, 0.8 and 1.5
respectively. Whereas, off design mass flow rate is
set as 30 kg/s, 40 kg/s and 90 kg/s for Mach
Grid Independence numbers 0.6, 0.8 and 1.5 respectively.
6.800 2.6 Turbulence Model Independence
6.700
6.600 Selection of turbulence model is as important as
6.500
selection of grid size for numerical analysis, hence,
NPR

6.400
6.300 this aspect cannot be ignored. For this research,
6.200 three different models (SA. k-epsilon and k-omega)
6.100 were analyzed keeping in view the complex flow
6.000
phenomenon involving both internal flow inside
Grid 1 Grid 2 Grid 3
exhaust nozzle and external flow over aircraft. SA
Fig. 2. Grid Independence Study is a single equation turbulence model while k-
epsilon and k-omega are two equation turbulence
It was observed that variation in Nozzle models (Kuntz and Menter 2004, Bulat and Bulat
Pressure Ratio was almost negligible for Grid 2 2013). Comparative results of NPR for these
(11.2 million) and Grid 3 (13.3 million), however, models are presented in Fig. 4 below:
for Grid 1, Nozzle Pressure Ratio was under
predicted. Based on this grid independence study, a
Grid 2 (11.2 million) was selected for further Turbelence Model Independence
6.800
analysis (shown in Fig. 3). 6.600
6.400
6.200
6.000
NPR

5.800
5.600
5.400
5.200
5.000
SA k-omega k-epsilon
Fig. 3. Final Meshed Geometries of Aircraft and Fig. 4. Turbulence Model Independence
Exhaust Nozzle

1514
I. Arif et al. / JAFM, Vol. 11, No. 6, pp. 1511-1520, 2018.

Based on the results, it was observed that the variation Pressure contours on aircraft and flow downstream
between the NPR obtained from S-A model and k- of exhaust nozzle is clearly visible in the Figs. 5a, b
omega model was almost negligible, however, the and c. At subsonic Mach numbers, a large plume
results obtained from k-epsilon were slightly under section can be observed downstream of exhaust
predicted. The convergence stability and residuals nozzle in the form of shock waves and expansion
from S-A and k-epsilon were not satisfactory as waves. From the plume structure, it is clearly
disordered sinusoidal behavior was observed evident that the nozzle is under expanded for all
throughout the simulations. Hence, the average of last flow conditions. From the pressure distribution
1000 iterations were taken into account for estimation contours at subsonic flow, it is observed that nozzle
of NPR. Hence, k-omega turbulence model was flow has no significant effect on aircraft major
selected for this research based on the accuracy of
surfaces such as fuselage, wing upper and lower
results and its convergence stability. K-omega is a two
surfaces, and nose section. However, there is a
equation turbulence model which includes extra
transport equations to evaluate turbulent properties of prominent effect of exhaust nozzle flow on
fluid (2016). The results were consistent and in horizontal stabilizers, vertical tail and rear fuselage
agreement with previous research (Hassan, et al. 2015, area of the aircraft. Variation in NPR at different
Masud, et al. 2017). Mach numbers are shown in Table 2 below.
Table 2 Nozzle Pressure Ratio Vairation
3. ANALYSIS OF FLOW with Mach number
CHARACTERISTICS
M# NPR
The analysis was carried out at three different Mach 0.6 6.6
numbers (0.6, 0.8 and 1.5) at five different angle of 0.8 6.72
attack (-4, 0, 4, 8, 12). Analysis on calculated 1.5 6.80
results is divided into two parts, namely, Flow
Characteristics Downstream of Exhaust and
Exhaust Nozzle Performance Analysis. Flow From the table above, it can be observed that
characteristics including turbulence, plume with an increase in Mach number, nozzle
structure, and pressure distribution at different flight pressure ratio slightly increases from 6.6 to 6.8,
conditions are observed and are presented in which shows that the nozzle becomes more under
subsequent sections. expanded with increase in speed. This is due to
3.1 Pressure Distribution fact that the pressure at the exit of nozzle is
unable to expand till free stream pressure, hence
The analysis at subsonic speeds was carried out at the exit pressure is higher than free stream
two different Mach numbers, i.e. 0.6 and 0.8, pressure at that particular flow condition
whereas the supersonic analysis was carried out at (Mattingly and Von Ohain 2006).
Mach number 1.5. Pressure contours at Mach
numbers 0.6, 0.8 and 1.5 are shown in Fig. 5: In supersonic flow, the plume section is not clearly
visible due to excessive pressures in surroundings
due to presence of shock waves. Although, the
nozzle is still under expanded but the pressure
gradients due to shock structure dominates the flow.
Formation of strong shock waves are dominant in
pressure contours on aircraft nose, fuselage and
wing area. However, at supersonic speeds, a plume
structure can be clearly observed if temperature
plots are generated.
Fig. 5a. Pressure Contours at Mach number 0.6 AoA 0 3.2 Flow Characteristics Inside the
Nozzle
Flow behavior inside the exhaust nozzle was
studied to analyze whether the nozzle is perfectly
expanded, under-expanded or over-expanded in test
flight conditions. A sectional view and plot of static
pressure variation along the length of nozzle at
Mach number 0.8 is shown in Fig. 6 below.

