Earthquake Induced Damage Mitigation from Soil Liquefaction Data ...

Earthquake Induced Damage Mitigation from Soil Liquefaction Data Report – Centrifuge Test CT7

March 2005 C-CORE Report R-04-095-145

This page is intentionally left blank

Earthquake Induced Damage Mitigation from Soil Liquefaction Data Report – Centrifuge Test CT7

Version 1

Prepared for: University of British Columbia

Prepared by: C-CORE

C-CORE Report: R-04-095-145

March 2005 C-CORE Captain Robert A. Bartlett Building Morrissey Road, St. John's, NL Canada A1B 3X5 T: (709) 737-8354 F: (709) 737-4706 [email protected] www.c-core.ca

Earthquake Induced Damage Mitigation from Soil Liquefaction

The correct citation for this report is: C-CORE. Earthquake Induced Damage Mitigation from Soil Liquefaction. Data report – Centrifuge Test CT7. Contract Report Prepared for University of British Columbia. C-CORE Report R-04-095-145, March 2005.

Project Team: Ryan Phillips (Project Manager) Minqiang Tu Stephen Coulter

R-04-068-145-CT5v1.doc December 2004

Earthquake Induced Damage Mitigation from Soil Liquefaction

EXECUTIVE SUMMARY A cooperative research project between the University of British Columbia (UBC), C-CORE, Memorial University of Newfoundland (MUN) and industrial partners was initiated in 2001 to optimize soil liquefaction treatment using numerical and centrifuge testing, and so reducing the cost associated with soil liquefaction. This project will be conducted with UBC researchers playing a leadership role and providing the numerical modeling procedures to be used. C-CORE will provide the centrifuge expertise and MUN will provide additional numerical modeling. The non-academic partners are all well experienced in earthquake remediation and will provide input on the soil conditions and structures that are of prime concern to them. They will have the major input on the tests to be carried out and will review the centrifuge data and analyses. Eight centrifuge tests with the test No. CT1 to CT8 will be conducted at C-CORE for this cooperative research project. The results of centrifuge tests CT1, CT2, CT3, CT4, CT5 and CT6 have been reported in C-CORE reports R-04-010-145, R-04-027-145, R04-029-145, R-04-030-145, R-04-068-145 and R-04-094-145, respectively. This report presents the results of the centrifuge test CT7 conducted on February 9, 2005 using the earthquake simulator (EQS) developed at C-CORE. The test CT7 had an embedded silt barrier layer with an input motion of 2A2475. The remaining tests should be completed by March 2005.

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

1

Earthquake Induced Damage Mitigation from Soil Liquefaction

TABLE OF CONTENTS Executive Summary ............................................................................................................ 1 1.1 1.2 1.3

Research Background...................................................................................... 3 Objective ......................................................................................................... 4 Scope of Work and Schedule for Remaining Work ........................................ 4

2

Soil Properties ............................................................................................................. 5

3

Earthquake Simulator and Data Acquisition System .................................................. 6

4

Model Configuration and Preparation ......................................................................... 7

5

Instrumentation and Model Container....................................................................... 10

6

The Test Results of CT7............................................................................................ 13 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9

Testing Procedure.......................................................................................... 13 The Record of a shock that the Model CT7 Experienced ............................. 14 The Test Results of CT7 for Swingup and Swingdown ................................ 18 The Test Results of CT7 for 2A2475 Input Motion...................................... 23 The Post-Test Excavation Observations in the Test CT7.............................. 33 The Model Profile of the Test CT7 before and after Test ............................. 36 P-wave Results of the Test CT7 .................................................................... 38 Temperature Observed in the Test CT7 ........................................................ 38 Viscosity Measurements in the Test CT7...................................................... 38

7

Suggested Improvements .......................................................................................... 39

8

REFERENCES.......................................................................................................... 40

Appendix A Testing Instrumentation Specifications

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

2

Earthquake Induced Damage Mitigation from Soil Liquefaction

1. Introduction 1.1

Research Background

This research is part of a NSERC collaborative research project between the University of British Columbia (UBC), C-CORE, Memorial University of Newfoundland (MUN) and industrial partners, which was initiated in 2001 to optimize soil liquefaction treatment and so reduce the costs associated with the mitigation of soil liquefaction. This collaborative research project arose as engineers from the major geotechnical consultants in British Columbia need reliable procedures for predicting liquefaction-induced displacements so that the mitigation could be optimized (www.civil.ubc.ca/liquefaction). There are a number of methods currently available for predicting liquefaction-induced displacements, ranging from empirical to numerical analysis. UBC has been a leader in this area for many years. These existing methods have been developed from post-earthquake observations and laboratory element test results. These methods have not yet been validated due to a lack of quality field data. Centrifuge tests, which started at Cambridge University in the late 1970s, have proven to be a powerful tool for seismic-induced liquefaction modeling (Schofield, 1981; Ko, 1994; Steedman and Ledbetter, 1994; Dobry et al., 1995). Centrifuge model testing offers the best opportunity for validating models by solving boundary value problems because the tests can be extensively instrumented, prepared under controlled conditions and shaken by prescribed inputs. Constitutive models, numerical procedures, and finite element models can be clearly tested by seeing how well the performance of centrifuge model tests can be predicted. Also, numerical models and procedures can be calibrated and improved or modified for phenomena that may not have been adequately accounted for in a model (Finn, et al., 1994). The multi-million dollar NSF VELACS project (Verification of Liquefaction Analysis by Centrifuge Studies) conducted between 1989– 1993 is a good example of research intended for this purpose. In this collaborative research project, UBC will conduct laboratory element tests, and UBC and MUN will carry out numerical analyses using different analysis methods. C-CORE will undertake the centrifuge model tests and the industrial partners will contribute their practical experience to the project.

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

3

Earthquake Induced Damage Mitigation from Soil Liquefaction

1.2

Objective

The objective of this research program is to investigate the effectiveness of mitigation measures used for slope stability against earthquake-induced pore pressure generation. The centrifuge tests will focus on (i) optimization of liquefaction treatment, including size of densified zone and densification effects, (ii) whether high pore pressures from liquefied loose zones will transmit into adjacent dense zones and result in significant movements under sloping condition, and (iii) if so, can drains in the densified zones prevent such movements? 1.3

Scope of Work and Schedule for Remaining Work

A total of eight centrifuge modeling tests will be conducted. The planned testing program is shown in Table 1-1. Table 1-1: Planned testing program Test Number CT1 CT2 CT3 CT4 CT5 CT6 CT7 CT8

Test Configuration Loose sand layer, No ground improvement Loose sand layer, No ground improvement Loose sand layer with dense dyke Loose sand layer with drainage dyke Loose sand layer with silt barrier layer and three drain dykes through silt barrier layer Loose sand layer without silt barrier layer Loose sand layer with silt barrier layer Loose sand layer with silt barrier layer and three drain dykes through silt barrier layer

Earthquake Event A475 followed by A2475 A2475 A2475 A2475 2A2475 2A2475 2A2475 2A2475

Note: Silt barrier layer is a low permeability layer with a slope of 1 vertical to 5.7 horizontal. Two earthquake input motions (A475 & A2475), which match the firm ground target spectrum of current building codes for earthquakes in Vancouver, will be used in this test program. The input motions A475 and A2475, respectively, have 10% and 2% exceedence limits in 50 years. According to the specification of the C-CORE earthquake simulator, the selected acceleration input motions A475 and A2475 have to be passed through a 40 to 200 Hz (0.57 to 2.86 Hz in prototype) band-pass filter. The filtered

