Determination of field dry density of sandy soils using a dynamic cone penetration test

Abstract

The degree of compaction plays a critical part in assessing the quality of soil layers during road and highway construction. However, conventional field density testing methods are time- consuming and may not fully ensure compaction compliance throughout construction. Thus, there is a need for efficient quality control and assurance procedures to maintain project progress while upholding construction quality. This study aims to establish relationships between the Dynamic Cone Penetration (DCP) Rate and field dry density, facilitating efficient field density assessment without solely relying on traditional methods like the sand cone test. Although previous researchers have developed correlations to determine field dry density with the DCP penetration index, regression analysis results indicate their incompatibility with local soil conditions due to variations in soil formations and environmental factors. Hence, reliability analysis and modification are necessary to tailor these correlations to local conditions. Therefore, this study develops numerical correlations between the DCP penetration index (PRI) and field dry density (FDD) of subgrade soil using data obtained from laboratory and field investigations related to road construction. Following this, the outcomes of tests including Atterberg limits, sieve analysis, DCP penetration, Proctor Compaction, and field density tests are subjected to further analysis. Soil classifications were conducted in accordance with the Unified Soil Classification System (USCS). The regression analysis is performed using MATLAB software, employing a curve fitting tool with robust regression options, such as Least Average Residual (LAR) and Bisquare, to mitigate the impact of outliers during model fitting. This study proposes to modify the existing correlations and two new correlations using regression analysis in the form of first and second- order numerical equations, with multiple variables representing the DCP Penetration Index (PRI) and the Field Moisture Content (FMC).