Pressure Variation Inside Nozzle

5.5E+05
Fig. 5b. Pressure Contours at Mach number 0.8 AoA 0
4.5E+05
Pressure (Pa)

3.5E+05

2.5E+05

1.5E+05

5.0E+04
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
x/L

Fig. 5c. Pressure Contours at Mach number 1.5 AoA 0 Fig. 6. Pressure Variation inside Nozzle

1515
I. Arif et al. / JAFM, Vol. 11, No. 6, pp. 1511-1520, 2018.

From pressure plot, it can be observed that the static ambient pressure, hence compression waves are
pressure at the inlet of nozzle is quite high and formed to increase the pressure upto ambient
slightly decreases till throat area, however, from pressure. Compression waves are formed at the
throat area till exit of nozzle there is a sharp interaction of expansion waves with jet boundary.
decrease in pressure which indicates sudden Merging of compression waves results in barrel
expansion and rise in flow velocity. However, the shock. Barrel shock is the separation line between
expansion of flow is not enough to match the the region inside the plume which is independent of
atmospheric pressure at nozzle exit which indicates the ambient pressure and the region outside the
that the nozzle is under-expanded at this flight plume which is dependent upon ambient pressure.
condition. Since the nozzle is under-expanded in Also, the expansion waves forming at the nozzle
this case, there is a certain loss in thrust potential of lips allows the gas velocity to increase which
engine. sweeps the barrel shock further away from nozzle
axis. From the contours, Mach discs can be easily
3.3 Flow Characteristics Downstream observed downstream of exhaust which are formed
of Nozzle (Plume Structure) due to shock interactions. Also, jet boundary pattern
Plume structure downstream of exhaust was studied and Mach discs are also clearly visible from the
in detail during the research. Jet plume structure at contour plot.
Mach number 0.8 at AoA 0o is shown in Fig. 7 The variation of flow along the centreline of nozzle
below. and downstream of nozzle is plotted in Fig. 8 to
observe the flow behaviour inside and downstream
of nozzle in combination.

Pressure Variations
5.0E+05

4.0E+05
Pressure (Pa)

3.0E+05

2.0E+05

1.0E+05
Fig. 7a. Plume Structure Downstream of Nozzle
0.0E+00
0 2 4 6
x/L
Fig. 8. Pressure Variations inside and
downstream of Nozzle
From pressure plot, it is clearly evident that the flow
pressure keeps decreasing inside the nozzle.
However, the expansion of flow is insufficient and
nozzle behaviour is under-expanded as pressure at
nozzle exit is well above the ambient pressure. In this
case, a complex jet plume is produced consisting of
regions of decreasing and increasing pressure (or
Fig. 7b. Plume Structure Velocity Contour at density) (Saddington, et al. 2004). Since the flow is
Mach number. 0.8 viscous in nature, turbulent mixing and viscous losses
with surrounding air causes the sinusoidal wave
pattern to decay after small number of shock and
expansion wave formations (Abbett 1971). This
phenomenon can be observed by the sinusoidal
variation in pressure plot. A mean centreline could
also be observed through the plot and number of
Mach discs formed due to under-expansion are also
calculated with the help of the pressure plot. Same
phenomenon can also be seen in Fig. 9 below:

Fig. 7c. Plume Structure Pressure Contour at

Mach number. 0.8
A typical jet plume phenomenon downstream of
under-expanded nozzle can be seen in figure above.
The horizontal line is the axis of symmetry of
exhaust. The flow at the nozzle lip is deflected
through an angle to expand the gas into free stream
pressure. Inside the jet flow, the flow is expanded
rapidly and causes the pressure to fall below Fig. 9. Flow Structure inside and downstream of
Nozzle

