An Evaluation of the Potential Use of Non-Nuclear Density Gauges for Asphalt Concrete Acceptance

This report describes the results of a study using non-nuclear density gauges (NNDGs) to measure the in-situ density of asphalt concrete (AC) material in Virginia. The study compared the NNDG results with those obtained from the use of two traditional AC density acceptance methods: the core method (AASHTO T 166) and the nuclear density gauge (NDG) method. Although these two methods are the most widely used methods of accepting AC density, the core method is time-consuming and destructive and involves bulky test setups and the NDG method, although portable and non-destructive, is associated with safety concerns related to the presence of radioactive materials in the gauge. The purpose of the study was to evaluate the use of two NNDGs as a potentially safe, portable, and expedient method of measuring AC density—a key indicator of pavement performance. The direct comparison of NNDG and core density and the acceptance rates with the use of NNDGs and NDGs were the focus of the study. Extensive field and laboratory tests were conducted to determine AC density using two models of NNDGs (i.e., the Troxler PaveTracker Plus and the TransTech Model PQI 301) and one model of an NDG (i.e., the Troxler Model 4640-B). Density measurements of AC cores/plugs taken at gauge testing locations were conducted in the laboratory in accordance with AASHTO T 166 for comparisons. The results of the field testing showed that NNDG measurements were not well correlated with core density or NDG measurements. However, there was good agreement between readings from NNDGs and NDGs in terms of identifying core cutting locations (67%), control strip acceptance (75%), and test section acceptance (95%). This apparent contradiction between the acceptance rate among the gauges and the poor correlation could be explained by the relatively low ranges in measured field density (0-5 lb/ft³), which is within the precision ranges of the gauges used. The results of additional laboratory testing of 10 AC slabs with air void contents ranging from about 3% to 20% confirmed the results of the field testing. Specifically, they demonstrated that compared with NDGs, NNDGs were less sensitive, with an average relative bias of 19.6 lb/ft³ and 9.6 lb/ft³ for the PQI 300 and the PaveTracker Plus, respectively, compared with 2.2 lb/ft³ for the NDG. The results also showed that results from use of the NNDGs were not well correlated with core density measured in accordance with AASHTO T 166, which is generally accepted as the most accurate method of measuring density. The study concludes that NNDGs of the types used in the study are not suitable for measuring AC density for acceptance purposes and thus are not recommended for use as density acceptance tools in Virginia.