Abstract
The three-dimensional braided composites, with intertwined fiber bundles in the through-thickness direction, have advantages of high interlaminar shear strength, fracture toughness, and excellent impact resistance, making them a promising material for applications in the field of aeroengine fan blades. As the bird impact behavior of the fan blade directly affects the safety of the aeroengines, it is of great significance to study the dynamic response and damage mechanism of 3D braided composites under bird strike load. In this paper, the bird impact tests on the 3D four-step braided composite targets were carried out using the gas gun system. The effects of impact velocity, impact location, and braiding angle on the bird impact behavior were studied. It is concluded that the damage and failure become more severe with the increasing impact velocity. The whole impact event could be divided into 3 stages, i.e., local deformation stage, postflow impact stage, and bending deflection stage. The braided composite presents flexible characteristics and could bear extraordinary deformation during the bird impact. One distinguishing feature of bird impact damage is the destruction of the clamping root due to bending load caused by cantilever construction. The internal damage form at the impact area was mainly the separation of the fiber bundles from the matrix while the breakage of the fiber bundles and the crushing of the matrix play the primary role at the root part. The target plate impacted at the 70% height had the largest bending angle and most serious damage, followed by those impacted at the 90% and 50% heights. Both the appearance damage and internal damage extent are smallest for 45° braiding composites.
1. Introduction
In the design of engine fan blades, the foreign object impact has been one of the greatest concerns [1]. Bird strike impacts frequently occur due to the suction at the air inlet of the engine during operation. According to the International Bird Strike Committee, 55 fatal accidents, in which 108 aircraft were damaged and 277 people died, were reported between 1912 and 2009 [2]. To manage the risk related to bird ingestions, Aviation Authorities developed safety regulations for foreign object ingestions by turbine engines [3–6]. Strict requirements on the impact-induced damage and residual deformation of the fan blade are imposed so that the engine maintains sufficient thrust when the fan blade is impacted by a bird.
Usually, full-scale engine bird ingestion tests, which are extremely expensive, are required by safety regulations to demonstrate their ability to withstand bird ingestion and following the ingestion to produce enough trust. In order to decrease the cost of engine development, a series of simulated testing method in laboratory scale and analysis methods applicable to bird strike simulation were developed [7–9]. It has now been extensively accepted that bird under high-velocity impacts could be modelled as a fluid because bird’s tissue strength is significantly lower than the stresses generated during the impact event [10]. Furthermore, the bird could be treated as hemispherical-ended cylinder with the similar mass, density, and compressibility [11]. Based on these theories, artificial birds were employed widely in the study of bird impact. Three most frequently used configurations are hemispherical-ended cylinder, straight-ended cylinder, and ellipsoid at various length-to-diameter aspect ratios. In addition, some bird models with realistic geometries proposed by Hedayati and Ziaei-Rad [12], Hedayati and Sadighi [13], and McCallum et al. [14] using CT images and SPH method showed more accurate results than the traditional folded configuration model. This model could also enable the investigation of the effect of orientation of bird on bird strike events to evaluate the most damaging scenario. The relative effects of the artificial bird geometry on the impact behavior impinging a flexible aeroengine fan blade were examined in LS-DYNA [15], and it is found that the initial contact area between the bird and target in the early phase of the impact event has a significant effect on the peak impact force while the aspect ratio has little influence. Although impact test using gelatin bird is an effective method to evaluate the impact resistance and damage mechanism of fan blades during a bird strike event, limited literatures were addressed due to the cost of bird impact test [16]. Alternatively, finite element analysis is the primary investigation method [17–19].
Carbon fiber-reinforced plastic (CFRP) composites have been recently applied as a material for the fan blades of high-bypass turbofan jet engines [20, 21], because of their light weight and high specific strength [22–24]. It is essential to understand the detailed characteristics of the bird strike impact resistance for these composite fan blades. To address the impact-induced deformation and damage of composite plates subjected to soft-body, high-velocity impacts of composite fan blades, Nishikawa et al. [25] presented a numerical simulation and found the damage in the composite changed from bending-induced matrix cracking to an intensive fiber breakage mode causing local shear perforation as the impact velocity increased. Kim et al. [26] and Siddens and Bayandor [27] investigated the initial degradation of individual fan blades struck by a bird and evaluated subsequent damage to other engine components using an arbitrary Lagrangian-Eulerian (ALE) method and a meshless Lagrangian particle method, respectively. Heimbs and Bergmann [28] presented the dynamic behavior of laminated composite plates exposed to tensile and compressive preloads under soft-body impact. Tensile and compressive preloading was found to exert an influence on the damage pattern.
