136 Ground improvement by deep vibratory methods
avoided by opting for an average depth z avg taken as representative of the
slip circle or plane considered, and establishing in this way the average stress
concentration factors. An example calculation of the stability of an embankment founded on stone columns is shown in Section 4.6.2.
When comparing results of slope stability analyses carried out by threedimensional finite element methods (Kirsch and Sondermann, 2003)—with
a good reproduction of the depth-dependent stress distribution between
column and soil with conventional methods using homogenization of the
shear parameters by the weighted area average or the Barksdale and Bachus
method with the β and n values developed after Priebe—it can be concluded
that the weighted area method underestimates safety considerably, that the
combined Barksdale and Bachus/Priebe methods overestimate safety, and
that a safety factor obtained from a calculation with the average of both
methods represents a reasonable result close to reality.
4.3.4 Drainage, reduction of liquefaction potential,
and improvement of earthquake resistance
It is obvious that stone or gravel columns installed in fine-grained and
cohesive soils represent elements of considerably greater permeability than
the surrounding native soils. In water-bearing soils their beneficial effect of
accelerating consolidation settlement and reducing the liquefaction potential of sandy soils (Seed and Booker, 1977) is generally appreciated.
As with sand drain construction, the wet installation of vibro replacement stone columns is superior to the displacement dry methods because
it avoids the development of smear zones along the column perimeter. For
standard, nonseismic applications, the choice of grading of the stone column material installed by the wet process in sand, silty sand, and sandy silt
is unproblematic even when using uniformly graded coarse backfill, since
the voids are generally filled with the remaining coarser particles of the
native soil which were not washed out during column installation, in this
way precluding material entry from the surrounding area.
When looking at the gradation of the column material for vibro replacement stone columns installed by the dry method—which is generally quite
uniform as we have seen in Section 4.1—the question of filter stability is
occasionally raised. It is suggested that the uniform grading of the column
material would allow surrounding cohesive material to be transported with
the seepage flow into their voids, thereby not only clogging them up but
also leading to additional settlement. This effect has never actually been
observed in practice, and dry vibro replacement stone columns generally
show similar load settlement behavior to wet columns. When analyzing
this problem, the Terzaghi filter rule cannot be used as it is only applicable for noncohesive soils. Stability of the boundary between the granular
column material and the cohesive soil is governed instead by the relevant
void diameter of the filter, the tensile strength of the cohesive soil, and the
avoided by opting for an average depth z avg taken as representative of the
slip circle or plane considered, and establishing in this way the average stress
concentration factors. An example calculation of the stability of an embankment founded on stone columns is shown in Section 4.6.2.
When comparing results of slope stability analyses carried out by threedimensional finite element methods (Kirsch and Sondermann, 2003)—with
a good reproduction of the depth-dependent stress distribution between
column and soil with conventional methods using homogenization of the
shear parameters by the weighted area average or the Barksdale and Bachus
method with the β and n values developed after Priebe—it can be concluded
that the weighted area method underestimates safety considerably, that the
combined Barksdale and Bachus/Priebe methods overestimate safety, and
that a safety factor obtained from a calculation with the average of both
methods represents a reasonable result close to reality.
4.3.4 Drainage, reduction of liquefaction potential,
and improvement of earthquake resistance
It is obvious that stone or gravel columns installed in fine-grained and
cohesive soils represent elements of considerably greater permeability than
the surrounding native soils. In water-bearing soils their beneficial effect of
accelerating consolidation settlement and reducing the liquefaction potential of sandy soils (Seed and Booker, 1977) is generally appreciated.
As with sand drain construction, the wet installation of vibro replacement stone columns is superior to the displacement dry methods because
it avoids the development of smear zones along the column perimeter. For
standard, nonseismic applications, the choice of grading of the stone column material installed by the wet process in sand, silty sand, and sandy silt
is unproblematic even when using uniformly graded coarse backfill, since
the voids are generally filled with the remaining coarser particles of the
native soil which were not washed out during column installation, in this
way precluding material entry from the surrounding area.
When looking at the gradation of the column material for vibro replacement stone columns installed by the dry method—which is generally quite
uniform as we have seen in Section 4.1—the question of filter stability is
occasionally raised. It is suggested that the uniform grading of the column
material would allow surrounding cohesive material to be transported with
the seepage flow into their voids, thereby not only clogging them up but
also leading to additional settlement. This effect has never actually been
observed in practice, and dry vibro replacement stone columns generally
show similar load settlement behavior to wet columns. When analyzing
this problem, the Terzaghi filter rule cannot be used as it is only applicable for noncohesive soils. Stability of the boundary between the granular
column material and the cohesive soil is governed instead by the relevant
void diameter of the filter, the tensile strength of the cohesive soil, and the