International Journal of Crashworthiness

ISSN: 1358-8265 (Print) 1754-2111 (Online) Journal homepage: http://www.tandfonline.com/loi/tcrs20

Bird strike assessment for a composite wing flap

S. Orlando, F. Marulo, M. Guida & F. Timbrato

To cite this article: S. Orlando, F. Marulo, M. Guida & F. Timbrato (2017): Bird strike
assessment for a composite wing flap, International Journal of Crashworthiness, DOI:
10.1080/13588265.2017.1342521

To link to this article: http://dx.doi.org/10.1080/13588265.2017.1342521

Published online: 03 Jul 2017.

Submit your article to this journal

Article views: 18

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=tcrs20

Download by: [University of Florida] Date: 21 July 2017, At: 18:43

INTERNATIONAL JOURNAL OF CRASHWORTHINESS, 2017
https://doi.org/10.1080/13588265.2017.1342521

Bird strike assessment for a composite wing flap

S. Orlandoa, F. Marulob, M. Guida b
and F. Timbratoa
a
SALVER, Brindisi, Italy; bDepartment of Industrial Engineering – Aerospace Branch, University of Naples, Napoli, Italy

ABSTRACT ARTICLE HISTORY

The paper presents the design, analysis and test of the CFRP flap to demonstrate its compliance Received 15 June 2016
with the regulation and safety standards toward the bird strike requirement. The purpose of this Accepted 8 June 2017
work is to define a methodology based on the finite element analysis to predict the damage for KEYWORDS
high-velocity impact of a full-scale aircraft component using the SPH bird impact model, 4 lb, at a CFRP pre-preg composite
velocity of 194 kts (100 m/s). material; aircraft certification;
In order to validate the numerical models, materials were initially characterised through an flap; wing track; bird strike;
extensive campaign of tests aimed to identify their static and dynamic properties. The bird strike explicit analysis
simulation was carried out, first on a rigid plate to correlate the experimental test performed by
Barber, and then the results were extended for a selected composite panel to investigate the
particle’s distance for the SPH approach and pressure path distributed on the target, and finally
these validated results were applied on the flap structure.
The numerical simulations have been used from the initial designing phase as well as for the
certification according to the aeronautical requirements, to be able to design a flap structure and
the relative fixture reproducing the real test condition. Finally, the flap has been manufactured,
tested, certified and installed on the large commercial aircraft.

1. Introduction In any case, it is not expected that the bird strike qual-
ification be obtained with an experimental ‘trial-and-
The bird strike is a collision with birds, and it is a phe-
error’ approach. The result would be extremely expen-
nomenon that involves the impact of a bird and an air-
sive and time-consuming.
craft; the collision, apparently inoffensive to the aircraft,
Consequently, the numerical simulation approach has
can produce considerable damages and consequences,
become a fundamental design tool to satisfy both time
not only to the structure, but also to the electronic sys-
and cost. Predictive tools that optimise thickness and
tems, mechanical systems, visual systems and guidance
size, layup and fitting of aeronautical components prone
and control systems. So, it can even affect the flight
to this kind of accident are not common, because it is a
safety and the proper functioning of the whole system.
highly transient problem, depending on many
The American and European aviation agencies
unknowns and parameters to take into account. This
require that an airplane must be designed to safely land
paper deals with the numerical simulations of a full-scale
after being impacted by a medium-size bird [9]. These
flap of an aircraft impacted by a 4 lb bird. For the leading
rules require that the aircraft design includes ‘bird-proof’
edges, the certification criteria require that even in case
qualification, especially for those parts and structural
of penetration of the leading-edge skin, no critical dam-
components at greater risk of impact such as the wind-
age may be introduced to the front spar elements or the
screen/canopy, cockpit windows, engines, and the lead-
wing box, assuring a continued safe flight and landing
ing edges, to be airworthy and flight certified. After the
after impact. This has to be proven for 4 lb (1.8 kg) birds
prototype has successfully showed compliance with the
impacting the wing (FAR 25.571 [8]).
safety requirements, a further test is carried out with a
During the bird strike test, the projectile behaves as a
high-speed impact to determine the ultimate limit of the
fluid, and the model of the bird is obtained by a highly
structure. As is easily understood, the achievement of
deformable projectile adopting the Lagrangian, Eulerian
safety standards together with the basic need of a light
approach or their combination in order to catch the
structure might require a large number of tests on the
deformation, as reported in [10]. The alternative is the
structure with strong undesired feedback on the devel-
combination between a meshless method as SPH
opment time and the overall cost of the aircraft.

