Field Instrumentation and Testing to Study Set-up Phenomenon of Driven Piles and Its Implementation in LRFD Design Methodology

Abstract

This research study investigates the pile set-up phenomenon for clayey soils and develops the models to predict pile set-up resistance at a certain time after the end of driving (EOD). The increase of pile resistance after EOD is known as “pile set-up”. To fulfill this objective, a total number of twelve prestressed concrete (PSC) test piles were driven in different soil conditions of Louisiana. In addition, dynamic load tests (DLT) and static load tests (SLT) were usually performed to verify the axial resistances of piles at specific times after EOD, as well as to quantify the amount of increase in resistance compared to the EOD (i.e., set-up). The focus of this research was to calculate the resistance of individual soil layers along the length of the pile. In order to implement this goal, all the test piles were instrumented with vibrating wire strain gages. Vibrating wire piezometers and pressure cells were also installed in the pile face in order to calculate the dissipation of excess pore water pressure, together with the corresponding increase in effective stress, respectively with time. The Case Pile Wave Analysis Program (CAPWAP) was performed in all the DLT data, in order to calculate the resistance of individual soil layers. Logarithmic set-up parameter “A” of individual soil layers were calculated using the unit side resistance (fs). The set-up parameter “A” was tried to correlate with different soil properties. With the aid of Statistical Analysis Software (SAS), detailed regression analyses were performed to develop models with incorporated soil properties. Five different levels of empirical models were developed in order to estimate the amount of set-up. In addition, one set-up model was developed directly from the in-situ test data (corrected cone tip resistance, qt). Load resistance factor calibration was performed in order to calibrate the set-up factor (ϕset-up). The developed models were implemented to predict the amount of resistance at four different time intervals after EOD. Finally, the statistical parameters of measured resistance to predict resistance were applied to calibrate the set-up factor (ϕset-up) and to incorporate that factor into the LRFD framework.