Design optimization of the A320 engine inlet cowl

Bruno Santos da Conceição

[email protected]

Instituto Superior Técnico, Universidade de Lisboa, Portugal

November 2016

Abstract
Aircraft operate in environments in which the components are subject to large temperature
and pressure variations. In structures as the engine nacelles, composed by several components and
materials, the presence of wear and corrosion becomes noticeable, due to the their operation in
environments as the one foremost described. Corrective actions must be employed to the components
which present this kind of problems. The acoustic panels of the inlet cowl of the Airbus A320/A321,
present corrosion problems on the aluminium doublers of the joints. In order to develop a corrective
action to the joint of the acoustic panels, the analysis of the mechanical behaviour and forces acting
on the joint must be made. In this work, a methodology involving Computational Fluid Dynamics
(CFD), Finite Element Method (FEM) and Computer Aided Design (CAD) tools is developed in order
to analyse the mechanical behaviour of the acoustic panels joint. The geometry of the inlet cowl and
of its components are in Solidworks. The assessment of the aerodynamic loads acting on the inlet cowl
is made with CFD tools, with STAR CCM+ software. The structural analysis of the members of the
joint of the acoustic panels is made with the use of FEM tools in ANSYS Workbench software. The
steps involved in the analysis are explained and the result are presented.

Keywords: Inlet cowl, acoustic panel, aerodynamic loads, mechanical behaviour, joints

1. Introduction
Commercial aircraft engines are invariably exter-
nal pod-mounted and they are usually attached to
the wing. The engines are enclosed in a structural
housing, called nacelle. The main goal of a nacelle
is to reduce the drag associated to airflow passing
around the engine, minimize engine noise propaga-
tion and to provide a smooth airflow to the engine
[1]. The nacelle configuration varies with the engine
type [2].
TAP, the Portuguese airline has a fleet of 80 air-
craft. From the 80 aircraft fleet, 19 aircraft are Air-
bus A320 and 3 are Airbus A321. At TAP, both the Figure 1: Airbus A320/A321 inlet cowl.
A320 and A321 aircraft share the same engine and
the same inlet cowl. The engine of these aircraft is
the CFM56-5B. Despite the use of the same engine, exposed to large pressures and temperature ranges.
the engine is installed on each aircraft type with Atmospheric temperatures can range from -65◦ C to
different rates. The inlet cowl of the A320/A321 is50◦ C, and some components of the inlet cowl can
composed of three acoustic panels, the upper, the reach even higher temperatures due to the anti-icing
lower and side panels. The acoustic panels form system. In flight operation the inlet is exposed to
a diffuser that provide an uniform airflow to the rain, hail and birds. On ground operation, debris
engine and also have acoustic properties, that par-can be suctioned with the airflow into the inlet. All
tially cancel the engine’s noise [3, 4]. The inlet cowl
these factors contribute to the inlet degradation [5].
of the A320/A321 is presented in Fig.1. When aircraft components start to degrade, cor-
During the aircraft operation, the inlet cowl is rective actions must be employed. The corrective

