Hughes, C. S.
Highway Research Engineer, Virginia Highway Research Council, Charlottesville, Va.
Pages: 28 Published: Jan 1969
Nuclear gauges for the express purpose of measuring the asphalt content of bituminous mixes are now commercially available. The purposes of this study were to determine the precision of a Troxler AC-200 asphalt content gauge and to evaluate its accuracy under such variables as aggregate type and gradation, and asphalt content, type, and penetration grade. A total of forty-four mixes were investigated. Linear regression analyses were performed on each aggregate from the first 36 mixes and on the pooled data of all aggregates. The high correlation coefficients found demonstrate the potentialities of the nuclear gauge for measuring asphalt content. The magnitudes of the standard errors indicate that there are still problems that have to be resolved. The results of the regression indicate that a calibration curve should be obtained for each aggregate. However, gradation does not appear to influence the results significantly. The question of the influence of producers and penetration of asphalt was not resolved. Several mixes were also extracted by the refluxmethod, and these data show a bias as compared to the “input” values. The tentative conclusions are that the nuclear gauge at this state offers a potentially rapid and accurate method for determining asphalt content, but that a gauge redesign, which is now in progress, should improve the accuracy.
asphalt content, reflux method, nuclear testing
Paper ID: STP48357S