D.T. Bergado

Soil-geogrid reinforcement interaction by pullout and direct shear tests

Pullout and direct shear tests have been conducted to investigate the soil-reinforcement interaction behavior in pullout and direct shear mechanisms. An apparatus—discussed in this paper—has been developed that is capable of performing both pullout and direct shear tests. Displacement measurements to monitor dilatancy on the backfill soil during pullout tests are incorporated in the apparatus. Test results for the geogrid type of reinforcement embedded in dense granular soil are discussed. The devised testing program provides valuable information on the appropriate interaction models and parameters that will be employed in the design and analysis of reinforced soil structures. The dilatancy measured in the laboratory may also provide useful information on the increase of interface shear resistance when dilatancy is restrained under field conditions.

INTERACTION BETWEEN COHESIVE-FRICTIONAL SOIL AND VARIOUS GRID REINFORCEMENTS

Abstract A total of 52 large-scale laboratory pullout and 24 large-scale direct-shear tests were conducted to investigate the interaction behavior between the different reinforcements and cohesive-frictional soil. The reinforcements used were steel grids, bamboo grids, and polymer geogrids. The backfill material used was locally available weathered Bangkok clay. The test results show that the inextensible reinforcements, such as steel grids, move approximately as a rigid body during the pullout test, and the maximum pullout resistance was reached within a relatively small pullout displacement. For extensible reinforcements, such as Tensar geogrids, the degree of resistance mobilization along the reinforcement varies, and the pullout-resistance achieved in the tests was controlled by the stiffness of the reinforcement. For steel grids, the friction resistance from the longitudinal member contributed only to about 10% of the total pullout resistance of the grids. The pullout of the bamboo and Tensar geogrids without transverse members yields 80–90% of the pullout resistance of the corresponding grids with transverse members, attributed to the nodes or ribs on longitudinal members. The bond coefficient as calculated for steel and bamboo grids demonstrated that the steel grids yielded a higher bond coefficient than that of the bamboo grids with the same grid size. However, for a polymer geogrid, the bond coefficient cannot be calculated from a pullout test because of the complicated pullout-resistance-mobilization mechanism along the reinforcement. The large-scale direct-shear-test results showed that, for the soil/grid-reinforcement interfaces, shear resistance can exceed the direct-shear resistance of the soil itself owing to the influence of the apertures on the grids. Finally, for compacted weathered clay, the strength parameters obtained from large-scale direct-shear tests were found to be substantially smaller than the results of triaxial UU tests. This may be because the failure plane in the large-scale direct-shear test was formed progressively, and the peak soil strength along the predetermined shear plane may not have been mobilized simultaneously.

IMPROVEMENT OF SOFT BANGKOK CLAY USING VERTICAL DRAINS

Abstract This paper presents case records demonstrating the use of vertical drains on soft Bangkok clay. The documented cases include four sand drains and two prefabricated vertical drains. The observed performances of the vertical drains are evaluated. The effects of soil flow parameters, including smear effects due to the installation of the drains, in particular the prefabricated vertical drains, are also evaluated in the light of the current practice of using vertical drains for ground improvement. It is concluded that the use of vertical drains is a viable ground improvement method for soft Bangkok clay.

PROPOSED CRITERIA FOR DISCHARGE CAPACITY OF PREFABRICATED VERTICAL DRAINS

Abstract Modified triaxial tests and ASTM-based discharge capacity tests were carried out to obtain specifications of discharge capacity. The test results indicate that, at straight condition, ASTM-based discharge capacity decreases with increasing lateral pressure, time and hydraulic gradient. With lateral pressure application, the filter jacket of a prefabricated vertical drain (PVD) is pressed into the channel system of the core, thereby reducing the flow area. Furthermore, the discharge capacity decreases at a high hydraulic gradient due to loss of flow energy as a result of turbulent flow. Moreover, discharge capacity has been investigated by simulating the different possible drain deformations in the field. The average percentage reductions of the discharge capacity from the straight condition to various PVD deformed conditions have been found as follows: 26, 32, 33, 43, 48 and 78%.for conditions of 10% bent, 20% bent, 90° twisted, 180° twisted, 20% bent with one clamp and 30% bent with two clamps, respectively. Subsequently, reduction factors for discharge capacity have been formulated. Moreover, a simple method is proposed to evaluate specifications of discharge capacity with some established relationship considering the length of PVD, .spacing, time required for consolidation and the magnitude of the horizontal coefficient of consolidation.

The interaction mechanism and behavior of hexagonal wire mesh reinforced embankment with silty sand backfill on soft clay

Abstract The pullout/direct shear mechanisms as well as the behavior of hexagonal wire mesh reinforced embankment with silty sand backfill had been investigated by numerical method. Finite element method under plane strain condition using SAGE CRISP software has been utilized in the numerical simulations. For the numerical simulation of full-scale reinforced wall, the reinforcement stiffness, backfill soil properties, soil/reinforcement interaction, properties of soft clay foundation, and consolidation period were significantly considered in the analysis. The equivalent interaction coefficients of the interface element were used in the simulations for the pullout and direct shear modes. The results have been favorably compared with the previous simulation using PLAXIS FEM software. The numerical technique has reasonably captured the actual behavior of the reinforced embankment on soft foundation. Most of the interaction mode obtained from the simulation is governed by the direct shear mechanism.

