| It is important to study the crack propagation under hydrostatic pressure because of the dam collapse by plenty of cracks in hydraulic concrete. At present the study of hydraulic concrete fracture includes two parts: one is the numerical simulation and the other is the experiment. And it is little work for the latter due to the difficulty and complexity in the test. Furthermore micro crack during either the construction period or the running period of concrete impacted on the dam safety and hydraulic fracture resulted in the unstable fracture with the increasing water pressure. Hence, it can be said that the most important problem is to test crack propagation and to predict its path, so that some methods can be deal with at time. As a result, the concrete fracture mechanics is a good facility to monitor the crack propagation, determinate the initial point and the unstable fracture stage. In order to solve the above problems, the following aspects are carried out in this thesis:1. Forty specimens, 1150mm×1104mm×200mm,are experimented using Wedge-splitting test on compact tension, in which twenty-two ones are studies successfully. The test included two parts: one is the no water pressure load numbered 10 , which are divided three group by restraint; the other is the water pressure load number 12, which loaded four different pressure, 0.1MPa,0.2MPa,0.3MPa and 0.4MPa respectively. In this thesis the sealing device and load method are demonstrated in detail.2. In order to simplify the calculation against reducing accuracy, the membrane, steel plate and concrete are considered a total specimen. Hence, the corresponding parameters such as compressive strength, tensile strength and Young's modulus, are recalculated in nonlinear softening model. Based on the above results, the initial toughness, unstable toughness and cohesive toughness are studies by Double-K fracture model. As a result, the evaluation initial fracture toughness is approximately similar to the test initial fracture toughness, which can approximately meet the requirement in the engineering practice.3. The length of crack propagation is compared to the experimental value by full-bridge strain gauges, which are widely used for the simplicity and facility. Moreover it is easy to determinate the exact location of crack propagation by strain back point, which can make up for the shorting of identify the turning point between the linear stage and the nonlinear stage. It is concluded that the experimental results are larger than the calculation because the crack in the interior is ahead of the surface. However the average error is less than 6%, and the experimental methods can be used in the water pressure for the limitation of calculation.4. The changes of water pressure are tested by silicon pressure transducers adopting on the side of specimen. Moreover the results is fitting by secondary curves, which is accordance with most scholar's conclusion, that is. the water front decreasing to zero by wedge shape, is behind of the crack propagation. Hence, the zero water pressure can be calculated by fitting results and the complete curve of propagation is determined according to the above result.5. The maximum load decreased with the increasing water pressure by the comparison of the specimens loaded different pressure using double-K model, and even at the 0.4MPa hydraulic pressure the value is less 50% than the one at 0MPa pressure. Therefore the water pressure is an important component of the maximum load. The results show that initial toughness from the test is the approximately similar to the one from the formula calculation, which indicated that the possibility of double-K fracture criterion is to study the hydraulic concrete. However the total load including water pressure is less the force on no water pressure because of the water softening.
|
PDF Full Text Request