Evaluation of new nuclear density gauges on asphalt concrete

This report documents an evaluation of the state-of-the-art capabilities of nuclear density gauges to monitor the density of asphalt concrete. In particular, providing immediate information on compaction by the use of roller-mounted gauges and measuring the densities of thin layers were addressed. The study included three phases: literature search, laboratory tests, and field trials. The first phase included a review of current literature on the theory and operating characteristics of commercially available equipment and a review of state highway agency procedures and specifications for monitoring asphalt concrete density. The second phase included a series of laboratory tests to verify factory calibrations and to determine the depth sensitivity, chemical composition errors, and thin-lift capabilities of the various gauges under laboratory conditions. The third phase consisted of five sets of field trials. The laboratory and field trials were carried out using five commercially available static gauges, two commercially available roller-mounted gauges, and one prototype roller-mounted gauge previously developed for the FHWA. Full-depth measurements were taken at one field site and thin-lift measurements were taken at two sites. At a fourth site, the three roller gauges were mounted on a compacting roller and used during paving operation. At a fifth site, an attempt was made to correlate surface roughness and the speed of the roller-mounted gauges with density measurement accuracy. The density measurement data and the correlation of these data with core data are presented. Under carefully controlled laboratory conditions, the accuracy and precision of all the gauges were well within manufacturers' specifications. When compensated for chemical composition, lift thickness, and density of the underlying material, the depth sensitivity and the thin-lift measurement capabilities of the gauges were impressive. In the field, however, the correlation of individual gauge density readings with core density measurements and with each other ranged from excellent to fair. The inability to precisely field calibrate the gauges prior to each use hampered their performance. The data show that, within limitations, static nuclear gauges can be used for acceptance testing of thinlifts, but only when all parameters affecting the measurements are precisely known. The dynamic gauges can be effectively used to monitor relative density growth.