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

4

Earthquake Induced Damage Mitigation from Soil Liquefaction

acceleration input motions will be further modified by Ernie Naesgaard using a 10th order polynomial for the baseline correction. After the test CT1, the project decided that only the earthquake input motion A2475 will be fired on the models of CT2 to CT4. The results of centrifuge tests CT1, CT2, CT3 and CT4 are reported in C-CORE Reports R-04-010-145, R-04-027-145, R-04-029-145 and R-04-030-145. After the test CT4, the project decided that an earthquake input motion 2A2475, the acceleration magnitude of which is 2 times of A2475, will be fired on the models of CT5 to CT8. The results of centrifuge tests CT5 and CT6 are reported in C-CORE Reports R-04-068-145 and R-04-094-145. The earthquake input motion 2A2475 was fired on the model CT7. The test CT8 should be completed in March 2005. 2

SOIL PROPERTIES

The Fraser River sand from seismically active Fraser River Delta in western Canada has been chosen for this study. The sand was screened using a 2mm sieve before pouring into the model container. The Fraser River sand is gray coloured medium-grained sand, and its average mineral composition is 40% quartz, quartzite, and chert, 11% feldspar, 45% unstable volcanic rock fragments, and 4% miscellaneous detritus (Vaid and Sivathayalan, 1996). The particle size distribution of this sand is shown in Figure 2-1. The sand has a mean particle size D50 of 0.26mm with fine content (passing #200 sieve) of 0.4%. The specific gravity of this sand is 2.71. The maximum and minimum dry densities are 1.67 g/cm3 and 1.40 g/cm3, respectively. The maximum and minimum void ratios are calculated as 0.94 and 0.62 (www.civil.ubc.ca/liquefaction). A coarse sand material has been selected to form the drainage dyke and drainage layer, and its particle size distribution curve is shown in Figure 2-1 for comparison. The specific gravity, maximum and minimum void ratios of the coarse sand are 2.67, 0.81 and 0.62, respectively. The hydraulic conductivity of coarse sand was estimated using Hazen’s formula from the particle size distribution curve to be 4.0 cm/s, which is about 100 times higher than the hydraulic conductivity of the surrounding Fraser River sand (4.3x10-2 cm/s at Dr= 40%). The silt that was used for the barrier layer consisted of U.S. Sil-Co-Sil 52 Fine Ground Silica silt. This material is uniform, white in colour, and consists of a mineral composition of primarily silicon dioxide quartz. The specific gravity of this material is 2.65 (http://www.u-s-silica.com/). R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

5

Earthquake Induced Damage Mitigation from Soil Liquefaction

Particle Size Distribution Curve 100 90 Drainage Gravel

80

60 Fraser River Sand

50 40 30 Sil-Co-Sil Silt

Percentage (%)

70

20 10

10

1

0.1

0.01

0 0.001

Particle Size (m m )

Figure 2-1. Particle size distribution curve 3

EARTHQUAKE SIMULATOR AND DATA ACQUISITION SYSTEM An earthquake simulator (EQS), Figure 3-1, with an integrated high-speed data

acquisition system, newly developed by Actidyn Systems, has been installed at C-CORE’s centrifuge facility (Perdriat et al., 2002). This earthquake simulator is capable of generating a 40g sine acceleration on a 400kg model at frequencies up to 200Hz or more, and it may generate larger accelerations up to 60g on smaller payloads. The fully computerized multi-degree-of-freedom control system can generate several modes of excitation such as: sine, broadband noise, or arbitrary transient waveforms with both amplitude and frequency control. This earthquake simulator is able to reduce the stress and strain induced in the centrifuge by model actuation through dynamic balancing, which is obtained by a combination of mechanical configurations, including balanced reciprocating motion, and advanced digital control as used in industry multiple shaker controls.

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

6

Earthquake Induced Damage Mitigation from Soil Liquefaction

Figure 3-1. Earthquake simulator The data acquisition system acquires data simultaneously with the operation of the EQS. The Matrix system includes eight analogue data inputs low pass filtered at 1kHz and sampled at 2.56kHz per channel using VXI hardware. This hardware has a further 24 channels of analogue inputs controlled by Data Physics 620 data acquisition software. These 24 inputs are typically filtered at 2kHz and sampled at 5.12kHz/channel for a 16 second period before, during and after the earthquake event. The P-waves were measured using two buried accelerometers connected in the test CT7 and the data for the P-wave tests were collected by Data Physics 620 data acquisition software. 4

MODEL CONFIGURATION AND PREPARATION

Figure 4-1 shows the model configuration with dimensions in prototype scale. The model CT7 is a loose sand slope with a embedded silt barrier layer. The model consists of a 2:1 slope (23.6m long and 11.8m high) with a crest width 28m. A silt barrier layer with a 1:5.7 slope and a thickness of approximately 1m was formed within the loose sand R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

7

Earthquake Induced Damage Mitigation from Soil Liquefaction

slope. A drainage layer with a thickness of 1.4m formed by coarse sand was placed on the bottom of the container in order to create a one-dimensional saturation front. The water level was 2.1m above the soil surface. Therefore, the model was fully submerged. The model was flown at 70g. The effective centrifuge radius is the distance from the central axis to two-thirds of the slope height measured from the top surface.

Figure 4-1. Model slope configuration The model slope was prepared by air pluviating Fraser River sand into the container through a funnel with one row of holes 3mm in diameter and spaced 8mm apart. The width of the funnel equals the width of the container. The funnel was manually moved back and forth along the longest dimension of the container, and the free-falling height was controlled to achieve the desired relative density. Once the sand layer of desired height is formed, the final slope geometry was created using a vacuum. Note that instrumentations, such as PPTs and accelerometers, were place at desired locations during the preparation of the sample. After the sand model below the silt barrier layer was prepared and saturated, a liberal amount of Vaseline was applied to the sidewalls where the silt and the sand on top of it will come in contact. Then the sil-co-sil silt was placed on the sand slope surface and the remaining sand was pluivated on the silt layer surface. A small amount of the fine R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

8

Earthquake Induced Damage Mitigation from Soil Liquefaction

fraction of Fraser River sand was placed along the walls of the container in each side during the placement of the silt barrier layer and the sand layer above it to reduce the side friction. The relative densities are 32% for loose sand, which have been achieved by adjusting the height of the funnel during pluviation. Some model compression will occur during model saturation and mechanical handling. The relative densities were increased to about 40% during self-weight compression of the model during the centrifuge test.

Figure 4-2. Vacuum saturation method A schematic diagram of saturation process is shown in Figure 4-2. A purge process with carbon dioxide (CO2) was conducted to displace less soluble air pockets prior to saturation of the model. Methylcellulose, instead of water, was used in this test in order to maintain scaling laws. The targeted viscosity of methylcellulose solution is 35 cst at 20°C, which is 35 times the viscosity of water. The methylcellulose powder with 1.8% to water in weight was mixed with water in an external container to achieve the required viscosity. The solution was then placed into a pressure vessel reservoir and kept under vacuum for 8 hours. A sheet metal former was placed on the slope surface to maintain its geometry during saturation and mechanical handling. Geotextile bags filled with gravel were placed on top of the former to hold it in position. De-aired methylcellulose mixture was then supplied from the bottom of the model container (Figure 4-2). The drainage layer, at the bottom of the container, provides a uniform saturation front and prevents any R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

9

Earthquake Induced Damage Mitigation from Soil Liquefaction

significant disturbances of the sand during saturation. A common vacuum source was applied to both the reservoir and the model container, with an equalization line to ensure a balance in the vacuum pressure in the two vessels. The whole saturation process takes about 2 days. In the case of the test CT7, the saturation was carried out twice. The sand model below the silt barrier layer was prepared and saturated first. Then the silt barrier layer was constructed and the remaining sand was pluviated after the vacuum was released. The model CT7 was pressurized at a pressure of 140 kPa for 3 days after the second saturation was accomplished.

