Compressibility-Based Interpretation of Cone Penetration Test Data
Advisor: Dr. Scott M. Olson
Abstract
This work seeks to develop correlations for the cone penetration test (CPT) that incorporate the compressibility of soil . The proposed compressibility-based correlations are intended to provide the practitioner with a simple, practical, and accurate means of forecasting important soil properties (such as certain critical state parameters and density) and providing a means to normalize CPT data. While the compressibility-based correlations and overburden normalizations are the final product of this work, the main efforts of this dissertation are:
- Development of the compressibility concept as a means to interpret soil behavior,
- A compressibility-based correlation for the CPT to estimate the slope and y-axis intercept of the critical state line,
- A compressibility-based correlation for the CPT to estimate density in terms of state parameter and relative density, and
- A compressibility-based correlation for the CPT to normalize CPT data for overburden stresses.
To accomplish these goals, this effort aggregates a database of 847 cone penetration tests performed in sand-filled calibration chambers. Using this database, this dissertation first reexamines assumptions used to interpret tests and boundary effects, analyzes the efficacy of published calibration chamber correction factors, and provides a novel framework for interpreting calibration chamber data based on soil compressibility. This first effort finds that (1) cone tip resistance is relatively unaffected by boundary effects; (2) factors used to correct calibration chamber cone tip resistance to an equivalent free-field value only marginally improved the correlations/trends; and (3) cone tip resistance is similar for soils of equally similar compressibility, density, and effective confining stress. The author suggests that calibration chamber data may not require corrections for chamber size or boundary conditions so long as a ratio of chamber diameter to cone penetrometer diameter of 20 is satisfied.
This dissertation then presents new correlations for sandy soil behavior and properties that are intended to be used solely with CPT data without additional in situ or laboratory testing, iteration, or numerical modeling. The correlations were developed for the ΔQ soil behavior index using a database of CPTs performed in sand-filled calibration chambers. These correlations relate the ΔQ soil behavior index to three parameters of particular importance in critical state soil mechanics: (1) slope of the critical state line, (2) intercept of the critical state line at 1 kPa, and (3) state parameter; in addition to (iv) relative density. The proposed correlations illustrate that the slope and intercept of the critical state line are related to ΔQ and that state parameter and relative density also are related to ΔQ in combination with the overburden stress-normalized net corrected cone tip resistance.
Lastly, this dissertation presents a comprehensive framework to normalize cone tip resistance. The framework is based on density (either relative density or state parameter) and compressibility or ΔQ. This dissertation also presents the methodology for developing the normalizations. Results from this effort suggest that the compressibility-based overburden normalization perform similarly to other published normalizations but has the benefit of being simple to calculate without the need for additional laboratory testing or iteration.
