Abstract

Engineering structures in aviation and marine experience complex loads and are often affected by repeated impact loads. The damage accumulated from deflection in the process of repeated impact will often seriously affect the safety of the structure. In this research, the low-velocity repeated impacts behavior of corrugated core structures was investigated through experimental methods. A series of low-velocity repeated impact tests were carried out to study impact resistance, taking into account the effects of varied impact sites and impact energy levels and the effects of different impact locations and different levels of impact energy. It is also observed from the test that the upper panel played a crucial role and experienced the coupling mode of the local indentations and global bending deformations during the repeated impacts tests. Three different failure modes were observed when changing the impact energies and impact locations. Moreover, the “pseudo-shakedown” phenomenon was also found when the energy of the impacts is 10 J on the short span. The present surveys provide a reliable method and insight into the dynamic response of the aluminum corrugated core structures when subjected to low-velocity repeated impacts, which could be a significant guideline for the investigation and lightweight design of shipping and aviation.

1. Introduction

The low-velocity repeated impact on engineering structure is a non-negligible factor for the safety of the structure. The ships and marine structures were subjected to repeated impacts caused by slamming, ice floe striking, and dropped objects [1]. The helicopter deck suffered from repeated impacts unavoidably when the helicopters take off and load, and the vehicle deck also has similar loadings when vehicles pass through. Meanwhile, airplanes and helicopters could be subjected to repeated impacts during a hard landing, heavy loading, or taking-off [2]. For metallic materials, plasticity could be accumulated with the repeated impacts, causing the change in structure stiffness and elastic energy in each impact. Therefore, it is of great significance to evaluate the impact resistance of the structure under repeated impacts.

Research on the dynamic behavior of the offshore structures under repeated impacts could date back to the 1970s. Jones [3] and Wei [4] conducted a rigid-plastic method for preliminary calculation of the plastic deformation of plates and beams under repeated identical slamming impulses. Soon after, theoretical research on the dynamic responses of plates under repeated mass impact loadings was extended, which considered the effect of the masses deposited on the plate surface [5]. Zhu and Faulkner [1] studied the dynamic behavior of rectangular plates under repeated impacts of a wedge impactor experimentally and theoretically, and a rigid perfectly plastic solution to estimate the dynamic response of the plate was provided. Huang et al. [6] discussed the phenomenon of pseudo-shakedown of aluminum alloy circular plates under repeated impacts based on the energy criterion, which had a good agreement with experimental investigations. Cho et al. [7] and Truong et al. [8] proposed experimental and numerical investigations on the mechanical property of metal beams subjected to repeated lateral impacts at room and subzero temperatures, which found that the incremental permanent deflections dramatically decreased for the constrained beam with the impact number increasing. Xu and Guedes [9] experimentally investigated the impact resistance of mild steel beams and aluminum alloy beams under repeated drop weight impacts. They observed the pseudo-shakedown phenomenon during the tests, which was confirmed by Wei [4] and Huang et al. [6]. Shi [10, 11] proposed an ideal elastic-plastic analytical method for predicting the deformation and the structural stiffness of rectangular plates and beams made of metal material under repeatedly uniform pressure loadings. Duan [12] and Zeng [13] investigated the mechanical behavior of aluminum alloy plates and mild steel plates with precast cracks under repeated drop weight impacts. They showed that the initial crack reduced the energy absorption of thin plates and leaded plates become more sensitive with the increase in impact number.

Because the composite structures have many excellent properties and are extensively used, the dynamic behavior of composite structures under repeated impacts is being studied more closely. Hosur et al. [14] experimentally researched the dynamic behavior of stitched/unstitched S2-glass/epoxy woven fabric composites under single and repeated low-velocity impacts. Giovanni et al. [15] used data obtained for repeated impact experiments on seven laminates of different layups and thicknesses and proposed the damage index (DI) to evaluate the damage of thick laminates in the low-velocity impact process. Atas et al. [16, 17] experimentally researched the repeated low-velocity impact response of E-glass/epoxy composites, considering the effects of thermal aging and presenting a power regression formula to predict the number of perforation for different thicknesses under lower impact energies. Dogan [18] experimentally investigated the dynamic behavior of thermoset and thermoplastic matrix-based composites under repeated low-velocity impacts by using conical impactor and hemispherical impactor. Katunin et al. [19] proposed a method to assess fracture mechanisms and damage development of glass/epoxy composites through repeated impact load tests and developed a degradation model based on the theoretical fundamentals of damage mechanics. Zhou [20] predicted the repeated impact response and propagation behaviors of cross-ply composite laminates through finite element simulations with ABAQUS/Explicit.

