Vibro compaction of granular soils 93
its density was not sufficient to carry the high load of 250 kN/m 2 of the steel
tanks of equal diameter of 39 m which were of the fixed roof type.
Foundation design required densification of the sand layer below the
tanks together with a 5 m wide strip outside to 80% relative density, as
represented by specified cone resistance target curve with q c -values of up to
10 MPa at 3 m depth, of 15 MPa between 3 and 5 m, and 20 MPa between
5 and 7 m. Maximum settlements were restricted to 50 mm at the center of
the tanks and 25 mm below the ring beam supporting the tank shell.
Trial compactions were carried out at a representative location within
the future tank farm using Keller S300 vibrators on three different triangular grids with probe spacings of 2.20, 2.50, and 2.75 m. Generally all
grids showed a significant increase of the CPT cone resistance after vibro
compaction, and a considerable ground surface subsidence of about 8%.
Although all grids met the target curve within the upper 7 m, CPT tip
resistance fell below in a layer situated just a few meters above bedrock of
about 1 m thickness where the friction ratio rose generally above 0.5% at a
cone resistance below 5 MPa, representing the presence of silty sand, which
generally can only be marginally compacted (see Figure 3.34). Therefore,
conservatively the 2.20 m grid was chosen for the vibro compaction work
for all tanks. In addition, a row of 0.90 m diameter vibro replacement stone
columns were installed below the tank shell at center-to-center distances of
2 m. These columns extended 6.50 m into compacted sand.
Settlement performance of the chosen design was demonstrated by a
finite element analysis based upon the constraint modulus of the sand as
developed from the postcompaction CPT results. For the maximum settlement in the tank center 44 mm were computed, and below the tank shell
24 mm. About 10% of these settlements derived from the bedrock material
below the zone of soil improvement.
Contract vibro compaction work was carried out using a pair of S300 vibrators suspended from a standard crawler crane. The stone columns construction followed, employing the wet vibro replacement method using the same
depth vibrator. Surface subsidence was measured after vibro compaction for
each tank ranging between 6.7% and 13.3% depending on the presence of
silty sand layers, with an average of 8.5%. Contract conditions required the
relatively large number of 21 postcompaction CPTs to be carried out for
each tank for quality control purposes, which is equivalent to one test for
every 100 m 2 of compacted ground. Figure 3.44 shows average CPT curves
developed for each tank, which indicate that the 80% relative density target
was generally met, except in cases where the presence of a high silt content
in the sand in zones close to bedrock reduced or hampered the effect of vibro
compaction and where vibro stone columns were additionally installed.
In addition to the CPTs, three zone load tests were also carried out using
a prefabricated reinforced concrete footing measuring 2 × 2 × 0.6 m placed
in the center of four vibro compaction probes. The load tests were set up and
performed in accordance with ASTM D1194, employing a loading platform
its density was not sufficient to carry the high load of 250 kN/m 2 of the steel
tanks of equal diameter of 39 m which were of the fixed roof type.
Foundation design required densification of the sand layer below the
tanks together with a 5 m wide strip outside to 80% relative density, as
represented by specified cone resistance target curve with q c -values of up to
10 MPa at 3 m depth, of 15 MPa between 3 and 5 m, and 20 MPa between
5 and 7 m. Maximum settlements were restricted to 50 mm at the center of
the tanks and 25 mm below the ring beam supporting the tank shell.
Trial compactions were carried out at a representative location within
the future tank farm using Keller S300 vibrators on three different triangular grids with probe spacings of 2.20, 2.50, and 2.75 m. Generally all
grids showed a significant increase of the CPT cone resistance after vibro
compaction, and a considerable ground surface subsidence of about 8%.
Although all grids met the target curve within the upper 7 m, CPT tip
resistance fell below in a layer situated just a few meters above bedrock of
about 1 m thickness where the friction ratio rose generally above 0.5% at a
cone resistance below 5 MPa, representing the presence of silty sand, which
generally can only be marginally compacted (see Figure 3.34). Therefore,
conservatively the 2.20 m grid was chosen for the vibro compaction work
for all tanks. In addition, a row of 0.90 m diameter vibro replacement stone
columns were installed below the tank shell at center-to-center distances of
2 m. These columns extended 6.50 m into compacted sand.
Settlement performance of the chosen design was demonstrated by a
finite element analysis based upon the constraint modulus of the sand as
developed from the postcompaction CPT results. For the maximum settlement in the tank center 44 mm were computed, and below the tank shell
24 mm. About 10% of these settlements derived from the bedrock material
below the zone of soil improvement.
Contract vibro compaction work was carried out using a pair of S300 vibrators suspended from a standard crawler crane. The stone columns construction followed, employing the wet vibro replacement method using the same
depth vibrator. Surface subsidence was measured after vibro compaction for
each tank ranging between 6.7% and 13.3% depending on the presence of
silty sand layers, with an average of 8.5%. Contract conditions required the
relatively large number of 21 postcompaction CPTs to be carried out for
each tank for quality control purposes, which is equivalent to one test for
every 100 m 2 of compacted ground. Figure 3.44 shows average CPT curves
developed for each tank, which indicate that the 80% relative density target
was generally met, except in cases where the presence of a high silt content
in the sand in zones close to bedrock reduced or hampered the effect of vibro
compaction and where vibro stone columns were additionally installed.
In addition to the CPTs, three zone load tests were also carried out using
a prefabricated reinforced concrete footing measuring 2 × 2 × 0.6 m placed
in the center of four vibro compaction probes. The load tests were set up and
performed in accordance with ASTM D1194, employing a loading platform