Sung Gyo Chung

A comprehensive evaluation of methodologies for reconstructing instantaneous consolidation settlement curves based on field-monitored data from clayey ground improved using prefabricated vertical drains is presented. Analyses of case histories from two distinct clay sites reveal that settlement records obtained following fill placement are most appropriate for accurate reconstruction. Among the updated observational methods assessed, the curve fitting approach consistently yielded more reliable predictions of consolidation behaviour compared with the linear rule and modified Asaoka techniques. Moreover, the selection of arbitrary initial settlements and non-fixed time points during loading had minimal impact on the estimation of ultimate settlement and the rate of settlement. When the recommended practices are implemented, the effects of staged loading schemes and variable geotechnical properties become negligible. These findings support the standardisation of reconstruction methodologies for instantaneous consolidation settlement curves, enhancing the consistency and reliability of back-analysed consolidation parameters.

This paper presents a closed form of the simplified three-dimensional (3D) (vertical and radial) consolidation solution, which has negligible errors for practical cases. Based on the simplified solution, two observational methods (namely, 3D modified curve fit and linear rule methods) that could predict the actual 3D consolidation settlement behaviors by reflecting the effect of sampling range were newly developed. The appropriateness of the proposed methods was verified by using 3D and two-dimensional (2D) theoretical data. Then, the proposed methods were applied to three case records. Back-analyzed results (i.e., the end of consolidation settlement S₁₀₀ and consolidation coefficient c_h(n)) were slightly affected by the differently approached methods and the inputted value of vertical consolidation coefficient (c_v). S₁₀₀ and c_h(n) values obtained from the 3D modified methods were close to those secured from the 2D modified methods due to the insignificant effect of vertical drainage for the cases. Owing to incomplete measurement (93%–98% of S₁₀₀, i.e., no consideration of sampling range effect), an existing 3D method and the original 2D Asaoka and exponential methods predicted a lower S₁₀₀ and higher c_h(n) than those predicted by the 3D or 2D modified curve fit methods. Therefore, the proposed 3D methods can be recommended for practical applications.

This study presents a novel method for determining the coefficient of vertical consolidation (cv) by combining the complete range of theoretical and experimental consolidation behaviors. The Taylor model, which nearly precisely simulated the theoretical solution of Terzaghi, was adopted for this purpose. The primary consolidation settlement (δ100) from the 24-h test data was determined based on the variation in the theoretical settlement rate. The estimated ultimate settlement and cv values were validated via inverse analysis. Thereafter, the proposed method and eight other existing methods were applied to the results of the oedometer tests on high-quality samples, which were obtained at a deltaic deposit. The proposed methods produced the approximately familiar variations in cv versus the stress increments along the entire depths. The δ100 values obtained using the existing methods were 70–140 % of those using the proposed method, whereas most of the cv values were inconsistently underestimated (i.e., 30–110 %). The irregular estimation from the existing methods is attributed to the use of specific points or segments of the experimental relationships, which are different from the theoretical one. Moreover, the applicability of the theoretical determination of cv was discussed.

This study presents a novel approach to determine the coefficient of consolidation (ch), which represents the complete behavior of primary consolidation while instantaneously loading consolidation tests with central drainage (CD) and peripheral drainage (PD). The entire consolidation range (i.e., ultimate settlement δult) was graphically determined using settlement rate variations, and then ch was obtained by best fitting such data range with 2-D theoretical solutions. The proposed method and 11 existing methods were applied to 15 existing test data. The applicability of the estimated δult and ch values was verified via back analysis. Results indicate that the errors between the back-analyzed and measured settlements are within a maximum of 7 % in the case of the proposed method but within a maximum of over 50 % in the case of existing methods. Thus, the ch values obtained using existing methods on the CD and PD test results varied within the range of 60–120 % and 50–70 %, respectively, of those of the proposed method. Such substantial underestimation or overestimation was mainly attributed to the variations between the theoretical and experimental time–settlement relationships, incorrectly determined δult, and limitations in graphical procedures.

The applicability of the ball penetration test (BPT) for identifying soil layering and evaluating several key geotechnical properties was studied. BPT and piezocone test data at four sites were evaluated and correlated with field and laboratory test results. Both ball and cone resistances were effective in evaluating soil layering. However, the larger transition zone induced by the ball deteriorated the linear variation of the effective overburden stress with depth. The ball factor effectively differentiated the depositional units of Korean clays better than the Japanese clays. The preconsolidation stress and the shear wave velocity Vs (or shear modulus G0) correlated well with the ball and cone resistances, which varied within a deviation of ±20% depending on the properties of the local soils. In the analyses, the total resistance gave a better correlation than the net resistance. Multiple correlations with two or three variables were found to predict the measured Vs (or G0) values more reliably than one-variable correlations. Accordingly, the BPT may be extended for geotechnical characterisation.

This study presents an enhanced hydraulic fixed-piston (HNS-N) sampler that reflects a local practice. This sampler was employed at a deltaic site along with two other types of samplers: a similar hydraulic (HNS-K) type with no vacuum breaker and an oil-operated (ONS) sampler that is equipped with a vacuum breaker and is fixed at the center of a casing. The generation of a slowly but erratically increasing pressure in the hydraulic samplers often does not ensure constant speed and full penetration; thus, the pressure in these samplers was increased rapidly after the drill rods were filled with water. By contrast, the pressure in the ONS sampler was increased instantaneously. The hydraulic samplers exhibited slightly discrete penetration behavior, whereas the ONS sampler induced constant penetration. The sample qualities of the samplers are ordered based on an evaluation via two nondestructive methods and a consolidation test, as follows: ONS > HNS–N > HNS–K. The sample quality was affected signifi...