Page 71 - MODELING OF ASPHALT CONCRETE

Modeling of Asphalt Binder Rheology and Its Application to Modified Binders 49

fundamental problem with this approach because it is unlikely to be useful in
representing the effect of repeated cyclic loading and the changes in binder properties
with accumulation of damage. The fatigue damage behavior was the subject of a
previous publication that showed the effect of modification on the nonlinear behavior
and specifically the damage behavior (Bonnetti et al. 2002).
The effort to develop a new test was focused on simulating the fatigue phenomenon
in a binder-only fatigue test. The DSR was used to conduct what is called a time sweep
test. The test provides a simple method of applying repeated cycling of stress or strain
loading at selected temperatures and loading frequency. The initial data collected were
very promising and showed that the time sweeps are effective in measuring binder
damage behavior under repeated loading in shear (Bahia et al. 1999).
To understand the test results and to establish the best testing conditions that could
lead to effective characterization of binder fatigue behavior, all nine binders used in the
production of mixtures were tested at conditions that match the mixture beam fatigue
conditions. The binders were aged in the RTFO to simulate the effect of mixing and
compaction and the testing was conducted at 10 Hz at temperatures as close as possible
to the mixture beam fatigue temperatures. The testing of binders was conducted in
strain-controlled mode, and to match the mixture strain level, an estimated strain of
3 percent was used for all binders. Figure 2-19 shows the results of the binder testing
∗
and indicates that although the initial G values are similar; these binders show
significantly different fatigue behavior. Some of these binders did not reach 50 percent
∗
of the initial G even after applying close to 1,000,000 cycles while others reached this
∗
level of G after only 10,000 cycles. In all these tests the maximum strain at edge of plate
was kept constant at 3 percent.
To see if the binder fatigue life measured in the strain-controlled binder test has any
relationship to the mixture strain-controlled fatigue life, Fig. 2-20 was prepared to show

FIGURE 2-19 Binder fatigue results at 10 Hz, 3 percent strain, and the temperatures selected
for the mixture beam fatigue testing (binders were RTFO aged).