Uji Triaxial
Uji Triaxial
Uji Triaxial
www.geo-slope.com
Feature highlights
GeoStudio feature highlights include: Using the axisymmetric option to simulate a triaxial test Displacement-type boundary conditions to simulate a strain-controlled test Using the MCC model with a Load/Deformation analysis with no pore-pressure changes due to the loading A fully coupled analysis with specified initial pore-pressures A fully coupled analysis with zero-flow boundary conditions to simulate an undrained test
Included files
Full details of this example and the GeoStudio files are included as: MCC-triaxial tests.gsz MCC-triaxial tests.pdf The specifics of each analysis are available in the GeoStudio data file.
General Methodology
The simulated shearing phases are preceded by the simulation of the consolidation phase of a triaxial test. Consolidation is isotropic with the confining pressure equal to 100 kPa or 150 kPa. The isotropic stress state is simulated by applying a normal stress on the top and on the right side of the sample equal to 100 kPa or 150 kPa. The consolidation stage is set as the Parent; that is, the initial condition, for the subsequent simulations involving shearing.
Page 1 of 13
www.geo-slope.com
60 50
metres (x 0.001)
40 30 20 10 0 -10 -10
10
20
30
40
metres (x 0.001)
Figure 1 Triaxial test configuration for establishing initial stress state
The shearing phase of the analysis is simulated as a strain-rate controlled test. The definition of the strain-rate involves defining the number of time steps and the displacement that occurs over each step. Although the time steps are being defined, it is more appropriate to think of the time steps as load steps. Absolute time has no meaning in the context of these analyses. The number of load steps defined in the shear stage simulations is generally 50 or 25 and the incremental y-displacement of the top of the specimen is defined as -0.0002 m (per load step), where the negative sign indicates downward displacement. Consequently, 50 and 25 load steps multiplied by a y-displacement of -0.0002 m per load step results in a total vertical displacement of 0.01 m and 0.005 m, respectively. Symmetry is assumed about the vertical and horizontal centre-lines; consequently, only of the specimen is simulated. The dimensions of the simulation portion of the specimen are 0.025 m by 0.05 m, which is half of the width and height of a conventionally sized triaxial specimen. Total vertical y-displacements of 0.01 m and 0.005 m produce axial strains of 0.2 (or 20%) and 0.1 (10%), respectively.
Notice that Linear-Elastic parameters are used when setting up confining stresses; non-linear models are not required for this and the value of E is not relevant
The Cam-Clay properties are in the data file. The material has a of 25.84 degrees and an OCR of 1.25. A of 25.84 is equivalent to an value of 1.0.
Page 2 of 13
www.geo-slope.com
The first step in using the Cam-clay models is to establish the yield surface created sometime in the past under some particular stress state condition. In the field this would be some past insitu condition. This is referred to as the past or initial yield surface. SIGMA/W uses the initial vertical stress specified together with a specified value and a specified OCR (over-consolidation ratio) value to establish the past or starting yield surface. The initial confining stress is 100 kPa. The past vertical effective stress then is, y = 100 x OCR = 100 x 1.25 = 125 kPa Ko = 1- sin = 1 0.44 = 0.56 (formula in SIGMA/W code) x = z = 125 x 0.56 = 70.52 mean = (125 + 70.52 + 70.52) /3 = 88.68 The shear stress q at the past mean stress is,
1 ( 2
)2 (
)2 (
) 2 = 54.48 kPa
This is one point on the yield surface created by the past stresses. Since the sample is slightly over-consolidated, the past stress state was higher than the current stress state. Currently, the sample is isotropically consolidated to 100 kPa (x = y = z). The past maximum-mean stress (pre-consolidation pressure) is,
pc
1
2
q2
p2
122.15
px is at the center of the yield surface where q is at its maximum value on the initial yield surface; px = 122.15 / 2 = 61.08 kPa. Now that pc is known, we can draw the initial yield surface for an assumed range of p values between 0.0 and pc using the equation,
pc p
p2
Figure 2 shows the initial or past yield surface. The three means stresses computed above are marked on the diagram. The maximum q value occurs where the yield surface passes through the CSL (critical state line), as it properly should. The qpast and ppast values form the starting point for establishing the initial yield surface.
