Comparison of measured resistance for drilled shafts socketed in weak rock in Oklahoma to previously established SPT and TCP-based predictive design relations

Abstract

The design of drilled shafts socketed in shale and other weak rock is challenging for a multitude of reasons. Sites may be difficult to characterize because the materials may quickly degrade upon sampling. The materials can also be highly variable on not only a site-specific basis, but on regional and even global scales as well. Shale and other weak rock also pose challenges for purely empirical calibrations because of the relative strength of these materials -- it is often difficult to achieve ultimate failure in load tests and as a result, many available field load test measurements are not completely suitable for calibration of resistance factors. Furthermore, the performance of drilled shafts in shale and other weak rock may be greatly affected by construction techniques which are difficult to account for in design. Due to the difficulties posed by shale and other weak rock, numerous alternative design methods based on in situ testing have been proposed and used (e.g. TCP, SPT, pressuremeter, point load index test, etc.); however the reliability of these methods is often unquantified or poorly characterized. Fortunately, more consistent reliability can be achieved by developing and adopting resistance factors that are taken to be a function of the variability and uncertainty for design input parameters. Currently, design of bridge foundations for transportation infrastructure in the U.S. is performed following LRFD methods; however ODOT does not presently employ such a design method. Therefore, the motivation of the works described herein is to facilitate and inform the development of rational and defensible methods for design of drilled shafts in weak rock based on previous experience developing design methods for federal and state transportation agencies. Eight drilled shafts were constructed to support the development of methods for the load and resistance factor design of drilled shafts in weak rock. Four drilled shafts with 4-foot diameters and 70-foot lengths were constructed in Minco, Oklahoma targeting the Fairmont Shale formations. Four more drilled shafts with 4-foot diameters and 45-foot lengths were constructed in Edmond, Oklahoma targeting the Garber Sandstone formation. All test shafts were loaded with a bi-directional load cell, the Osterberg Cell (O-CellTM), to establish ultimate unit tip and ultimate unit side resistance values. The results of the load tests were compared to previously established SPT and TCP-based predictive design relations to facilitate and inform the development of an Oklahoma-specific load and resistance factor design method with the expectation that this developed method would produce a more economical drilled shaft foundation design for use by the Oklahoma Department of Transportation while fulfilling established target reliabilities.