CONTACT M. Guida [email protected]

© 2017 Informa UK Limited, trading as Taylor & Francis Group
2 S. ORLANDO ET AL.

technique, adopted to describe the projectile, and the strike test. This simulation has allowed a strong reduc-
finite element method to characterise the structure defor- tion of the number of experimental shots, ending with
mation, as reported in [19]. This paper focuses on SPH only two of them, and allowing the possibility of a
technique for the projectile, paying attention on the defi- unique test specimen, therefore saving time and cost.
nition of the governing equations of the impact problem, For the two shots, strain gauge and force reaction
and analysing the entire impact process. This analysis is time-history are compared with the numerical predic-
enhanced by validating the bird numerical model with tions. The paper ends with a dedicated comparison
the experimental data presented in the literature [7]. Bar- between test results and the relevant analysis carried out.
ber performed an extensive campaign of tests on the rigid
flat disk, where the focus was dedicated to the projectile
and real fluid behaviour during the impact. 2. Description of the test article
Some emphasis has been devoted to the materials, Large commercial aircraft install a single slotted-fowler
which are deputed to absorb high level of kinetic energy. configuration for the outboard flap, which has a length
These materials, as in the case of bullet impact, can dissi- of 21 feet (6.4 m) and a weight of 223.4 lb (101.3 kg).
pate the kinetic energy involving a small portion of the The composite flap offers different layups and thick-
target. Composite materials are widely used in the air- nesses distributed in several areas of its surface, and it is
craft industry, both for primary and secondary struc- engineered using autoclave curing instead of pre-preg.
tures; the materials devoted to absorb high-level energy, The flap structure consists of the following primary sub-
in particular the impacts due to rain, birds and hail- components (Figure 1):
stones, are differently dealt in correspondence of the
involved structural element.
Upper and lower composite skins;
Different solutions are discussed throughout the scien-
Front and rear composite spars;
tific literature in which the structures are made out of: (1)
Upper and lower stringers;
laminated composite materials [16,17]; (2) fibre metal
Composite ribs;
laminates (FML) [12,24]; (3) a tensor skin laminate [18];
Metal ribs, inboard and outboard side.
(4) a sandwich construction with an FML outside skin,
an aluminium honeycomb core, and an aluminium inside The outboard flap surface consists of upper and lower
skin [13], or other sandwich structures [14]. In these skins with integrally co-cured spars and stringers. For
materials, impacts are known to induce damage that will this reason, the design of the composite upper skin
significantly affect the structure’s load-carrying capability. includes the front and rear spars and two integral T-sec-
The challenge of the simulation is to predict the damage tion stringers. Similarly, the design of the composite
induced by the bird strike and in particular to predict the lower skin includes the co-cured six integral T-section
size and location of ply debonding [1]. stringers. This approach helps to minimise the structural
Before introducing the impact simulations on the weight, number of parts and assembly time.
entire flap, a dedicated model calibration, which fully The outboard flap is supported at two flap track
reflects ASTM procedure of tensile and compression tests, beams, which provide connection to the wing torque
has been prepared and simulation results have been com- box. Figure 2 shows two different attachments; the first
pared with typical experimental tests used for a Carbon one, labelled ‘track 3’, shows the linkage of the forward
reinforced fiber plastic (CFRP) material. Furthermore, the drive rod, aft linkage and roller carriage.
dimensioning of flap body and flap-to-flap track attach- The attachment to the outer track of the flap, ‘track 4’,
ment fittings has represented a crucial aspect for the is achieved by means of a forward drive rod, aft and
selection of most critical impact condition. By using an swing linkage.
iterative approach, all requested failure parameters have
been properly tuned. Then, the correct model of the pro-
jectile impact has been introduced to study the interac-
tion between the bird (modelled as equivalent gel) and
the target, including the equations of state and contact.
The flap model is a single slotted flap fixed to the
main wing structure with two tracks, respectively, staring
from the wing root.
The results of current work have allowed to establish
robust criteria and to define the most critical impact
points needed to be investigated by means of a bird Figure 1. Outboard flap overview.
 INTERNATIONAL JOURNAL OF CRASHWORTHINESS 3