Page 3 of 13
www.geo-slope.com
80 70 60
Shear stress - q
50 40
CS
0 10 20 30
30 20 10 0
L
1
40
50
60
70
80
90
100
110
120
130
A drained test can be done with a Load/Deformation analysis, which does not involve any changes in pore-pressures due to the loading (straining). During drained loading, the yield surface continues to increase in size as Figure 3. The total stress path (which is equal to the effective stress path in this case) on a q-p plot for a triaxial test will have a slope of 1h:3v. This being the case, the stress path intersects the CSL at 150 kPa.
Page 4 of 13
www.geo-slope.com
160
140
120
3
Shear stress - q
100
CS
80
60
40
20
Figure 3 Total stress path and final yield surface under drained loading
The final deviatoric stress (q) will be 150 kPa. The final vertical (y) stress will be the confining stress plus the deviatoric stress; that is, 100 + 150 = 250 kPa. Moreover, this being a drained test, the sample will undergo some volumetric strain. The next three figures from SIGMA/W confirm this behavior.
Stress path
150
100
50
0 100
110
120
130
140
150
Page 5 of 13
www.geo-slope.com
Y stress - strain
250
200
150
100 Y-Strain
0.08
Volumetric Strain
0.06
0.04
0.02
0 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 Y-Strain
This analysis is a repeat of the previous drained test simulation, but now using the fully coupled formulation with pore-pressure specified as a constant equal to zero pore-pressure. The specified porepressure becomes a specified hydraulic boundary condition. Notice in Figure 7 how the hydraulic boundary condition is specified at all nodes and outside edges.
Page 6 of 13
www.geo-slope.com
60 50
metres (x 0.001)
40 30 20 10 0 -10 -10
10
20
30
40
metres (x 0.001)
Figure 7 Hydraulic boundary conditions for coupled drained simulation
q versus strain
150
100
50
0 Y-Strain
The results from the Coupled analysis are identical to the previous Load/Deformation analysis. This verifies that two different formulations give matching results. This example also illustrates that drained conditions in a Coupled analysis can be simulated by fixing the pore-pressures with a specified boundary condition.
The previous coupled analysis is now repeated, but the hydraulic boundary conditions are specified as no flow. Actually, leaving the hydraulic boundary conditions undefined has the effect of zero flow across the
Page 7 of 13
www.geo-slope.com
perimeter. By not allowing flow to exit from the sample, this analysis simulates a consolidated-undrained test with pore-pressure measurements. Based on theoretical consideration for the MCC model, the effective stress path should be vertical until it meets the past maximum yield surface as in Figure 9. Once the effective stress path meets the past yield surface, the path bends to the left and continues to rise slightly until it hits the CSL. The total stress path again is straight line rising at a 1h:3v slope. The different between the total and the effective stress paths is the excess pore-pressure. The final total mean stress is 122 kPa. The effective mean stress where the effective stress path hits the CSL is 66 kPa. The final pore-pressure therefore is 122 66 = 56 kPa.
90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 90 100 110 120 130 Mean stress - p'
u = 56 kPa
Shear stress - q
Two SIGMA/W output graphs below confirm these values. The SIGMA/W q-p plot ends at p = q = 66 kPa (Figure 10). The SIGMA/W pore-pressure versus y-strain plot indicates the maximum pore-pressure is 56 kPa (Figure 11).
Page 8 of 13
www.geo-slope.com
Stress path
70
Deviatoric Stress (q) (kPa)
60 50 40 30 20 10 0 65 70 75 80 85 90 95 100 105
pore-pressure:axial strain
60
50 40 30 20 10 0
0.02
0.04
0.06
0.08
0.1
0.12
Y-Strain
The volumetric strain for this test is zero, as it properly should be. The sample, however, undergoes some plastic strain, which results in some strain-hardening and the yield surface consequently expands such that it passes thorough the point where the effective stress path meets the CSL.
Analysis 4 Load-unload-reload
The MCC model treats the soil as elastic when the stress state is under the past maximum yield surface. In an undrained test, the effective p:q stress path is vertical inside the yield locus whether the loading path is one of unloading or loading. Figure 12 reveals that this is indeed the case: the stress path resumes its non-linear behavior once the yield locus is crosses upon re-loading.
