 
          3331
        
        
          Technical Committee 210 + 201 /
        
        
          
            Comité technique 210 + 201
          
        
        
          absolute values and not expressed as a percentage. The average
        
        
          water content in the peat increases with depth from 6.45 (645%)
        
        
          to 12.4 (1240%). The organic clay layer shows lower values,
        
        
          with a large scatter. In the non-organic clay layer the water
        
        
          content reduces to 0.74 (74%).
        
        
          A series of CPTs, including measurements with different
        
        
          cone types and ball penetrometer tests, finalises the subsoil
        
        
          characterisation. Figure 1 shows a typical CPT for the test field.
        
        
          In this CPT, the succession of subsoil layers can be recognized.
        
        
          According to Eurocode ISO/DIS 22476-1:2005,IDT the
        
        
          measurements represent a class 2 type CPT. For a class 2 CPT
        
        
          , q
        
        
          c
        
        
          , equals 100 kPa. Figure 1
        
        
          resistance in peat and are therefore used for
        
        
          co
        
        
          Simple Shear (DSS)
        
        
          tes
        
        
          stress conditions. Since
        
        
          aches the
        
        
          su
        
        
          low, in the order
        
        
          ents for each
        
        
          , a ball
        
        
          ately 0.5
        
        
          point in the
        
        
          
        
        
          -
        
        
          
        
        
          sample
        
        
          or each
        
        
          sam
        
        
          istance
        
        
          is led
        
        
          to th
        
        
          va
        
        
          cavation, the ditch bottom rose
        
        
          due to
        
        
          ining depth during the test
        
        
          were stepwise filled with
        
        
          water.
        
        
          0.25 m increase in water
        
        
          level.
        
        
          t step if no failure had
        
        
          rmation rates slowed
        
        
          l in the ditch was stepwise
        
        
          decre
        
        
          of a 0.25 m lowering of the water
        
        
          lowering the water table in
        
        
          instrumentation placed in the second measurement row is
        
        
          shown. At three depths in the peat layer and at one depth in the
        
        
          clay layer, the pore pressure development was measured with
        
        
          Vibrated Wire Probes, VWPs and horizontal displacements
        
        
          were with a SAA unit, see Abdoun (2007), placed at the front of
        
        
          the container row. For measurement rows 1 and 2 the pore
        
        
          pressure development was measured at only two locations in the
        
        
          peat layer. Besides the instrumentation in the measurement rows
        
        
          the settlement under the middle two containers was measured
        
        
          with automated settlement plates, the water level in the
        
        
          container was measured for each container and the heave of the
        
        
          ditch after excavation was measurement with settlement plates.
        
        
          the accuracy of the tip resistance
        
        
          indicates for the peat a tip resistance of the same order of
        
        
          magnitude as the accuracy for which the CPT is conducted. This
        
        
          is typical for peat areas and also reported for other locations
        
        
          Den Haan & Kruse (2007) and Boylan et al. (2011). The low
        
        
          accuracy for these measurements makes the conventional class
        
        
          2 CPT inadequate for accurate correlations with strength
        
        
          parameters. Ball penetrometer tests give a more accurate
        
        
          reading of the
        
        
          rrelation purposes.
        
        
          occurred in the previous step and defo
        
        
          3 LABORATORY TESTING
        
        
          A large series of triaxial tests and Direct
        
        
          ts were conducted. Discussion of all the test results is beyond
        
        
          the scope of this paper. This paper focusses on the DSS tests for
        
        
          which the sample was consolidated at approximately field
        
        
          stresses. The results of these tests were used to correlate the
        
        
          undrained shear strength,
        
        
          
            s
          
        
        
          
            u,
          
        
        
          to ball penetrometer tests according
        
        
          to equation (2).
        
        
          (2)
        
        
          in which
        
        
          
            q
          
        
        
          
            ball
          
        
        
          represents the penetration resistance of the ball
        
        
          and
        
        
          
            N
          
        
        
          
            ball
          
        
        
          represents the resistance factor.
        
        
          In total 5 samples were tests at field
        
        
          the density of the peat is low and the water table re
        
        
          rface the vertical effective stresses are also
        
        
          of 2 – 7 kN/m
        
        
          2
        
        
          . In the set-up of the field measurem
        
        
          boring, used to retrieve samples for DSS testing
        
        
          penetrometer test is executed at a distance of approxim
        
        
          m. The DSS-strength is defined as the deflection
        
        
          ’  diagram, representing the shear stress where the
        
        
          behaviour changes from compression to dilation. F
        
        
          ple the DSS-strength is compared to the measured res
        
        
          in the nearby ball penetrometer test at the same depth. Th
        
        
          e correlation of
        
        
          
            N
          
        
        
          
            ball
          
        
        
          = 17.9
        
        
          
        
        
          1.2.
        