(Belytschko–Lin–Tsay shell formulation). In order to

increase the level of the accuracy of the model, where
the impact occurs, a very detailed mesh of the interested
area, including the lower skin and the metallic attach-
ment, joined together by an appropriate contact formu-
lation, has been prepared.
Regularly, the skins have been meshed using the mid-
plane as reference and the element normal point
inwards. The web ribs, in CFRP material, have been
meshed using the mid-surface as reference, while the cap
rib has been modelled with a beam element with appro-
priate area to simulate the inertia of the upper and lower
caps. The front and rear spar have been modelled using
the same approach of the composite rib. Finally, the rod
elements have been used to properly model upper caps
of spar, rib cap or feet, while on lower skin, where the
Figure 2. Outboard flap tracks. detailed mesh is applied, the stringers are modelled by
composite shell elements.
The movement of the flaps is achieved through an In the numerical model, two different materials were
actuator drive unit which drives the flaps along their introduced to define the metallic and composite parts.
inner and outer flap tracks from a retracted position at 0 The Ls-Dyna MAT54 material card, which is designed
degrees, through deployment angles of 10, 15, 20 and 37 specifically to handle orthotropic materials such as uni-
degrees, as reported in Figure 3. The test specimen con- directional tape composite laminates, was used as repre-
sists of the flap body and the flap-to-flap track interface sentative of composite material. The MAT24 material
components only, whereas, the test rig includes dummy card, which defines an isotropic, piecewise linear elastic-
tracks and flap-to-flap track interfaces. plastic with failure material model, was used for the
metallic components and in particular for the trailing
edge involved in bird impact event [22].
3. FE model description All the input parameters for both cards are properly
defined in the next sections, where referenced.
3.1. Flap structure Finally, the contact condition was introduced to
The analytical model of the full-scale flap (including define the contact between lower skin, ribs and spars
tracks attachment) has been accomplished by a finite ele- (see Figure 4). The tied contact option with a gap was
ment analysis, using the nonlinear explicit solver LS-
Dyna. This software was considered appropriate to
investigate the so complex scenario of the bird strike,
where the bird and structure models were defined by a
meshless and Lagrangian approach, respectively.
The numerical Lagrangian model has been built using
a 2D-element mesh for all parts employing specific for-
mulation for composite parts (Hughes–Liu shell formu-
lation) and a different one for the metallic parts Figure 4. Contact definition between lower skin and ribs.

Figure 3. Outboard flap kinematic.

4 S. ORLANDO ET AL.

introduced between the slave and master nodes, so any Table 1. Material properties for epoxy resin with fabric 2 £ 2
slave nodes that meet the criteria to be tied will be twill.
Density 1.425E¡4
moved to the master surface and a physical gap between
Axial Young’s modulus (psi) 8.6E6
the tied surfaces is to be retained; during this formula- Transverse Young’s modulus 8.3E6
tion, each slave nodes rotations and translation are Minor Poisson’s ratio 0.053
Shear modulus (psi) 5.6E5
transferred to the master nodes based on relative master Maximum strain for matrix straining in tension or 0.023
and node distance [22]. compression (psi)
Maximum strain for fibre tension (in./in.) 0.013
Maximum strain for fibre compression (in./in.) 0.0013
Longitudinal compressive strength (psi) 9.9E4
3.2. Constraint and joints Transverse compressive strength, y-axis (psi) 9.6E4
Longitudinal tensile strength (psi) 1.1E5
Flap body is free to move along two track guides for each Transverse tensile strength, b-axis (psi) 1.006E5
Shear strength (psi) 1.27E4
wing during the landing and take-off manoeuvre. Dur-
ing the sliding along the guides, the flap moves and
rotates simultaneously. In order to simulate the actual Table 2. Layup of the laminates with different ply number; thick-
ness of the single ply is 0.013 in.
constraint, different rigid joints have been used to ensure
Ply number Stacking sequence Thickness
the real attachments on tracks. The simulation of the
5 [45/0/0/0/45] 0.0628
actual constraints is defined by a translational joint to 8 [45/0/45/0]s 0.1005
allow any translation of flap along track main axis, while 10 [45/0/45/0/45]s 0.1256
16 [45/0/45/0/0/45/0/45]s 0.2010
the rotational joint was adopted to simulate the relative
rotation of the link along the carriage. Then, the intro-
duction of a spherical joint, between swing link and model. As required, a big number of coupons have been
roller carriage as shown in Figure 5, does not allow the tested to measure the material properties by static tensile
moment to be transferred between track and roller and compression test, notched and un-notched, and so
attachment. Finally, the node in correspondence of the on. Test coupon (tension, compression including open
track attachment has been fixed with reference to both hole) has been modelled to simulated standard test pro-
translational and rotational motion. cedure as required by ASTM [3–5], in order to tune the
material properties for the specific behaviour.
The material card adopted for the model is MAT54 in
3.3. Composite material Ls-Dyna, which is a progressive failure model specifically
designed to handle orthotropic materials such as unidi-
Flap is designed using a qualified material as epoxy resin rectional tape composite laminates, and implementing
with carbon fabric 2 £ 2 Twill. the Chang–Chang matrix failure criterion.
In order to calibrate the FE model, specific material An iterative model calibration process has defined the
testing is required for identifying input parameters for material characteristics as listed in Table 1.
highly nonlinear transient impact loads, and several test The composite material has been modelled as multi-
procedures have been used to prepare the dedicated FE layer composite panel using shell elements with an inte-
gration point for each lamina; the number of integration
points through the section thickness has been set equal
to the total number of plies +1. When failure has
occurred in all the composite layers (through-thickness
integration points), the element is deleted. The flap body
consists of several stacking sequences with a distribution
of ply orientations through the laminate thickness. Typi-
cal layups are reported in Table 2.
Orthotropic material constants are defined in the
material coordinate system. The material coordinate sys-
tem was initially established for each through-thickness
integration point as well.

