84 Ground improvement by deep vibratory methods
It is therefore recommended to correlate the CPT results with direct in-situ
density measurements or to conduct special laboratory calibration tests
between density and penetration resistance, whenever the carbonate content resulting from shell debris is in excess of 10%, unless enough experience and data exist for the referenced soil.
Wehr (2005a) presented a shell correlation factor f s as a function of the
relative density D r for Dubai sand (carbonate content in excess of 90%, D 60 /
D 10 = 3) by which the CPT results measured in the calcareous sand need to
be multiplied to arrive at corresponding values for silicate sand:
f
q
q
D
s
c
c
r
(silica)
(shell)
0.0046
1.3629
=
=
+
(3.32)
The study revealed that the influence of the in-situ density was in all tests
so dominant compared with the overburden pressure that the latter could
be neglected. Accordingly this correlation factor ranges between 1.4 and
1.8 and amounts to 1.64 for a relative density of 60%. It was stated that the
values found represented lower limits since all shell particles larger than
4 mm in diameter were removed from the test specimens representing about
8% of the total weight.
In general it is always recommended, in addition to keeping an accurate
record of all imported backfill material, to also consider the measured surface settlement after vibro compaction as an indicator and as supporting
evidence of the increase in density achieved comparing the pre and postcompaction volumes.
When interbedded sand layers with cohesive soils above and below
required compaction, it is often found that the result of the standard deep
compaction is below expected levels. The reduced compaction is especially
noticeable at the upper and lower interface in the granular soil with the
cohesive material. It is surmised that the vibratory motions are dampened
to a certain extent by these cohesive layers. The proper functioning of the
sand backfill is important in these cases to compensate the volume reduction in the sand layer which will come to the fore at the upper interface.
An extended compaction time and, if necessary, also reduced compaction
probe spacing will help to avoid this problem.
Natural cementation of sand can be found in many parts of the world. It
occurs predominantly in arid climatic conditions in calcareous sand deposited
above the water table. Cementation can also occur when hydrous silicates or
iron oxides act as a kind of glue at the contact point between the sand particles. Cementation may indicate a greater density or strength than actually
exists and in-situ density measurements and their standard correlation may
result in misleading interpretations. Weakly cemented soils with actual low
relative densities may respond to vibratory compaction by collapsing and
yet producing lower post- than pretreatment test results, although the compacted material is now in a much more stable condition. Strong cementation
It is therefore recommended to correlate the CPT results with direct in-situ
density measurements or to conduct special laboratory calibration tests
between density and penetration resistance, whenever the carbonate content resulting from shell debris is in excess of 10%, unless enough experience and data exist for the referenced soil.
Wehr (2005a) presented a shell correlation factor f s as a function of the
relative density D r for Dubai sand (carbonate content in excess of 90%, D 60 /
D 10 = 3) by which the CPT results measured in the calcareous sand need to
be multiplied to arrive at corresponding values for silicate sand:
f
q
q
D
s
c
c
r
(silica)
(shell)
0.0046
1.3629
=
=
+
(3.32)
The study revealed that the influence of the in-situ density was in all tests
so dominant compared with the overburden pressure that the latter could
be neglected. Accordingly this correlation factor ranges between 1.4 and
1.8 and amounts to 1.64 for a relative density of 60%. It was stated that the
values found represented lower limits since all shell particles larger than
4 mm in diameter were removed from the test specimens representing about
8% of the total weight.
In general it is always recommended, in addition to keeping an accurate
record of all imported backfill material, to also consider the measured surface settlement after vibro compaction as an indicator and as supporting
evidence of the increase in density achieved comparing the pre and postcompaction volumes.
When interbedded sand layers with cohesive soils above and below
required compaction, it is often found that the result of the standard deep
compaction is below expected levels. The reduced compaction is especially
noticeable at the upper and lower interface in the granular soil with the
cohesive material. It is surmised that the vibratory motions are dampened
to a certain extent by these cohesive layers. The proper functioning of the
sand backfill is important in these cases to compensate the volume reduction in the sand layer which will come to the fore at the upper interface.
An extended compaction time and, if necessary, also reduced compaction
probe spacing will help to avoid this problem.
Natural cementation of sand can be found in many parts of the world. It
occurs predominantly in arid climatic conditions in calcareous sand deposited
above the water table. Cementation can also occur when hydrous silicates or
iron oxides act as a kind of glue at the contact point between the sand particles. Cementation may indicate a greater density or strength than actually
exists and in-situ density measurements and their standard correlation may
result in misleading interpretations. Weakly cemented soils with actual low
relative densities may respond to vibratory compaction by collapsing and
yet producing lower post- than pretreatment test results, although the compacted material is now in a much more stable condition. Strong cementation