5

INSTRUMENTATION AND MODEL CONTAINER

Figure 5-1 shows the planned instrumentation layout. The expected locations of the transducers are listed in Table 5-1. Pore pressure transducers and accelerometers were employed to monitor the responses of the model slope during the test. Linear Variable Differential Transformers (LVDTs) were used to measure vertical displacements at three locations as shown in Fig. 5-1 (L1, L3 and L4) and lateral displacement of the soil mass above the silt barrier layer at one location as shown in Fig. 5-1 (L2). The horizontal displacement of the model container was monitored using a laser transducer, L5. The specifications of various instruments used in this test are listed in Appendix A. The instruments have different frequency responses as listed in Appendix A. The miniature PPTs have a frequency response up to 2 kHz with no filter present; however, when placed in soil they should be fitted with a porous stone. Lee (1990) calculations show that this frequency response is not attenuated below 2 kHz for a 35 cst pore fluid for the type of sintered bronze stone used at C-CORE.

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

10

Earthquake Induced Damage Mitigation from Soil Liquefaction

Figure 5-1. Instrumentation layout (P: pore pressure transducers, A: accelerometer)

Table 5-1. Planned coordinates of transducers (unit: m) Transducer

X

Z

Transducer

X

Z

A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 P1 P2

25.7 9.8 5.6 25.0 9.8 25.1 40.8 17.2 25.1 32.6 10.0 22.5

3.7 14.2 12.6 10.5 21.7 18.5 15.6 23.5 22.5 21.1 10.5 10.5

P3 P4 P5 P6 P7 P8 P9 L1 L2 L3 L4 L5

36.8 4.4 14.2 24.0 33.8 43.6 24.0 3.0 30.8 25.0 45.0 51.6

10.5 22.8 20.8 17.5 17.0 14.9 22.1 24.0 20.0 24.0 15.5 10.2

The actual coordinates of transducers obtained from post-test excavation are shown in Table 5-2. R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

11

Earthquake Induced Damage Mitigation from Soil Liquefaction

Table 5-2. Actual coordinates of each transducer after test (unit: m) Transducer

X

Z

Transducer

X

Z

A1 A2 A3 A4 A5 A6 A7 A8 A9 A10

25.7 9.8 5.6 25.2 10.0 25.4 41.1 19.3 28.8 35.3

3.6 14.1 12.3 10.4 20.7 18.1 15.3 22.1 20.8 19.2

P1 P2 P3 P4 P5 P6 P7 P8 P9 L2

9.9 22.5 36.8 4.6 24.4 35.1 46.5 27.6 35.0

10.4 10.4 10.4 21.4 17.2 16.6 13.9 20.5 18.4

In the test CT7, the heads of the accelerometers of A2, A3, A5, A6, A7, A8, A9 and A10 were installed to face the down slope direction, and the A1 and A4, were in the opposite orientation. The accelerations in the datafile sent to UBC have been corrected such that all of the acceleration results are in the same direction and the positive accelerations correspond to down slope accelerations. Figure 5-2 shows the equivalent shear beam container (ESB container) developed at C-CORE. This container has been modified into a rigid container by installing 14 steel bars in the aluminum rings. These steel bars prevent any movement in the direction of shaking. The inside of the container was lined with stainless steel sheets to reduce the friction between the container and soil, which prevents water leakage.

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

12

Earthquake Induced Damage Mitigation from Soil Liquefaction

Figure 5-2. Rigid model container

6 6.1

THE TEST RESULTS OF CT7 Testing Procedure

Once the saturation was completed, the package was carefully moved onto the centrifuge arm by a forklift. All instrumentation was then connected to the signal box, and the model surface profile and the temperature of soil were measured on arm. The rotational speed of the centrifuge was increased gradually from 0 to 113 rpm. The actual inertia acceleration was calculated as 69.96g. This steady acceleration was maintained for 15 minutes for consolidation. During this period, all instruments were checked to ensure they were functioning properly. The earthquake input motion 2A2475 was fired. All data was collected using a high-speed data acquisition system. The centrifuge acceleration was maintained for another 15 minutes after the actuation to allow for the dissipation of excess pore pressure.

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

13

Earthquake Induced Damage Mitigation from Soil Liquefaction

6.2

The Record of a shock that the Model CT7 Experienced

The model CT7 experienced a shock before firing the earthquake input motion 2A2475 due to an unexpected shutdown of the EQS’s hydraulic pump. This shock was recorded by the data acquisition system. Figures 6-1 to 6-4 show the measured base horizontal acceleration and horizontal accelerations at various locations caused by the shock in the test CT7. Figures 6-5 to 6-7 show the measured time histories of excess pore pressure caused by the shock at various locations in model CT7 test. The excess pore pressures generated by the shock at the location of P4 reached the initial effective overburden stress, indicating the liquefaction of the soil near the surface on the slope top. However, the excess pore pressures generated by the shock were below the initial effective overburden stress at other locations. The measured histories of surface settlements caused by the shock in the test CT7 is shown in the Figure 6-8. As can be seen from the Figure 6-8, the shock caused a vertical settlement of between 100 and 200mm at the locations of L1, L2 and L3. The centrifuge was stopped after the model CT7 experienced the unexpected shock. The earthquake simulator and the model were carefully checked and then the test restarted after it was confirmed that the model had not failed. 1 A1 Depth = 20.3m, in loose sand layer

Acceleration (g)

0.5

0

-0.5

-1

0

1

2

3

4

5 Time (sec)

6

7

8

9

10

0.02 A1 Depth = 20.3m, in loose sand layer

0.015

Power

0.01

0.005

0

0

0.5

1

1.5 Frequency (Hz)

2

2.5

3

Figure 6-1. Measured base horizontal acceleration caused by the shock in model CT7 test

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

14

Earthquake Induced Damage Mitigation from Soil Liquefaction

1 A1 Depth = 20.3m, in loose sand layer 0

-1

0

1

2

3

4

5

6

7

8

9

10

9

10

9

10

9

10

1 A2 Depth = 9.8m, in loose sand layer 0

Acceleration (g)

-1

0

1

2

3

4

5

6

7

8

1 A3 Depth = 11.4m, in loose sand layer 0 -1 0

1

2

3

4

5

6

7

8

1 A4 Depth = 13.5m in loose sand layer 0 -1 0

1

2

3

4

5 Time (sec)

6

7

8

Figure 6-2. Measured horizontal acceleration caused by the shock in model CT7 test 1 A5 Depth = 2.3m, in loose sand layer

0.5 0 -0.5

Acceleration (g)

-1

0

1

2

3

4

5

6

7

8

9

10

9

10

1 ) g( n oi t ar el e c c A

A6 Depth = 5.5m, in loose sand layer

0.5 0 -0.5 -1 0

1

2

3

4

5

6

7

8

1 A7 Depth = 2m, in loose sand layer and under slope

0.5 0 -0.5 -1

0

1

2

3

4

5 Time (sec)

6

7

8

9

10

Figure 6-3. Measured horizontal acceleration caused by the shock in model CT7 test