1
actions depend on many factors, such as the severity 2.1. CAD Modelling - A320/A321 Nacelle
of the problem and its location on the component. Both aircraft models, A320 and A321, share the
The corrective actions can be found in the repair same inlet cowl and engine, the CFM56-5B. In order
manuals provided by the component manufacturer. to correctly simulate the flow behaviour around the
With regard to the acoustic panels, there are sev- engine’s nacelle, a CAD model was created in Solid-
eral repairs that are foreseen by the manufacturer. works environment. Information gathered from Air-
From time to time, aluminium corrosion appears bus website, Airbus A320 manual and measure-
on the doublers of the joint of the acoustic panels. ments performed at TAP facilities, were used to cre-
Fig.2, shows a typical A320 acoustic panel and its ate the CAD model, presented in Fig.3. A very im-
components. A repair for the aft doubler of the portant input for the CAD model was the mapped
panel (please see Fig.2) is foreseen by the manufac- geometry of the lower acoustic panels. With the
turer, however, no repair for the internal doubler is data of the mapped acoustic panel it was possible
present in the repair manual. to determine the inlet curvature and length.
TAP is certified with a Design Organization Ap-
proval (DOA). The certification is granted by EASA
(European Aviation Safety Agency). This certifi-
cate grants to TAP the authorization to design
changes for repairs, for some aircraft areas.
Since the repair of the corroded internal doubler
of the acoustic panel is not foreseen in the repair
manuals of the aircraft, making use of the certifica-
tion to design a repair for this component can be Figure 3: A320/A321 nacelle CAD model.
beneficial for when such problems arise.
Hereupon, the work presented aims to support
2.2. Simulated Flight Conditions
the approval of the repair with the analysis the me-
In order to structurally analyse the joint of the
chanical behaviour of the joint where the doublers
acoustic panel, the critical aerodynamic load must
are applied.
be determined. The takeoff and cruise phases were
considered the most critical flight phases. In the
takeoff phase engines are pushed close to their max-
imum power, in turn, the maximum air mass flow is
suctioned by the engine. In the cruise phase maxi-
mum aircraft speed is attained. The landing phase
(a) wasn’t considered critical due to the low speeds of
the aircraft and low power employed.
Four flight conditions were analysed. Three
for takeoff with different angles of attack, at sea
level. The fourth condition, for cruise at 11000 m
(36000ft). The flight conditions are summarized in
Table 1.

Table 1: Simulated flight conditions.

Flight Altitude Mach Mass Flow
Conditions [m] Number [kg/s]
◦
(b) Takeoff 0 0 0.438 400
Takeoff 9◦ 0 0.438 400
Figure 2: Acoustic panel side section and main Takeoff 16◦ 0 0.438 400
Joint. (a) Acoustic panel side section view;(b) Joint Cruise 11000 0.847 110
main components.

2.3. CFD Approaches

2. Aerodynamic Loads Determination Two different methodologies were implemented in
CFD Methodology order to obtain the aerodynamic load on the engine
In order to determine the aerodynamic loads acting nacelle. As the engine fan blades are constantly ro-
on the engine’s nacelle of the Airbus A320/A321, a tating, real boundary conditions are complex and
CFD methodology was developed. difficult to reproduce. Some simplifications to the

2
real model had to be done. Both implemented 2.4.1 Computational Domain
methodologies intended to reproduce the engine’s
A rectangular prism with typical dimension was
operation.
used to defined the computational domain. All the
The first approach intended to simulate the en- prism dimensions were set as function of the fan di-
gine fan operation. In order to simulate the fan a ameter Df an , starting at a referential with origin
STAR-CCM+ inbuilt fan simulator was used. Some at the center of the fan. Both the computational
simulations were performed in order to understand domain and the boundaries used in the CFD simu-
the fan simulator’s reaction to different inputs. A lations are presented in Fig.5.
porous media was used at the engine core in order to
simulate flow dissipation and to generate a uniform
flow at the engine exit. Due to poor results, unreal-
istic flow behaviour and lack of information about
the engine’s fan, this methodology was dropped.
To simplify the complexity of the real engine
model and to avoid the problems associated to the
fan modelling, an alternative methodology was in-
troduced. This methodology was the adopted one
to perform all CFD calculations. In order to sim-
plify the real engine model the following simplifica-
tions were made: the fan geometry was replaced by
Figure 5: Computational domain and boundaries.
a section with the same diameter of the real fan.
Boundary conditions were applied to the fan sec-
tion to simulate the fan flow suction; No engine core
2.4.2 Generated Mesh
components were modelled. A ”black box” mecha-
nism was used to simulate the engine core; as the The mesh used in the present work was composed
focus of this work is to analyse the effect of the by a trimmed mesh combined with a prismatic
flow at the engine inlet, no exhaust flow simulation layer. The prismatic layer was only generated on
was performed. Instead, a ”Sting” was created at the surfaces of interest, that is, on the surfaces
the end of the nacelle. With the introduction of a where the wall shear stress was to be calculated. A
sting the interaction of both hot and cold jets was few layers of prismatic cells are needed, typically 5
avoided, reducing the simulation complexity. The to 8 to determine the above properties. Ten layers
sting has a length 5 times the fan diameter to mini- were used in the computational simulations. The
mize the wake effect on the area closer to the engine, number of layers was also taken into account in or-
as made in [6]. der to obtain a y+ <200. Due to computational
The final engine configuration for the CFD sim- limitations, a mesh with a maximum number of 1.61
ulations is presented in Fig.4. million cell was created. In order to obtain better
accuracy in the regions of interest a block refine-
ment was created. The mesh and its different levels
of refinement can be seen in Fig.6.