Prediction of vertical-band-drain performance by the finite-element method

Abstract This paper presents the predictions of the performance of vertical-band drains (i.e. Alidrains) on soft Bangkok clay under full-scale embankment loading in the field and during large-scale consolidation with a drain by using a reconstituted sample in the laboratory. A finite-element model of transient- and axisymmetric-flow problems considering equal strain was used. The numerical model also included the smear effects. The predicted data were compared with the laboratory- and field-test results. Laboratory consolidation tests using a Rowe cell and an oedometer apparatus were performed to obtain the compressibility parameters. By using a value of kh/ks of together with ds=2dm, the predicted results agreed with the field performance of the vertical drains. In the laboratory, by using a reconstituted clay specimen, better predictions were obtained with kh/ks=2 and ds=2ds. Thus, the prediction by means of an equal-strain finite-element model was found to be satisfactory for the performance of vertical-band drains on soft Bangkok clay provided that the disturbance due to its installation was taken into account.

Interaction of nonwoven needle-punched geotextiles under axisymmetric loading conditions

Abstract Geotextiles have been successfully used for reinforcement of unpaved roads on soft subgrade to improve the performance of a reinforced fill layer placed on soft ground. The tension–strain behavior of a nonwoven needle-punched geotextile under axisymmetric loading condition as well as the mechanism and effects of the different grades of geotextile on the increase in bearing capacity of reinforced unpaved roads over weak subgrade under traffic load were considered. The strain energy capacity concept is proposed to describe the tension–strain of geotextile under an axisymmetric loading condition. Modified CBR tests on soft and weathered clay overlain by compacted sand as well as on soft and weathered clay overlain by compacted sand reinforced with fix- or free-end nonwoven needle punched geotextile were carried out. Finite element method (FEM) using the PLAXIS software was utilized to back-analyze the results of the modified CBR tests. No significant difference between in-air and in-soil stiffness has been found for geotextile reinforcement of unpaved road. The calculated results indicate an additional load capacity due to the presence of the geotextile using an axisymmetric stiffness which demonstrated a significant contribution of membrane action by the different types of geotextile on the increase in bearing capacity of soil–geotextile system. The effects of the different types of geotextile obtained in this study can be used to preliminary select the appropriate grades of nonwoven needle-punched geotextile corresponding to the allowable rut depth in the design of the reinforced unpaved roads under traffic load.

FE analysis of grid reinforced embankment system on soft Bangkok clay

Abstract The behavior of a reinforced embankment on soft Bangkok clay has been analyzed by plane strain finite element method. The finite element analysis considers the selection of proper soil/reinforcement properties according to the relative displacement pattern of upper and lower interface elements. The large deformation phenomenon is simulated by updating the node coordinates, including those of the embankment elements above the current construction level, which ensures that the applied fill thickness simulates the actual field value. A full scale test reinforced embankment with a vertical face (wall) on Bangkok clay has been analyzed by the proposed finite element method, and the numerical results are compared with the field data. The response of a reinforced embankment on soft ground is principally controlled by the interaction between the reinforced soil mass and soft ground and the interaction between the grid reinforcement and the backfill soil. The tension in reinforcement and lateral displacement of the wall face varied during consolidation of foundation soil. The maximum tension force occurred in the reinforcement layer placed at the base of reinforced mass, due to bending of the reinforced mass resulting from differential settlements. It is considered necessary to account for the permeability variation of the soft ground foundation in the finite element analysis.

Innovations and performances of PVD and dual function geosynthetic applications

Abstract Full-scale test embankments, which were constructed on separate locations but within the site of the Second Bangkok International Airport and at the Campus of the Asian Institute of Technology, confirmed that prefabricated vertical drain (PVD) installation not only accelerated and facilitated uniform consolidation settlements but also aided in recharging the subsoil. Thus, PVD can mitigate the negative piezometric drawdown of the subsoil caused by the excessive pumping of groundwater down to the depth of PVD installation. The compressibility and flow parameters were back-calculated from the performance of the full-scale tests. Further back-analysis proved that every stage of preloading corresponded with certain rate of consolidation, and hence the variation of compressibility parameters of PVD-improved soft clay. Another full-scale test embankment employing vacuum-assisted preloading revealed that the rate of settlement has increased by 60% and the period of preloading was reduced by 4 months. Moreover, the use of electro-conductive PVDs could further shorten the required time of consolidation. In addition, dual function geosynthetic had been proposed to prevent excavation slope failures caused by the drawdown of water level along irrigation/drainage canals.

PERFORMANCE OF REINFORCED EMBANKMENT ON SOFT BANGKOK CLAY WITH HIGH- STRENGTH GEOTEXTILE REINFORCEMENT

Abstract In order to study the improvement of embankment stability on soft ground, two test embankments were constructed to failure. One test embankment was reinforced with high-strength, nonwoven geotextile as base reinforcement. The reinforcement consisted of one layer of high-strength geotextile placed directly on the natural ground surface at the bottom of the embankment fill. For comparison, another full-scale, unreinforced embankment using the same fill material as that of the reinforced one was also constructed to failure at the adjacent site. The height at failure of the reinforced embankment was 6 m while that of the unreinforced embankment was 4 m. This paper presents the instrumentation program, construction procedure, monitored data, and stability analysis of the test embankments. The study indicated that the high-strength, nonwoven geotextile as base reinforcement, can considerably increase the ultimate height of embankments on soft clay. It was also shown that the rapture occurred at large deformation of foundation subsoils.