The newest generation of composite fan blade is manufactured using 3D braiding techniques in combination of resin transfer molding (RTM) process. With advantages of low delamination tendency, high strength, out-of-plane stiffness, and impact tolerance, 3D braided composites have recently been recognized as one of the most promising structural composite materials [29]. A number of research have been carried out on the mechanical properties of 3D braided composites [30–34]. However, few studies were conducted on its impact behavior, which is quite different with that under quasistatic loadings. It has been found that the uniaxial tensile modulus and maximum tensile stress of 3D braided composites increase linearly with the strain rate [35, 36]. The energy absorption also increases with the strain rate, corresponding to the different damage and energy absorption mechanisms under quasistatic and high strain rate compression [37]. Similarly, Li et al. [38, 39] reported the strain rate strengthening effects and dynamic toughness phenomenon of 3D braided composites under longitudinal and transverse compression. Researchers have found that carbon fiber-braided composites have good fiber/matrix adhesion and impact resistance during high-velocity impact. Tang et al. [40] concluded that the primary damage is cracking at bundle interface and intrabundle delamination near bundle crimping locations while the failure mechanisms are mainly fiber debonding, delamination, and cracking propagation in the resin matrix. Xu and Gu [41] indicated that the major failure mechanisms of 3D braided composites under high-velocity impact are shear and compression failure in the impact side and extensive tension failure in the distal side. Jenq and Mo [42] and Jenq et al. [43] conducted ballistic impact tests with incident velocities ranging from 70 to 170 m/s and found that the major impact damage patterns are indentation, matrix failure, and fiber breakage with axial and braider fiber yarn pullout. The high-velocity impact tests in the research mentioned above all employ the metal material as projectile, which is quite different with that in a bird strike event.
In the current paper, bird impact tests were carried out on 3D carbon/epoxy braided composite targets using the gas gun system. The effects of impact velocity, collision position, and braid angle on the bird impact behavior were studied with the help of a high-speed camera and X-ray image system. The results of this research help the anti-bird impact design of composite fan blade.
2. Bird Impact Testing
2.1. Testing Device
The bird impact test on 3D braided composites was conducted with gas gun system of 37 mm caliber in the velocity ranges of 50 m/s~200 m/s. The schematic diagram and photo of the gas gun testing device is shown in Figure 1. The projectile impact velocity is obtained by a velocimeter system installed between the gun muzzle and the target. When the projectile passes through the velocimeter system, the laser beams are obstructed successively, generating the transformation of voltage trigger signal. With time interval of two trigger signals and the distance of the velocimeter, the velocity could be determined. High-speed videos with rate of 12,000 frame/s were taken for all tests. These photos provide the observation of target deformation, projectile trajectory, and estimation of the residual velocity of the projectile.
2.2. Gelatin Bird Projectile
It is generally considered that cylindrical gelatin birds with an aspect ratio of 2 : 1, a density of 950 kg/m [3] and a porosity of 10%, are suitable to simulate the fluid characteristics of real birds at high-speed impact. Gelatin birds used in this test were prepared in accordance with the above requirements. The testing birds were 25 mm in diameter and 50 mm in length. The length and diameter ratio is 2.0 as suggested by Reference [44]. The weight of a gelatin bird is 23.3 g. The small size and weight of gelatin bird are chosen to match the size of a gas gun device and the shrunken flat target with respect to real composite fan blades of high-bypass ratio turbofan engine. Furthermore, the main goal of this research is to dig into the damage and deformation of braided composite blades under soft-body impact and evaluate the effect of impact velocity and collision position, helping the anti-bird impact design of composite fan blade.