Figure 4: Engine configuration for CFD simula-

tions.

Figure 6: Mesh at the geometry mid-plane.

2.4. The CFD Methodology
The methodology used to simulate the flight condi-
2.4.3 The Problem Physics
tions was based on the second approach previously
presented. All the CFD simulations were made us- In order to correctly represent the engine a tri-
ing an isolated STAR CCM+ engine model. dimensional model was used to account for the en-

3
gine asymmetries and to allow the determination of vergence. Adiabatic wall boundary conditions were
the pressure and wall shear stress on the engine’s used on the rigid sections. A pressure outlet BC
nacelle. All simulations were performed in steady was used at the fan face with a defined pressure
state. difference relative to the reference pressure.
To define both air dynamic viscosity and thermal
conductivity Sutherland’s law was used. Since all 2.5.1 The Fan BC Calculation
simulations are performed for Mach numbers above
0.3, compressible effects had to be taken into ac- In order to determine the desired ∆p, to define
count. The air density was modelled with the ideal the fan boundary condition, the isentropic relations
gas model. were used together with the mass flow function.
RANS equations were used along with SST K- The information available to perform these calcula-
Omega turbulence model to perform the simula- tions is: the mass flow (ṁ), the atmospheric static
tions. SST K-Omega model utilization is recom- pressure (patm ), the atmospheric static temperature
mended by the software for transonic simulations (Tatm ), the free stream velocity (V∞ ) and the free
and simulations in which recirculation regions are stream speed of sound (a∞ ). The free stream Mach
present. number can be simply obtain with the expression
The Coupled Flow Model was used. This model represented in Eq.(1).
solves simultaneously the conservation for mass,
V∞
momentum and energy. The model uses an implicit M∞ = (1)
spatial integration along with the multigrid method. a∞
A second order upwind discretization scheme was Combining this information with the isentropic re-
used. lations and mass flow function (Eq.(2)), a starting
2.5. Boundary Conditions value for the ∆p was defined. From this point the
Boundary conditions (BC) have a very important ∆p was corrected until the desired mass flow was
role in CFD simulations as they have a direct im- achieved in the simulation. The isentropic relations
pact on the flow behaviour and on its interactions can be use in this approximation since no shock
with the components involved in the simulation. waves are expected to occur in the inlet duct.
Boundary conditions were divided into domain BC
and engine BC. All BC are assign to the boundaries (γ+1)
√ ! 2(γ−1)
presented in Fig.5. ṁf an p0 γ 1
For the domain boundary conditions, three types =√ Mf an (2)
Af an RT0 1 + γ−1 2
2 Mf an
of BC were used: free stream, pressure outlet and
symmetry. Free stream BC were used to simulate In Eq.(2), ṁf an , Af an and Mf an , are respectively
the incoming airflow into the domain, simulating the mass flow at the fan, the area of the fan and
the aircraft velocity and angle of attack. The free Mach number at the fan; p0 and T0 , are the stag-
stream BC was assigned to CV INLET boundary, nation pressure and temperature, respectively. R
SYMMETRY 2 and SYMMETRY 4. The last two, is the universal gas constant and γ is the ratio of
only for angles of attack other than zero. A pres- specific heats for a perfect gas [4].
sure outlet BC was assigned to the CV OUTLET
boundary. The symmetry plane BC was used in all 2.6. Methodology Suitability
SYMMETRY boundaries, except for the mentioned Since no experimental data of the A320/321 was
case explained above, where free stream conditions available to figure out the suitability of the compu-
are used. tational result, a different engine with experimental
When selecting the BC for the engine model spe- data was analysed. The NAL-AERO-02-01 T.P.S.
cial care must be taken, since a complex model is (Turbine Powered Simulation) wind tunnel experi-
intended to be simulated. In [7] simulations of pow- mental model was used. This model represents an
ered engines are performed and different types of axissymmetric turbofan engine from the Japanese
boundary conditions are presented. Three types of Aerospace Technology Research Institute [9]. With
boundary conditions can be specified at the fan. the available information, it was possible to cre-
The mass flow, the pressure or the velocity can ate a three-dimensional CAD model of the T.P.S..
be defined for a specific engine working condition. In order to analyse the suitability of the computa-
In the present work a pressure boundary condition tional model, all simulation parameters used in the
was used i.e. a pressure variation ∆p, relative to A320/A321 model were maintained for the T.P.S.
the reference pressure value, is applied to the fan analysis. Despite the difference in sizes between
face. In [8], both mass flow and pressure boundary models, the proportions in the computational do-
conditions are analysed. Results showed that pres- main were maintained, since the computational do-
sure based boundary conditions have better con- main is a function of the fan diameter. The mesh

