Page 74 - MODELING OF ASPHALT CONCRETE

52 Cha pte r T w o

creep stiffness. The creep and recovery response measured with the DSR could be used
to estimate the G value and the accumulated permanent strain for any selected
v
combination of loading and unloading times.
This finding implies that the accumulated permanent deformation is a function of
viscosity, load, and loading time.

γ = f(, , (2-11)
η τ t)
1
f ητ
S = (, , t) (2-12)
By selecting the appropriate testing stress (t) and the appropriate time of loading (t)
the viscous component of the stiffness G could be directly related to the rate of
v
accumulation of permanent deformation S and thus used as a fundamental indicator of
rutting resistance of asphalt binders.

Binder Fatigue Parameter
Although there are different loading modes that could be used in fatigue testing, a
reliable indicator of fatigue failure should be independent of the loading mode. It should
provide a consistent indication of the level of damage and progression of damage in the
material in terms of changes in mechanical behavior under any loading conditions.
The most commonly used definition of fatigue failure in asphalt mixtures is a
decrease in the initial stiffness by 50 percent, as was indicated in the previous sections.
This arbitrary definition, however, does not allow evaluation of the distinctly different
mechanism by which a material would respond to the energy input during a loading
history for the different loading modes. Researchers have, therefore, focused on using
the concept of dissipated energy to explain fatigue behavior of asphalt mixtures. For
many decades researchers have used the loss modulus as an indicator of fatigue resistance
∗
because of the relationship between this modulus (G sind) and the energy dissipated per
cycle. The success of this approach has been questioned, however, in many studies
because this parameter tends to give different results at different loading conditions.
Recent advancements in fatigue research have indicated that a better indicator of fatigue
is the rate of change of dissipated (distortion) energy per load cycle.
There are several approaches to present the criterion of fatigue based on the rate of
change in the dissipated energy. The most promising approaches are presented by
Carpenter and coworkers (Carpenter and Jansen 1996; Ghuzlan and Carpenter 2000),
and by Pronk and coworkers (Pronk 1995; Pronk and Hopman 1990).

Rate of Change of Dissipated Energy
Ghuzlan and Carpenter (2000) defined the ratio of dissipated energy as
ΔDE W − W
= i i+1 (2-13)
DE W i
where W is the total dissipated energy at cycle i calculated by area within hysteresis
i
loop and W is the total dissipated energy at cycle i+1.
i+1
Plotting the values of this ratio versus loading cycles gives a curve that can be
used to determine the fatigue life (N ) by identifying the sudden change in the rate.
p
The problem with this approach is that the data points, especially for the constant
stress tests, are scattered widely, which makes it difficult to determine an accurate N
p
value.