COMPARISON OF ANALYSIS TECHNIQUES TO OBTAIN MODULUS AND PHASE ANGLE FROM SINUSOIDAL TEST DATA

Stiffness of hot-mix asphalt can be measured in various ways. One method used in the US is the compressive dynamic (complex) modulus test, first introduced as ASTM D-3497 test method in 1979. A revised test protocol is in preparation for the new 2002 Guide for the Design of Pavement Structures currently under development at Arizona State University. The new test protocol deviates from the old one in terms of having a larger range of test temperatures and frequencies. However, testing at high temperatures introduces strains that may not be recoverable during the period of the test, and "noise" characteristics often vary over the range of test conditions. Both of these issues can be addressed with post-processing techniques. This paper compares statistically different analysis techniques applied to stress and strain data in the determination of modulus and phase angle. These methods are based on relatively simple data smoothing and parameter estimation algorithms. Although the algorithms may be similar, it is demonstrated that significant differences in the reported modulus and/or phase angle can result from fundamentally similar analysis techniques and even from relatively subtle differences in the data-sampling scheme used within a single technique. In practice the measured test data often deviates from a perfect sinusoidal wave due to one or more factors arising from material response or test machine response. This paper shows the importance of standardizing the analysis method used in the test protocol. Dynamic modulus test data of a dense graded asphalt mix obtained at five different temperatures, 9, 4.4, 21.1, 37.8, and 54.4 degrees C, and six different frequencies, 25, 10, 5, 1, 0.5, and 0.1 Hz, were analyzed in this study. This type of a data set is needed to construct a master curve of the mix required in the new pavement design guide. Testing was conducted in load control and load amplitude was decreased as temperature increased. In the frequency sweep, the number of load applications was reduced as the frequency decreased. Data acquisition was set to record only the last five cycles of data for each temperature and frequency in the analysis. For the covering abstract see ITRD E117423.