Geotechnical bearing capacity of MV piles

Abstract

Almost 4000 MV piles are used as anchorage of quay walls in the port of Rotterdam. The current design method of MV piles in Dutch practice is a CPT-based method that correlates the cone resistance to the shaft friction by a factor α_t. In the port of Rotterdam, the cone resistance is restricted under the assumption that no shaft resistance higher than 250 kPa is mobilised. Consequently, α_t is used in combination with a limiting value for the cone resistance of 18 MPa. Moreover, the value of α_t that is currently used in the port of Rotterdam (1.4%) is derived after limiting the cone resistance at 18 MPa. Since this design method was developed in the 1980s', many more MV piles were tested in the port of Rotterdam. The maximum test load was generally at least two times the characteristic value of the required anchor force. None of these full-scale tests were loaded up to failure and no significant creep effects were observed. Consequently, the design standard was never updated as ultimate bearing capacity remained undetermined. Recently, the Port of Rotterdam has executed failure load tests in the Maasvlakte area. The tests allowed detailed strain readings along the full length of the test piles and offer the possibility to accurately determine the local mobilisation of shaft friction in addition to the ultimate failure load. Apart from describing the successful instrumentation with BOTDA fiber optical sensors (Brillouin Optical Time Domain Analysis), this study addresses the assessment of the obtained data in detail. Multiple relations are considered in this thesis. Analysis of CPT's indicates an increase in cone resistance due to pile installation. Investigation of pile driving data shows that installation energy correlates well with the cone resistance. Soil-structure interaction is analysed and a mobilisation curve is composed. This curve illustrates that mobilisation of shaft friction as a function of displacement of MV (tension) piles is similar to the Dutch standard for small and non-displacement compression piles. This thesis presents proof that limiting the cone resistance based on a maximum shaft friction of 250 kPa is not correct. The derivation of α_t without limiting the cone resistance results in a value of 1.2%. Installation energy proves to be a good indicator of the soil conditions. The piles that give good predictions for the bearing capacity with α_t = 1.2% present similar ratio's between the installation energy and the cone resistance. Future research may establish a consistent relation between CPT-based bearing capacity and pile driving energy to reduce uncertainty. This thesis will contribute to an update on the design method for MV-piles in dense sand layers of the Maasvlakte area.