Figure 2 shows a gelatin bird and its placement in sabot. The gelatin birds are wrapped with a plastic film in order to maintain its shape. The gelatin bird projectile is supported by several foam rings in an aluminum sabot. The sabot and projectile are pushed and accelerated by compressive air in gun barrel. The projectile launch velocity is determined by the air pressure. The sabot is captured at the end of gun barrel by a sabot separator, while the projectile moves forward and impacts the target. The fragments, spallation from the target and projectile with residual velocity, are arrested by the fragment collector behind.
(a)
(b)
(c)
2.3. Composite Target
The three-dimensional four-step braided composites (3D braided composites) used as the testing specimens are made up of 12 k Toray T700 carbon fiber and the matrix material of TDE86 epoxy resin (provided by Tianjin Jingdong), with a fiber volume content between 60% and 62%. Composite specimens with three different braiding angles, i.e., 20°, 30°, and 45°, are fabricated. The braid composite plates are rectangular with dimension of 380 mm × 180 mm × 2 mm. The 3D reinforcement structures of different braiding angles are presented in Figure 3. The partially enlarged view shows the height of two pattern knots of each kind. The height of the pattern knots decreases as the braid angle increases.
To avoid the edge effect, marginal area was cut off from both sides of the samples. Then, the large composite plates were divided equally into 4 pieces of targets at intervals of 2 mm. The composite target for bird impact tests is in size of 189 mm × 84 mm × 2 mm, which is in rectangular shape to represent the features of composite fan blade. Being limited to technique and cost, the structural characteristics of a fan blade, including variable cross section, 3D configuration, and titanium edge, are not considered in this study. As the fan blades are always in the form of cantilevered structure, the testing specimens were clamped at the root as suggested in References [44, 45], as shown in Figure 4. The black part represents the testing composite specimens and the yellow part is the clamping fixture.
(a)
(b)
(c)
It is well known that rotating bird strike tests requires large-scale and expensive test facilities and procedures, which result in extremely high cost, making it difficult to assess new materials for fan designs. By contrast, ballistic impact testing on a static flat panel or blade is an effective, convenient, and cost-efficient method which is widely used to investigate the dynamic response and damage threshold and evaluate different designs such as thickness and ply configuration. Friedrich [44] conducted bird impact tests on the flat specimens mounted in a cantilevered fashion in a fixed end support. In our study, similar testing procedures were employed. During ballistic impact tests, the relative velocity of real bird impacting to the blades is the vector sum of the axial velocity of the bird and the tangential velocity of the blade at the impact radius. As mentioned in an open literature, the deformation and damage mechanisms in the static and dynamic cases could be quite different. Particularly, NASA Glenn Research Center has been working to identify a static blade test procedure that would be effective at reproducing similar results as seen in rotating tests. Howard et al. [19] compare the simulation results of a bird strike on various nonrotating blades with those on a rotating blade. In the beginning, the prestress effect was assumed to be the major influencing factor. As the stress field in a rotating blade varies radially, no method for duplicating a radial variation equivalent to the rotating case could be achieved experimentally. Hence, the techniques to load the blade in a manner that would match the stress conditions in the rotating blade at the point of impact through applying a force through cables attached to the blade tip or applying force through mass attached in some way to the blade tip. Although various degrees of success were attained in matching the prestress conditions at the impact point, the blade damage was similar and always more than the rotating case. The results verified that matching the prestress is not sufficient to guarantee similar damage. The additional attempt with a rotating blade spinning about its center of gravity impacted by a bird model shot with the same relative velocity as the full rotating case indicated that rotating effects are significant and likely overshadow the prestress effects in stiffening the blade. In spite of promising findings were obtained from numerical method, more research and experimentation still need to be done to validate it. In addition, the rotation of a single blade, though much easier than the rotation of a full fan stage, still needs special rotating element in the gas gun system. The timing of shooting a bird to hit the blade at just the right time and place would be more difficult. It would cost much more effort and money to modify the testing device and induce great risk of failure. Considering that it is not the goal of this work to replace the bird strike certification tests, which consider aspects such as the ability to continue operating at a reduced power level, but rather the goal is to determine the direct soft-body impact damage of 3D braided composites with different braiding angles and impact locations. Such a test could be used as a cost-effective screening test for new materials and fan designs that could then go further in the development cycle and eventually be tested in a rotating test.
