Failure analysis of bolted and bonded carbon composite joints

Since the beginning of the civilization, humankind has been tried to utilize different type of materials simultaneously. As the result of these trials, it is quite obvious that composite materials are the most significant epitome. Areas of usage of composite materials are increasing day after day, just because of its lightness, design flexibility and so forth. The increase of areas of usage of composite materials created a need for researching their behavior under load with other composite materials either metallic materials. With the studies in last fifty years, accumulation of knowledge in this subject is spiked up. There are two diversified methods for the jointing of composite materials. First method consists of different types of mechanical joints and this method can be obtained using bolts, rivets and pins. Alternatively, there is mechanical joints obtained using butterfly shaped joints. However, these joints are not possible to apply for commercial products. On the other hand, joints obtained using bolts and rivets are commonly held in our day. Just because the process of jointing is easy and effortless in these joints. At the same time, it is easy to obtain the components for these processes and components which are used in these joints are compatible for the environmental conditions. In spite of that, material handling such as drilling holes on the composite material is an extra cost in these processes. Also extra weight of joint components is a disadvantage factor. Bonded joints are a lot better than mechanical joints considering that lower costs of material handling and lightness of joint. Another important advantage is that different adhesives can be developed for different mechanical properties. Because of these features, bonded joints are a good choice for jointing composite materials. Composite materials are used with aluminum and titanium because of their lightness. Joint strength between composite materials and metallic materials is limited with the minimum strength of jointing component or material. Component or material with the lower strength may fail such as adhesive, bolt, metallic material or composite material. As the literature research shows, there are multiple analytical models used for calculating joint strength. Strength calculation of mechanical joints is relatively easy while the strength calculation of adhesive joints is more complex and difficult. Moreover, when the irregularity and complexity of the composite material are added on top of it, calculation becomes more difficult. Therefore, more numerical solution methods are used to solve such problems. After the elastic zone of the materials, many material failure models were introduced to examine the plastic behavior. Failure models such as Von Mises, Mohr - Coulomb, and Drucker - Prager for adhesives have been developed, while Tsai - Wu, Hashin and Rotem have been developed for composites. Despite these theoretical studies, the most reliable methods of strength calculation in composite and adhesives are tests and experiments. xxviii In this study, mechanical and bonded joint strengths were calculated with experimental and numerical methods. Then these results were compared to each other. At the designing process of bonded joints, there were two different typical adhesives used. By this means, different adhesives were modelled. Double lap joint was chosen as the joint type. Literature review also shows that in composite bonding studies, it is generally preferred to use double lap or single lap joints. For the manufacturing of the specimens, 90 ° plain woven 200 gsm carbon fiber fabric was used. After the carbon fiber fabrics were cut to the appropriate size, carbon fiber plate manufacturing was carried out using epoxy based general purpose resin. Eleven layers of fiber were used to achieve the required thickness. Plates obtained as a result of hand lay-up were cut in accordance with ASTM D 3528 standard. Metal, another component of the connection, has been chosen as aluminum because it is often used with composite materials. Aluminum materials with 6063 T5 code were cut in accordance with the standards and drilled holes on the materials which will be connected with. M4 bolts made of A-2 quality stainless steel are used for bonding the materials with bolts. Fastened with M4 nuts by using washers. The bolts were tightened only to keep the joint, no extra torque is applied. Araldite 2011 and Hysol 3450 adhesives -which are used in structural bonding-, were preferred for bonded joints. After the bonding surfaces of the composite and aluminum were prepared, the materials were stuck each other in laboratory conditions. The specimens were allowed to cure for 24 hours. A total of nine different specimens were prepared for three different cases. Tensile tests were then carried out at the speed of 1.27 mm / min in the conditioned environment. For the experiments to be simulated in a computer environment, 3D Cad models of the specimens were created in SolidWorks 2017 program. HyperMesh 13.0 and HyperCrash 13.0 programs, which are commercial FEA programs, are used to create finite elements according to 3D Cad models. Regardless of the program used, it is essential that modeling is done correctly to observe the fail conditions of the materials and to obtain the correct results as much as possible. To ensure these conditions and to be able to observe the effect of delamination in the event of possible delamination, all composite layers, interlayer resin layers, adhesives and bolts are modeled using three dimensional hex elements. After the material properties that obtained from the manufacturers and similar studies are entered into the program, material failure criterias are defined for all materials. Applied failure criterias were selected by studying current studies. Radioss 13.0 was used to solve the generated models and HyperView 13.0 was used to evaluate the results. The results were compared with each other and the experimental results. When the previous studies are examined, it has been seen that the joint strengths are compared with the joints strength obtained by dividing the maximum tensile force by the overlap area. In this study, the cases mentioned in the study were compared according to the joint strengths. As a result; bolted joints and bonded joints are compared each other according to their experimental and numerical results of joint strengths. Experimental results showed that the joint strength and the elongation at break in the bonded joints vary widely from specimen to specimen. On the contrary, these values are more stable in the mechanical joints. In the modeling of adhesives with finite elements, different characters of adhesives prevent the adhesives from being modeled with a single fail criteria. While the ductile adhesive gives more accurate results with the energy criteria, it gives more accurate results with the brittle adhesive strain criteria. xxix Specimens that bonded with Araldite 2011, which is a more ductile and strong adhesive, were almost as durable as bolted specimens. On the other hand, the specimens bonded with Hysol 3450 could not show the same durability. Both adhesives were found to have a very high rate of damage progression after the beginning of the fail. This leads to a sudden breakdown of the joints and jeopardizes the system integrity. When the distribution of stresses in the joints is examined, it is observed that there is a uniform stress distribution in the bonded joints and the notch effect and stress concentrations in the bolted joints due to the bolt All these results show that the bonded joints are as rigid as at least bolted joints in case of selecting the appropriate adhesive. Although there are advantages and disadvantages to each other, both types of joints will continue to be used in composite systems. However, the sudden fracture behavior of adhesives, manufacturing processes and bond strengths which are very sensitive to environmental conditions will continue to limit the use of adhesives.