1516
I. Arif et al. / JAFM, Vol. 11, No. 6, pp. 1511-1520, 2018.

4. EXHAUST NOZZLE PERFORMANCE Table 3 Variation of NPR, EPR and ETR with
ANALYSIS change in AoA
M# AoA NPR EPR ETR
The performance of exhaust nozzle is critical in 0.6 0 6.60 3.25 6.46
overall efficiency of engine and the flow behaviour
inside and downstream of exhaust. This section 0.6 4 6.61 3.26 6.46
presents the exhaust nozzle characteristics under 0.6 8 6.60 3.28 6.47
different conditions. In this part of research, nozzle
performance is analyzed when it is integrated with 0.6 12 6.61 3.26 6.46
the aircraft. The complete aircraft geometry included 0.6 -4 6.61 3.25 6.47
exhaust nozzle and intake duct along with aircraft
external surfaces. Nozzle performance was predicted
with high accuracy which was verified through the It is observed that the NPR, ETR and EPR does not
results and a plume structure was observed. Some of change with varying AoA. This is due to the fact
the most important parameters which identifies the that the inlet duct is able to deliver the design mass
nozzle performance includes Exit Jet Velocity, flow rate at all AoA effectively. This in turn keeps
Nozzle Pressure Ratio (NPR), Engine Pressure Ratio the engine operations normal at all AoA. Hence,
(EPR), and Engine Temperature Ratio (ETR). NPR, ETR and EPR are not much effected by the
change in AoA.
4.1 Nozzle Pressure Ratio (NPR)
The nozzle pressure ratio calculated from CFD
Nozzle pressure ratio (NPR) is the ratio of nozzle analysis was differs slightly from analytical results.
exit total pressure to static pressure. NPR of nozzle An average difference of 18% was observed between
in isolation can be analytically calculated using the the results at all flow conditions. The variations
following expression (Mattingly and Von Ohain between the results are due to the fact that the
2006). analytical calculations are based on perfectly
NPR = Ptotal exit / Pstatic exit (6) expanded nozzle, whereas from the CFD analysis it is
evident that the nozzle is under expanded at these
The calculated values of NPR from numerical flight conditions. The magnitude of thrust loss due to
analysis and analytical calculations were plotted at under expansion in the exhaust nozzle cannot be
different Mach number is shown in Fig. 10 below: estimated with analytical equations which results in
slight deviation from actual results. Hence, results
Nozzle Pressure Ratio from numerical analysis using this methodology
7 proved to be quite effective and realistic as it
6.8 approximates the under expansion in actual scenario
6.6
at these flight conditions.
6.4 18% Difference
NPR

6.2
6 CFD 4.2 Engine Pressure Ratio (EPR)
5.8 Analytical
5.6 Engine pressure ratio is the ratio of total pressure at
5.4 exit to total pressure at compressor inlet. This
5.2
5
parameter is directly dependent on engine
0.4 0.6 0.8 1 1.2 1.4 1.6 performance and therefore does not depend on free
Mach number stream pressure. Also, pressure losses in aircraft
intake are not catered in engine pressure ratio. EPR
Fig. 10. Nozzle Pressure Ratio
of nozzle in isolation can be analytically calculated
From the Fig. 10, it is observed that at subsonic using the following expression (Mattingly and Von
speeds, nozzle pressure ratio does not change, since Ohain 2006).
the total pressure at the exit vary linearly with
speed. Hence, the ratio of total pressure at exit to EPR = Ptotal exit / Ptotal compressor (7)
static pressure remains constant. However, at The calculated values of EPR from numerical
supersonic speed there is a slight increase in NPR. analysis and analytical calculations were plotted at
This is due to the fact that total pressure is governed different Mach number is shown in Fig. 11 below:
by exit velocity which is more dominant in
supersonic regime than in subsonic speeds.
Engine Pressure Ratio
Therefore, the nozzle pressure ratio at supersonic 4
speed is higher than subsonic speeds. A significant
3.5 12% Difference
change in appearance of external flow field is also
observed near the nozzle area in supersonic case as 3
EPR

compared to subsonic case for the same NPR. There

2.5
is a less dominant formation of shock and CFD
expansion waves near the exit as compared to 2 Analytical
subsonic Mach number. This aspect results in the
increase of NPR, as the increase in freestream 1.5
0.4 0.6 0.8 1 1.2 1.4 1.6
velocity has equivalent effect on NPR. Variation of Mach No
NPR, EPR and ETR at Mach number 0.6 with
change in AoA is shown in Table 3 below. Fig. 11. Engine Pressure Ratio