Sandwich structures have a wide application prospect range in the field of aerospace, automotive, construction, and marine structures, for example, honeycomb [2, 21, 22], foams [23], lattice core [2426], and cellular metals [27, 28]. Among all sandwich-type panel structures, the corrugated core sandwich structure is a valuable industrial solution because of the high stiffness-to-mass ratio and outstanding ventilation characteristics [29]. Scholars have recently investigated the mechanical characteristics of sandwich constructions subjected to repeated impacts. Akatay et al. [2] studied the dynamic behavior of honeycomb sandwich structures after multiple impacts and contributed to evaluating the residual compressive capabilities of honeycomb sandwich structures that had been exposed to repeated impacts testing. Balcı [21] proposed experimental investigations on the performance of repaired honeycomb sandwiches under repeated impacts. With experimental measurement and analytical technology, Ozdemir et al. [23] conducted the repeatedly impact resistance of sandwich structures made up of E-glass/epoxy face sheets and balsa cores. Guo et al. [27, 28] examined the repeated impact resistance of aluminum foam sandwich plates experimentally and computationally. The results show good agreement and point out that the thickness of the front panel is the key to the impact resistance of aluminum foam sandwich structures. Meanwhile, the mechanical behavior of corrugated core sandwich structures is mainly concentrated on the single impact dynamic [3032], few investigations were made into the dynamic behavior of repeatedly impacted corrugated core sandwich structures.

In this study, the dynamic behavior of repeatedly impacted sandwich structures made up of aluminum alloy panel and aluminum alloy corrugated core was investigated experimentally. The drop hammer tests under three different impact energy levels were performed. The long span and the short span were chosen to gain a more comprehensive understanding of their impact resistance property. Then, the time history of impact force, time history of energy absorption, impact number of peak force, and the energy absorption curves were compared and analyzed. The conclusions about the effects of energy levels, impact locations, and impact numbers were summarized.

2. Specimens and Experimental Procedure

2.1. Specimens and Materials

Two aluminum alloy panels and an aluminum alloy corrugated core make up the sandwich samples with three unit cells. Using 2A12-T4 aluminum alloy sheets as material, the trapezoidal corrugated cores are prepared by using a press with a controlled pressure system and the folding technique. Then, the corrugated core is adhered to the panel at room temperature under constant pressure for at least 24 hours using a two-part epoxy adhesive. The detailed manufacturing process of the sandwich structure is depicted in Figure 1. The actual picture of the corrugated core sandwich structure is shown in Figure 2(b).

The basic forms and structural parameters of corrugated sandwich panels are depicted in Figure 2(a). The in-plane dimension of the sandwich plate was 100 mm × 100 mm. The core parameters are L1 = 7 mm, L2 = 25 mm, t = 0.5 mm, and h = 15 mm, and Both upper and lower face sheets’ thicknesses were calculated as 2.0 mm.

The uniaxial tensile response of aluminum alloy sheets of different thicknesses was tested with the universal tensile test machine at a displacement rate of 2 mm/min, according to the regulations in Chinese standard GB/T 228-2010. The mechanical properties of the aluminum alloy sheets are listed in Table 1. The relationships between nominal stress-strain and true stress-strain were obtained by tensile test, as shown in Figure 3. It is observed from Figure 3 and Table 1 that aluminum alloys of different thicknesses have a similar stress-strain curve and yield strength, and the thicker one has a higher fracture strain. The adhesive is the ductile Araldite® 2015, which is widely used in engineering, and its properties are given in Table 2.

2.2. Experimental Procedure

The repeated impact experiment of corrugated core sandwich samples was conducted by instrumented drop hammer impact tower, as shown in Figure 4. The indenter consists of the additional mass, the force transducer, and the impactor. The test tower has two guide rails, which ensure that the indenter can fall vertically. The specimens were fixed by the rigid clamper, and a rebound catcher was installed to catch the impactor preventing the rebound hammer from being affected by two unexpected impacts in the process of a drop hammer.

In order to obtain the dynamic behavior of corrugated sandwiches under multiple impact loads, a series of impact energy levels (10, 20, and 40 J) was achieved. Two representative impact sites are selected on the short span and long span respectively to explore the impact resistance of corrugated sandwiches, as shown in Figure 5. During impact tests, the total drop hammer weight of all impact tests is constant and is 13.268 kg, and the impact energies are changed by adjusting the height of the drop hammer. The impactor has a hemispherical steelhead and is 16 mm diameter, which could be assumed to be a rigid body due to the use of rod machining and surface hardening manufacturing technology. The data acquisition system with a sampling frequency of 1.25 MHz was used to collect the force-time data from the force transducer, and other measured results including the velocity-time, displacement-time, and absorbed energy-time were computed by it.

