International Symposium on Strong Vrancea Earthquakes and Risk Mitigation Oct. 4-6, 2007, Bucharest, Romania

LABORATORY INVESTIGATION FOR ESTIMATION OF SEISMIC RESPONSE OF

THE GROUND

Cristian Arion1, Cristian Neagu2

ABSTRACT

Laboratory measurements of soil properties can be used to supplement or confirm the results of field measurements. They are necessary to establish values of damping and modulus at strains larger than those that can be obtained in the field. The Dynamic Deformation Characteristics of the soil are used in order to calculate seismic response of ground, earth structures and structure-ground response. The cyclic triaxial equipment installed at CNRRS is used, when the dynamic proprieties of the soil must be obtained. Comparison of the obtained results with the well knows international one is presented.

INTRODUCTION

In designing of engineering structures, when a seismic consideration is required, the dynamic proprieties of the soil must be obtained. A large number of laboratory tests for obtaining dynamic properties have been developed, and research continues in this area. These tests can generally be classified into two groups: those that apply dynamic loads and those that apply loads that are cyclic but slow enough that inertial effects do not occur. Laboratory measurements of soil properties can be used to establish values of damping and modulus at strains larger than those which can be obtained in the field or to measure the properties of materials which do not now exist in the field, such as soils to be compacted. The most widely used of the cyclic loading laboratory tests is the cyclic triaxial test. In this test a cyclic load is applied to a column of soil over a number of cycles slowly enough that inertial effects do not occur. Cyclic load is usually applied as cyclic axial load by mechanical, hydraulic or pneumatic actuator. The Dynamic Deformation Characteristics of the soil are used in order to calculate seismic response of ground, earth structures and structure-ground response. They are also used to express phenomenons that make soil to fail under seismic loading. Table 1 shows the appropriate strain ranges for respective soil test methods.

TRIAXIAL EQUIPMENT AT CNRRS

In July 2003 at National Center for Seismic Risk Reduction Bucharest Romania (CNRRS) was installed the triaxial equipment. Seiken Inc. Japan made the equipment and the commissioning.

TESTS USING the DTC -367, SEIKEN, INC

Static test Cohesive soil and sand

Dynamic Cohesive soil and sand

Cyclic Sand sample

Deformation Cohesive soil and sand

1 Dr. eng., Technical University of Civil Engineering, Bucharest & National Center for Seismic Risk Reduction,

Bucharest, Email: arion@utcb.ro 2 Eng., National Center for Seismic Risk Reduction, Bucharest, Email: cristi@cnrrs.ro

International Symposium on Strong Vrancea Earthquakes and Risk Mitigation

223

Table. 1) Dynamic Mechanical Property tests

Deformation Characteristics Strength METHOD

Compression Modulus,

E

Shear Modulus,

G

Poisson’s Ratio,

ν

Damping Factor, h

Dynamic Shear

Strength

Range of Strain

Cyclic triaxial compression 5*10-4

- 10-1

Cyclic simple shear test 10-4

- 10-1

Ultra sonic Pulse test Very small

Resonant column test 10-6

- 10-4

Torsion Shear test 10-4

- 10-2

Laboratory Test

Ring Shear test 10-4

- 10-2

Seismic Survey Refraction, Reflection, Crosshole, Downhole, Surface Wave Methods

Resonant footing

Cyclic Pressuremeter

Field survey

Standard Penetration Test

: The property is directly determined : The property is indirectly determined : The property is estimated based on many experimental data.

The equipment fulfills all the requirements of The Japanese Geotechnical Society, 2000. The CNRRS triaxial equipment can solve the following types of problems:

- Static problems with strain level at 10-3 or greater (the main concern regarding the static

problems is used to evaluate the degree of safety of foundations or soil structure against the failure);

- Dynamic problems with soils subjected to a strain levels as small as 10-6 (used to

evaluate the soil strength in comparison with stresses induced by external loading and the settlement of ground or structures associated with the deformation of soils. The difference between static and dynamic loading conditions is in the term of time of loading and is expressed in terms of speed of loading or rate of straining (speed effect or rate effect). If load application last more than 0.1 sec then we have “static problems” and if load application have a shorter time of application we have “dynamic problems”. The cyclic loading test is commonly used to determine the strength of soils under seismic loading conditions. A specimen is consolidated first and subjected to the initial stress σs. A sequence of prescribed shear stress cycles is then applied to the test specimen with relatively small amplitude. When the points reached at the end of each sequences are connected in the plot of the stress-strain diagram.

