| The work leading up to this thesis dealt in particular with the fabrication of composite sandwich panels for use in marine structures. Methods for making complex sandwich panels and the forming of the foam cores for complex panels were experimentally studied.;There is scarce information on hydrostatically loaded doubly curved composite sandwich panels. In particular their fabrication and the affect curvature has on their performance.;Doubly curved composite sandwich panels loaded by evenly distributed pressure were designed, analyzed, manufactured and tested. Quick and cost effective methods for making molds for vacuum infused doubly curved sandwich panels were studied and implemented. Several different manufacturing techniques for making doubly curved panels and doubly curved foam cores were investigated. Tests were performed using a hydrostatic water tank.;Doubly curved composite sandwich panels with a thin light foam core and stiff skins in comparison to a flat panel is much more affected by an increase in curvature than a panel with a thick, denser foam core and less stiff skins. A time and cost effective method for fabricating a doubly curved panel is to first CNC route the shape from a high density foam, then CNC route the foam core flat and thermoform it to shape in the mold, then cover the mold with either epoxy or vacuum bag and layup the panel and vacuum infuse. For the panel to be watertight care must be taken during vacuum infusion so there are no air leaks. The predicted results showed good correlation to experimental results, within 1.2% difference for deflection.;Previous work has been done on a hybrid hull concept for a five-sided ship hull cross section, having a steel frame designed to take the global bending loads, and bonded fiberglass sandwich panels designed to take shear and water pressure loads. Wave induced loads were calculated using the American Bureau of Shipping (ABS) guidelines for a 142m (water line length) destroyer class ship, and applied to a 1:35 scale model. This model had a constant cross section and it was tested in a six-point bending setup. The model exhibited excellent global ductility, something that would not be expected from an all-composite structure. The beam was tested to 36% above the design load, which resulted in yielding within the steel truss and permanent global deformation but otherwise no damage. In particular there were no failures within the composite panels or bonded joints. This work has been extended to build a much larger 1:8 scale hybrid hull and study its fatigue performance.;Both conventional (flat) and corrugated skin sandwich panels with the size 1.34 m x 1.59 m were designed and manufactured for a large hybrid ship hull specimen. This specimen had the geometric shape of an existing bridge girder for a magnetically levitated train, which was retrofitted and tested as a 1:8 scale model of a destroyer class ship. A three point bending fatigue testing fixture was developed, and the specimen was fatigue tested for 325,000 cycles. Cracks were found in the steel of the specimen, but almost no damage occurred in the bonded joints or the composite panels. The flat sandwich panels were very much lighter than the steel they replaced, and the corrugated panels were even 15% lighter than the flat panels.;A steel/composite hybrid hull for an 8.75m, high-speed boat is currently being manufactured. This craft will serve as an instrumented slamming load test facility. A method has been developed to measure strains on the outer skins without having strain gages attached to the outside; gages on the outside could be damaged during testing. The strain gages were placed between the outer skin and the foam core of the sandwich panels. The manufacturing process of embedding strain gages inside composite sandwich panels is discussed. Two different test panels were fabricated, one to test the method of embedding strain gages inside sandwich panels and the other to compare the readings of the strain gages inside the panel verses outside. The difference between the embedded strain gages and strain gages mounted to the outside of the specimen were within 3%. |