3. Results and Discussion

3.1. Impacts on the Long Span
3.1.1. When Energies Are 10 J

When the energy of the repeated impact is the lowest level of 10 J, the typical impact force-time, absorbed energy-time, impact force-impact number, and energy absorption ratio-impact number of corrugated core sandwich structures are drawn in Figure 6. During the 1st impact event, the peak force is 5.8 kN and the contact time of impact is about 6.8 ms, according to the impact force-time curves. A minor decline in the impact force-time curve occurs after the force reaches its greatest value from the 2nd to the 7th impact event. According to the analysis, the repeatedly impacted corrugated core being more sensitive to buckling led to this phenomenon, and the corrugated core trend was steady after the 8th impact event. With the increasing number of impacts, the peak impact force also tends to increase, and the maximum impact force is about 10 kN at the 8th impact event. Then, the impact force-time curves tended to be stable until the upper panel fractured at the 18th. Finally, the peak force suddenly decreased to 9.6 kN and the contact time is about 6.5 ms at the 18th impact event. By observation and analysis of Figures 6(b) and 6(c), the energy absorption ratio continues to decline with the increasing number of impacts. After the 10th impact event, the energy absorption ratio tends to be around 15%, indicating that the indenter's energy is absorbed by a rebound in each impact event. At the 18th impact event, the failure of the upper panel causes more energy to be absorbed, which makes the energy absorption rate rise to about 45%.

With the increase of impact times, indentation appears on the upper panel at the location of impact and corrugated cores begin to plastic buckling the cells under the impactor, which is stable after the 10th time. When the number of impacts reaches 18, an obvious crack was found on the upper face and local plastic buckling for corrugated cores, as shown in Figure 7. However, no damage was found on the bottom panel.

3.1.2. When Energies Are 20 J

The typical impact force-time, absorbed energy-time, impact force-impact number, and energy absorption ratio-impact number of corrugated core sandwich structures are shown in Figure 8 when the energy of the repeated impact increases to 20 J.

As shown in Figures 8(a) and 8(c), the peak force increases from 7.2 kN in the 1st impact event to 11 kN in the 5th impact event and decreases to 10.3 kN in the 7th impact event. According to Figures 8(b) and 8(c), the energy absorption ratio of the corrugated sandwich structure decreases from 90.7% in the 1st impact event to 60.7% in the 5th impact event and then increases to 67.4% in the 7th impact event.

As displayed in Figure 9, due to the groove on the upper panel caused by the impactor, the impact force will have a short period of stability from the 2nd to the 4th impact event. After the 5th impact event, the impact force will first slowly decline for some time and rapidly decline. Because before the 5th impact event, no obvious adhesive layer debonding occurs, there is mainly the upper panel to absorb the kinetic energy caused by drop-hammer impacting; during the 5th impact event, the adhesive layer begins to debone and the corrugated core appears obvious global buckling. The corrugated cores deformation led to absorption energy increases in the 6th to 7th impact event.

3.1.3. When Energies Are 40 J

When the energy of the repeated impacts increases to 40 J, the impact force-time and the energy absorption-time of corrugated core sandwich structures are shown in Figure 10. The peak force is 8.9 kN and the contact time of impact is about 7.2 ms in the first impact event, and the peak force is 8.8 kN and the contact time of impact is about 12 ms in the second impact event, which could be obtained from the impact force-time curves.

The force would decline abruptly after reaching the initial peak point and then a load plateau region with stable fluctuations would appear, as illustrated in Figure 10. The impact rebounded and about 85% of the kinetic energy is absorbed in the first impact event. It was easily observed that the impactor penetrated through the upper panel in the second impact event. However, no obvious fracture has been found on the lower face sheet and corrugated cores, as shown in Figure 11.

3.2. Impacts on the Short Span
3.2.1. When Energies Are 10 J

Figure 12 shows the impact force-time, absorbed energy-time, impact force-impact number, and energy absorption ratio-impact number of corrugated core structures when the energy of the repeated impacts is 10 J. The peak impact force is 6.1 kN and the contact time of impact is about 7.5 ms in the first impact event, and then it gradually increases to about 10.2 kN at the 21st impact event. The “pseudo-shakedown” phenomenon was also observed, and the energy absorption ratio remained at about 15%, as depicted in Figure 12(b).

As shown in Figure 13, the local pit of the upper panel and the plastic buckling of corrugated cores rise dramatically as the low-velocity impact continues. After 20 times of impact events, the local pit of the upper panel and the plastic buckling of corrugated cores remain stable after each impact.

3.2.2. When Energies Are 20 J

As the energy of the repeated impacts increases to 20 J, the peak impact force is 6.1 kN and the contact time of impact is about 7.5 ms in the first impact event, and then it gradually increases to about 10 kN at the 4th impact event. Finally, the upper panel fractured and the peak impact force is about 7.6 kN at the 7th impact event, as shown in Figure 14.