1517
I. Arif et al. / JAFM, Vol. 11, No. 6, pp. 1511-1520, 2018.

At subsonic speeds, numerical analysis results show upon material limitations as well. Therefore,
that EPR slightly increases from Mach number 0.6 temperature at combustion chamber, turbine and
to 0.8, whereas EPR calculated with analytical afterburner section is dependent upon material
methodology slightly decreases. This variation is temperature limit. This aspect limits the ETR to a
due to the fact that the analytical model is based on certain value and hence only a slight decrease in
ideal expansion of nozzle and cannot predict the ETR is observed at subsonic speeds. However, at
extent of under expansion or over expansion of flow supersonic speed, a significant decrease in EPR is
through nozzle. A significant change in EPR was observed as compared to subsonic speeds. Due to
observed in supersonic case as compared to formation of shock waves near aircraft nose,
subsonic case for the same NPR. There was a less fuselage, wing and intake area, there is significant
dominant formation of shock and expansion waves temperature gradient at these surfaces. Similarly,
near the exit as compared to subsonic Mach the total temperature at compressor inlet is also
number. It is observed that there is a large decrease higher as compared to total temperature at
in EPR at supersonic speeds. Due to formation of compressor inlet at subsonic speeds. The Total
shock waves near aircraft nose, fuselage, wing and temperature at nozzle exit is restricted due to
intake area, there is significant pressure change at material temperature limitations and hence, the ratio
these surfaces. Similarly, the total pressure at decreases.
compressor inlet is also higher as compared to total
pressure at compressor inlet at subsonic speeds. The The variation in calculated values of engine
Total Pressure at nozzle exit also increases but to a temperature ratio (ETR) is also less than that of
smaller extent as compared to the Total Pressure at NPR. An average difference of 9% was observed.
compressor inlet. A reduction in difference between the two results
was due to the fact that temperatures at
The variation in calculated values of engine combustion chamber, turbine and afterburner duct
pressure ratio (EPR) is less than that of NPR. An are restricted by material limitations. Hence the
average difference of 12% was observed between analytical results are in good agreement with CFD
the results at all flow conditions. A reduction in results.
difference between the two results was due to the
fact that compressor total pressure can be estimated 4.4 Exit Velocity Calculations
accurately at compressor inlet at design mass flow Velocity calculations from CFD results along
rate. Hence the impact of nozzle exit conditions for with comparison with analytical results are
the calculation of engine pressure ratio is less than presented in this section. The exit velocity for a
that for nozzle pressure ratio. convergent-divergent nozzle can be evaluated by
4.3 Engine Temperature Ratio (ETR) following expressions (Mattingly and Von Ohain
2006).
Engine Temperature Ratio (ETR) is the ratio of
ɤ
nozzle total temperature to compressor inlet 9
total temperature. ETR of nozzle in isolation
can be analytically calculated using the and
following expression (Mattingly and Von Ohain
2006). ṁ ṁ

ETR = Ttotal exit / Ttotal compressor (8)
10
The calculated values of ETR from numerical
analysis and analytical calculations were plotted
2
at different Mach number as shown in Fig. 12 . . 1 11
below: 1

Engine Temperature Ratio Comparison between calculated values of exit

7 velocity with analytical calculations are presented in
6.5
Table 4 below. A percentage difference of 10% was
9% Difference
observed between the CFD calculation and
6
analytical calculation at subsonic Mach number
5.5
whereas the difference reduced to only 2% at
ETR

5
supersonic flow condition.
4.5
CFD
4 Table 4 Comparative Analysis of Velocity
Analytical Exit
3.5 Exit Percent
Velocity
3 M# Velocity Difference
(Analytical)
0.4 0.6 0.8 1 1.2 1.4 1.6 (CFD) (m/s) (%)
(m/s)
Mach No
0.6 1296.1 1136.9 14.0 %
Fig. 12. Engine Temperature Ratio
0.8 1299.48 1168.7 11.2 %
From Fig. 12, it is observed that at subsonic speeds,
1.5 1300.43 1276.1 1.9 %
engine temperature ratio slightly decreases from
Mach number 0.6 to Mach number 0.8. This is due
to the fact that the temperature limit is dependent Variation in exit velocity at different flight speeds