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

15

Earthquake Induced Damage Mitigation from Soil Liquefaction

1 A8 Depth = 0.9m in loose sand layer

0.5 0 -0.5

Acceleration (g)

-1

0

1

2

3

4

5

6

7

8

9

10

9

10

9

10

0.5 0 -0.5 -1

A9 Depth = 2.5m, in loose sand layer 0

1

2

3

4

5

6

7

8

1 A10 Depth = 1.8m, in loose sand layer and under slope

0.5 0 -0.5 -1

0

1

2

3

4

5 Time (sec)

6

7

8

Figure 6-4. Measured horizontal acceleration caused by the shock in model CT7 test

150 100

P1 Depth = 13.5m, in loose sand layer initial vertical effective stress

50

Excess Pore Pressure (kPa)

0 -50 ) a P k( er u s s er P er o P s s e c x E

0

100

200

300

400

500

150 100

P2 Depth = 13.5m, in loose sand layer

50 0 -50

0

100

200

300

400

500

150 P3 Depth = 9m, in loose sand layer and under slope

100 50 0 -50

0

100

200

300 Time (sec)

400

500

Figure 6-5. Measured time histories of excess pore pressure caused by the shock in model CT7 test R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

16

Earthquake Induced Damage Mitigation from Soil Liquefaction

40 P4 Depth = 1.2m, in loose sand layer initial vertical effective stress

20 0

Excess Pore Pressure (kPa)

-20 0

100

200

300

400

500

40 20 P5 Depth = 3.2m, in loose sand layer 0 -20 0

100

200

300

400

500

40 P9 Depth = 1.9m, in loose sand layer

20 0 -20 0

100

200

300 Time (sec)

400

500

Figure 6-6. Measured time histories of excess pore pressure caused by the shock in model CT7 test

80 60 P6 Depth = 6.5m, in loose sand layer initial vertical effective stress

40 20

Excess Pore Pressure (kPa)

0 -20 -40

0

100

200

300

400

500

50 0 P7 Depth = 4.1m, in loose sand layer and under slope -50 0

100

200

300

400

500

100 P8 Depth = 1.3m, in looae sand layer and under slope

50 0 -50 0

100

200

300 Time (sec)

400

500

Figure 6-7. Measured time histories of excess pore pressure caused by the shock in model CT7 test R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

17

Earthquake Induced Damage Mitigation from Soil Liquefaction

200 LVDT1 surface settlement on slope top

0 -200 -400

Settlement (mm)

0 ) m m ( t n e m el tt e S

100

200

300

400

500

100 0 LVDT2 lateral displacement of soil mass above silt layer

-100 -200

0

100

200

300

400

500

0 -200

LVDT3 surface settlement at top of soil mass above silt layer

-400 0

100

200

300

400

500

0 -200 LVDT4 surface settlement at slope toe -400 0

100

200

300 Time (sec)

400

500

Figure 6-8. Measured histories of surface settlement caused by the shock in model CT7 test 6.3

The Test Results of CT7 for Swingup and Swingdown

Figure 6-9 to 6-11 show the static pore pressure at various locations during swingup. The transducer P5 malfunctioned in the test CT7. The other pore pressures increase with time from zero to constant values. The static pore pressures recorded match well with calculated static pore pressures using the planned depths of PPTs (Table 5-1) except those recorded by P4 and P9. The static pore pressures recorded by P4 and P9 are slightly higher than the calculated value, presumably due to the settlement of the transducers of P4 and P9 caused by the shock. Figure 6-12 shows the vertical settlements measured at various locations. The LVDT L2 was installed upside down and the tip of the spindle was connected with a string to measure the lateral displacement of the soil mass above the silt layer. The string was tightened by the self-weight of the spindle of L2 when the centrifugal acceleration increased to 70g, hence, the value recorded by L2 during the swingup was presumably due to the tightening of the string rather than a lateral displacement to the up-slope direction.

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

18

Earthquake Induced Damage Mitigation from Soil Liquefaction

150 100 P1 Depth = 13.4m calculated static pore pressure

Pore Pressure (kPa)

50 0 0.5 ) a P k( e r u s s e r P e r o P ci t at S

1

1.5

2

2.5

3

3.5

4 4

x 10 150 100 50

P2 Depth = 13.4m

0 0.5

1

1.5

2

2.5

3

3.5

4 4

x 10 150 100 P3 Depth = 9m

50 0 0.5

1

1.5

2 2.5 Time (sec)

3

3.5

4 4

x 10

Figure 6-9. Time histories of static pore pressure measured in model CT7 (swingup) 60 40 20 P4 Depth = 1.2m calculated static pore pressure

Pore Pressure (kPa)

0 0.5

1

1.5

2

2.5

3

3.5

4 4

x 10

80 60 40 P5 Depth = 3.2m

20 0 0.5

1

1.5

2

2.5

3

3.5

4 4

x 10 100

50 P6 Depth = 6.5m 0 0.5

1

1.5

2 2.5 Time (sec)

3

3.5

4 4

x 10

Figure 6-10. Time histories of static pore pressure measured in model CT7 (swingup)

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

19

Earthquake Induced Damage Mitigation from Soil Liquefaction

100

50 P7 Depth = 4.1m calculated static pore pressure

Pore Pressure (kPa)

0 0.5 ) a P k( er u s s er P er o P ci t at S

1

1.5

2

2.5

3

3.5

4 4

x 10 100

50 P8 Depth = 1.3m 0 0.5

1

1.5

2

2.5

3

3.5

4 4

x 10

60 40 20 P9 Depth = 1.9m

0 0.5

1

1.5

2 2.5 Time (sec)

3

3.5

4 4

x 10

Figure 6-11. Time histories of static pore pressure measured in model CT7 (swingup) 0

LVDT1 surface settlement on slope top

-100 -200

0.5

1

1.5

2

2.5

3

3.5

4

200

4

x 10

Settlement (mm)

LVDT2 lateral displacement of slope 0

-200

0.5

1

1.5

2

2.5

3

3.5

4 4

x 10

0 LVDT3 surface settlement on slope top -100 -200

0.5

1

1.5

2

2.5

3

3.5

4 4

x 10

0 -100 LVDT4 surface settlement at slope toe -200

0.5

1

1.5

2 2.5 Time (sec)

3

3.5

4 4

x 10

Figure 6-12. Time histories of surface settlement measured in model CT7 (swingup)

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

20

Earthquake Induced Damage Mitigation from Soil Liquefaction

Figures 6-13, 6-14 and 6-15 show the static pore pressure at various locations measured during the swingdown in the model CT7. Figure 6-16 shows the vertical settlements measured at various locations in the model CT7. The post-test observations showed that the extension bars of L1, L3 and L4 were off the pads. Therefore, no reasonable rebounds were recorded by L1, L3 and L4. The LVDT L2 was installed upside down and the tip of the spindle was connected with a string to measure the lateral displacement of the soil mass above the silt layer. The string tension by the self-weight of the spindle of L2 when the centrifugal acceleration decreased to 1g from 70g and the value recorded by L2 during the swingdown was presumably due to the untensioning of the string rather than a lateral displacement to the down-slope direction.