4
cells density was also maintained between simula- also converged. In all flight conditions the residu-
tions, please see Fig.7. als are below 10−4 and the mass flow converged to
A simulation to recreate the experimental data the desired values, approximately 400 kg/s for the
was performed, simulating the cruise condition. In takeoff conditions and approximately 110 kg/s for
Fig.8, a graphic with the experimental and com- cruise.
putational pressure coefficient (Cp), is presented.
From analysis of the velocity field it could be ver-
In Fig.8 the good agreement between the experi-
ified that flow is uniform for the lower takeoff angle
mental and computation result can be seen. Errors
conditions. For the takeoff 16◦ , flow separation is
bellow 3% can be found in the region of interest
visible, accompanied with flow recirculation, that
(Inlet cowl). Since good results were obtained with
can affect the engine performance, please see Fig.9.
this computational setting, it is expected to obtain
At cruise condition, the flow is uniform in the inlet
result for the A320/A321 as suitable as the ones
region, but the exterior of the nacelle is affected by
obtained with the T.P.S. model.
shock waves.
The analysis of the Mach number at the fan, for
the takeoff conditions, shows that the Mach number
is in the typical range for turbofan engines, 0.4-0.7
[4]. For the takeoff 16◦ condition, the influence of
the flow separation is visible on the Mach number
distribution, please see Fig.10. The result for the
cruise condition are slightly below the typical val-
ues, maybe due to the mass flow approximations.
Figure 7: T.P.S. model and mesh.
The analysis of the pressure distribution shows
that the maximum pressure has about the same
magnitude for the three takeoff conditions. The
minimum pressure increases in magnitude as the an-
gle of attack increases, due to a further acceleration
around the lip leading edge. For the cruise condi-
tion large pressure variations appear in the shock
wave regions. The pressure result are presented in
Table 2.
The analysis of the wall shear stress (WSS) shows
that its magnitude is higher in regions of higher ve-
locity, the inverse is also verified. Comparing the
WWS magnitude to the pressure magnitude it can
be seen that the WSS corresponds to about 1% of
the pressure magnitude. The WSS results are pre-
sented in Table 3.

Figure 8: T.P.S. Cp distribution for computational

and experimental results.

2.7. CFD Results and Discussion

Four flight conditions were simulated in order to
determine the critical flight conditions to which the
maximum flight aerodynamic load is presented.
All simulations ran until the stopping criteria was
attained. The control parameters in the present
work were the Residuals, the Mass Flow and the y + .
The y + was used to control the mesh quality. Both
the residuals and mass flow were used to control
the solution convergence. In the present work the
solution was considered to be converged when all
the residual had values bellow 10−4 . Although, the Figure 9: Velocity field for takeoff 16◦ .
mass flow variation was analysed in parallel with the
residual to ensure that this important quantity had

5
 pressure and wall shear stress were imported to
ANSYS. The imported data correspond to the four
fight conditions, takeoff 0, 9, 16 and cruise. Atten-
tion was made to verify if the imported coordinate
system was coincident with the one found in AN-
SYS, and also to verify if the same units are used
in both systems.

