A revisit of the use of nuclear densometer gauge (NDG) method for soil density measurements

Abstract

Soil compaction is necessary to improve the performance and stability in the construction of reclamations, embankments and slopes. Relative Compaction is commonly used as a quality control of the degree of soil compaction, which is defined as the ratio of the in-situ dry density to the maximum dry density of the fill. The in-situ dry density is calculated from the in-situ bulk density with moisture content measured. In the Hong Kong soil testing standard (Geospec 3), there are two methods to measure the in-situ bulk density of soil, namely the Sand Replacement Test (SRT) method and the Nuclear Densometer Gauge (NDG) method. SRT is a traditional and more accurate method but is time-consuming and more labour intensive. In contrast, the NDG method has the advantages of ease of use and shorter measurement time. However, there are also some well-known limitations in the use of NDG. To address the limitations of the NDG method, this study aims to explore ways of extending the applicability of NDG, for example, to cover coarse-grained soils, to relax the strict requirements of comparison tests with SRT, to further examine the applicability in moisture content measurements which is now prohibited in Hong Kong, and to review the possible use of a Low-activity NDG in Hong Kong. In this study, a total of 116 pairs of field NDG and SRT measurements from 5 different construction sites were obtained. It was found that NDG was accurate to within 5% of the values obtained by the SRT method for fine- and medium-grained soils with a Root Mean Square Error (RMSE) of about 60 kg/m3. However, for coarse-grained soils, the bulk density measured by NDG exhibited significant deviation with a RMSE of about 120 kg/m3. In this study, moisture content measurements by NDG showed a promising result but this finding was not consistent with previous studies. A possible explanation and further study were proposed. The application of a site-specific calibration method based on BS 1377:1990 was adopted resulting in a significant improvement in the accuracy of the NDG measurements, irrespective of soil grain size. The Low-activity NDG is an ideal substitute for the conventional one, given its comparative accuracy and low radioactive risk. However, after checking with the Radiation Ordinance of Hong Kong, this novel Low-activity NDG still cannot be exempted from the stringent control of the said Ordinance. Follow-up actions may be taken to explore further the applicability of this equipment in Hong Kong.