150

P1 Depth = 13.4m

100

Pore Pressure (kPa)

50 0 1

2

3

4

5

6 4

x 10 150 P2 Depth = 13.4m 100 50 0 1

2

3

4

5

6 4

x 10 150 P3 Depth = 9m

100 50 0 1

2

3 Time (sec)

4

5

6 4

x 10

Figure 6-13. Time histories of static pore pressure measured in model CT7 (swingdown)

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

21

Earthquake Induced Damage Mitigation from Soil Liquefaction

60 40

P4 Depth = 1.2m

20

Pore Pressure (kPa)

0 1

2

3

4

5

6 4

x 10

80 60 P5 Depth = 3.2m

40 20 0 1

2

3

4

5

6 4

x 10 100 P6 Depth = 6.5m 50

0 1

2

3 Time (sec)

4

5

6 4

x 10

Figure 6-14. Time histories of static pore pressure measured in model CT7 (swingdown) 100 P7 Depth = 4.1m 50

0 1

2

3

4

5

6

Pore Pressure (kPa)

4

x 10 100 P8 Depth = 1.3m 50

0 1

2

3

4

5

6 4

x 10

60 40

P9 Depth = 1.9m

20 0 1

2

3 Time (sec)

4

5

6 4

x 10

Figure 6-15. Time histories of static pore pressure measured in model CT7 (swingdown)

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

22

Earthquake Induced Damage Mitigation from Soil Liquefaction

0 LVDT1 surface settlement on slope top

-100

Settlement (mm)

-200 ) m m ( t n e m el tt e S

1

2

3

4

5

6 4

x 10

600

LVDT2 lateral displacement of slope

400 200 0

1

2

3

4

5

6 4

x 10

0 LVDT3 surface settlement on slope top

-100 -200

1

2

3

4

5

6 4

0

x 10

-200 LVDT4 surface settlement at slope toe

-400 -600

1

2

3 Time (sec)

4

5

6 4

x 10

Figure 6-16. Time histories of surface settlement measured in model CT7 (swingdown) 6.4

The Test Results of CT7 for 2A2475 Input Motion

Figure 6-17 shows the comparison between the planned horizontal earthquake input and the actual base horizontal acceleration for the test CT7. The triaxial accelerometer on the shake table was not monitored in the test CT7. The base motion was assessed from the average of the accelerations measured at the table edges by Endevco 5216-100 accelerometers, which have a frequency response of 10-1,000Hz. The target motion was reasonably well replicated. The measured horizontal accelerations by each accelerometer are shown in Figures 6-18 to 6-20. The accelerometer of A1 was placed on the bottom of the loose sand layer at a depth of 20.3m; hence, the acceleration response at this location should be close to the input motion. The time history of the acceleration recorded by A1 is very similar in shape and magnitude with the input motion. The accelerometer of A4 was placed in the loose sand layer at a depth of 13.5m. The time history of A4 is similar with the input motion with neither amplification nor attenuation.

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

23

Earthquake Induced Damage Mitigation from Soil Liquefaction

The accelerometers of A2 and A3 were placed in the loose sand layer at a depth of 9.8 and 11.4m, respectively, and A2 and A3 were used to measure the P-wave velocity in the soil. The time history of A2 and A3 are similar with the input motion with neither amplification nor attenuation. The accelerometers of A5, A6 and A7 were placed in the loose sand layer and under silt barrier layer at a depth of 2.3m, 5.5m and 2.0m, respectively. The time histories of the accelerations recorded by A6 and A7 showed a directional bias, namely large acceleration spikes occurred exclusively in the negative direction, accompanied by low amplitude acceleration response in the positive direction. This observed bias in acceleration magnitude was presumably due to the movement of the soil towards the down-slope side during shaking. The post-test excavation revealed that the accelerometers of both A6 and A7 moved 4mm towards down-slope side. However, there is no such a directional bias in the time history of the acceleration recorded by A5. The accelerometers of A8, A9 and A10 were placed in the loose sand layer above the silt barrier layer at a depth of 0.6m, 1.5m and 0.6m, respectively. The time histories of the accelerations recorded by A8, A9 and A10 showed a directional bias, namely large acceleration spikes occurred exclusively in the negative direction, accompanied by low amplitude acceleration response in the positive direction. This observed bias in acceleration magnitude was presumably due to the movement of the soil towards the down-slope side during shaking. The post-test excavation revealed that the accelerometers of A8, A9 and A10 moved 30mm, 53mm and 38mm towards down-slope side, respectively.

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

24

Earthquake Induced Damage Mitigation from Soil Liquefaction

1 Planned horizontal earthquake input actual base horizontal acceleration Tz

Acceleration (g)

0.5

0

-0.5

-1

0

5

10

15

20

25

Time (sec)

1 0.8

Planned horizontal earthquake input actual base horizontal acceleration

Power

0.6 0.4 0.2 0

0

0.5

1

1.5 Frequency (Hz)

2

2.5

3

Figure 6-17. Comparison between planned horizontal earthquake input and actual base horizontal acceleration in model CT7 test 1 A1 Depth = 20.3m, in loose sand layer 0

-1

0

5

10

15

20

25

20

25

20

25

20

25

1 A2 Depth = 9.8m, in loose sand layer 0

Acceleration (g)

-1

0

5

10

15

1 A3 Depth = 11.4m, in loose sand layer 0

-1

0

5

10

15

1 A4 Depth = 13.5m in loose sand layer 0

-1

0

5

10

15 Time (sec)

Figure 6-18. Measured horizontal acceleration in model CT7 test

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

25

Earthquake Induced Damage Mitigation from Soil Liquefaction

1 A5 Depth = 2.3m, in loose sand layer

0.5 0 -0.5

Acceleration (g)

-1

0

5

10

15

20

25

20

25

1 ) g( n oi t ar el e c c A

A6 Depth = 5.5m, in loose sand layer

0.5 0 -0.5 -1

0

5

10

15

1 A7 Depth = 2m, in loose sand layer and under slope

0.5 0 -0.5 -1

0

5

10

15

20

25

Time (sec)

Figure 6-19. Measured horizontal acceleration in model CT7 test 1 A8 Depth = 0.6m in loose sand layer

0.5 0 -0.5

Acceleration (g)

-1

0

5

10

15

20

25

20

25

0.5 A9 Depth = 1.5m, in loose sand layer 0 -0.5 -1

0

5

10

15

1 A10 Depth = 0.6m, in loose sand layer and under slope

0.5 0 -0.5 -1

0

5

10

15

20

25

Time (sec)