3.1.2 Geometry and Geometry Importation

In order to generate the structural analysis, a CAD
model was created and imported into the ANSYS
environment. All the geometries used in the analy-
sis were created using Solidworks. Since the objec-
tive of the present work is to analyse the joint of
Figure 10: Fan Mach number distribution for take- the acoustic panel, special attention was made to
off 16◦ . correctly simulate the components in that region.
In Fig.11 ,the external region of the inlet cowl are
Table 2: Pressure variation for the simulated flight presented. The three sections corresponding to the
conditions. acoustic panels are visible with three different col-
Takeoff Takeoff Takeoff Cruise ors. In Fig.12 the internal view of the inlet cowl
Property 0◦ 9◦ 16◦ with the joint components is presented. In addition
Maximum 14815 14998 14798 13541 to the components represented, there is an attach-
Pressure [Pa] [Pa] [Pa] [Pa] ment ring that creates the connection between the
Minimum -11757 -38840 -49922 -14071 inlet cowl joint and the engine.
Pressure [Pa] [Pa] [Pa] [Pa]
3.1.3 Material and Mechanical Properties
Several materials are used in the construction of the
Table 3: Wall Shear Stress variation for the simu- inlet cowl and acoustic panel. All material proper-
lated flight conditions ties were introduced in the software. The proper-
Takeoff Takeoff Takeoff Cruise ties were obtained form CES EduPack 2015 Soft-
Property 0◦ 9◦ 16◦ ware, except those of the honeycomb core, which
WSS [Pa] 94.41 153.14 166.69 66.53 were calculated based on the geometrical parame-
WSS [Pa] 0.15 0.22 0.13 0.16 ters [10]. The Honeycomb was modelled as an or-
thotropic material.
3. Analysis of the Joints of the Acoustic
Panel - FEM Methodology
The methodology used to determine the mechan-
ical behaviour of the joint of the acoustic panel and
the fasteners of the joint is presented. The different
approaches to the problem are explained. Results
from the CFD methodology chapter were used, in
order to correctly simulate the aerodynamic loading
on the structure.
3.1. Preparing the FEM Simulations
The joint of the acoustic panel intended to be anal-
ysed is the joint presented in Fig.1. In order to un-
derstand the effect of the aerodynamic load on the Figure 11: Inlet cowl isometric view.
acoustic panel’s joint and fasteners, several FEM
simulations were performed.
3.1.4 Contact Between Components
3.1.1 Aerodynamic Load
Whenever two or more components surfaces touch
All aerodynamic load obtained with Star CCM+ each other, these components are said to be in con-
software, were imported to ANSYS Workbench. In tact. In the simulations of the present work the two
order to preform the structural analysis both the types of contact used were Bonded and Frictionless.