Figure 1. Construction of a shear stress versus residual curve from multistage loading test results

Sh

ear

stre

ss A

σd

A’

B

B’

C

C’

σs P

Shear strain

C. Arion & C. Neagu 224

The automatic stress strain path control and monitoring for triaxial test is as follows, Figure 2:

Figure 2. Example of monitoring of axial pressure and axial cyclic deformation at CNRRS

1. For lateral loading is use a pneumatic pressuring system. An air pump and a servo EP transducer make possible the accurate control of the air-pressure.

2. For axial loading it is important to have a high resolution as well as a large stroke of axial displacement. We use a servo-pneumatic loading system.

3. For very small vertical displacements a pair of gap sensors are used

LOW AMPLITUDE MODULI OBTAINED FROM LABORATORY TEST

Despite the fact that soil deformation under seismic loading is relatively small, its modulus is dependent on dynamic stress or strain level. Soil modules such as: Young's modulus and shear modulus decrease as the level of stress or strain increases. The deformation characteristics of soil are highly nonlinear and this is manifested in the shear modulus and damping ratio. The G- γ, E- γ and h- γ relationships obtained in DDC (dynamic deformation characteristics tests) are generally recognized to express the shear deformation characteristics of soil. The shear modulus is called maximum shear modulus, initial shear modulus, or low amplitude shear modulus and is noted by Gmax or G0. To obtain a 10

-4 – 10

-5

axial strain we must use a very small deviator stress. Its value may vary between fairly wide limits, and its influence on the results is very large. For very small strains the shear modulus may be a factor 10 or even 100 larger than it is for large strains. Examples are shown in Fig. 4.

RESTULTS FROM THE TESTS

During last two years we conduct series of dynamic triaxial tests especially on the Bucharest clay soils. In the Fig. 4 are represented relations of shear modulus ratio G/GO versus shear strain and the strain dependent damping for the laboratory samples. Also we represented the strain-dependent modulus and damping curves quoted in the literature such as Vutecic and Dobry (1991), Seed, Sun.

Figure 3. Sampling, preparation and testing of the soil samples

International Symposium on Strong Vrancea Earthquakes and Risk Mitigation

225

0

0.2

0.4

0.6

0.8

1

0.0001 0.001 0.01 0.1 1 10

Single amplitude shear strain, γ (%)

G/G

ma

x

Vutecic G/Gmax IP=0

Vutecic G/Gmax IP=100

Vutecic G/Gmax IP=200

Gravel Seed G/Gmax

Sun G/Gmax PI=80+

Sand Seed lower bound G/Gmax

Sand Seed average G/Gmax

Clay Layer 1

Clay Layer 2

Clay Layer 3

Clay Layer 4

Gravel Seed G/Gmax

Vutecic IP=200 G/Gmax

0

5

10

15

20

25

30

35

0.0001 0.001 0.01 0.1 1 10

Single amplitude shear strain, γ (%)

Hy

ste

reti

c d

am

pin

g r

ati

o h

(%)

Vutecic h (%) IP=0

Vutecic h (%) IP=100

Vutecic h (%) IP=200

Sun h(%) lower bound

Sun h(%) average

Sun h(%) upper bound

Clay Layer 1

Clay Layer 2

Clay Layer 3

Clay Layer 4

Vutecic IP=200 h(%)

Sun upper bound h(%)

Figure 4. Test results for Bucharest clay layers and comparison with analytical model curves

ACKNOWLEDGMENTS

The authors would like to acknowledge the funding provided by Japan International Cooperation Agency (JICA).

REFERENCES

Arion, C., 2004: Report on soil tests and investigations (laboratory and field experiments), JICA/CNRRS/BRI.

Ishihara, K. 1982: Evaluation of soil properties for use in earthquake response analysis Proc., Int. Symp. on Numerical Models in Geomechanics, Zurich: 237-259. SEIKEN, Inc. (2003). Technical documentation of triaxial testing apparatus

Tatsuoka F., Teachavorasinskun S., Dong J., Kohata Y. and Sato T. 1994, Importance of measuring local strains in cyclic triaxial tests on granular materials, Proc. of ASTM Symposium Dynamic Geotechnical Testing, ASTM, STP 1213, 288-302. The Japanese Geotechnical Society (2000). Standards of Japanese Geotechnical Society for Laboratory Shear Test (English Version).

Vutecic M. and Dobry R. Pore pressure build-up and liquefaction at level sandy sites during earthquakes, research report No. CE-86-3 Rensselaer Polytechnic Institute, Troy, NY 12181.