Figure 6-20. Measured horizontal acceleration in model CT7 test

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

26

Earthquake Induced Damage Mitigation from Soil Liquefaction

Figures 6-21, 6-22, and 6-23 show the short-term time histories of the excess pore pressures measured at different locations in the test CT7. The transducer P5 malfunctioned in the test CT7. The pore pressure transducers of P1, P2 and P3 were installed in the loose sand layer at the depths of 13.5m, 13.5m and 9.1m, respectively. The maximum excess pore pressure recorded by P1, P2 and P3 reached a level that is approximately 0.82, 0.61 and 0.53 of the initial effective overburden stress at P1, P2 and P3, indicating that liquefaction did not occur at the elevation of 13.5m measured from the slope top. The pore pressure transducers of P4 and P8 were installed beneath the silt barrier layer and in the loose sand layer at the depths of 1.2m and 1.3m, respectively. The excess pore pressures recorded by P4 and P8 reached the initial effective overburden stress, indicating the liquefaction of the surrounding soil at these locations. The pore pressure transducers of P6 and P7 were installed beneath the silt barrier layer at the depths of 6.5m and 4.1m, respectively. The maximum excess pore pressures recorded by P6 and P7 reached a level that is 0.86 and 0.4 of the initial effective overburden stress at P6 and P7, respectively, indicating that liquefaction did not occur at these locations. The pore pressure transducer of P9 was placed above the silt barrier layer at a depth of 1.9m. The maximum excess pore pressure recorded by P9 reached a level that is approximately 0.35 of the initial effective overburden stress at P9. Figures 6-24, 6-25 and 6-26 show the long-term time histories of the excess pore pressures measured at different locations in the test CT7. The earthquake motion took place at 1.7 seconds in prototype scale after starting data recording. Full dissipation of the excess pore pressure at P3 and P8, which were placed at the depths of 9.0m and 1.3m finished in about 763 and 254 seconds after the earthquake motion took place. The long-term excess pore pressure recorded by other pore pressure transducers did not drop back to zero, indicating there were some settlements of these transducers. The pore pressure transducers of P4, P6 and P7 were installed beneath the silt barrier layer. As can be seen from Figures 6-25 and 6-26, the excess pore pressures measured by P4 and P7 continued to rise after the earthquake motion ceased. This is a clear indication of the accumulation of water beneath the silt layer with a corresponding increase in void R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

27

Earthquake Induced Damage Mitigation from Soil Liquefaction

ratio (dilation), and consequently resulted in a post-shaking movement of the soil mass above the silt barrier layer.

150 100

Excess Pore Pressure (kPa)

50 P1 Depth = 13.5m, in loose sand layer initial vertical effective stress

0 0 ) a P k( e r u s s e r P e r o P s s e c x E

5

10

150

15

20

25

30

35

40

35

40

P2 Depth = 13.5m, in loose sand layer

100 50 0 0

5

10

15

20

25

30

150 P3 Depth = 9m, in loose sand layer and under slope 100 50 0 0

5

10

15

20 Time (sec)

25

30

35

40

Figure 6-21. Short-term time histories of excess pore pressure measured in model CT7 40 20 0 P4 Depth = 1.2m, in loose sand layer initial vertical effective stress

Excess Pore Pressure (kPa)

-20 0

5

10

15

20

25

30

35

40

35

40

35

40

40 20

P5 Depth = 3.2m, in loose sand layer

0 -20 0

5

10

15

20

25

30

40 P9 Depth = 1.9m, in loose sand layer 20 0 -20 0

5

10

15

20 Time (sec)

25

30

Figure 6-22. Short-term time histories of excess pore pressure measured in model CT7 R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

28

Earthquake Induced Damage Mitigation from Soil Liquefaction

80 60 40 20 0

P6 Depth = 6.5m, in loose sand layer initial vertical effective stress

Excess Pore Pressure (kPa)

-20 -40

0

5

10

15

20

25

30

35

40

P7 Depth = 4.1m, in loose sand layer and under slope

50 0 -50 0

5

10

15

20

25

30

35

40

100 P8 Depth = 1.3m, in looae sand layer and under slope 50 0 -50 0

5

10

15

20 Time (sec)

25

30

35

40

Figure 6-23. Short-term time histories of excess pore pressure measured in model CT7 150 100

P1 Depth = 13.5m, in loose sand layer initial vertical effective stress

50

Excess Pore Pressure (kPa)

0 -50 ) a P k( e r u s s e r P e r o P s s e c x E

0

100

200

300

400

500

600

700

800

700

800

150 100 P2 Depth = 13.5m, in loose sand layer

50 0 -50

0

100

200

300

400

500

600

150 P3 Depth = 9m, in loose sand layer and under slope

100 50 0 -50

0

100

200

300

400 500 Time (sec)

600

700

800

Figure 6-24. Long-term time histories of excess pore pressure measured in model CT7

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

29

Earthquake Induced Damage Mitigation from Soil Liquefaction

30 P4 Depth = 1.2m, in loose sand layer initial vertical effective stress

20 10 0

Excess Pore Pressure (kPa)

-10 -20

0

100

200

300

400

500

600

700

800

700

800

700

800

40 20 P5 Depth = 3.2m, in loose sand layer 0 -20 0

100

200

300

400

500

600

20 0 P9 Depth = 1.9m, in loose sand layer

-20 0

100

200

300

400 500 Time (sec)

600

Figure 6-25. Long-term time histories of excess pore pressure measured in model CT7 80 60 P6 Depth = 6.5m, in loose sand layer initial vertical effective stress

40 20

Excess Pore Pressure (kPa)

0 -20

0

100

200

300

400

500

600

700

800

50

0 P7 Depth = 4.1m, in loose sand layer and under slope -50 0

100

200

300

400

500

600

700

800

50

0 P8 Depth = 1.3m, in looae sand layer and under slope -50 0

100

200

300

400 500 Time (sec)

600

700

800

Figure 6-26. Long-term time histories of excess pore pressure measured in model CT7 Figures 6-27 and 6-28 show the short and long-term time histories of the vertical displacements of the soil surface measured by LVDT L1, L3 and L4 and of the lateral R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

30

Earthquake Induced Damage Mitigation from Soil Liquefaction

displacement of the soil mass above the silt barrier layer measured by LVDT L2. An acrylic bar with a diameter of 6mm that was buried on the surface of the silt barrier layer was connected to the tip of the spindle of LVDT L2 by a string to measure the lateral displacement. The 30mm by 30mm LVDT pads for L1, L3 and L4 exerted a bearing pressure of less than 6 kPa. The horizontal displacement of the model container was not recorded by L5 in the test CT7. The post-excavation results revealed that the extension bars of L1, L3 and L4 were off their pads and consequently the extension bars penetrated into the soil 110mm, 11mm and 16mm (model scale), respectively. The vertical displacement recorded by L1 shows that the soil surface started to settle as the earthquake motion took place, and recorded the most of its settlements with large settlement oscillations during the earthquake motion. The vertical displacement recorded by L1 also indicates that the reading started to rebound at about 25 second and then reached to a constant value. This is because the extension bar of L1 penetrated into the soil 110mm after the shaking and consequently resulted in the LVDT L1 being out of working range. The vertical displacement recorded by L3, which was located on the top of the slope, shows that the soil surface started to settle as the earthquake motion took place, and recorded the most of its settlements during the earthquake motion. The vertical displacement recorded by L3 also shows that there was a little settlement after the earthquake motion ceased and then the soil surface at location of L3 started to settle again at about 60 seconds. This is caused by the post-shaking movement of the soil mass above the silt layer The vertical displacement at the slope recorded by L4 shows that the soil surface started to settle as the earthquake motion took place, and then it began to heave and settle again after the first 4 and 8 cycles of vibration, respectively. The post-test profile measurement results showed that the soil surface heaved at the location of L4 and was different from the measurement result by L4. The reason for this difference is due to the movement of the pad of L4 that allowed the LVDT’s extension rod to penetrate into the soil 16mm and prevent it from recording the soil heave. The LVDT L2 did not properly function in the test CT7. The lateral displacement of the soil mass above the silt layer at the location of L2 was estimated to be 4.49m through comparing the coordinates of L2 before and after the test. However, the reading recorded by L2 indicates that there was a large movement of the soil mass above the silt layer R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

31

Earthquake Induced Damage Mitigation from Soil Liquefaction

occurred at about 60 seconds and this coincided with the observation in the L3’s record that the soil mass above the silt layer moved to the down-slope direction after the earthquake motion ceased.

