Combination of Power Function and Log-Linear Models to Estimate Pile Setup

This paper presents a new approach of estimating pile setup starts from the end-of-drive (EOD) resistance by the use of a power function model followed by a log-linear function. Often pile setup is estimated using the Skov and Denver(in: Proceedings of the 3rd international conference on the application of stress-wave theory to piles, Canada, 1988) model, which requires knowing the pile resistance at a reference time (t_o). This requires additional effort for testing the pile at the reference time (t_o). This effort may cause delay in the foundation construction and thus increasing the cost of construction. Pile load testing program was conducted on seven 914 (36 in.) square close-ended prestressed concrete (PSC) test piles at the Caminada Bay bridge project in coastal Louisiana to develop a methodology to estimate pile setup effectively starting from EOD resistance. Several dynamic load tests (DLTs) were performed on each test pile, with waiting periods of 60 min to 55 days after installation, to measure the magnitude and rate of setup. Static load tests (SLTs) were also performed at the end of the load testing program to validate the results of dynamic load tests. The load testing results showed that the total resistance increased up to 12 times of the EOD resistances after 28 days from EOD. The Skov and Denver (1988) setup parameter “A” was calculated for each test pile using different initial reference times (t_o). The results showed that the setup parameter “A” was highly variable and uncertain for t_o less than 1 day. This paper proposes a new power pile setup model that can be used to estimate pile setup immediately after EOD to the initial reference time, t_o, which is usually 1 day for a log-linear model. The proposed model was validated using results from published case studies for various geological conditions, which shows that the results of the model effectively match the setup test results within a small tolerance.