• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.3C
• /
• pp.201-207
• /
• 2006

Soil tests are generally conducted using a membrane to measure a pore water pressure. However, it has also been recognized that the membrane penetrates into the specimen by the change of the confining pressure, and it results in the erroneous measurement in the pore water pressure and the volumetric strain. This study examined the effectiveness of the correction equation of the membrane penetration on the basis of the experimental data acquired during the isotropic unloading and the cyclic shear process using the hollow cylindrical shear test equipment. The results showed that the membrane penetration by the correction equation could be overestimated when the mean effective stress was lower than 20kPa in this study. The limitations originated from the sudden increase near the zero effective stress, and in order to prevent the overestimation in low effective stress condition, the use of the constant a was proposed in this study. Furthermore, the correction equation for the membrane penetration had to be applied carefully when the initial relative density was high and the density changes were occurred by the relocation of the soil particle by the liquefaction.

• Journal of the Korean GEO-environmental Society
• /
• v.11 no.12
• /
• pp.57-62
• /
• 2010

In this study, the effect of different silt contents on the shear characteristics of silty sands was evaluated. Two series of triaxial compression tests were performed on the cylindrical specimens of compacted Nakdong river sand with 10% and 30% silt contents under unconsolidated undrained condition. All identical specimens were prepared to compact with same initial water content for five layers and saturated using control panel and then sheared under initial effective confining pressure, 100 to 400kPa. All specimens exhibited a strain softening tendency after failure in stress-strain curves and deviator stresses of specimens with 10% silt content were greater than those of specimens with 30% silt content. Pore water pressures of specimens with 10% silt content were observed negative(i.e. swelling) due to increasing void ratio after failure but those of specimens with 30% silt content were shown only positive. The behavior of compacted cylindrical specimens with low silt content was more dilative than that of high silt content. Peak deviator stresses decreased as increasing silt content and peak pore water pressures increased as increasing silt content.

• Journal of the Korean Geotechnical Society
• /
• v.21 no.4
• /
• pp.65-70
• /
• 2005

Many case histories of re-liquefaction phenomena seem to support the idea that sand deposits, if they once have been liquefied, could be reliquefied again by a subsequent earthquake even though the earthquake is smaller than the previous one. The magnitude of the strains induced in the initial liquefaction has a significant influence on the resistance of the sample to re-liquefaction. The deposits undergoing liquefaction experience large shear strain during liquefaction. And this previous strain changes the microstructure into highly anisotropic structure such as columnlike structure and connected voids. This type of anisotropy is so unstable that it can reduce re-liquefaction resistance. It is blown that the extent of anisotropic structural change depends on the gradation characteristics of ground. The purpose of this study is to estimate the correlation between the gradation characteristics of the sand and the ratio of re-liquefaction resistance to liquefaction resistance. In this study, 1-g shaking table tests were carried out on five different kinds of sands. During the tests the values of excess pore pressure at various depths and surface settlements were measured. Re-liquefaction resistances were not affected by the initial void ratio and the effective confining pressures, and the deposits of all test sands which had once been liquefied were reliquefied in the cyclic loading number below 1 to 1.5. The ratio of re-liquefaction resistance to liquefaction resistance linearly decreased as $D_{10}/C_u$ increased, and was constant as about 0.2 above the value of $D_{10}/C_u$, 0.15 mm.