500 0 -500

LVDT1 surface settlement on slope top

-1000

Settlement (mm)

0 ) m m ( t n e m el tt e S

20

40

60

80

100

120

140

160

140

160

100 0 -100 -200

LVDT2 lateral displacement of soil mass above silt layer 0

20

40

60

80

100

120

0 LVDT3 surface settlement at top of soil mass above silt layer -500 -1000

0

20

40

60

80

100

120

140

160

140

160

100 LVDT4 surface settlement at slope toe

0 -100 -200

0

20

40

60

80 Time (sec)

100

120

Figure 6-27. Short-term time histories of surface settlement measured in model CT7

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

32

Earthquake Induced Damage Mitigation from Soil Liquefaction

500 0 -500

LVDT1 surface settlement on slope top

-1000

Settlement (mm)

0 ) m m ( t n e m el tt e S

100

200

300

400

500

600

700

800

100 0 -100 -200

LVDT2 lateral displacement of soil mass above silt layer 0

100

200

300

400

500

600

700

800

0 LVDT3 surface settlement at top of soil mass above silt layer

-500 -1000

0

100

200

300

400

500

600

700

800

100 0

LVDT4 surface settlement at slope toe

-100 -200

0

100

200

300

400 500 Time (sec)

600

700

800

Figure 6-28. Long-term time histories of surface settlement measured in model CT7 6.5

The Post-Test Excavation Observations in the Test CT7

The photos of the model CT7 before and after the test are shown in Figures 6-29 and 6-30. Eight white markers and one blue marker with a size of 2-3mm were installed in the loose sand layer and the silt layer in the model CT7, respectively, to investigate the lateral displacement of the slope through comparing their coordinates before and after the test. Figure 31 shows the layout of the markers installed in the model CT7. However, the markers of M5 and M7 were missing in the post excavation. The table 6-1 shows the coordinates of the markers before and after the test. The pore pressure transducers of P6, P7, P8 and P9 were placed in loose sand layer and their cables were fixed on the sidewalls, thus, they are free to move along the longitudinal direction. The movements of these transducers can be used to identify the soil lateral displacement. Figure 6-32 shows the positions of these transducers and markers installed in the model CT7 before and after the test.

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

33

Earthquake Induced Damage Mitigation from Soil Liquefaction

Figure 6-29. The model CT7 before the test

Figure 6-30. The model CT7 after the test R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

34

Earthquake Induced Damage Mitigation from Soil Liquefaction

Figure 6-31. The layout of markers installed in the test CT7 Table 6-1. The coordinates of markers before and after the test Coordinates of markers before the test Marker X Z M1 M2 M3 M4 M5 M6 M7 M8 M9

20.6 20.6 20.6 30.0 30.0 30.0 40.8 40.8 20.6

23.1 17.5 14.2 22.5 17.0 13.5 18.1 13.5 21.7

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

Coordinates of markers after the test Marker X Z M1 M2 M3 M4 M5 M6 M7 M8 M9

23.1 20.9 20.7 34.1 30.2 41.7 23.2

22.0 17.0 13.8 20.7 13.2 13.0 20.7

35

Earthquake Induced Damage Mitigation from Soil Liquefaction

30

Elevation (m)

25

20

15

10

Original soil surface Post-test soil and silt layer surface Positions of transducers or markers before test Positions of transducers or markers after test

5

0

0

5

10

15

20

25 30 Slpoe length (m)

35

40

45

50

Figure 6-32. Observed lateral displacement below silt barrier layer in model CT7 6.6

The Model Profile of the Test CT7 before and after Test

The vertical distance of the model surface to the top of the container was measured by a scale before and after the test at different locations along the longitudinal direction of the model. The measurement values (in model scale) are in Table 6-2. The model profiles of CT7 after pluviation, and before and after the test are shown in Figure 6-33. Selfweight compression from 32% to around 40% relative density of a 320mm deep loose sand accounts for about 5mm of the observed 12mm of slope crest settlement, Z2.

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

36

Earthquake Induced Damage Mitigation from Soil Liquefaction

Table 6-2. The measured model profiles X (mm) Y1 (after poured mm) Y2 (before flying mm) Settlement Z1 (mm) Y3 (after test mm) ettlement Z2 (mm) X (mm) Y1 (after poured mm) Y2 (before flying mm) Settlement Z1 (mm) Y3 (after test mm) Settlement Z2 (mm)

0 230 233 3 245 12 400 232 235 3 253 18

50 230 233 3 245 12 450 255 258 3 260 2

100 230 233 3 245 12 500 280 282 2 278 -4

150 230 232 2 245 12 550 305 307 2 303 -4

200 230 233 3 246 13 600 330 333 3 327 -6

250 230 233 3 246 13 650 355 358 3 348 -10

300 230 233 3 247 14 700 380 379 -1 358 -21

350 230 234 4 248 14 737 399 392 -7 365 -27

Note: X: the longitudinal distance measured from the container end wall on upslope side; Y: the vertical distance measured to the model soil surface from top of container; Settlement Z1: the vertical settlement caused by the model saturation and mechanical handling; Settlement Z2: the vertical settlement caused by self weight compression of the model during the centrifuge test and earthquake shaking.

30

25

Elevation (m)

20

15 Original soil surface soil surface before test soil and silt layer surface after test

10

5

0

0

5

10

15

20

25 30 Length (m)

35

40

45

50

Figure 6-33. Profile of the model CT7 before and after test R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

37

Earthquake Induced Damage Mitigation from Soil Liquefaction

6.7

P-wave Results of the Test CT7

P-wave measurements were attempted in-flight to evaluate whether the models were adequately saturated during the test CT7. The P-wave was generated by hitting a bolt that passed through the wall of the model container by a small electric solenoid hammer. The P-waves were captured by accelerometers A2 and A3 at a distance of 64 mm. The Pwave velocity is determined from the difference in the first wave arrival time at the two accelerometers (A2 and A3). The estimated P-wave velocities at different g level in the test CT7 are shown in Table 6-2. According to the study by Ishihara et al. (2001), generally, P-wave velocity of Vp greater than 750 m/s indicates a degree of saturation in excess of 99%. The Vp value measured from the test CT7 at 70g was very close to 750 m/s and it indicated that a degree of saturation about 99% was achieved for the soil model of CT7. Table 6-2. The P-wave velocity measured in the test CT7 g-level 1g 70g 6.8

P-wave velocity (m/s) 500 744

Temperature Observed in the Test CT7

In the test CT7, a thermometer was installed in the loose sand layer at a depth of 17.6m to monitor the temperature change in the soil model. The shaking table of the EQS, on which the model container is installed, is heated by the oil under the shaking table during test. The temperatures were measured to be 21.5 oC and 22.5 oC, respectively, before and after the test in the test CT7. The temperature change during the test CT7 was smaller than that observed in the previous tests. This improvement on controlling the temperature change in the test CT7 is due to a major modification of the EQS completed recently. 6.9

Viscosity Measurements in the Test CT7

The designed viscosity is 35 cst for this research project. The actual viscosity for the test CT7 was measured to be 37.3 cst at a temperature of 18.7 oC after the saturation of the model.