Page 9 of 13
www.geo-slope.com
60 50 40 30 20 10 0 65 70 75 80 85 90 95 100 105
10
In this analysis, the sample is subjected to a cell pressure of 150 kPa and the back pressure is set to 50 kPa. The cell pressure is simulated with normal stress boundary conditions equal to 150 kPa. The back pressure is specified as a material activation PWP equal to 50 kPa. This makes the effective confining consolidation stress equal to 100 kPa. The next step is to simulate a consolidated undrained test with pore-pressure measurements for an overconsolidated soil. In this test, we will set the confining stress to 150 kPa with a back pressure of 50 kPa. The effective confining stress is again 100 kPa. This is discussed further in the next section. With and initial effective confining stress at 100 kPa, the past vertical effective stress is, y = 100 x OCR = 100 x 5.0 = 500 kPa Ko = 1- sin = 1 0.44 = 0.56 (formula in SIGMA/W code) x = z = 500 x 0.56 = 282.07 mean = (500 + 282.07 + 282.07) /3 = 354.71 The shear stress q at the past mean stress is,
1 ( 2
)2 (
)2 (
) 2 = 217.93
This is one point on the yield surface created by the past stresses. Since the sample is slightly over-consolidated, the past stress state was higher than the current stress state. Currently, the sample is isotropically consolidated to 100 kPa (x = y = z). The past maximum-mean stress (pre-consolidation pressure) is,
pc
1
2
q2
p2
488.6
Page 10 of 13
www.geo-slope.com
px is at the center of the yield surface where q is at its maximum value on the initial yield surface; px = 488.6 / 2 = 244.30 kPa. Now that pc is known, we can draw the initial yield surface for an assumed range of p values between 0.0 and pc using the equation,
pc p
p2
Figure 13 shows the initial or past yield surface. The three p stresses computed above are marked on the diagram. The maximum q value occurs where the yield surface passes through the CSL (critical state line) as it properly should. The qpast and ppast values form the starting point for establishing the initial yield surface.
350
300
250
qpast = 217.93
q - kPa
200
150
100
L CS
50
0 0 50 100 150 200 250 p' - kPa 300 350 400 450 500
Now the effective stress path starts at 100 kPa and rises vertically until it hits the initial yield surface. The soil behaves in a elastically up to this point, as illustrated in Figure 14. After meeting the yield surface, the effective stress path bends to the right and rises slightly until it intersects the total stress path. This is the point at which the excess pore-pressure is zero. After this point the excess pore-pressure diminishes until the effective stress path intersects the CSL. Beyond this point the soil behaves in a plastic manner with no further change in load or pore-pressure.
Page 11 of 13
www.geo-slope.com
350
250
q - kPa
200
150
0 0 50 100 150 200 250 p' - kPa 300 350 400 450 500
Of significance in this case is the fact that the effective stress path remains below the initial yield surface. This is in response to dilation that occurs once the stress path meets the initial yield surface.
11
The following graphs from SIGMA/W confirm this behavior. The effective stress path is vertical until meets the initial yield surface. Then it bends over to the right and continues to the right until the path meets the CSL.
Stress path
250
Deviatoric Stress (q) (kPa)
100 120 140 160 180 200 220 Mean Effective Stress (p') (kPa)
CS
300
u = -39 u = 0.0
Page 12 of 13
www.geo-slope.com
pore-pressure:axial strain
120
100 80 60 40 20 0
0.05
0.1 Y-Strain
0.15
0.2
0.25
The pore-pressure starts at 50 kPa, which is the starting back pressure. The pore-pressure then rises until the effectives stress path meets the initial yield surface. After that, the pore pressure diminishes due to the tendency for dilation until it approaches the CSL. The ending pore-pressure is around 10 KPa. Without the initial back pressure, the ending pore-pressure would be around -40 kPa. In SIGMA/W the MCC model is actually formulated only for saturated conditions and in the SIGMA/W formulation, this means the pore-pressure must be positive. This is the reason for the back pressure. The pore-pressure in this case can fall below the initial value and yet remain positive, so the SIGMA/W results can be compared with the hand-calculated values in Figure 14.
Page 13 of 13