3.4. Bird
Bird has been modelled as a fluid which is represented
Figure 5. Roller carriage model of outboard track. by a set of moving particles, each one representing an
 INTERNATIONAL JOURNAL OF CRASHWORTHINESS 5

interpolation point, with known fluid properties, as

described in ref. (11). Then, with a regular interpolation
function called smoothing length, the solution of the
desired quantities can be calculated for all the particles.
Points in SPH formulations are similar to nodes in a
mesh. The difference is that these points are continu-
ously deformable and distort automatically to address
the computational effort in regions of relatively high
Figure 6. Bird representation, LS-DYNA.
density.
The equation of state (EOS) is a pressure–density
allowed to identify the following values for the numerical
relation for the bird, and there are several formulations
simulation, as reported in Table 3.
for it. In most studies [15,20,23,29], the polynomial
C is defined as water shock wave velocity related to an
form of the EOS was used to model the bird; an option
impact velocity of 198 m/s with 10% of water porosity.
is adopting a simpler Murnaghan EOS equation as
Shock velocity is 1430 m/s (56,300 in./s), while S1
reported in [18,21,25]; another application is the
parameter, defined for water without porosity, was tuned
Gr€uneisen EOS [2,28].
by an iterative approach to properly fit the expected
For our studies, the constitutive relation used is the
impact force. Its most appropriate value resulted to be
Gr€uneisen equation given by
equal to 2.48.

  The bird dimensions have been chosen according to
r0 C 2 m 1 þ 1  g20 m  a2 m2
p¼h i Equations (2) and (3) that model the body of the bird
m 2 m 3
1  ðS1  1Þm  s2 mþ1  s3 ðmþ1 Þ2 like a cylinder with both ends as two hemispheres, like
the sketch shown in Figure 6.
þ ðg 0 þ amÞE: (1)
L ¼ 2D; (2)
The Gr€uneisen EOS with cubic shock particle-velocity  13
2M
defines the pressure for compressed material, where E is D¼ ; (3)
pr
the internal energy per initial volume, C is the intercept
of the us-up curve, S1, S2, and S3 are the coefficients of where L is the length of the bird and D its diameter.
the slope of the us-up curve [28], g 0 is the Gr€ uneisen Particle mass has been calculated once the number of
gamma, and a is the first-order volume correction to g 0. points, SPH nodes, was known. The number of SPH nodes
Constants C, S1, S2, S3, g 0, and a are all input was set using data of Figure 6, and reported in Table 4.
parameters. Afterward, a sensitivity analysis was carried out on a
The compression is defined in terms of the relative composite panel, in order to investigate the tuning of the
volume, V, as finite element model and how the particle distance influ-
ences the expected impact force occurring during the
m ¼ 1 6 v  1:
interaction between the projectile and the target. The
target has a square planform with an edge of 20 in., and
In order to validate the bird model, the experimental
it employs the same material come out from the calibra-
test performed from Barber in 1975 was replicated. In
tion process. The velocity of the impact was fixed at 194
the experimental shot labelled 5126A, a circular flat
kts with 4 lb bird weight and normal impact.
panel in 4340 steel material with a thickness of 1 mm
The following formula has been used to estimate
and diameter of 152.5 mm was assumed to be rigid, so
impact loading:
the response of the target can be neglected, whereas the
mass and velocity of the bird were, respectively, 0.074 kg

Empirical expression from [26]
and 198 m/s.
Comparison of the pressure distribution vs. time 2
=3
Pi ¼ 1:18M V 2 cos a (4)
obtained by Barber with the numerical simulation

Table 4. Geometric and mass of the bird model.