6
 components. Since only reaction forces and mo-
ments are to be obtained, the use of Bonded contact
doesn’t compromises the results.
All simulations were performed with the same ge-
ometry and mesh in order to have a base of com-
parison between simulations. The mesh used was of
the Hex Dominant Type, with 305136 elements.
For each simulation, the respective aerodynamic
loads, were imported into the model.
To ensure that the geometry was held in place,
Figure 12: Inlet cowl internal view. a constraint was imposed at the Attachment Ring.
The constrain was defined as a Fixed Support.
A simulation with the structural weight was also
When using bonded contact it is assumed that com- performed in order to understand its influence.
ponents in contact are glued together. No sliding or
separation is allowed between components. Bonded
3.2.2 Results and Discussion
contact was used in the acoustic panel components.
The frictionless contact allows components to slide Reaction forces and moments corresponding to the
and to separate from each other. Unlike the bonded analysed loading conditions are presented in Table
contact type, frictionless contact has nonlinear for-4. As the angle of attack increases, both forces and
mulation that results in longer solutions time. This moments reaction increase. It is possible to con-
type of contact is used to simulate contact between clude that the loading corresponding to the Take-
the acoustic panel doublers and the attachments off 16 flight condition corresponds to the critical
ring. It is also used to simulate the contact betweenloading condition. From the analysis of the results
the fasteners shank and the Joint’s holes. corresponding to the structural weight condition, it
is visible that all components of the reaction force
3.1.5 Defining the Mesh and moment reaction, have the opposite sign when
compared to the other analysis made. This means
ANSYS Workbench presents several meshing meth- that the addition of the structural weight reduces
ods to mesh solids components. Depending on the the global loading.
type of method selected, different building blocks
will be used in the construction of the mesh. In the
present work, the hex-dominant method was used. Table 4: Moment and and reaction forces results.
The hex-dominant method creates a mainly hexa- Force Moment
hedral element mesh. When the hexahedral mesh Reaction [N] Reaction [N.m]
is compared to the tetrahedral it can be verified Loading
that the hexahedral mesh requires much less ele- Condition Total Total
ments than the tetrahedral one to obtain the same Takeoff 0 947.53 2128
solution accuracy. Quadratic elements were used in Takeoff 9 21154 14476
the analysis. As the computational resources are Takeoff 16 34695 20236
limited, a global mesh size was defined and local Cruise 7371.7 6168.4
refinements were used to increase the mesh density Structural 1380 1490.4
in locals of interest. Weight
3.2. Determination of the Critical Load Con-
dition
In order to determine the critical loading condition, 3.3. Approach to Analyse the Acoustic Panel
the pressure and wall shear stresses, determined in Joint
STAR CCM+, were applied to the structures in In order to obtain enough resolution to capture the
ANSYS environment. An analysis of the Force Re- desired properties, both for the fastener of the joint
action and Moment Reaction is made to determine and the interface between the internal honeycomb
the critical loading conditions. and internal doubler, some simplifications had to be
done to the model.
3.2.1 Model Considerations
3.3.1 Model Simplifications
The complete model of the inlet cowl was used in
the analysis. In order to reduce the simulation com- In order to simplify the model, the equivalent stress
plexity and the computational effort, no fasteners of the complete bonded model was analysed. The
were simulated and bonded contact was used in all loading conditions used in the simulations were the

7
critical ones, Takeoff 16. From these analysis it was refined. For the convergence analysis only the fas-
possible to identify the region with higher stress lev- teners were refined.
els. This region corresponds to the interface region
between the Lower and Side Acoustic Panels. A 3.4.2 Convergence Analysis
section of the model comprising that region and a
fastener pattern on each side of the interface was For the convergence analysis the equivalent stress
created. Please see Fig.13. (von Mises) of the fasteners was analysed. In order
As the inlet cowl was sectioned, new boundary to always analyse the convergence of the fastener at
conditions had to be implemented in order to sim- the same point, a fixed referential was created for
ulate the complete structure behaviour. A spring each fastener, at the maximum stress node. The
model was used to simulate the rigidity of the com- same referential was used at each refinement.
plete model. Please see Fig.14. As it can be seen from Table 5, there still exist
a fluctuation on the stress values. Although, the
error between consecutive refinements is very small.
Refinement 5 was used to obtain the result for the
bolts analysis.

Table 5: Bolts convergence analysis

N◦ Elements von Mises Error
Refinement [Pa] [Pa] %
1 132150 1.206E7 1.074
2 134838 1.219E7 6.084
3 143908 1.298E7 0.038
4 208108 1.299E7 0.964
Figure 13: Inlet cowl most stressed region. 5 274881 1.287E7 -

3.4.3 Result and Discussion of the Fasten-

ers Analysis
Analysis of the Tension and Shear Stress
The results obtained from the fasteners analysis
allowed to compare the maximum tension stress
σmax , and the maximum shear stress τxymax , of
each Hi-Lok, with the material yield strength limits.
From the analysis of the results a minimum safety
factor of 47 and 246 were obtained for tension and
shear respectively.

