Investigating the effect of the third dimension (model length) on wall inclination and settlement of a vertical trench using small-scale physical modeling

Due to the high land volume in urban areas, maximum-slope excavation has become an inseparable part of the construction of a building. The sensitive nature of the excavation operation in terms of cost and safety requires an accurate knowledge of the soil’s behavior, making it an important research field. Most of studies investigate the soil’s behavior using numerical methods, which are accompanied by a certain level of inaccuracy. To sidestep the unavoidable errors associated with numerical methods, this study uses a small-scale physical modeling methodology to investigate soil behavior. The phased excavation of an unsupported vertical sandy soil trench has been experimentally modeled using three small-scale physical models with/without an externally applied load. Three models with different lengths and constant cross-sections were constructed to assess the effects of length (the third dimension) on the trench’s deformation and stability. The results show that during excavation with no external loading, increasing the model’s length does not significantly increase horizontal and vertical deformations. However, in tests in which the system was subjected to an applied load, decreasing the length caused substantially larger vertical and horizontal deformations in the soil. Increasing the models’ length-to-height ratio decreased the discrepancies among the deformations created during loading. In other words, increasing the length caused the deformations of the models to converge toward a similar value. It was observed that the 3D-to-2D deformation ratios are dependent on the load applied to the trench. These ratios decreased as the model approached failure.