The volume of expansive soil is varied when its moisture condition is changed. The results of these soil volume variation are led to cracking or uplift in light structures. One type of these vital structures is water-conveyance-lining canals. The damage of these canals impose heavy cost on irrigation and drainage network projects. In this paper two ways are studied to control and reduce the effect of swelling soil on canal lining. The first way is the optimization of number of joints on canal lining and the second is the study of location of joints on soil-lining interaction behavior. These proposed approaches have been evaluated with two methods: physical and numerical modeling. The irrigation and drainage network of Tabriz plain canal that is under construction on expansive soil is selected as a case study in this paper. The physical models are constructed as small scale (1/10) in laboratory. Also SIGMA/W 2007 software which is one of the programs in Geo-Studio software package is used for numerical modeling. The results of physical and numerical modeling show the effect of joints to control and distribute of expansive soil-canal lining interaction forces. It can be inferred from results, the relative displacement of panels and destructive bending moments of lining are decreased by considering joints on location of maximum internal forces in the canal section. The number and arrangement of joints should be chosen based on canal section characteristics such as geometric properties and swelling potential of bed soil. Based on obtained results for current condition, considering two series of joints are recommended in both of the canal section walls, first series near the canal floor, the intersection point of wall and floor, and second series about middle height of canal section, the concentration point of interaction forces.
The manner in which oedometer tests are performed will influence the test results greatly. One important factor is the time allowed for each phase of the test to be completed. Not uncommonly, after inundation of the sample, the next load, and all subsequent increments of loading, are applied after a period of 24 hours. This may not be adequate for the heave to be fully developed. The results of oedometer tests are presented in which the times required for the various phases of the test to be completed were measured. After adding water, there is a period of time before swelling begins. This time period is influenced by the inundation stress. The amount of time over which swelling occurred varied significantly with applied stress. It was also shown that the manner by which the specimens in the oedometer were wetted influenced the time over which swelling occurred. When the specimens were wetted from the bottom of the sample with the top left open, swelling occurred for a longer period of time, and in greater amount, than when the entire sample was submerged.
A series of direct shear tests are carried out on the interfaces between expansive soil and sand liner during wetting-drying cycles. Several factors impacting the shear behavior are discussed, including wetting-drying cycles and saturation methods. Upto 5 times of the wetting-drying cycles are applied on expansive soil-sand liner samples before shearing. Saturation methods comparison is carried out under confined and unconfined conditions, which means whether the samples are fully restricted in a saturator or are able to swell freely during saturation. Test results show that the interface behaves strain-hardening and the shear zone extents to a certain depth in sand liner. The shear strength is higher under confined saturation condition than that under unconfined saturation condition,the friction angles under confined saturation condition are about 0.4 to 9.7 degrees greater than those under unconfined saturation condition at the same wetting-drying cycle. As wetting-drying cycling, the friction angle increases under confined saturation condition, while it increases at beginning and then decreases under unconfined saturation condition. The mechanism of the interface friction angles behavior is revealed as the combination result of two aspects: sand particles merge into expansive soil and samples compacted by expansive force, which increases the density of the “shear band” and sample, and fissures grow, which destroys the integrity
In this paper, unconsolidated siltstone is tested with static and cyclic loading that represents the dynamic force during an earthquake, under K0 and isotropic conditions. Meanwhile, anisotropy of magnetic susceptibility (AMS) is conducted to check the changes in the internal textures of the specimen after these tests. As the results, no major fabric variation within the siltstone was confirmed based on AMS, in spite of substantial residual deformation occurred due to cyclic loading in all test cases. Furthermore, a phenomenon similar to fatigue fracture of metal was observed when the siltstone subjected to cyclic loading under K0 condition. These results are exactly the same as the mechanical properties of décollement zone in which the internal texture remained unchanged but the density increased significantly. In other words, cyclic loading, such as the effective-pressure fluctuation, may become a possible reason for the formation of décollement zone.
Determination of rock engineering properties is important in civil, mining and geotechnical engineering. Uniaxial Compressive Strength (UCS) is one of the most important properties of rocks. Point Load Test (PLT) is practically used in geotechnical engineering to determine rock strength index. Despite that the PLT is fast, economical and simple in either field or laboratory, Uniaxial Compressive Test (UCT) is time-consuming and expensive. UCS can be estimated using Point Load Index (PLI). So, implementation of correlation between results of PLT and UCT is of interest. In this research correlation between the results of point load test and uniaxial compressive test are presented for rock samples from three sites in Iran. Two rock types including Shale and Marlstone have been utilized in this research. Correlations between UCS and PLI in this study are verified with proposed equation by pervious researchers.
In recent past, effects of blast loads on structures/foundations have gained considerable attention due to increase in threat from various man-made activities. Protecting foundation systems from such threat has become a major concern for designers. Terrorist attack as a source of blast may be in the form of missile attack which will penetrate the ground and then explode. Mining, quarrying, construction activities and all other operations which involve the use of explosives for rock breaking are some of the other sources of blast load. Blast phenomenon can lead to excessive settlement or distortion of foundation systems which in turn may collapse the building. Full-scale experiments of blast are expensive and model tests are found to be unrealistic many a times, numerical simulation of blast is required to be done to understand the complex response of foundation subjected to blast. The finite difference method is a powerful numerical method to solve geo-mechanical problems which can be used to analyze soil foundation interaction. In the present paper, finite difference software Fast Lagrangian Analysis of Continua in 3 Dimensions (FLAC3D) is used to model ground shock wave propagation in rock. In order to study the response of foundations to blast loads through numerical modeling, many factors such as fixity conditions, constitutive models, mesh size, time step need to be calibrated. Blast load is applied and Particle velocity time-history is obtained. Peak Particle Velocity (PPV) is estimated at various points away from blast. The FLAC3D result has been compared with field data to arrive at suitable constitutive model, mesh size, time step etc. Calibration of FLAC3D model is achieved. Effects of Young’s modulus and Poisson’s ratio have also been presented. The results will be very useful for evaluation of response of foundation subjected to blast.