1518
I. Arif et al. / JAFM, Vol. 11, No. 6, pp. 1511-1520, 2018.

for CFD and analytical results are shown in Fig. 13. and pressure contours. Nozzle performance was
predicted with high accuracy which was verified
Exit Velocity through the results and plume structure observed. A
1500 difference of 18% in NPR, 12% in EPR, and 9% in
ETR was observed between integrated nozzle and
1300
isolated nozzle which further signifies the
importance of integrating exhaust nozzle in aircraft
Velocity

1100
analysis. The magnitude of thrust loss due to under
900
CFD expansion in the exhaust nozzle cannot be estimated
700 Analytica with analytical equations, which results in slight
l deviation from actual results. Hence, numerical
500
0 0.5 1 1.5 2 analysis results using this methodology proved to be
Mach number quite effective and realistic as it approximates the
Fig. 13. Comparative Analysis of Thrust magnitude of under expansion in actual scenario at
Variation these flight conditions.

From the CFD results, it is observed that the exit

ACKNOWLEDGEMENTS
velocity almost remains constant at all Mach number.
This is due to the fact that the nozzle is under expanded
at all flow conditions and pressure at nozzle exit The authors acknowledge the use of Numerical
remains higher than ambient pressure. However, exit Analysis Lab (NAL) of College of Aeronautical
velocity from analytical calculations increases when Engineering, Risalpur, Pakistan.
Mach number is increased as these calculations are
carried out for perfectly expanded nozzle. REFERENCES

5. CONCLUSION ANSYS (R) (2016) from

http://www.ansys.com/Products/Fluids/ANSYS
We have presented a unique approach of analyzing -Fluent.
jet exhaust nozzle integrated to aircraft and Abbett, M. (1971). Mach disk in underexpanded
propulsion system. The methodology requires exhaust plumes. AIAA Journal 9(3), 512-514.
external flow analysis over the aircraft, internal
flow analysis inside exhaust nozzle and Abdol-Hamid, K. S., K. Uenishi, B. Keith and J. R.
determination of boundary conditions at nozzle Carlson (1993). Commercial turbofan engine
through analysis of propulsion system. A case of exhaust nozzle flow analyzes. Journal of
supersonic aircraft with integrated exhaust nozzle propulsion and power 9(3), 431-436.
was analyzed using the same methodology. Flow Adamson Jr, T. (2012). On the structure of jets from
characteristics of exhaust nozzle were analyzed and highly underexpanded nozzles into still air.
jet plume structure was studied / validated with Journal of the Aerospace sciences, AIAA, Vol
literature. The propulsion system of the aircraft 1.
under study is a dual-duct, dual-rotator turbofan
engine, which has an afterburner shared by ANSYS, I. (2013). CFD User’s Manual. Ansys Inc.
inner/outer channel and a fully adjustable
supersonic nozzle. In order to determine the flow Arif, I., J. Masud, Z. G. Toor, S. Salamat and A.
properties at certain engine components specially Javed (2018). Analytical Modelling and
exhaust nozzle for further numerical analysis of Validation of RD-93 Turbofan Engine at Design
aircraft with integrated propulsion system, an in- Conditions. AIAA Aerospace Sciences Meeting.
house analytical model was developed and Boyce, M. P. (2011). Gas turbine engineering
validated. The analysis was divided into two major handbook, Elsevier, 4th Edition. ISBN:
parts. In the first part, flow characteristics inside 9780123838438.
and downstream of nozzle were studied. Later,
nozzle performance was evaluated, compared and Bulat, M. P. and P. V. Bulat (2013). Comparison of
validated with nozzle performance in isolation. turbulence models in the calculation of
supersonic separated flows. World Applied
It is observed that nozzle flow has no significant Sciences Journal 27(10), 1263-1266.
effect on aircraft major surfaces such as fuselage,
wing upper and lower surfaces, and nose section. Butt, A. H. and A. Arshad (2015). Design and analysis
However, there is a prominent effect of exhaust of a clustered nozzle configuration and
nozzle flow on horizontal stabilizers, vertical tail comparison of its thrust. Student Res. Paper Conf.
and rear fuselage area of the aircraft. The results Chuech, S., M. C. Lai and G. Faeth (1989).
revealed that the nozzle under study was under- Structure of turbulent sonic underexpanded free
expanded at all test conditions. At the nozzle lip, jets. AIAA Journal 27(5), 549-559.
flow deflected through an angle to expand the gas
into free stream pressure. Compression waves were DalBello, T., N. J. Georgiadis, D. A. Yoder and T.
also formed at the interaction of expansion waves G. Keith (2004). Computational study of
with jet boundary. Merging of compression waves axisymmetric off-design nozzle flows. AIAA
resulted into barrel shock and Mach Discs. These paper 530.
phenomenon were observed through pressure plots

