Moon Joo Lee

Preparation of a Large Size Granular Specimen Using a Rainer System with a Porous Plate

In this study, a rainer system capable of forming a large homogeneous granular specimen is introduced. A series of laboratory tests is carried out in order to study the performance of the proposed system. The features of the rainer system used in this study are the adoption of a porous plate and the air-pluviation without changing the deposition intensity. Without a porous plate, the rainer induces an insignificant density increase with increasing drop height, providing a dense to very dense specimen. However, the rainer with a porous plate produces a medium dense to dense specimen, with a drastic density increase at drop heights of 10–40 cm and a progressively reduced rate of density increase at larger drop heights. It is shown that the density obtained by using a porous plate rainer with a 70 cm drop height is similar to that achieved by using a conventional rainer with a 10 cm drop height. It is concluded that the use of a conventional rainer that adopts a porous plate significantly widens the range of density achievable even without an alteration in the deposition intensity. It is also concluded that using the rainer with a porous plate significantly improves the vertical and horizontal homogeneities of the specimen. The relative density evaluated from the measured cone resistance appears to provide a reasonable estimation at depths of 40–80 cm. The profile of shear wave velocity measured using bender elements seems to accurately reflect the vertical non-uniformity of the specimen.

Time Dependent Well Resistance Factor of PVD

As the conventional equations of radial consolidation with PVD assume a constant well resistance with time, an appropriate in-situ consolidation curve cannot be predicted. In this study, small cylinder cell tests are performed to evaluate the discharge capacity of PVD. In addition, a radial consolidation test is performed on a large block sample to observe the change in the drainage capacity of PVD with time. The consolidation curves predicted using small cylinder test results significantly overestimate the degree of consolidation. It is shown from test results of a block sample that the drainage curves normalized with respect to the initial drainage of each PVD are similar, regardless of the PVD type. It is also found that the well resistance factor, L(t), back-calculated from the consolidation curve of a large block sample is not a constant value, but is linearly proportional to the dimensionless time factor, Th. Finally, the usage of L(t) is verified by analyzing the field settlement data.

Evaluation of State Parameter of Crushable Jeju Sand Using Cone Resistance

This study was performed to evaluate the critical state parameter of crushable Jeju sand and to suggest the relationship between the state parameter and cone resistance of Jeju sand. It is observed from a drained triaxial test that Jeju sand mainly shows contractive behavior due to high void ratio and large compressibility. Although strain localization is not evident, the particle crushing results in a lower void ratio than the critical state void ratio of some Jeju specimens. Critical state parameters of Jeju sand are similar to those of calcareous sands, but significantly larger than those of common sands. The relationship between the normalized cone resistance and state parameter of Jeju sand significantly differs from that of other materials, and this relation appears to be affected by the stress level. This is because the variation of stress level affects the state parameter of Jeju sand more significantly than that of common sand due to the high compressibility of Jeju sand.

Evaluation of K0 of Granular Soils using DMT Indices

A series of CPTs and DMTs are performed in the calibration chamber in order to develop methods that estimate the coefficient of at-rest earth pressure (K0), which is an indirect measure evaluating the stress history of granular soil. Among DMT indices and the qc value, KD is observed to be the most sensitive to the stress history while qc and ED show the similar trend of change as the vertical stress and stress history. The relation of Baldi et al. (1986) appears to be applicable for only NC sands as it does not include the term representing the effect of horizontal effective stress. The relations suggested in this study, KD/K0 = χ(ED/ )δ and KD/K0 = a0(ED/pa)a1( /pa)a2, are observed to provide comparable accuracy in K0 values with those of Jamiolkowski and Robertson (1988) and Mayne (1995, 2001), regardless of the stress history. Suggested relations use only DMT indices while Jamiolkowski and Robertson (1988) require both CPT and DMT results, and Mayne (1995, 2001) needs both the CPT result and the OCR value. Although the adoption of OCR in Mayne (1995, 2001) offers a considerable improvement in the accuracy of the K0 value, it is not considered practical since the inclusion of both soil property (OCR) and index parameters in an empirical correlation provides excessive redundancy.