15 May 2001
used in CRD-C 649 and CRD-C650. Plots of data from table 2-6 for stability, flow, unit weight, percent
voids total mix, and percent voids filled with asphalt should be made, as shown in table 2-6. The
average actual specific gravity is obtained for each set of test specimens, as shown in column G of
table 2-6. Each average value is the core density in grams per cubic meter (g/m3) at 10EC, 50EF. At this
temperature the average values are multiplied by 62.4 to obtain density in pounds per cubic foot (pcf).
These data are entered in column K. The density values thus obtained are plotted as shown in
figure 2-7, and the best-fit smooth curve is then drawn. The data from columns I and J are used to plot
curves for percent voids total mix and voids filled with asphalt, respectively, in figure 2-7. The corrected
stability values in column M and the flow values in column N of table 2-6 are plotted on figure 2-7 to
evaluate stability and flow properties of the mixture.
(7) Relationship of test properties to asphalt cement content. Test property curves, plotted as
described above, have been found to follow a reasonably consistent pattern for mixes made with
penetration and viscosity grades of asphalt cement. Trends generally noted are outlined as follows:
(a) Flow. The flow value increases with increasing asphalt content at a progressive rate
except at asphalt contents significantly below optimum.
(b) Stability. the Marshall stability increases with increasing asphalt content up to a point,
after which it decreases.
(c) Unit weight. The curve for unit weight of total mix is similar to the curve for stability,
except that the peak of the unit-weight curve is normally at a slightly higher asphalt content than the
peak of the stability curve.
(d) Voids total mix. Voids total mix decreases with increasing asphalt content. The void
content of the compacted mix approaches a minimum void content as the asphalt content of the mix is
(e) Voids filled with asphalt. Percent voids filled with asphalt increases with increasing
asphalt content and approaches a maximum value in much the same manner as the voids total mix
discussed above approaches a minimum value.
(8) Requirement for additional test specimens. The curves in figure 2-7 are typical of those
normally obtained when penetration or viscosity grades of asphalt cement are used with aggregate
mixes. Aggregate blends may be encountered that will furnish erratic data such that plotting of the
typical curves is difficult. In most of these cases, an increase in the number of specimens tested at each
asphalt content will normally result in data that will plot as typical curves.
c. Optimum asphalt and design test properties.
(1) Selection of asphalt content. Previous testing has indicated that the optimum asphalt
content is one of the most important factors in the proper design of an asphalt paving mixture. Extensive
research and pavement behavior studies have resulted in establishment of certain criteria for
determining the proper or optimum asphalt content for a given blend of aggregates. Criteria have also
been established to determine whether the aggregate will furnish a satisfactory paving mix at the
selected optimum asphalt content.
(2) Determination of optimum asphalt content and acceptability of mix by Marshall method.
Data plotted in graphical form in figure 2-7 are used to determine optimum asphalt content. In addition,
optimum asphalt content and acceptability of the mix are determined based on table 2-7. Separate
criteria are shown for use where specimens were prepared with 50- and 75-blow compactive efforts. As