Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE credits
During construction of large sandwich structures one of or perhaps the greatest difficulty lies increating good joints. The joint is often the first component to fail and constitute a large part of theweight of the structure and production workload. The study is focused on investigating if noveljoining methods could affect how military vessels are able to handle an internal blast withoutcatastrophic damage to the sections surrounding the area of the explosion. If the conventionalmethod of joining sandwich panels was to be used the result of an internal blast would mostprobably be a complete failure of the surrounding joints. This is based on that the conventional Xjointslacks continuous fibers through the joint in at least one direction and are therefore weak whensubjected to tensile forces in that direction. The belief is that a joint with continuous fibers wouldhave a smaller risk of failure or at least the failure would be less severe than with the conventionaljoints and could therefore maintain its structural integrity until reparations can be made.The purpose of this study is to manufacture small scale samples of novel X-joint design concepts.These concepts are then compared by manufacturability, tensile strength in either direction of thejoint and the joints flexibility. How well the promising novel concepts handle an internal blast cannotbe seen in this study, since full scale samples and advanced blast trials would be required.Six novel methods for joining of composite sandwich panels and a reference conventional joint,referred to as the standard joint, were manufactured and evaluated through three tensile tests. Outof the six concepts, all except the 3D-woven joint was found to have advantages over the standardtype, which only surpassed the novel joints in the aspect of bulkhead strength, and the lath and thebundle joint concepts were found to be the most promising. The tensile strength tests gave that theultimate strength of lath joint was 86% of the reference value in its deck direction and at least 75%in the bulkhead direction when reinforced with laths (49% unreinforced), where the reference value(100%) was the ultimate strength of the standard joint bulkhead. The tensile test results for thebundle joint, which did not have a reinforced bulkhead, reached 77% and 66% (deck respectivelybulkhead) of the reference strength value. These results supports the theory that the circular holesof the bundle concept has less impact on the strength of the bulkhead than the rectangular holes ofthe lath concept. The fiber bundles, however, are difficult and time consuming to work with,especially since the fiber bundles needs to be flattened and spread out to increase the area ofattachment when adhered to the sandwich panels. In contrast, the lath pieces are easy tomanufacture and work with during the final assembly and therefore more suitable to use whenreinforcement of the bulkhead is required.The joint flexibility test indicated that the most flexible joint was, as might be expected, the taperedfinger joint. If the tapered panel could be designed to counteract or avoid delamination of thetapering area and if the panels of the joint could be prevented from bending to the point fracture ofthe single skin laminates, this concept could be a suitable solution for confining a blast in terms ofrapidly developing membrane forces. The lath and the bundle joints were once again the strongestand reacted very similar. Unfortunately no difference could be found between the two jointconcepts because the hinges used to attach the samples in the test broke before the samples werebroken.Since the fiber properties and amounts were not compared it is not definite that either of the lath orthe bundle concept is better than the other in terms of mechanical properties. The conclusion is thatboth these two concepts work, are relatively easy to produce and have a far greater potential thanthat of the standard joint, in the internal blast situation and as a joining method in general.
2013. , 26 p.