| Ever since the high-performance composite materials were widely used in the aerospace, composite fastened joints had been investigated as the key point of researches. Numerous studies showed that80%of the composite structures failure occurred at the joints, and the joints accounted for about25%of the total structure weight, so the connecting technology with fasteners has always been a major concern of composite structural design. As the application of the overall structure of composite materials, the usage of aircraft components and fasteners has been reduced greatly. However, whether it is common joint design in the past or modular design at present, bolted joints as the dominant fastening mechanisms are still needed to ensure the reliability and stability of the structure. The technology of fastened joints has been constantly innovated and developed. Interference-fit blind fastener, a new type of composite bolted joints, is ready to be used in aerospace field currently because of its excellent fatigue properties. However, its research on static strength, fatigue performance and failure mechanism of blind bolt are compatatively lagging behind the application. In order to maturely apply this in composite connecting structures, it is very important to carry out deep and wide researches.Today aerospace structural design not only requires high security, but also constantly seeks the maximum of economic efficiency. If the new fastener (blind bolt) is used in the designing progress of composite bolted joints in aircraft structures, in addition to meet the security of the structures, the fatigue life of the structures has to be considered. Thus it is theoretical important and practically valuable to carry out the researches on static strength, fatigue life and failure mechanisms of composite bolted joints. The main contents of this paper are organized as follows:1. In this paper, analytical method, finite element method and experimental method were conducted to study the stress distribution and static bearing strength at the edge of composite hole.(1) Analytical method. Under plane stress conditions, the model of anisotropic material fastened joints with interference fit was established, in which the complex potential function was solved by complex variable function method, and the stress components σx,σy,σxy at the edge of hole under different amount of interference fit was derived, while redial stress σrr, circumferential stress σθθ and shear stress σrθ at the edge of hole were derived by taking advantage of transforming corner formulas. According to the loading situation of connecting structures with interference fit, this paper establishes the contact equation of interference fit and loading, which solve the contact problem in the process of loading after interference fit between hole and bolt. In boundary condition of loading by bolt’s displacement, a complex potential function in the model was deduced, after solving the function the stress distribution at the edge of hole was obtained. According to the stress situation obtained from former model, the initial failure load and bearing failure strength of structure was calculated by introducing Tsai-Hill failure criterion and Hoffman failure criterion.(2) Finite element method. By utilizing finite element software ABAQUS, a three-dimensional progressive damage model of loading after interference fit and a analysis flowchart of model with interference fit were established. The shear nonlinearity of composite and the geometry nonlinear of contact were taken into account in this model, and a second-order reduction criteria based on single failure mode and mixed failure mode was established by introducing Hashin and Yamada-sun strength criterions. The second-order damage factor D1st,D12nd,D2,...,D6in the three-dimensional model were determined by the test data of clearance fit joint. The bearing response of0.5%interference fit structure was studied by finite element model, and comparing with the experimental results verified the rationality of the model and the analysis process.(3) Experimental method. According to the static bearing experiments of composite bolted joints with three sizes of interference fit,4%D bearing strength and ultimate strength obtained can be a basis to verify analytical method and finite element method. The results showed that the ultimate strength of composite bolted joint with0.5%interference fit is higher1.51%than the strength of the high-lock bolted joints with H9/h9clearance fit; however, the ultimate strength of composite bolted joint with3%interference fit is lower1.85%than the strength of the high-lock bolted joints.According to these three research methods (analytical method, finite element method, and experimental method), the bearing responses of bolted joints with three sizes of interference fit were comparatively analyzed. The predicted results from analytical method are generally higher than experimental results, just because it does not take into account the element failure at the edge of hole and the loss of material properties after the element failure; In finite element method, considering the progressive failure theories, the predicted results of4%D bearing strength and ultimate strength were in good agreement with the experimental results, and the error was within±3%.2. According to fatigue testing under several levels of cyclic stress, both fatigue properties of composite joints with high lock bolt and blind bolt were comparatively analyzed, and the effects of different interference-fit size on structural fatigue life of waere studied, and other impact factors, such as different sequences of main laminates and different materials of laps were also considered.Fatigue results showed that fatigue performance of0%interference fit fastened joints was better than that of high lock bolted joints. Because, at the same cyclic stress level, the former was1-2orders of magnitude higher than the latter and had more than50MPa ultimate fatigue stress. Different sizes of interference fit highly influence the fatigue life of composite bolted joints. At low level of cyclic stress (±45%σbru), the fatigue life of connecting structures increased with the size of interference fit (in the range from0%to3%), and reached