        
          Figure 6. Su from ball penetrometer test S15e and DSS test results.
        
        
          Red square indicates DSS test at nearby location, red triangles other test
        
        
          results.
        
        
          The undrained shear strength from the DSS tests ranges from
        
        
          6.7 tot 7.7 kN/m
        
        
          2
        
        
          with an average value of 7.1 kN/m
        
        
          2
        
        
          . The
        
        
          riation in the test results is small and the average value for
        
        
          N
        
        
          ball
        
        
          found for the individual DSS test results fits well to the
        
        
          overall strength profile.
        
        
          4 TEST SET-UP
        
        
          In total 5 field tests are conducted. In the first two tests the peat
        
        
          is loaded in a few days to failure. The other 3 field tests include
        
        
          a two-periods loading procedure with several weeks of
        
        
          preloading before failure. This paper focusses only on the first
        
        
          two tests. For testing reproducibility the test loading is identical.
        
        
          Figure 7 shows the 3 loading phases of these tests.
        
        
          In loading phase 1, a row of containers is placed and a ditch
        
        
          is excavated. The containers have the dimensions of 7.25 m
        
        
          (length) × 2.5 m (width) × 2.2 m (height). The ditch has a depth
        
        
          of 2.5 m and slope 1:1. After ex
        
        
          swelling of the peat. The rema
        
        
          was approximately 2 m.
        
        
          In loading phase 2, the containers
        
        
          Each load step consisted of a
        
        
          It was decided to start the nex
        
        
          down significantly.
        
        
          In loading phase 3, the water leve
        
        
          ased. Each step consists
        
        
          table. Failure was found after
        
        
          loading phase 3.
        
        
          a) loading phase 1
        
        
          container
        
        
          Concrete slab
        
        
          Figure 7. Planned test set-up
        
        
          The instrumentation was concentrated in three measurement
        
        
          rows, see Figure 8. In figure 9 the location of the
        
        
          
            ball
          
        
        
          
            u
          
        
        
          
            ball
          
        
        
          
            q s
          
        
        
          
            N
          
        
        
          
        
        
          excavation
        
        
          
        
        
          2.5 m
        
        
          1 m
        
        
          b) loading phase 2
        
        
          c) loading phase 3
        
        
          were wi
        
        
          the con
        
        
          pressur
        
        
          peat lay
        
        
          the sett
        
        
          with a
        
        
          contain
        
        
          ditch af
        
        
          Figure 6. Su from ball penetrometer test S15e and DSS test results.
        
        
          Red square indicates DSS test at nearby location, red triangles other test
        
        
          results.
        
        
          The undrained shear strength from the DSS tests ranges from
        
        
          6.7 tot 7.7 kN/m
        
        
          2
        
        
          with an average value of 7.1 kN/m
        
        
          2
        
        
          . The
        
        
          absolute values and not expressed as a percentage. The average
        
        
          water content in the peat increases with depth from 6.45 (645%)
        
        
          to 12.4 (1240%). The organic clay layer shows lower values,
        
        
          with a large scatter. In the non-organic clay layer the water
        
        
          content reduces to 0.74 (74%).
        
        
          A series of CPTs, including measurements with different
        
        
          cone types and ball penetrometer tests, finalises the subsoil
        
        
          characterisation. Figure 1 shows a typical CPT for the test field.
        
        
          In this CPT, the succession of subsoil layers can be recognized.
        
        
          According to Eurocode ISO/DIS 22476-1:2005,IDT the
        
        
          measurements represent a class 2 type CPT. For a class 2 CPT
        
        
          , q
        
        
          c
        
        
          , equals 100 kPa. Figure 1
        
        
          resistance in peat and are therefore used for
        
        
          co
        
        
          Simple Shear (DSS)
        
        
          tes
        
        
          stress conditions. Since
        
        
          aches the
        
        
          su
        
        
          low, in the order
        
        
          ents for each
        
        
          , a ball
        
        
          ately 0.5
        
        
          point in the
        
        
          
        
        
          -
        
        
          
        
        
          sample
        
        
          or each
        
        
          sam
        
        
          istance
        
        
          is led
        
        
          to th
        
        
          va
        
        
          cavation, the ditch bottom rose
        
        
          due to
        
        
          ining depth during the test
        
        
          were stepwise filled with
        
        
          water.
        
