Available at: https://digitalcommons.calpoly.edu/theses/428
Date of Award
MS in Aerospace Engineering
This thesis involves the development of a fiberglass-foam composite sandwich structure with the introduction of delamination arrestment keys; therefore, a study of an initially delaminated composite sandwich structure was the experimental analysis on multiple configurations in how the arrestment keys are placed.
The first part of this thesis research was to the experimental design and manufacturing of the composite sandwich plates. These plates were later cut down to the specific test dimensions and manufacturing processes for the composite sandwich plates and test specimens were created. The composite sandwich plates were manufactured using a vacuum resin infusion process. The dimensions of the composite layup are 14 inches in length with a width of 10.75 inches. The width size has margin to account for machining. The actual dimensions of the test specimen after it is prepared are 14 inches by 0.75 inches. The test anvil length is 11 inches and is used to perform tests to determine mechanical characteristics of the structures under buckling loading. These plates provide approximately 9 to 13 specimens per each case. All the test specimens have 4 plies of 18 oz fiberglass woven roving fabric from Jamestown Distributors, a LAST-A-FOAM FR-6710 foam core, and 5 to 1 ratio of West Systems 105/206 epoxy. Also, a non-porous material was integrated into the structure to create an initial delamination in some of the case studies. The integration of the delamination arrestment keys involve milling the foam core to provide the necessary grooves for key placements before the structure is vacuumed and epoxy is flowed. The arrestment keys are made of unidirectional fiberglass strand and the West Systems 105/206 epoxy using a wet layup process. In addition, fiberglass woven roving specimens were created to see the material characteristics under compression and tensile loading. The same is created to determine the material properties of the foam core, wood boundary core, and arrestment keys under compression loading.
The second part to this thesis investigation is the experimental testing of the test specimens with all different variables considered. Those variables includes determining the final solid cure duration of the fiberglass skin, the geometric lengths between pure compression and pure buckling, behaviors of different initial delamination size, effects of continuous and discontinuous arrestment keys parallel and perpendicular to the in-plane loading, and material properties. The final solid cure duration differ from what the manufacturer gave on their epoxy. This experiment testing followed ASTM D-3039 standard to see the differences in elastic modulus over duration of 15 days. The resulting data shows that the test specimen fully cures after 13 to 14 days. The test specimens in search of the geometric buckling length for this investigation did not follow ASTM C-364 standard in full, but follows a variation of the ASTM C-364 standard in order to support buckling loading condition and the limited accessibility of the test equipments. Instead, the modifications are found with a different test jig design and test specimen configuration. The test jig was created to provide a pinned condition with a 0.25 inch diameter. The test specimen is laid up with a foam and wood cores. Two wood cores are laid at each edge of the foam core to increase loading capacity and holes are drilled through the wood cores to create a pinned-pinned case for the optimum buckling condition. The results detailing the geometric buckling show that after 9 inches anvil length there is no compression; only buckling occurs with a cross-sectional dimension of 0.75 inch by 0.575 inch. The 11 inch foam length was chosen for convenience of machining. This modified setup was also used for testing the different configuration with the embedded arrestment keys. The multiple different configurations completed for these test specimens under unstable loading, the experiment results show that a continuous arrestment key embedded significantly improve the loading capacity over a perfectly sound non-delaminated specimen and maintain the majority of loading capacity even with an introduced delamination. The embedded continuous key also provided a higher horizontal displacement capability before fracture in comparison to the initially delaminated test specimens. As for the test specimens used to determine the material characteristics, ASTM D-3410 and modified ASTM C-364 standards were followed. The test specimens had a fiber volume fraction of approximately 0.60, which details the brittle failure under tensile and compression loading. The results also show that the fibrous fiberglass test specimens have a higher ultimate strength in compression or buckling then in tension.
All of the experimental testing was completed in the Aerospace Engineering Structural/Composite Lab at California Polytechnic State University at San Luis Obispo, California. Therefore, an introduction of a continuous arrestment key parallel to the in-plane loading and embedded into the composite sandwich structure provided a significant increase in loading and buckling capabilities in comparison to the control test specimens with and without an initial delamination and no embedded key. The continuous key placed parallel to the load vector increased the structural strength with an increase of 126% from a 1-inch delaminated structures and only an 11% drop from non-delaminated structures. That is, 1-inch and 2-inch delaminated structures showed a 61% drop and 81% drop from non-delaminated structures. Some configurations have reduced or arrested of the delaminated region.