Table 3. EOS parameters for the Gr€
uneisen equation. Diameter 2.1 in. (0.0533 m)
Gr€uneisen Length 8.2 in. (0.208 m)
Density 0.0343 lb/in.3 (950 kg/m3)
C S1 S2 S3 V0 g0 Mass 4 lb (1.81 kg)
5.63E4 2.48 0 0 1 0.1 SPH nodes 27,020
6 S. ORLANDO ET AL.

where Pi is the impact force expressed in Newton, M is Table 5. Results comparison among theoretical and LS-DYNA
the bird mass in kg, V is the bird velocity in Knots, a is impact force (composite plate).
Momentum Empirical Pressure Numerical impact
the angle between the bird and the vector normal to the transfer (lbf) formula (lbf) integration (lbf) force (lbf)
target. 48,596 37,790 21,950 51,228

Momentum equilibrium equation [27]

degree of scatter. This means that the force calculated
Pi t ¼ MVcos a (5)
from equations is not conservative but an average value.
Moreover, the results are strongly dependent on the
where M is the bird mass in kg, V is the bird velocity in
mesh dimension and SPH particles distance. Therefore,
m/s, Pi is the impact force expressed in Newton, t = L/V
the calibration process included a necessary mesh sensi-
is the impact duration.
tivity analysis. The SPH particle numbers were varied to
Dividing Equation (5) by the impact duration, the
understand the pressure variation due to the density par-
average force acting on the target is obtained.
ticles. The panel mesh was fixed while changing the
MV 2 cos a number of SPH particles. This variation allowed to study
Pi ¼ (6) its influence on the stability of the model, the deforma-
L
tion and the resulting force plotted vs. the simulation

Pressure distribution integration [29] time. The results are shown in Figure 7.
The model with 96700 SPH points gave cleaner
For a normal impact, Equation (7) provided the results with respect to the model with fewer nodes
impact force. (Table 6). The steady-state pressure is more visible but
the peak force is lower.

D  2
Pi ¼ 3:137rV 2 2
; (7)

where Pi is the impact force in Newton, r is the bird den- 4. Description of the test
sity in kg/m3, V is the bird velocity in m/s, and D is the Flap is made up of upper skin co-cured assembly
bird diameter in meters. (including stringers, spars and leading edge), lower skin
The numerical results of the impact on a composite co-cured assembly (including stringers), ribs and metal-
plate are compared with the empirical and theoretical lic attachments.
formula (Table 5). It should be noted that the forces dis- The lower skin is composed of eight plies with the
played in actual bird strike demonstrate a considerable thickness of 0.01 in., quasi-isotropic orientation, with

Table 6. Impact force varying the SPH points.

Momentum transfer Empirical formula Pressure integration Model with 27,020 SPH Model with 35,920 SPH Model with 96,700 SPH
(lbf) (lbf) (lbf) nodes (lbf) nodes (lbf) nodes (lbf)
48,596 37,790 21,950 51,228 46,620 42,591

Figure 7. Impact force for different SPH points.

 INTERNATIONAL JOURNAL OF CRASHWORTHINESS 7

four plies at 45 degree and four plies at 0 degree. Same The envelope of the results is listed in Table 11 where
layup is applied for the stringers. the letters in the ‘Damage’ column have the following
The inboard flap consists of two spars: front spar and meaning:
rear spar. Front spar again is composed of eight plies
with an orientation of four plies at 45 degrees and four
F: Failure
at 0 degrees with a thickness of 0.01 in. for each ply.
P: Penetration
Ribs are composed of five plies, three plies oriented at
N: Not penetration
45 degrees and the remaining two at 0 degrees.
T: Partially penetrated

S: Stringer failure

D: Diverted
4.1. Simulation results
The main goal of the simulation is to provide a robust From such analysis, two impacts have been chosen as
method to perform sensitivity analysis of bird impact the most dangerous:
event on a flap fully assembled on wing track attach-
ments. It is expected from the simulation to provide con-
Impact 18 – the most critical for the rear spar.
fidence on the selection of the most critical condition to
Impact 7 (37-deg flap angle deployment) – the
be experimentally tested, based on the most severe skin most critical for the interface load; compared to
damage resulting from the model simulation impact and Impact 6, it shows a partial penetration and a
the critical one identified for the metallic attachments. potential damage on the upper skin.
The bird strike analysis allows to simulate a bird
impact on different areas of the outboard flaps (Figure 8), The impacts at 25-deg flap angle deployment, at the
in order to identify, based on the flight condition and end, have not been considered because the kinetic energy
flap angle deployment, the most critical condition for level is lower than the 37-deg flap angle position. There-
the flap itself and its wing attachments. By considering fore, 25-deg flap angles exhibit a lower damage of the
the wing geometry and deployed flap geometry, the sur- panel and a lower interface load is expected, compared
faces that represent the potential impact zones for the to 37-deg flap angle.
flap were identified, and a parametric analysis, based on
the maximum and minimum angles of attack for flight
conditions, has been performed, as shown in Figure 8.
5. Description of the test rig
The expected result, at different time-steps, is shown
in Figure 9, where the bird flows on the impacted sur- The test rig includes dummy tracks 3 and 4 and column
face, sliding or perforating it, depending on the condi- supports, and it has been arranged in order to properly
tion under analysis. reflect the actual flap attitude according to the aircraft
The selection criteria of the test impact point are global coordinate system. Furthermore, the supporting
based on the most severe skin and metallic attachments structure has been designed to withstand the loads
damage resulting from the simulation, as reported in expected from the test impacts providing structural stiff-
Tables 7–10. ness and rigidity to the test article constraint.