1519
I. Arif et al. / JAFM, Vol. 11, No. 6, pp. 1511-1520, 2018.

Dash, S., H. Pergament and R. Wilmoth (1978). Composite Analysis and Characterization of
Prediction of nearfield jet entrainment by an Exhaust Effects on Aerodynamic Behavior of a
interactive mixing/afterburning model. AIAA Supersonic Aircraft. 55th AIAA Aerospace
Paper 78-1189. Sciences Meeting.
Dash, S. M., D. E. Wolf and J. Seiner (1985). Mattingly, J. D. and H. Von Ohain (2006). Elements
Analysis of turbulent underexpanded jets. I- of propulsion: gas turbines and rockets,
Parabolized Navier-Stokes model, SCIPVIS. American Institute of Aeronautics and
AIAA Journal 23(4), 505-514. Astronautics Reston, Va, USA.
Gamble, E., D. Terrell and R. DeFrancesco (2004). Mc Ghee, R. J. (1970). Jet-plume-induced flow
Nozzle selection and design criteria. AIAA separation on a lifting entry body at Mach
Paper 3923, 1-11. numbers from 4.00 to 6.00. Technical Report
NASA-TM-X-1997, L-6980.
Hassan, S., J. Masud and O. Khan (2015). Intake
and Airframe Characterization through Mikhail, A. G., W. L. Hankey and J. S. Shang
Composite CFD. 53rd AIAA Aerospace (1980). Computation of a supersonic flow past
Sciences Meeting, AIAA SciTech Forum, an axisymmetric nozzle boattail with jet
(AIAA 2015-1669). exhaust. AIAA Journal 18(8), 869-875.
Hirsch, C. (2007). Numerical computation of Pandya, S. A., S. M. Murman and M. J. Aftosmis
internal and external flows: The fundamentals (2004). Validation of inlet and exhaust
of computational fluid dynamics, Butterworth- boundary conditions for a Cartesian method.
Heinemann. AIAA Paper 4837, 2004.
Jassim, E. I. (2016). CFD study on particle Rao, G. V. R. (1958). Exhaust nozzle contour for
separation performance by shock inception optimum thrust. Journal of Jet Propulsion
during natural gas flow in supersonic nozzle. 28(6): 377-382.
Progress in Computational Fluid Dynamics,
16(5), 300-312. Robinson, C. and M. High (1974). Exhaust plume
temperature effects on nozzle afterbody
Korst, H., R. White, S. E. Nyberg and J. Agrell performance over the transonic Mach number
(1981). Simulation and Modeling of Jet Plumes range, Arnold Engineering Development Center
in Wind Tunnel Facilities. Journal of Spacecraft Arnold Afb Tn.
and Rockets 18(5), 427-434.
Saddington, A. J., N. Lawson and K. Knowles
Kuntz, M. and F. Menter (2004). Simulation of (2004). An experimental and numerical
fluid-structure interactions in aeronautical investigation of under-expanded turbulent jets.
applications. European Congress on The Aeronautical Journal 108(1081): 145-152.
Computational Methods in Applied Sciences
and Engineering, Eccomas Jyväskylä, Finland. Stitt, L. E. (1990). Exhaust nozzles for propulsion
systems with emphasis on supersonic cruise
Li, G. and E. J. Gutmark (2005). Effect of exhaust aircraft. Technical Report NASA-RP-1235.
nozzle geometry on combustor flow field and
combustion characteristics. Proceedings of the Virdi, P. S., M. S. Khan, N. Pereira, K. Suresh and
combustion institute 30(2), 2893-2901. R. S. D’Silva (2017). Design and Fabrication of
Major Components of Turbojet Engine. Energy
Li, W., R. Campbell, K. Geiselhart, E. Shields, S. and Power 7(5), 130-135.
Nayani and R. Shenoy (2009). Integration of
Engine, Plume, and CFD Analyzes in Woodmansee, M. A., J. Dutton and R. P. Lucht
Conceptual Design of Low-Boom Supersonic (1999). Experimental measurements of
Aircraft. AIAA Paper 1171. pressure, temperature, and density in an
underexpanded sonic jet flowfield. AIAA Fluid
Masud, J., A. Sayani and Z. Toor (2017). Dynamics Conference, 30 th, Norfolk, VA.

1520