313
number of layers in the pavement structure. The possible effects of
subgrade stratification on NDT deflections and pavement response are
briefly discussed below.
The CPT log for SR 26A, as shown in Figure 7.3, was used to assess
the variability of the subgrade layer. This test section had a clay
layer (approximately 0.6 m thick) at a depth of about 2 m. The water
table was at a depth of 1.58 m. The CPT log was used to divide the sub-
grade layer into 4 sublayers and an average qc value was assigned to
each layer. The Dynaflect modulus-qc equation for the combined data in
Table 7.5 was used to predict the respective layer modulus. The number
of pavement layers was assumed to vary from 4 to 7, depending on the
number of layers in the subgrade. BISAR was then used to predict Dyna
flect deflections for the various layers. In all cases, the moduli
values for the AC, base and subbase layers were kept constant.
Table 7.9 shows the results of Dynaflect deflections predicted by
BISAR for the various numbers of layers. The actual field measured
deflections are also listed in Table 7.9. It is seen from the table
that significantly different deflections were predicted by BISAR
depending on the number of layers and the modulus value of the semi-
infinite subgrade layer. For example, when the weak clay layer was
considered as the semi-infinite layer in a 6-layer system, the predicted
deflections were more than twice the measured values. However, when
this weak clay layer was underlain by a relatively stiff subgrade in the
7-layer system, predicted and measured deflections were comparable.
The above illustration tends to support the argument that founda
tion layers with highly variable moduli significantly influence the
response characteristics of nondestructive tests. The presence of a