Figure 8. Impact zones on the outboard flap.

8 S. ORLANDO ET AL.

Figure 9. Bird path during an impact with complete deviation.

The CAD view of the test rig is shown in Figure 10. It (flap), resulting in a consistent deformation of the centre
was designed to support the flap allowing for the correct of the structure, in correspondence of the impacted zone.
orientation of projectile to each of the identified impact The experimental test was considered positively
locations. passed at first shot. As expected from the numerical sim-
The gun trajectory [6] has been aligned parallel to the ulation, the number of tests was dramatically reduced
aircraft main longitudinal global axis to satisfy the taking advantage of the simulations. The test confirmed
requirement of angle of attack and slip angle for the bird the perfect integrity of the lower skin in terms of poten-
impact, as shown in Figure 11. tial delamination with absence of foreign objects
The reverse installation of the flap, as shown in penetration.
Figure 12, guarantees the best target impact avoiding The destructive inspections have been performed by
interferences of the flap tracks. The height of supporting the boroscopic probe across the impact area, showing a
structure was adapted for the airgun height. Figure 13 failure of the rib’s trailing edge with a damage in corre-
shows the final solution with both impact positions. spondence of the fastners (Figure 15).
Figure 14 shows a sequence of the bird strike against Figure 16 shows the simulation results, with a stress
the flap, from the moment of the impact until the bird is concentration located on the portion of the lower skin in
completely squashed into the panel. It is clearly visible correspondence of the rib’s trailing edge, resulting in the
the deformation of the bird during the impact and its failure of composite lower skin.
squashing into the flap. The deformation behaviour of Furthermore, the lower skin has been cut in corre-
the impacted panels is at t = 0.001 s from the beginning spondence of the impact zone (Figure 17) in order to
of the impact phenomenon. During the first 2 ms, measure the delamination area of parts involved during
almost 90% of the bird load is transferred to the plate the strike event. All the inspections confirm the
 INTERNATIONAL JOURNAL OF CRASHWORTHINESS 9

Table 7. Comparison between design interface load and interface load generated by bird strike event at aft link 3 for all the impact
positions.

Table 8. Comparison between design interface load and interface load generated by bird strike event at aft link 4 for all the impact
positions.
10 S. ORLANDO ET AL.

Table 9. Comparison between design interface load and interface load generated by bird strike event at push rod track 3 for all the
impact positions.

Table 10. Comparison between design interface load and interface load generated by bird strike event at push rod track 4 for all the
impact positions.
 INTERNATIONAL JOURNAL OF CRASHWORTHINESS 11

Table 11. Summary of the bird impact results. Interface load for each impact and the panel damage result.

Figure 10. Iso-view of the outboard flap test article configuration and the impact position.

Figure 11. Air pressure gun – experimental facility.

12 S. ORLANDO ET AL.

Figure 12. Outboard flap experimental facility.

Figure 13. Outboard flap dummy track attachment (left view).

Figure 14. Sequence of the impact.

 INTERNATIONAL JOURNAL OF CRASHWORTHINESS 13

Figure 15. Post-inspection by scan camera.

conclusion predicted by FE analysis; in particular, as The numerical simulation, as reported in Figure 20,
shown in Figure 18, the trailing edge rib which fails with shows a critical zone in correspondence of the impact
a crack runs from corner radius up to middle web. showing a good correspondence in terms of the exten-
Figure 19 shows a significant delamination region sion of the damaged area.
measuring approximately 3.900 £ 3.400 (99 mm £ 86 During the setup phase and the definition of the test
mm) on the lower skin close to the impact location. plan, particular attention was dedicated to define the

Figure 16. Detail of the rear stringer.

14 S. ORLANDO ET AL.

Figure 18. Failure of the stringer.

Figure 17. Inspection of the lower skin.

Figure 19. Impact zone on the lower skin.

