JEC COMPOSITES MAGAZINE - Issue #37 - December 2007 - 57

The damage progression module utilizes composite micro-and macro-mechanics, laminate analysis, as well as cyclic loading durability analysis, and calls a finite element analysis module that uses anisotropic thick shell elements to model laminated composites. For impact simulation, the computational approach uses an explicit dynamic solution including the determination of the time history response (stresses, strains, displacements, contact force and impact energy). By imposing failure criteria locally within each micromechanics sub-volume, the methodology calls for time-dependent progressive damage and fracture simulations. At each time/load step, the stresses and strains from the finite element analysis are decomposed to assess failure in the material constituents. The data flow in a typical simulation cycle is depicted in figure 3. the rigid body to move towards the composite panel. Simulations were conducted in GENOA using LSDYNA as an explicit stress solver incorporating the methodology discussed earlier. The simulation predicted the total duration of the impact event as 20 ms and the impact energy as 24.54 lbf-ft (33.27 J). Figure 4 shows snapshots at two intervals (2.1 and 9.6 ms) obtained from the FE analysis. It shows that the damage in the composite panel initiated at 2.1 ms solely because of transverse out-of-plane shear stress and further accumulated because of longitudinal out-of-plane shear stress until 9.6 ms. Fig. 4: Snapshots of dynamic FE simulations showing damage initiation and propagation along with damage modes at (a) 2.1 ms and (b) 9.6 ms Fig. 3: Data flow in a typical simulation cycle The damaged properties are progressively updated to the timemarching algorithm until the structure collapses. Hence, the complicated behaviour of composite structures, which usually involves large deformation and extensive material degradation, can be accurately predicted with this module. If damage is too great at a particular load step or if convergence is not achieved, the time step size is automatically reconfigured. Figures 5a and 5b compares the simulation and experimental contact force versus time and impact energy versus time curves, respectively. The damage initiated at 2.1 ms and propagated until peak at 9.6 ms; thereafter, the rigid body bounced back and the contact force was reduced to zero at 20 ms. According to both the experiment and simulation results, the maximum contact force reached approximately 1,530 lbf (6,805 N). Generally, the predicted impact energy was consistent with that of the test. Results and discussion A foam core composite panel (279.4 x 254 x 14.1 mm) is subjected to impact loads from a high-speed, rigid hemispherical (diameter = 25.4 mm) piece followed by in-plane compression until catastrophic failure of the panel occurs. The composite panel is made of G30-500/R3676 face-sheets that are 1.42 mm thick and a core made of Rohacel 200W foam core that is 12.7 mm thick. Each face-sheet is made up of four plies [0/90/-45/+45], where each ply has a thickness of 0.35 mm. A total of 3,520 shell elements were used to form the mesh for the composite panel FE model. The rigid hemisphere, weighing 0.014 kg, was assigned with stiffer material properties and a velocity of 1,652 mm/sec towards the centre of the constrained panel. A total of 192 shell elements were used to form the mesh for the rigid body FE model. For the impact analysis, both translational and rotational degrees of freedom were initially constrained ~63.5 mm away from the edges, while all rotational and two translational degrees of freedom were constrained for the rigid body. The unconstrained translational degree of freedom allowed Fig. 5: (a) Load versus time; (b) Impact energy versus time No37 December 2007 / JEC Composites Magazine

JEC COMPOSITES MAGAZINE - Issue #37 - December 2007

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #37 - December 2007