When the force reaches its first peak point, the force-time curve drops sharply and a load plateau region follows this in each impact event. Because the corrugated cores begin to plastic buckle the area under the impactor. It indicates that the upper panel and the corrugated core are both played the important role in energy absorption. At each impact, the energy absorption rate of corrugated structures can reach more than 77%, and these data are shown in Figures 14(b) and 14(c).

Indentation appears at the impact position of the upper panel. Meanwhile, the corrugated cores begin to plastic buckle the cells under the impactor, as shown in Figure 15. After the 5th impact event, there is an obvious local plastic buckling for corrugated cores, and three visible cracks were found in the region of the top panel in the 7th impact event, as shown in Figure 15.

3.2.3. When Energies Are 40 J

When the energy of repeated impacts increases to 40 J, a single impact could cause a fracture to appear on the upper panel and a rebound emerges. The peak impact is 8.9 kN and the absorption of energy is 90%, as shown in Figure 16.

When the force hits its initial peak point, the force-time curve has a small decline. Then, the force curve drops sharply from the highest impact force, followed by a load plateau and little fluctuation. As illustrated in Figure 17, the reductions are mostly due to complicated damage in the impact region of the top panel and the plastic buckling of the core members. Visual results show that there is an obvious circular indentation in the impact area of the top panel. Under the action of the impactor, not only the top panel has been damaged, but also the corrugated core has plastic buckling, while no core members are penetrated.

4. Discussion

Through the repeated impact tests on aluminum corrugated core sandwich structures at different positions and energies, the number of impact failures, maximum impact force, and failure modes are summarized in Table 3.

Impacting on the long span of samples, when the impact energies are 10 J, the impact force increases with the increase of the impact number, which increases to 10.42 kN after the 12th time and then tends to be stable until the 18th time of crack appears on the upper panel. When the impact energies are 20 J, the number of impact damage is reduced to 8 times, and the failure mode is changed into corrugated core global buckling caused by debonding of the adhesive position on both sides of the impactor; there is a pit on the upper panel and no crack was found. When the impact energy increases to 40 J, the ideal rebound occurs during the first impact event, and the impactor penetrates through the upper panel during the second impact event.

Impacting on the short span of samples, when the impact energies are 10 J, the impact number reached 15 times, and the force and absorbed energy of each impact remain stable. The energy absorbed at each time is very small, and the structure is considered to be in “pseudo-shakedown.” When the impact energies are 20 J, the corrugated core layer can participate more in the impact energy absorption, so the impact energy absorption is slightly more than that of the long span. Due to the local stiffness, cracks appear on the upper panel at the 7th impact. When the impact energies are 40 J, due to the local stiffness, a local crack on the upper panel appeared during the first impact.

Significantly, the impact energy level plays the most critical role when the punch strikes the aluminum corrugated core sandwich structures. The core buckling and the fracture on the upper panels were identified as the predominant damage mechanisms at lower impact energy levels. The punch penetrates directly into the upper panels at the higher impact energy levels. The bending deformations of the upper panels and corrugated core continued to increase as the number of repeated impacts increased. The energy absorption ratio declined significantly, and they tend to be stable until the local crack appears on the upper panels.

5. Conclusions

In this study, the dynamic behavior of aluminum corrugated core structures under repeated impact loadings was experimentally investigated. The mechanical properties of the aluminum alloy have been tested, and repeated impact experiments for corrugated core structures with three energy levels in two different locations have been carried out. The results of impact force, impact number, and absorbed energy, among others, are given and evaluated in depth. The following conclusions have been reached based on the results of repeated impact experiments:(1)The stiffness of corrugated plates will be significantly improved after the first impact with the maximum energy absorption. As the number of repeated effects increases, it will continue to harden in subsequent impacts until they are damaged or stabilized.(2)The upper panel experiences the coupling mode of the local indentations and global bending deformations, which play the most important part, whereas the bottom panel displays only the global bending deformation model, which is almost negligible compared to the upper panel. The deformation mode of corrugated cores is embodied as plastic buckling in the cells under the impactor and global buckling under any circumstances.(3)Different energies and impact locations would lead to three different failure modes: crack on the upper panel, penetration, and global buckling of corrugated core. The “pseudo-shakedown” phenomenon was also found when the energy of the impacts is 10 J on the short span.

The results obtained in this study provide a useful analysis of the damage mechanism of aluminum corrugated core sandwich structures under repeated impacts, which would lead to a guideline for composite structures used in shipping and aviation.

Data Availability

The authors confirm that the data supporting the findings of this study are available within the article.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

The authors acknowledge the financial support received from the National Natural Science Foundation of China (grant no. 52071150).