Page 111 - MSAM-3-4
P. 111

Materials Science in Additive Manufacturing                           Bistable 3D-printed compliant structure




            A                                                B




















            C
















            Figure 7. Structural response of specimens in Group 1 under quasi-static compression. (A) Force-displacement curves obtained from the experiment for
            three designs in Group 1. Two sets of experiments were conducted, including only the bottom surface of the specimens taped to the compression plate and
            both the top and bottom surfaces taped. The shaded area denotes the standard error among three repeating tests. Shaded areas represent the deviations
            among three tests for each design. (B) Comparison between experimental results and FE results in terms of force-displacement curves, with only the
            bottom surface taped to the compression plate. (C) Deformations and corresponding stress distributions of Design No. 6 (l’ = 30, h’ = 5, g’ = 1) under
            quasi-static compressive loading, obtained from the FE simulation. A: First snap-through starting point, B: First snap-through ending point, C: Second
            snap-through staring point.
            suggesting the pulling-down force from the sudden snap-  shown in Figure 7B. Overall, the trend of the simulation
            through. This implies that the snap-throughs were strong   results agrees with the experimental results. However, the
            enough to generate transitions between the structures and   simulation could not capture the zero force. This could be
            another stable configuration.                      attributed to the simplification of the constitutive model
                                                               used in  the FE simulation.  As the material model  was
              The force-displacement curves obtained from the   simplified to be elastic-perfect plastic, the non-linearity
            second set of experiments could also explain the partial   of onyx material could not be accurately considered.
            bi-stability of the structures in Group  1. According to   Even though the snap-through instability is mainly about
            Figure 1E, the presence of the negative force indicates the   structural non-linearity, material non-linearity could
            bi-stability of the deformed element. With both pairs of   affect the overall behavior to some extent. As shown in the
            beams snapping through, two negative-stiffness phases   stress distribution (Figure 7C) of the specimen from the
            were captured in the force-displacement curves. However,   FE simulations, regions with high stress (higher than yield
            a negative force segment was only observed in one of the   strength) exist. This indicates the presence of non-linearity
            snap-throughs. In other words, only one pair of the curved   of the material during snap-through in the FE. Compared
            beams achieved bi-stability, and the other pair exhibited   to the stress distribution of the specimens in Group 2, in
            reversibility. This is consistent with the observation of the   which the material experienced linearity, the snap-through
            structures on the removal of the loading.          for specimens in Group 1 could not be captured accurately.
              To compare the results obtained from experiments and   Furthermore, the peak forces obtained in the FE model
            simulations, force-displacement curves were plotted, as   were greater than that of the experiment for all three cases.

            Volume 3 Issue 4 (2024)                         10                             doi: 10.36922/msam.4960
   106   107   108   109   110   111   112   113   114   115   116