2.4. Characterization of Damage Morphology
Because the bird impact samples under bending load did not break completely, X-ray 3D microscopy (diondo D5 type high-resolution X-rays system and service) imaging was adopted to observe the material internal structural damage for bird impact damage characterization. The mesostructure of composites was inspected with the spatial resolution of 28 μm, and the 2D images of the specimens at different sections were obtained. The scanning area was chosen to be square with size of 50 mm.
3. Dynamic Behavior and Damage Mechanism
A series of bird impact tests were conducted on the 3D braided composites with a braiding angle of 20°. Particularly, the influences of bird impact velocity and bird impact location on the dynamic behavior and damage characteristics of this material were studied. The impact velocity is adjusted by changing the launch pressure of the gas gun, including four launch pressures (0.4 MPa, 0.6 MPa, 0.8 MPa, and 1.0 MPa), corresponding to projectile velocity between 100 m/s and 200 m/s. To address the influence of the impact location, three different locations are chosen, i.e., at the center of the target plate along the longitudinal 50%, 70%, and 90% heights, respectively. The test conditions of a 20° braiding angle are listed in Table 1. B20-01 and B20-09 tests were both tested for the same braiding angle (20°) at the same impact location (50% height). They are impacted with the same launch pressure (0.4 MPa), but the resultant impact velocity has some differences due to the slight differences of projectile mass and impact postures. In fact, B20-01, B20-07, B20-08, and B20-09 tests have the same testing conditions with close impact velocities (117.5 m/s, 139.1 m/s, 142.2 m/s, and 132.4 m/s, respectively). These 4 tests were conducted to investigate the effect of impact velocities. Besides, as the bird impact tests were carried out with the same launch pressure for 30° and 45° braiding composites, these tests were also a preparation for the next tests.
3.1. Effects of Impact Velocity
To identify the effect of impact velocity, four tests with different impact velocities at the same location (50% height) were chosen for comparison. Figure 5 shows the maximum deformation of the target plate photographed by the high-speed camera at different impact velocities. The high-speed image was measured through the Camera Measure software provided by e2eSoft company. The horizontal and vertical displacements of the back surface are plotted in Figure 5(e), and the variation of maximum horizontal displacement with time is plotted in Figure 5(f). As shown from the high-speed photo, the target plate will have obvious deformation when impacted by the bird body. A bulge will appear in the impact area, and the left and right sides of the impact area will bend to the middle. The maximum horizontal displacements for the selected 4 tests shown in Figures 5(a)–5(d) were 25.4 mm, 48.55 mm, 74.74 mm, and 80.21 mm, respectively. As the impact speed increases, the bulging in the center of the panel becomes larger. In particular, results of the two groups of birds with the highest speed (B20-03 and B20-04) showed that the material presented characteristics of the woven composites during the bird impact and had large tensile deformation along the thickness direction, covering the bird body like a net. From Figure 5(f), it is found that the maximum horizontal displacement increases almost linearly with time. The displacement grows faster for tests with higher impact velocity while the deformation grows much slower in those tests with smaller incident velocity.
(a) B20-09
(b) B20-02
(c) B20-03
(d) B20-04
(e) The displacement of the back surface
(f) The maximum displacement versus time
Figure 6 shows the final deformation and failure morphology of the composite target under different impact velocities. The clamping root is placed in the right side. It is shown that the bird impact damages of the braided composites are in the form of global deflection, local bulging in the impact area, falling off, the fracture of fiber, interface debonding, and fiber pullout. The damage and failure become more severe as the impact velocity increases. The deformation of the impact area of the target in B20-09 test basically restores, and the surface remains flat with no obvious signs of damage, as shown in Figure 6(a). In test B20-02 (Figure 6(b)), there are obvious global deflection and bulge deformation at the impact point. There are two bending curves at the corner of the blade top, where fiber fractures and fiber/matrix interface debonding occur at both ends of the curves. In test B20-03, due to the increase of the bird impact velocity, the damages along the bending curves become more severe. The corner part fractures and falls off completely, and the yarns in the fracture surface unravel from braiding preform as shown in Figure 6(c). The failure modes are similar in test B20-04 with the highest impact velocity as shown in Figure 6(d). The target was seriously damaged at the impact point with small pieces of fragment.