to106at3%interference; However, at high level of cyclic stress (±55%σbru and±60%σbru), the fatigue life increased as the size of interference fit (in the range from0%to1.8%), and then decreased when the size of interference fit varied from1.8%to3.0%. It could be concluded that composite bolted joint with1.8%interference fit had the best fatigue properties. At the conditions of55%σbr cyclic stress level and1.8%interference fit, the ply sequence had little impact on fatigue life of composite bolted joints; At the conditions of55%σbr cyclic stress level, bolted joints with aluminum alloy laps had the same trend of fatigue life as composite laps when the size of interference fit varied from0%to3%, however the joints with steel laps had a different trend that the fatigue life generally increased with the size of interference fit ranging from0%to3%.3. Hole-elongation failure mode was occurred in three types of bolted joints, including high lock bolt with clearance fit, blind bolt with0%interference fit and blind bolt with0.5%interference fit. However at the same cyclic stress level their fatigue lives were quite different. Based on this result, different failure mechanisms of three connecting structures were studied in the paper, and the reason why interference fit could increase the structural fatigue life was discussed. The elongation failure of composite hole was divided into two stages, including fretting wear and impact damage. The amount of fretting wear appeared a linear relationship with the contact angle, and the impact damage presented a power function with the gap size. Under cyclic loading, the contact angle of interference-fit structure was(-ψ,ψ),ψ>>π2, and the gap between bolt and hole became rather small so that the impact damage could be ignored. Therefore, at the first stage of failure there only existed a linear relationship between fretting wear and contact angle. But only if the amount of wear surpasses the initial interference fir size, the impact damage in the form of power function appears. However in clearance-fit structures, there existed rather a mount of gap at the initial fatigue failure stage, so impact damage in a power relationship with the gap could not be neglected. Wear debris generated between bolt and hole, and accumulated at the surface of the composite hole. Under cyclic bearing stress, wear debris began to form a dense layer, which decreased the friction coefficient between bolt and hole, reduced the amount of fretting wear, and delayed the time of hole-elongation, thereby increasing the fatigue life.According to different failure mechanisms of three types of joints, fatigue damage evolution equations were proposed correspondingly. Linear segments of damage evolution equation represented the cumulative model of fretting wear, while the power function model of damage evolution equation can express the he cumulative impact damage. Fatigue damage model of interference fit structure was obtained from the weight function superimposing linear damage equation and power function equation, while the model of clearance fit was replaced by the power function equation. Fatigue life model was then derived by integrating the above damage accumulation equations. According to the fatigue test results, two widely-used fatigue life models (exponential and power function) were compared with the failure mechanism model. The results showed that the power function model has high fitting degree in the condition of clearance fit or0%interference fit, indicating low fitting with interference; the exponential model was just the opposite, high in interference fit and low in clearance fit; while the derived hybrid fatigue model was perfectly matching with both interference and clearance fit.4. Fatigue fracture failure mode was mostly occurred in composite bolted joints with high interference (1.8%and3%). Based on fretting wear, fretting fatigue theory and stress concentration theory, fatigue crack initiation and propagation mechanism of fasteners were studied in the paper. Fretting wear at the surface of fasteners caused material wear, and fretting fatigue leads to the appearance of white layer at the maximum threshold T20of contact area. Micro cracks were generated in the white layer. The threshold T20was irrelevant to the number of cycles, but related to the amount of interference fit and the level of cyclic stress. That was why fracture failure of fasteners occurred in high interference, while hole-elongation failure took place in low interference. Meanwhile, the initiation and propagation of mode-I fatigue crack was generated when the stress concentrated on the root of TC4nut. Under cyclic loading, the internal and external micro-crack of nuts expanded together to form a smooth fracture zone on the nut section. When stress intensity factor K1at the crack tip reached the mode-I fracture toughness KIC, brittle fracture occurred to generate a rough zone. Then the screw bore the cyclic loading alone, and stress intensity factor increased sharply at stress focus region of external thread, in which brittle fracture happened. Fatigue fracture took place across the entire fastener in the end.According to the above two failure mechanisms, both fatigue damage evolution equations of micro-crack initiation and expansion were proposed, and models of crack initiation life and crack propagation life were derived by integration. By superposing those two models above, fatigue model of fracture failure could be obtained. Based on the fatigue test results, two widely-used fatigue life models (exponential and power function) were compared with the failure mechanism model. The results show that the exponential and power function models just fitted the test data, without considering the different failure modes of the connecting structure, and the fitted ultimate fatigue strength had not a definite physical meaning. Whereas the model based on failure mechanism agreed well with the experimental results. By considering the threshold of cracks generation in white layer, the fitted ultimate strength was just the inflection point strength when mode of structure began to change from bolt fracture to hole-elongation. And it would be meaningful and valuable in the design of fatigue structure. |
PDF Full Text Request