Best Use of Cone Penetration Testing

Project summary:

The development of correlations between CPT soundings and resilient modulus has direct application to pavement design. Quick and repeatable testing afforded with the digital instrumentation in the CPT system can replace traditional solid stem augering and provide assessment of soil stiffness and strength. Over the past few years, the Minnesota Department of Transportation (Mn/DOT) has been using cone penetration testing (CPT), which has several advantages over conventional drilling: fewer personnel required, increased productivity, and generation of essentially continuous, real-time data. Through its use of CPT equipment, Mn/DOT has gained a level of expertise in using the technology; however, they realized a need to optimize the technology for use in the transportation construction field, as well as to develop enhanced capabilities for subgrade characterization. The objective of this work was to show that cone penetration testing (CPT) can be used for pavement applications, specifically estimating resilient modulus and organic content. A series of undisturbed samples were obtained from borings directly adjacent to CPT soundings, which then underwent both laboratory resilient modulus and bender element testing. A statistical analysis was performed on these results in conjunction with the data obtained from the CPT soundings to determine the feasibility of developing correlations between field and laboratory measurements of moduli. A relationship was developed between Young's modulus determined by bender element testing and that determined by resilient modulus testing. However, the correlation did not apply to the field-based seismic measurements of stiffness from the CPT soundings. The analysis presented regarding identification of highly organic soils via CPT testing shows that, at this point, the model identified using the discriminate analysis method is not currently sufficient to use in practice. The 10 percent increase in correctly classified soils, however, holds promise for the future, and the introduction of additional independent parameters within a significantly larger data set can be easily analyzed using the methods and tools presented here.