(a) B20-09 (132.4 m/s)
(b) B20-02 (169.6 m/s)
(c) B20-03 (192 m/s)
(d) B20-04 (211.5 m/s)
One of the distinguishing features of bird impact damages is the destruction of the clamping root. In four tests, all plates fractured at the clamping root except the B20-03 test, in which the target deflects severely with fiber breakage and matrix cracking. From the photos caught by a high-speed camera, it is found that the 3D braided composite target bears extremely bending load at the root part when impacted at the 50% height due to the cantilever construction. In test B20-3 with medium high impact velocity, the composite target fractures and falls off at both corners, making part of the bird projectile flow away, releasing the bending load exerted on the target root to a certain extent. Therefore, the B20-3 target presents severe bending damage with fiber breakage, but is not broken completely. It is also worth noticing that the fracture surface is quite clean and neat in B20-04 with the highest impact speed. This is because the bird body with the maximum speed has the shortest contact time with target plate; the target root stress achieved the ultimate strength in a shorter period of time and was destroyed altogether. In other tests with longer contact time, the roots of the fiber bundles were not destructed at the same time, leading to a zigzag fracture surface. Furthermore, there was no obvious bulging deformation in the target in test B20-04, in which test the impact load at the impact area was released due to the premature fracture of the target root.
The impact process of bird impact on 3D braided composites was taken by a high-speed camera, as shown in Figure 7(a). The horizontal displacement of the tip and contact point versus time was also plotted in Figure 7(b). As the impact process is similar in all test, B20-02 test was taken as an example. It can be seen that the bird projectile extended obviously after separated from the sabot due to its soft feature. The bird impact process on composite target generally corresponds to typical four stages of bird impact: initial impact, pressure decay, constant flow, and flow termination. The bird body spread around after impact with the target plate, showing obvious fluid property. It is found that the whole impact event could be divided into 3 stages. The first stage is the local deformation stage, from the contact moment to 0.41 ms. In this stage, the bulge deformation area and bulging depth develop with the moving of integrated bird projectile. It is also found that in this stage, there are barely horizontal displacements for target tip form Figure 7(b). The second stage lasts from 0.41 ms to 1.50 ms, namely, the postflow impact stage. In this stage, the bird projectile flows like liquid striking the target in an extensive scope. The local bulge deformation depth continues to grow to a maximum level. Both the horizontal displacement of the tip and contact point increase approximate linearly with time. The third stage lasts from 1.50 ms to the end, called bending deflection stage. In this stage, the structural response plays the primary role. The whole target bends over 90° from the root. It can be seen that soft-body impact varies significantly from that impacted by hard object. The bird projectile flew and spread quickly after contacting with the target with impact load exerting on a relative large area, while the hard object acts as a rigid body with the impact force focused on the impact point.
(a) The impact process taken by high-speed camera
(b) Horizontal displacement of the tip and impact location
The internal damage of B20-08 target was detected before and after the impact, respectively, as shown in Figure 8. In these tests, the bird impact velocity is relatively low (142.2 m/s); thus, the damage of the target is almost invisible with the naked eyes. Two sections paralleled to the braiding direction were chosen from the impact surface. Different luminance and shades in the X-ray image represent different densities, and higher brightness means larger densities. As the density of the carbon fiber (1760 kg/m3) and the epoxy resin (1190 kg/m3) does not have much differences, the internal structure of the target before the impact in the above two figures was relatively dense, with no obvious boundary between the two. After the impact, the area within local area clearly showed the boundary of fiber bundles, and the color between the fiber bundles was also darker than that before the impact, indicating the fiber/matrix interface debonding and textile loose construction. It is demonstrated that although the impact of the gelatin bird does not cause any damage to the surface of the target, the internal structure of the impact area is changed. The damage form of the material is mainly the separation of the fiber bundles from the matrix.