        
          0.25 m increase in water
        
        
          level.
        
        
          t step if no failure had
        
        
          rmation rates slowed
        
        
          l in the ditch was stepwise
        
        
          decre
        
        
          of a 0.25 m lowering of the water
        
        
          lowering the water table in
        
        
          instrumentation placed in the second measurement ro is
        
        
          shown. At three depths in the peat layer and at one depth in the
        
        
          clay layer, the pore pressure development was measured with
        
        
          Vibrated Wire Probes, VWPs and horizontal displacements
        
        
          were with a SAA unit, see Abdoun (2007), placed at the front of
        
        
          the container row. For measurement rows 1 and 2 the pore
        
        
          pressure development was measured at only two locations in the
        
        
          peat layer. Besides the instrumentation in the measurement rows
        
        
          the settlement under the middle two containers was measured
        
        
          with automated settlement plates, the water level in the
        
        
          container was measured for each container and the heave of the
        
        
          ditch after excavation was measurement with settlement plates.
        
        
          the accuracy of the tip resistance
        
        
          indicates for the peat a tip resistance of the same order of
        
        
          magnitude as the accuracy for which the CPT is conducted. This
        
        
          is typical for peat areas and also reported for other locations
        
        
          Den Haan & Kruse (2007) and Boylan et al. (2011). The low
        
        
          accuracy for these measurements makes the conventional class
        
        
          2 CPT inadequate for accurate correlations with strength
        
        
          parameters. Ball penetrometer tests give a more accurate
        
        
          reading of the
        
        
          rrelation purposes.
        
        
          occurred in the previous step and defo
        
        
          3 LABORATORY TESTING
        
        
          A large series of triaxial tests and Direct
        
        
          ts were conducted. Discussion of all the test results is beyond
        
        
          the scope of this paper. This paper focusses on the DSS tests for
        
        
          which the sample was consolidated at approximately field
        
        
          stresses. The results of these tests were used to correlate the
        
        
          undrained shear strength,
        
        
          
            s
          
        
        
          
            u,
          
        
        
          to ball penetrometer tests according
        
        
          to equation (2).
        
        
          (2)
        
        
          in which
        
        
          
            q
          
        
        
          
            ball
          
        
        
          represents the penetration resistance of the ball
        
        
          and
        
        
          
            N
          
        
        
          
            ball
          
        
        
          represents the resistance factor.
        
        
          In total 5 samples were tests at field
        
        
          the density of the peat is low and the water table re
        
        
          rface the vertical effective stresses are also
        
        
          of 2 – 7 kN/m
        
        
          2
        
        
          . In the set-up of the field measurem
        
        
          boring, used to retrieve samples for DSS testing
        
        
          penetrometer test is executed at a distance of approxim
        
        
          m. The DSS-strength is defined as the deflection
        
        
          ’  diagram, representing the shear stress where the
        
        
          behaviour changes from compression to dilation. F
        
        
          ple the DSS-strength is compared to the measured res
        
        
          in the nearby ball penetrometer test at the same depth. Th
        
        
          e correlation of
        
        
          
            N
          
        
        
          
            ball
          
        
        
          = 17.9
        
        
          
        
        
          1.2.
        
        
          Figure 6. Su from ball penetrometer test S15e and DSS test results.
        
        
          Red square indicates DSS test at nearby location, red triangles other test
        
        
          results.
        
        
          The undrained shear strength from the DSS tests ranges from
        
        
          6.7 tot 7.7 kN/m
        
        
          2
        
        
          with an average value of 7.1 kN/m
        
        
          2
        
        
          . The
        
        
          riation in the test results is small and the average value for
        
        
          N
        
        
          ball
        
        
          found for the individual DSS test results fits well to the
        
        
          overall strength profile.
        
        
          4 TEST SET-UP
        
        
          In total 5 field tests are conducted. In the first two tests the peat
        
        
          is loaded in a few days to failure. The other 3 field tests include
        
        
          a two-periods loading procedure with several weeks of
        
        
          preloading before failure. This paper focusses only on the first
        
        
          two tests. For testing reproducibility the test loading is identical.
        
        
          Figure 7 shows the 3 loading phases of these tests.
        
        
          In loading phase 1, a row of containers is placed and a ditch
        
        
          is excavated. The containers have the dimensions of 7.25 m
        
        
          (length) × 2.5 m (width) × 2.2 m (height). The ditch has a depth
        
        
          of 2.5 m and slope 1:1. After ex
        
        
          swelling of the peat. The rema
        
        
          was approximately 2 m.
        
        
          In loading phase 2, the containers
        
        
          Each load step consisted of a
        
        
          It was decided to start the nex
        
        
          down significantly.
        
        
          In loading phase 3, the water leve
        
        
          ased. Each step consists
        
        
          table. Failure was found after
        
        
          loading phase 3.
        
        
          a) loading phase 1
        
        
          container
        
        
          Concrete slab
        
        
          Figure 7. Planned test set-up
        
        
          The instrumentation was concentrated in three measurement
        
        
          rows, see Figure 8. In figure 9 the location of the
        
        
          
            ball
          
        
        
          
            u
          
        
        
          
            ball
          
        
        
          
            q s
          
        
        
          
            N
          
        
        
          
        
        
          excavation
        
        
          
        
        
          2.5 m
        
        
          1 m
        
        
          b) loading phase 2
        
        
          c) loading phase 3