Abstract

Field and laboratory testing was carried out for the construction of a 100 m long cable stayed bridge situated in the foothills of the Alps in the south of Germany, a region dominated by deep post-glacial, fine-grained sediments. Due to the sensitivity and associated challenges with retrieving undisturbed soil samples in situ tests and their evaluation proved to be essential for the geotechnical design of the bridge foundation. This contribution focuses on the analytical and numerical interpretation of Cone Pressuremeter Tests (CPM). The non-linear 𝐺-𝛾 relationship and undrained shear strength (using both the limit pressure and reverse plasticity contraction analysis) were determined analytically. Numerical investigations were carried out as verification using both the Finite Element method (FEM) using both 1D (cavity expansion) and 2D simulations (where the penetration of the probe was modelled) as well as using the Finite Difference (FD) method. 2D simulations demonstrated that the assumption of the cylindrical cavity expansion is appropriate for modelling the CPM tests. The interpreted undrained shear strength showed good agreement with other field tests, including CPTu, vane shear (FVT) and seismic cone penetration (SCPT) tests, as well as with the results of laboratory tests on disturbed samples. CPM tests with strain rate jumps were conducted during pressuremeter expansion, wherewith it was possible to quantify the viscous response of the soil. Based on the holistic interpretation of the field and laboratory results involving both numerical simulations and analytical methods the parameters for the material model Viscohypoplasticity were calibrated.

Abstract

Residual soils are non-textbook materials that are hard to be modelled by traditional soil mechanics, which creates serious difficulties in the in-situ test interpretations and the consequent applications to geotechnical design. This is due to the presence of a cementation structure that is responsible for a cohesive-frictional behaviour of these soils, meaning that two strength parameters must be derived to represent the overall strength. Furthermore, cementation structure also deeply affects the stiffness behaviour, deviating from typical response of transported soils. The common interpretation of in-situ tests usually considers extreme behaviours represented by only one parameter, namely undrained cohesion for clays and angle of shearing resistance for sands, which naturally do not work in these cohesive-frictional materials. For this purpose, only tests that take two or more independent measurements can be used to solve this problem, as it is the case of SCPTu, SDMT and PMT tests, while SPT and DPSH tests cannot be effective in this determination. Portuguese research institutions have been looking over the granitic residual soil characterization through specific research works, from which resulted several publications on the subject. Following previous research works of calibration with (S)DMT (Cruz, 2010) and (S)CPTu (Cruz et al. 2018) tests performed in Polytechnic Institute of Guarda (IPG), Portugal, a new research frame was developed to settle a methodology for obtaining strength parameters of granitic residual soils from pressuremeter tests (PMT), which is presented and discussed herein.

Abstract

Self-boring pressuremeter (SBPM) tests are widely used in site investigations, due to their distinct advantage to measure the shear stress-strain-strength properties of the surounding soil with minimum disturbance. The measured pressuremeter curve can be interpreted using analytical solutions based on the long cylindrical cavity expansion theory with relatively simple constitutive models. However, SBPM tests are strongly affected by the soil behavior and details of installation procedure. In addition, the derived parameters for clays (e.g. undrained shear strength, and shear modulus) are affected by a number of state variables such as overconsolidation ratio, and stress level. In this paper, SBPM tests are investigated using finite element analysis and the MIT-S1 model, to consider complex soil behavior more realistically. SBPM tests in K0-consolidated Boston Blue Clay at different OCRs are simulated in axial symmetric and plain strain conditions, consistent with the assumptions used in analytical solutions. The derived undrained shear strength from both contraction and expansion curves are compared with theoretical values from stress-strain curves, to evaluate the reliability of the derived parameters from the SBPM tests.