(a) 0.75 mm away from the impact surface
(b) 1.25 mm away from the impact surface
Particularly, in test B20-08, the clamping root of the target was not fractured. The internal damage in the root part was detected using X-ray system as well. Figure 9 shows a cross-sectional and three-dimensional reconstruction of the roots of B20-08 targets after impact. As mentioned above, the root of the target plate was fixed on the fixture and undergone obvious bending load during the impact. It can be seen from Figure 9 that the breakage of the fiber bundles and the crushing of the matrix were both on the surface of and inside the target.
(a) Plane view
(b) 3D view
3.2. Effects of Different Impact Positions
Figure 10 shows the bird impact process in tests (B20-1, B20-05, and B20-06) impacted with similar velocity at different collision positions photographed by the high-speed camera. With relatively lower impact velocity, the global deflection of the target plays the primary action during the bird impact event. It can be seen that the largest bending angle of the target plate is the largest at the 70% height (almost 90°), followed by the 90% height (80°) and 50% height with the minimum bending angle (roughly 50°). High impact position exerts greater bending moment to the root of the target, making bending angle of target impacted at the 70% height larger than that of 50% height, whereas for too high impact location which is close to the top edge of the target, a great portion of the bird projectile flows away from the target, leading to smaller contact force and bending angle.
The targets impacted at these three collision positions all rebounded after bending to the maximum angle, swinging back and forth, and finally stopped in the longitudinal direction. Figure 11 shows the damage morphology of the target at 90% height. The damage morphologies of targets impacted at different locations were basically similar. The bulging deformation at the impact area recovered with no obvious damage on the surface, and the bending deformation at the root was basically restored. There were just creases, with fiber bundle fracture and matrix cracking on their surfaces. The difference is that the creases only occurred on the clamping ends in the root for 50% and 90% height cases, while additional crease in the upper part appeared in the target impacted at 70% height due to the largest bending angle.
Figure 12 shows internal X-ray images on the cross sections of the target perpendicular to the braiding direction at the center of impact. Figure 12(a) shows a nonimpacted target (target used in B20-09 tests) with a regular rectangular cross section. The target plates impacted at different heights all had a slight deformation after bird impact but no fiber bundle fracture. The main damage area of the target was in the fixed ends with fiber bundle rupture and matrix cracking due to substantial bending. As the braided composite has extraordinary deformation characteristics, the root was not completely broken and remained as a whole instead.
(a) No impact (B20-9 target before impact)
(b) 50% height (B20-01 test)
(c) 70% height (B20-05 test)
(d) 90% height (B20-06 test)
4. Influences of Braiding Angle
In this section, bird impact tests were performed on composites with braiding angles of 30° and 45° to figure out the influence of braiding angle on the anti-bird impact capability of 3D braided composites. The testing conditions are listed in Table 2.
4.1. Dynamic Behavior and Damage
Figure 13 shows the final global deflection deformation and the local root damages of the target plates with different braiding angles. It is observed that the 20° braiding angle target plate fractured at the clamping end; meanwhile, the root fracture surface was rough and the length of the fiber bundles pulled out at the fracture surface was different. On the contrary, the target plates with the 30° and 45° braiding angles deflect severely but not broken in the clamping ends. The deflection angle at the root is about 17.3° and 13° for 30° and 45° braiding composite targets, respectively. The folds on both sides of the 30° braiding angle target were different. The crease on the impacted surface was caused by the tensile damage, mainly with fiber fracture on the surface, while the crease on the reverse side was made by the extrusion damage, mainly with the matrix crush, fiber bundles, and matrix debonding. The crease of the 45° braiding angle target plate was similar to 30°, but with less damage. It can be seen from impact damages that the damage extent is smallest for 45° braiding composites, followed by 30°. This is because the unit cell size decreases with the braiding angle, making more unit cells arranged within the same area. The relative tight braided structures of the targets with larger braiding angle could alleviate the damages induced by impact, enabling the composite target bear larger impact load. This is keeping with the punch shear test results reported in Reference [46], in which it comes to a conclusion that the composite with larger braiding angle exhibited higher punch shear performance and increased energy absorption. The reason was attributed to more compact reinforcement structures in terms of greater length/width ratio in the elliptical cross section and enhanced interactions between braiding yarns.