Figure 14: Spring model. Pretension Results Hi-Lok fastener with two
different diameters were used in the main joint. The
preload analysis was made to the fasteners, of each
3.4. Analysis of the fasteners of the Joint diameter, which presented highest normal stresses.
In the present section the analysis of the fasteners The normal stress of each of these bolts, was con-
of the joint is made in order to understand the me- verted into tension force in order to make a com-
chanical behaviour of the fasteners in the Joint. parison between the resultant normal force and the
theoretical preload. The preload Fi , was calculated
with the Eq.(3) [11].
3.4.1 Components Contact and Mesh Re-
finement T
Fi = (3)
The aerodynamic loads tend to bend the inlet, gen- Kd
erating tension on the Hi-Loks. Bonded contact was T corresponds to the tightening torque applied to
used to connect the Hi-Lok head to the doubler and the fastener (obtained from the fastener data sheet)
the collar to the attachment ring. Frictionless con- and d correspond to the major diameter of the fas-
tact was used between the attachment ring and the tener. A torque coefficient, K, of 0.16 was used.
doublers and between the Hi-Lok shank and holes. The results are presented in Table 6. As the maxi-
Due to computational limitation bolts were locally mum tension force on the fasteners is only about 5%

8
of the theoretical preload, it is possible to conclude shear stress are made. Formulation presented in
that the joint is safe against flange separation. standards ASTM C297-94 [12] and C273-00ε1 [13]
were used. The forces used in the analysis are pre-
sented in Table 7.
Table 6: Comparison between the theoretical
preload and resultant tensile force on the fasteners.
Hi-Lok Theoretical Tension % of Table 7: Joints forces.
Diameter Preload[N] Force[N] Preload Force [N] Force [N]
Component
Lower Joint Side Joint
4.76 [mm] 5148.44 227 4.41
(Hi-Lok65 4) X 160.15 -150.6
6.35 [mm] 7784.03 375 4.81 Y 912.97 571.29
(Hi-Lok32 1) Z 450.22 365.37
Total 1030.5 694.66

3.5. Analysis of the Interface between inter-

nal honeycomb core and internal doubler
In the present section the forces acting on the The Flatwise Tensile Strength For the flat-
bonded interface between the internal honeycomb wise tensile strength analysis the maximum Y com-
core and the internal doubler are determined. ponent of the force was used, corresponding to
the lower joint. Both joints have the same cross-
sectional areas, A = 9946mm2 . The flatwise tensile
3.5.1 Components Contact and Mesh Re-
strength was calculated with Eq.(4).
finement
P
Bonded contact was used in all the acoustic panel σ= = 92kP a (4)
components. The contact between the attachment A
ring and the doubler was frictionless contact. Fric- In Eq.(4), σ represents the flatwise tensile
tionless contact was also used between the fasteners strength and P the load. Comparing the obtained
shank and the holes of the fasteners. result with the experimental test results of the
In order to capture the forces acting between the LOCTITE EA 9658 AERO[14], it can be verified
doubler and the honeycomb, a fixed joint connec- that the obtained tension are equivalent to 4.38%
tion was used between the components. Two joints of the minimum experimental strength. A force
are analysed, one joint correspond to the Lower 22.82 times the force obtained in the simulation
Panel section and the other to the Side Panel sec- would be needed to attain the minimum experimen-
tion. The two joints and respective referential are tal strength limit.
presented in Fig.15. In this analysis the refinement
was focus on the component of the joint. The Shear Stress The shear stress resulting
A convergence analysis was made and the most from the aerodynamic load on the joint corresponds
refined results were used for the analysis. to the X and Z components of the joint force. Since
the Z component of the lower joint is the largest
force component, this force was used to calculate
the shear stress in the joint. The calculation is made
as follows, Eq.(5).

P
τ= = 45kP a (5)
Lb
In Eq.(5), τ represents the shear stress, P the
load, L the length of specimen, and b the width of
specimen.
As in the previous analysis, the results were com-
pared with experimental data. The simulation re-
Figure 15: Joint’s interfaces and referential. vealed that results correspond to about 0.4% of the
limit strength.
4. Conclusions
3.5.2 Results and Discussion of the Joints
The overall results of this work allowed to reach the
Analysis
following conclusions:
In order to determine the Joints properties, calcu- • Two CAD models of the A320/A321 inlet cowl
lations to determine the flat tensile strength and were created for the analysis performed using CFD