 INTERNATIONAL JOURNAL OF CRASHWORTHINESS 15

Figure 20. FEM simulation of the damaged area.

flap’s portions that macroscopically are involved in the shift is 0.1 ms. Moreover, the impact duration is about
strike, and to estimate the force intensity and the impact 0.6 ms, allowing the bird to transfer all its kinetic energy.
duration, with the aim of correctly dimensioning the fix- The damage analysis on the rear spar has been per-
ture and the test rig, and possibly to reduce the number formed being it one of the most important structural
of tests. For this reason, Figure 21 shows the interface components of the outboard flap. A spar failure can lead
loads and the resultant impact force vs. time predicted to the all flap failure.
by the numerical simulation where the maximum peak The post-process analysis is focused on the strain
is 1.8E4 lbf (80 kN). The time of the maximum (or mini- measurement of the rear spar, in both X and Y local
mum) force at the aft link or push rod is not the same of directions, induced by a bird strike event. As shown in
the maximum resulting impact force. This time delay is Figure 22, the maximum strain, output from the numeri-
due to the inertia of the system that generates a phase cal simulation, is reached for the upper surface of the
shift between interface load and impact force. The phase elements in the Y-direction. Specifically, the maximum

Figure 21. Interface load and resultant impact force vs. time.
16 S. ORLANDO ET AL.

Figure 22. Plot of the upper surface Y-strain of the investigated spar elements.

level of mstrain is equal to 8750 mstrain for the upper The idea to validate the flap structure and bird strike
surface. scenario using the certification by analysis is far away,
since that in any testing, there is some inherent uncer-
tainty and variation in the measured responses that
6. Conclusions
affect the baseline comparison data, but the simulation
In this paper, an in-depth focus has been addressed to and the calibration of the finite model element and the
study a flap configuration compliant with EASA and bird strike scenario allowed engineers to use only one
FAA bird strike requirements. The importance and the flap configuration and only one shot for the experimen-
validity of the numerical simulations aimed at designing tal test. As a final result, the test was passed and the flap
and analysing a novel flap configuration of a commercial structure showed compliance with the requirements.
aircraft have been shown. The numerical and experi- The most critical condition was numerically identified
mental correlation has been fully validated and approved and approved for the experimental test, consistently sav-
by the certification agencies. ing time and money.
Preliminary tests on small components have been first
validated with experimental test from the literature on
fully rigid target. Then, the results have been extended to
Disclosure statement
a flexible composite target to investigate the parametric No potential conflict of interest was reported by the authors.
tuning process to be used for the full-scale structure.
The obtained results showed that the bird impact loca-
ORCID
tion significantly affects the structural damage, and
therefore an innovative composite configuration was M. Guida http://orcid.org/0000-0002-4116-5624
identified able to satisfy the minimum weight
requirement.
The comparison between numerical and experimental References
data showed very good correlation in terms of global [1] S. Abrate, Soft impacts on aerospace structures, Prog.
structural behaviour and relevant damage of the test arti- Aerosp. Sci. 81 (2016), pp. 1–17.
cle, confirming the validity of the approach. In particu- [2] M. Anghileri, L.M.L. Castelletti, and V. Mazza, Bird-
lar, the model was accurate to predict the internal strike: A pproaches to the analysis of impacts with pene-
tration, in Impact Loading of Lightweight Structures, M.
damage across trailing edge rib and the absence of for- Alves and N. Jones, eds., WIT Press, Southampton, 2005,
eign object penetration, showing that the adopted CFRP pp. 63–74.
material was able to protect the primary loading path [3] ASTM Standards: D3039/D 3039M-08. Standard Test
(e.g. attachment and spars) and to satisfy certification Method for Tensile Properties of Polymer Matrix Com-
requirements. posite Material Tensile Test, 2008.
[4] ASTM Standards: D5766/D5766M – 11. Standards
These results confirmed that advanced numerical
Method for Open-Hole Tensile Strength of Polymer
simulation techniques could significantly help to design Matrix Composite Laminates, 2011.
safer and more efficient, in terms of weight and [5] ASTM Standards: D695. Standard Test Method for
manufacturing costs, aircraft structures. Compressive Properties, 2010.
 INTERNATIONAL JOURNAL OF CRASHWORTHINESS 17

