1078
Proceedings of the 18
th
International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013
41.2% (stage 1). In all cases cracks initiate at the vertices
located in the change of geometry of the moulds.
10
35
20
40
20
Small Mould (SM)
h = 5
Figure 1. Small moulds (SM) used in the desiccation tests.(from Ávila,
2004). Dimensions are in mm.
The second sequence represents an intermediate stage and
the picture was taken 4h:35min after the beginning of the test.
Moisture content ranged between 36% and 37.2%. It is clear the
complete formation of primary cracks and in all cases their
location and form are similar. The volumetric soil contraction
can be observed as a separation of the sample from the walls of
the moulds.
The third sequence represents the final stage and the picture
was taken after 22 h of the test initiation. The primary cracks
were completely open and a secondary crack, located near the
central part of the sample appears in all samples oriented in
parallel direction with respect to the primary cracks. It is
remarkable the great similarity observed in the three tests. Note
that cracking test repeatability is not frequent due to the
multiple variables involved, as previously mentioned.
Similar results were obtained in other tests sequences with
different initial water content, as observed in Figure 3. In these
tests also tertiary cracks appear in the extremes of all the
moulds directed perpendicular to the primary cracks and located
in the middle of the extreme areas of the moulds.
During the desiccation process, moisture content of the
samples were controlled by weighting them carefully at
different times. The relation between initial water content of the
samples and water content at initial cracking is presented in
Figure 4. It is clear that, the higher the initial moisture content
the higher the moisture content at cracking. The relation
between both variables implies a greater potential volume
change for initially wetter samples due to moisture reduction
during desiccation.
3 SIMPLE ONE-DIMENSIONAL CONCEPTUAL MODEL
TO EXPLAIN SYSTEMATIC CRACKING
When a soil sample is subjected to a homogeneous drying
process, volumetric contraction tends to occur. In free shrinkage
conditions, with not friction restrictions in the base or laterally,
the sample cracks are not expected because no tensile forces act
on it. This condition is sketched in Figure 5 and represents the
common case of shrinkage limit tests in which a lubricant is
used to reduce friction between the sample and the mould.
However, if the sample composition or if the drying conditions
are not totally homogeneous the sample may crack due to
tensile forces generated inside it.
Figure 2. Sequence of cracking of three samples in SM moulds. Primary
and secondary cracks are generated in a systematic and homogeneous
way in each case.
Figure 3. Homogeneous pattern cracking observed at the final stage of
three tests similar to those of Fig. 2 but starting from higher water
content.
water content at cracking (%)
Figure 4. Relation between initial water content and water content at
cracking for SM samples.