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

38

Earthquake Induced Damage Mitigation from Soil Liquefaction

7

SUGGESTED IMPROVEMENTS The following improvements are suggested for the future tests:

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

39

Earthquake Induced Damage Mitigation from Soil Liquefaction

8

REFERENCES

C-CORE (2004). “COSTA-A Centrifuge Test Data Report.” C-CORE Report R-04-009075, June 2004. C-CORE (2004). “Earthquake Induced Damage Mitigation from Soil Liquefaction. Data Report – Centrifuge Test CT1.” Contract Report Prepared for University of British Columbia. C-CORE Report R-04-010-145, June 2004. C-CORE (2004). “Earthquake Induced Damage Mitigation from Soil Liquefaction. Data Report – Centrifuge Test CT2.” Contract Report Prepared for University of British Columbia. C-CORE Report R-04-027-145, July 2004. C-CORE (2004). “Earthquake Induced Damage Mitigation from Soil Liquefaction. Data Report – Centrifuge Test CT3.” Contract Report Prepared for University of British Columbia. C-CORE Report R-04-029-145, July 2004. C-CORE (2004). “Earthquake Induced Damage Mitigation from Soil Liquefaction. Data Report – Centrifuge Test CT4.” Contract Report Prepared for University of British Columbia. C-CORE Report R-04-030-145, July 2004. C-CORE (2004). “Earthquake Induced Damage Mitigation from Soil Liquefaction. Data Report – Centrifuge Test CT5.” Contract Report Prepared for University of British Columbia. C-CORE Report R-04-068-145, December 2004. C-CORE (2005). “Earthquake Induced Damage Mitigation from Soil Liquefaction. Data Report – Centrifuge Test CT6.” Contract Report Prepared for University of British Columbia. C-CORE Report R-04-094-145, February 2005. Dobry, R., Taboada, V. and Liu, L. (1995). “Centrifuge modeling of liquefaction effects during earthquakes.” Proc., IS-Tokyo ’95, The First International Conference on Earthquake Geotechnical Engineering, A.A. Balkema; 1291-1324. R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

40

Earthquake Induced Damage Mitigation from Soil Liquefaction

Finn, W. D. Liam, Ledbetter, R. H. and Marcuson, W. F. (1994). “Seismically induced deformation and remediation design.” Proc., the 4th US-Japan workshop on soil liquefaction, Tsukuba Science City, Japan, July 4-6, Ko. H. Y. (1994). “Modeling seismic problems in centrifuges.” Centrifuge 94, Leung, Lee & Tan, eds., Balkema, Rotterdam, pp. 3-12. Lee, F. H. (1990). “Frequency Response of Diaphragm Pore Pressure Transducers in Dynamic Centrifuge Model Test.” ASTM Geotechnical Testing Journal. Vol. 13, pp. 201-207. Perdriat, J., Phillips, R. Nicolas Font, J. & Hutin, C. (2002). “Dynamically balanced broad frequency earthquake simulation system.” Proceedings of International Conference on Physical Modeling in Geotechnics, July 2002, St John’s, Newfoundland, Canada. Schofield, A. N. (1981). “Dynamic and earthquake geotechnical centrifuge modeling.” International Conference on Recent Advances in Geotechnical Earthquake Engineering and Soil Dynamics, pp.1081-1100. Steedman, R. S. and Ledbetter, R. H. (1994). “Centrifuge modeling of earthquake problems.” Proc. 4th US – Japan Workshop on Soil Liquefaction, Tsukuba Science City, Japan, July 4-6, pp.353-386. Vaid, Y.P. and Sivathayalan, S. (1996). “Static and cyclic liquefaction potential of Fraser Delta sand in simple shear and triaxial tests.” Canadian Geotechnical Journal, Vol. 33, pp. 281-289.

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

41

Earthquake Induced Damage Mitigation from Soil Liquefaction

APPENDIX A Testing Instrumentation Specifications

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

42

Earthquake Induced Damage Mitigation from Soil Liquefaction

DRUCK PDCR 81 Miniature Pore Pressure Transducer

Dimensions: Operating Pressure Ranges: Excitation Voltage: Output Voltage: Zero Offset: Span Setting: Output Impedence: Load Impedence: Resolution: Operating Temperature: Mechanical Shock: Weight:

6.5 x 11.7 mm 100 and 200 psi 5 volts 6 ma nominal 75 mV ± 10 mV maximum ± 20% of nominal output 1000 ohms Greater than 100 kohms Infinite -5 O to 250oF 1000 g for 1 ms in each axes will not affect calibration 1.05 oz with 15 feet of cable

For additional information consult: http://www.druck.com/usa/products/MiniatureSeries.pdf

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

43

Earthquake Induced Damage Mitigation from Soil Liquefaction

PCB Piezotronics 353B18 Miniature High-Frequency Quartz ICP Accelerometer

Mass: Dimensions: Voltage Sensitivity: Measurement Range: Frequency Range: Mounted Resonance Frequency: Broadband Resoultion: Operating Temperature Range: Sensing Element: Electrical Connector: Mounting Thread:

1.8 grams 7.1 x 18.8 mm 10 mV/g ± 5% ± 500g peak 1 to 10000 Hz ± 5% > 70 kHz 0.005 g rms -65 to 250oF Quartz Shear 10-32 Coaxial/Top 5-40 Male

For additional information consult: http://www.pcb.com/products/svs/svs353b18.html

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

44

Earthquake Induced Damage Mitigation from Soil Liquefaction

Trans-Tek Series 240 General Purpose CV LVDT

Working Range: Maximum Working Range: Input: Nominal Output: Input Current: Non-Linearity: Internal Carrier Frequency: % Ripple: Output Impedance: Frequency Response: Temperature Range: Resolution:

± 25.4 mm ± 38.1 mm 6 to 30 VDC 4.6 to 24.8 VDC 8.3 – 52 mA ± 0.5% over working range, ± 1% over usable range 3200 Hz 0.8 5600 Ohms DC to 100 Hz -54 to 121oC Infinite

For additional information consult: http://www.transtekinc.com/Catalog_PDFs-01/LVDTs/Ser240_01F.pdf

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

45

Earthquake Induced Damage Mitigation from Soil Liquefaction

Baumer OADM 2014460/S14C Laser Distance Sensor

Dimensions: Measuring Range: Resolution: Linearity Error: Response Time: Sensing Element: Output: Switching Current: Indicators: Voltage Supply: Maximum Supple Current: Light Source: Laser Class: Wavelength: Operating Temperature Range: Laser Beam Diameter: Connectors:

20.4 mm x 50 mm x 65 mm 30 to 130 mm < 0.06 mm ± 0.2 mm < 10 ms Photoelectric Array Analog / 4-20 mA / 0-10 VDC < 100 mA LED Green (Power On) & LED Red (Soiled Lens) 12 to 28 VDC < 120 mA Pulsed Red Laser Diode 2 675 nm 0 to 5oC 2 … 1 mm ES 34C

For additional information consult: http://www.baumersensorsolutions.com/product.html?id=fee12f83fd3a8a221b206c359a0 8c629&lang=en&product=34336&category=33&sub=222

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

46

Earthquake Induced Damage Mitigation from Soil Liquefaction

GS Sensors Amplified Triaxial Accelerometer GSA3206

Dimensions: Mass: Excitation: Offset at Zero: Output Impedence: Linearity: Transverse Sensitivity: Operating Temperature Range: Frequency Response:

30 mm x 30 mm x 25 mm 30 grams 10 to 36 VDC 2.5 VDC 10 ohms nominal ± 2% < 3% -40 to 80oC DC - 500 Hz in z-axis, DC - 100 Hz in x-axis, DC 1000 Hz in y-axis

For additional information consult: http://www.gssensors.com/catalogue/index_prod.php3

R-04-068-145-R-04-030-145-CT7v1.docv1.doc December 2004

47