[6] ASTM Standards: DF33-10 Standard Test Method for [18] T. Kermanidis, G. Labeas, M. Sunaric, and L. Ubels,
Bird Impact Testing of Aerospace Transparent Enclo- Development and validation of a novel bird strike resis-
sures, 2010. tant composite leading edge structure, Appl. Compos.
[7] J.P. Barber, H.R. Taylor, and J.S. Wilbeck, Characterization Mater. 12 (2005), pp. 327–353.
of bird impacts on a rigid plate: Part 1, Technical Report [19] Lacome, Smooth particle hydrodynamics- Part II, FEA
AFFDL-TR-75-5, 1975, University of Dayton, Ohio. information international news for the world-wide engi-
[8] EASA and FAA Certification for Large Aeroplanes, CS- neering community, 6–11 (2001).
25.571 Damage-Tolerance (Discrete Source) Evaluation. [20] M.A. Lavoie, A. Gakwaya, M. Nejad Ensan, and D.G.
FAR part 25 “Airworthiness Standards: Transport Cate- Zimcik, Review of existing numerical methods and val-
gory Airplanes, paragraph 25.571 Damage Tolerance idation procedure available for bird strike modeling.
and fatigue evaluation of structure. Amendment 45.” Int. Conf. Comput. Exp. Eng. Sci. 2 (2007), pp. 111–
[9] EASA and FAA Certification for Large Aeroplanes, CS- 118.
25.631 Bird strike Damage, FAR part 25 “Airworthiness [21] J. Liu, Y.L. Li, and F. Xu, The numerical simulation of a
Standards: Transport Category Airplanes, paragraph bird-impact on an aircraft windshield by using the SPH
25.631 Bird strike Damage. Amendment 45. method. Adv. Mater. Res. 33–37 (2008), pp. 851–856.
[10] V.K. Goyal, C.A. Huertas, and T.J. Vasko, Arbitrary L [22] LS-Dyna Keyword User’s Manual, Livermore Software
agrange eulerian approach for bird-strike analysis using LS- Technology Corporation, c. 971, 2007, Livermore,
DYNA, Amer. Trans. Eng. Appl. Sci, 2. (2013, 109–132). California.
[11] M. Guida, F. Marulo, M. Meo, A Grimaldi, and G. Oli- [23] R.H. Mao, S.A. Meguid, and T.Y. Ng, Effects of incidence
vares, SPH - Lagrangian study of bird impact on leading angle in bird strike on integrity of aero-engine fan blade.
edge wing, Compos. Struct. 93 (2011), pp. 1060–1071. Int. J. Crashworthiness. 14 (2009), pp. 295–308.
doi:1016/j.compstruct.2010.10.001 [24] M.A. McCarthy, J.R. Xiao, C.T. McCarthy, A. Kamoula-
[12] M. Guida, F. Marulo, M. Meo, and M. Riccio, Analysis of kos, J. Ramos, J.P. Gallard, V. Melito, Modeling bird
bird impact on a composite tailplane leading edge, Appl. impacts on an aircraft wing – Part 2: Modeling the impact
Compos. Mater. 15 (2008), pp. 241–257. doi:10.1007/ with an SPH bird model, Int. J. Crashworthiness. 10
s10443-008-9070-6 (2005), pp. 51–59.
[13] M. Guida, F. Marulo, M. Meo, and S. Russo, Certification [25] M.A. McCarthy, J.R. Xiao, C.T. McCarthy, A. Kamoula-
by birdstrike analysis on C27J fullscale ribless composite kos, J. Ramos, J.P. Gallard, and V. Melito, Modeling of
leading edge, Int. J. Impact Eng. 54 (2013), pp. 105–113. bird strike on an aircraft wing leading edge made from
doi:10.1016/j.ijimpeng.2012.10.002 fiber metal laminates – Part 2: Modeling of impact with
[14] A.G. Hansen, Y. Girard, L. Olovsson, T. Berstad, and M. SPH bird model. Appl. Compos. Mater. 11 (2004),
Langseth, A numerical model for bird strike of aluminum pp. 317–340.
foam-based sandwich panels, Int. J. Impact Eng. 32 [26] I.I. McNaughton, The design of leading edge and intake
(2006), pp. 1127–1144. wall structures to resist bird impact, RAE TR72056, Royal
[15] A.G. Hanssen, Y. Girard, L. Olovsson, T. Berstad, and M. Aircraft Establishment, April 1972.
Langseth, A numerical model for bird strike of aluminium [27] R.L. Peterson and J.P. Barber, Bird impact forces in air-
foam-based sandwich panels. Int. J. Impact. Eng. 32 craft windshield design, National technical Information
(2006), pp. 1127–1144. Service, Ohio, AFFDL-TR-75-150, March 1976.
[16] S. Heimbs, Computational methods for bird strike simula- [28] S.Z.-R. Reza Hedayati, A new bird model and the effect of
tions: A review, Comput. Struct. 89 (2011), pp. 2093– bird geometry in impacts from various orientations, Aero-
2112. doi:10.1016/j.compstruc.2011.08.007 space Science and Technology, 28, 1 2012, 9–20.
[17] A.F. Johnson and M. Holzapfel, Numerical prediction of [29] J.S. Wilbeck, Impact Behavior of Low Strength Projectiles,
damage in composite structures from soft body impacts, J. Air Force Materials Lab Wright-Patterson Afb Oh, Ohio,
Mater. Sci. 41 (2006), pp. 6622–6630. July 1978.