(a) 20° (B20-09)
(b) 30° (B30-02)
(c) 45° (B45-02)
(d) Displacement curves
Figure 14 shows the damages of 45° braiding composite target plate after a typical bird impact. The target plate bent over the direction of the impact. The enlarged view of damage in the root indicates that the main failure modes are the fiber bundle breakage and matrix cracking.
Figure 15 shows the bird impact process of the 30° braiding composite plate taken by the high-speed camera. During impact, the maximum bending angle of the target plate almost exceeded 90°. Even in this condition, the root of the target could recover from extreme deformation state and did not break, indicating that the braided composite material has excellent flexibility and fracture toughness.
(a) 0 ms
(b) 2.3 ms
(c) 4.1 ms
(d) 9.3 ms
(e) 20.2 ms
(f) 36.1 ms
4.2. Internal Damage
As there is no visible damage at the impact area, X-ray images were employed to detect the internal damage of target with different braiding angles, as shown in Figure 16. It can also be seen from the figure that there were some small cracks in the 30° and 45° target plates but not in the 20° target plate. This is because the internal structure of the target plate with the larger braid angle is tight; therefore, the resin is more difficult to penetrate completely into gaps between the fiber bundles during resin injection in the manufacture process, forming small pores. In accordance with the damages observed from Figure 13, the internal damage extent is most severe for 20° braiding composite, followed by 30° braiding composite. The fiber bundle rupture was observed in the 20° braiding composite target. The fiber bundle and the matrix interface debonding happens on the 30° braiding composite target while the 45° braiding composite target showed no obvious impact damage. This also proves that the bigger the braid angle, the smaller the damage of the target, whether in the root fixed ends or the impact area.
(a) 1.25 mm away from the impact surface
(b) 0.38 mm away from the impact surface
5. Conclusions
In this paper, the gelatin bird impact tests were conducted on 3D braided composites of different braiding angles. The effects of the impact velocity, impact location, and braiding angle on the bird impact damages were studied. The main conclusions are drawn as follows: (1)For 20° braiding composites, the damage and failure become more severe with the increasing of impact velocity. The fiber fractures and fiber/matrix interface debonding damages mainly occur along bending curves at the corner of the blade top, which could fall off from the target completely. The braided composite presents flexible characteristics and could bear extraordinary deformation during the bird impact like a net. The maximum horizontal displacement at the impact point increases from 25.4 mm to dramatic 80.21 mm when the impact velocity lifts from 132.4 m/s to 211.5 m/s(2)The bird body acts as a fluid material during high-speed impact process and spreads around after contacting the target plate. The whole impact event could be divided into 3 stages, i.e., local deformation stage, postflow impact stage, and bending deflection stage. The horizontal displacement of the tip of the target lags for 0.41 ms in comparison to the impact point. One distinguishing feature of bird impact damage is the breakage of the fiber bundles and the crushing of the matrix of the clamping root due to bending load(3)The target plate impacted at the 70% height had the largest bending angle (almost 90°), followed by 90% height (80°) and 50% height (roughly 50°). Higher impact position exerts greater bending moment to the root of the target, making bending angle increase. However, too high impact location results in a great portion of the bird projectile flew away, leading to smaller contact force and bending angle. The bulging deformation at the impact area and the bending deflection at the root almost recover completely, leaving creases with fiber bundle fracture and matrix cracking on their surfaces(4)Both the appearance damage and internal damage extent are smallest for 45° braiding composites, followed by 30° braiding composites. The deflection angle at the root is about 17.3° and 13° for 30° and 45° braiding composite targets, respectively. Because the unit cell size decreases with the braid angle, more unit cells could be arranged within the same area of action. Furthermore, the more compact braided structures enable the composite target bear larger impact load; thus, composites with larger braiding angle present the least damages
Data Availability
The test data used to support the findings of this study are included within the article.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This work was supported by the Fundamental Research Funds for the Central Universities (Grant number NS2016029).