9
and FEM. The correct modelling on the nacelle can [2] Shmyrov, V.F., other (2010). Airplane power
be very challenging. plants systms designing. Kharkiv National
• A methodology using Computational Fluid Dy- Aerospace University.
namics to determine the aerodynamic loads acting
on the inlet cowl on of an aircraft’s engine was de- [3] Airbus. Component maintenance manual of the
veloped. Airbus A320.
• The aerodynamics loads, pressure and wall [4] Philip, H., & Carl, P. R. (1992). Mechanics and
shear stress, were obtained for four flight conditions. thermodynamics of propulsion. Addison Wesly.
• The suitability of the methodology employed to
determine the aerodynamic loads was verified. [5] Watson Jr, H. A. (1970). Structural and En-
• A methodology to analyse the joints of the vironmental Studies of Acoustical DuctLining
acoustic panels using FEM was developed. Materials. The Journal of the Acoustical Society
• The critical loading condition was determined. of America, 48(3C), 795-814.
• It was possible to conclude that the joint and [6] Hall, Z., Ahuja, V., Hartfield, R., Shelton, A., &
the fasteners are subjected to combined loads. Ahmed, A. (2009). Optimization of a Turbofan
• Stresses in the fasteners are far from their yield Inlet Duct using a Genetic Algorithm and CFD.
limits, both for tension and shear. This indicates AIAA Paper, 3775, 22-25.
that the fasteners do not present risks for the joint
safety. [7] Zhaoguang, Tan and Yinchun, Chen and Wei,
• Loads on the fasteners far from the preload Huang (2012). Research for jet flow influences to
ones. wing-mounted civil aircraft aerodynamic char-
• Forces acting on the interface between the inter- acteristics. International Congress of the Aero-
nal honeycomb core and the internal doubler were nautical Sciences.
determined.
[8] LUCIOLI, J. (2011). CFD analysis of a nacelle
• The analysis of the flatwise tensile strength and
at high angle of incidence.
shear stresses allowed to conclude that L0CTITE
EA 9658 AERO can be used to connected both the [9] HIROSE, Naoki, et al. Transonic 3-D Euler
internal honeycomb core and internal doubler, with Analysis of Flows around Fanjet Engine and
a large safety margin. TPS (Turbine Powered Simulator)-Comparison
• Finally, the development of this project allowed with Wind Tunnel Experiment, Evaluation of
to create a methodology that allows the analysis of TPS Testing Method and 3-D Flow. 1989.
aerodynamic and structural parameters of the inlet
cowl of an aircraft engine. This methodology could [10] Gibson, L. J., & Ashby, M. F. (1999). Cellu-
be used and adapted to analyse other models in- lar solids: structure and properties. Cambridge
volving aerodynamic loads and structural analysis. university press.

5. Future Work [11] Budynas, R. G., & Nisbett, J. K.(2011).

In the future it would be interesting to perform a Shigley’s mechanical engineering design. Mc
similar analysis with more computational power. It Graw Hill.
would be interesting to analyse larger sections of the
[12] ASTM International (1999). Standard Test
inlet, or even the complete model. An experimental
Method for Flatwise Tensile Strength of Sand-
analysis of the joint of the acoustic panel could be
wich Constructions. Designation: C 297 94.
done in order to compare experimental and numer-
ical results. Finally, the developed model could be [13] ASTM International (2000). Standard Test
used to approach different problem. Method for Shear Properties of Sandwich Core
Materials. Designation: C273-00.
Acknowledgements
I would like to express my gratitude to Prof. Luis [14] LOCTITE, LOCTITE EA 9658
Reis and Vitor Anes for their support and recom- AERO Epoxy Film Adhesive. From:
mendations through the development of this work. https://tds.us.henkel.com/NA/UT/HNAUTTDS
I would like to thank TAP Engines Engineering de- .nsf/web/22C510C43FA6A5B58525715C001BD519/
partment for their availability and for the opportu- Accessed: 2016-09-30.
nity of having direct contact with the components
at their facilities.
References
[1] Kundu, A. K. (2010). Aircraft design (Vol. 27).
Cambridge University Press.

10