PSI - Issue 52

Satrio Wicaksono et al. / Procedia Structural Integrity 52 (2024) 438–454

449

12

Satrio Wicaksono et al. / Structural Integrity Procedia 00 (2023) 000 – 000

lost. Therefore, the time for the failure to reach the vertical sandwich area is longer. This phenomenon is also seen in the kinetic energy curve against time. Additionally, in the test results, the maximum contact force occurs when the foam begins to solidify to a certain thickness and is pushed along with the impactor towards the vertical sandwich.

Kinetic Energy [J]

Contact Force [kN]

Time [ms] (a)

Time [ms] (b)

Figure 10 Comparison of (a) kinetic energy history plot and (b) contact force history for variations of element deletion on the core component of horizontal sandwich structure.

The other noticeable difference between the numerical model result in comparison to the experimental results is the area marked by green circle in Figure 9. It can be seen that in the numerical results, the contact force reached 0 kN at 2.5 ms before finally rising again, which is due to the vibration when the crippling deformation in the vertical sandwich begins to occur. When crippling started to occurred in the vertical sandwich, there will be a significant reduction in the strength of aluminum skin, which causes vibration in the T-joint. This vibration causes the impactor to experience a contact force of 0 kN just before finally touching the T-joint again. In the case of the experimental result, the T-joint is sufficiently stiff and not deformed much so that it does not experience vibration, causing the contact force not to drop to 0 kN. This vibration effect is influenced by the deformation process that occurs. If the deformation is large enough in a short time, vibration will occur in the T-joint. The last noticeable difference between the numerical model and experimental results is marked with double-sided red arrows shown in Figure 9. The numerical results show longer time for the impactor to be detached from the T joint. This difference is due to the assumption that the materials in the numerical model was weaker than the actual ones used in the experiment, especially in the material properties which directly related to the stiffness of T-joint structure. This is also supported by the results of the referred research in Table 4 that compares visualizations of failure at kinetic energies of 20 J, 40 J, and 60 J. The results show that the numerical model has succeeded in predicting the failure modes such as core crushing, core shearing, and delamination in certain areas. Additionally, the maximum contact force obtained from the numerical data also shows a relatively small difference of 2.03% at 2 ms in comparison to the experimental result. However, the numerical results show greater deformations and failure in comparison to the test results. Again, these phenomena may occur due to the assumption of material properties used in the simulation. Subsequently, those differences can be minimized by conducting a deeper study on material properties that should be used in the model.

3.2. Variation of PVC 70.75 mechanical properties

3.2.1. Variation of yield strength ratio Yield strength ratio ( k ) is the ratio between the initial yield stress from the axial compression test to the initial yield stress from the hydrostatic compression test. The value of k must be in the range 0 < k < 3 [11]. Based isotropic hardening principle, materials with a Poisson ’s ratio of around 0.3 will have k value of 1.1.