Investigation of Deep Foundations at the Spring Villa National Geotechnical Experimentation Site

Abstract

The Spring Villa test site is a National Geotechnical Experimentation Site located near Opelika, Alabama and managed by Auburn University. The test site is located at the southern end of the Piedmont geologic province and is underlain by residual silts and sands which grade with depth into partially weathered rock. The site has been extensively characterized through borings, penetration tests and geophysical tests. Several types of deep foundations, for instance pipe piles, H-piles and drilled shafts, have previously been installed at the Spring Villa test site and many have been load tested. This combination of characterization and foundation data at a single site makes the Spring Villa site a unique location to test new characterization methods and develop correlations to load test data in residual soils. However, there is no comprehensive database to summarize the foundation and in-situ test locations and the results from the various tests. This thesis will provide a new publicly available GIS database which contains the locations of foundation and in-situ tests and a summary of results from the in-situ and load tests. The site history and geology of Piedmont residual soils are briefly described. The final map with the location for each deep foundation and in-situ test of the GIS database and the previous load test results in the attribute table of the GIS database are provided. The second part of this thesis describes the three-dimensional numerical modeling of a new non-destructive testing method for the drilled shafts at the Spring Villa test site. The preliminary models consider a single drilled shaft with the same material properties and dimensions as the drilled shafts located at the Spring Villa test site. The shaft is modeled using the commercial finite difference software FLAC 3D. The non-destructive test is simulated by applying a sinusoidal force to the side of the upper portion of the drilled shaft and the acceleration time history of four different points are recorded. Simulations are performed using different loading frequencies and three different shaft lengths. The properties of the concrete and soil are also varied to see the effect on the response. The results from these models will be analyzed by future researchers in order to identify correlations and assist with the design of a field testing procedure. The database and the numerical results described in this thesis should make it possible for future researchers to better utilize the Spring Villa site to develop and test methods for evaluating deep foundations in residual soils.