ABSTRACT
Dynamic modulus is a key parameter used in mechanistic response and performance models to characterize the linear viscoelastic properties of asphalt concrete. Typically, only the initial mixture condition during mix design phase or plant production is characterized. However, it is known the modulus of the asphalt layer varies during service as the mixture experiences traffic and environmental loading as well as other physio-chemical changes (such as, oxidative aging). Although such information is available in the Long Term Pavement Performance (LTPP) database in terms of modulus back-calculated from Falling Weight Deflectometer (FWD), the challenges associated with the determination and normalization of temperature and frequency inhibits its use in the performance models requiring viscoelastic characterization. This paper presents an approach to tie and normalize in-service back-calculated modulus values to a specific temperature and frequency. The study focuses on a test section in Oklahoma and information needed for the study is obtained from the LTPP database. The modulus back-calculated from the first FWD test on pavement section is compared to the as-built built complex modulus master curve to yield the equivalent loading frequency which is assumed the same for subsequent FWD measurements. The time temperature superposition principle is applied to modulus measurements for temperature correction. Modulus indices determined by comparing initial reference modulus to subsequent measurements are used to understand the evolution of in-service complex modulus master curves of mixtures over time. The feasibility and limitations of the proposed analysis method and data collection scheme featured in LTPP are also investigated. Finally, the variation in in-service modulus is correlated to field distress data. The proposed method was found promising to establish a reference frequency and normalize the temperature difference among FWD measurements and subsequently develop modulus indices. The results also showed that the variation in environmental changes (temperature and moisture) can have a profound effect on the degree to which the layered-elasticity and time temperature superposition principle is a valid model for the pavement system based on the deflection encountered. To mitigate these limitations associated to validity of the time temperature superposition principle, the authors recommend establishing a test temperature range that results in a linear relationship during FWD testing.
