Development of a pluviator device for preparation of sand beds for shake table test

Abstract

The dynamic behavior of a wrap-around wall has been carried out experimentally through a physical model test using Shake Table. The main objective of this research is to develop a pluviator device for the preparation of sand beds for shake table experiments. This research is divided into two parts. First one is the development of the pluviator and second part is its application in conducting Shake table testing. A portable travelling pluviator device has been developed by following the air pluviation techniques. The performance of the pluviator has been tested using six different gradation of two types of sand samples. Relative density of sand increases with the increment of height of fall of a pluviator. Deposition intensity time of sand decreases with the increase of height of fall of the pluviator. An inverse relationship between relative density and deposition intensity time has been found.

Portable travelling pluviator is used to prepare different relative density sand bed for seismic experiments on a 2m X 2m computer-controlled servo-hydraulic single degree of freedom shaking table facility in Bangladesh University of Engineering and Technology (BUET). Sinusoidal waves of different frequencies and three different earthquakes (Kobe, Loma and Kocaeli) are chosen in this thesis in order to apply on different densities wrap-faced wall to observe their seismic characteristics from the recorded data of accelerometers, LVDTs and strain gauges. A 408 mm (16 inch) height wrap-faced retaining wall model has been constructed in a Plexiglas container of 1.79m X 0.46m X 0.57m by considering the prototype to model scale factor, N=10 where, the applicable height of prototype wall is 4 m (13.39 ft). 48%, 64% and 80% relative densities of Sylhet sand and 26%, 45% and 57% relative densities of Local sand (Collected from Savar) have been used in this research. Tests are performed by applying three different surcharge pressures (0.7KPa, 1.12KPa and 1.72KPa). Sinusoidal tests are implemented for three base accelerations (0.1g, 0.15g and 0.2g) and for eight different frequencies (1Hz, 2Hz, 3Hz, 5Hz, 8Hz, 10Hz, 12Hz and 15Hz). In total five hundred ninety-four tests are performed on wrap-faced sand retaining wall in this research. From these tests, it has been observed that acceleration amplification is inversely proportional to both the surcharge load and the relative density and proportional to base acceleration during sinusoidal testing. For example, acceleration amplifications of 0.7 kPa and 1.12 kPa surcharge load test at normalized height 0.5 is 5.86% and 2% higher than the 1.72 kPa surcharge load test respectively for 48% relative density Sylhet sand model. Moreover, face displacement at different elevations has been decreased with an increase of the surcharge pressure and the relative density but, has been increased with the increase of the base acceleration. For example, normalized face displacements for 57% and 45% relative density sample are 77.3% and 64.6% lower than the 26% relative density sample respectively at normalized elevation 0.625 for Local sand sinusoidal test. Again, strain changes have been reduced at higher surcharge pressure and have been increased at lower surcharge pressure. It has been also observed that strains of 0.2g and 0.15g base acceleration are 9.75% and 5% higher than the strain of 0.1g base acceleration respectively at normalized elevation 0.5 for 48% relative density Sylhet sand sample, which indicates the change of strain has been increased with the rise of base acceleration.

From the earthquake tests on wrap-faced retaining wall, it has been seen that Acceleration amplification is inversely proportional with the increase of both the surcharge load and the relative density for three types of earthquakes (Kobe, Loma and Kocaeli). On the other hand, acceleration amplification has been arisen with the increment of base accelerations. For example, acceleration amplifications of 0.1g and 0.15g base acceleration are 6.5% and 2.7% lower than 0.2g base acceleration respectively at normalized elevation 0.5 for 48% relative density Sylhet sand sample under Kobe earthquake testing. Moreover, face displacement at different elevations have been decreased with an increase of surcharge pressure and relative density. For example, face displacements of 80% and 64% relative density sample are 12.9% and 8.2% lower than 48% relative density sample respectively at normalized elevation 0.625 for Loma earthquake experiment. Besides, face displacement has been risen with the increase of the base acceleration at the same normalized elevation. Again, strain changes have been reduced at higher surcharge pressure. It has been seen that, strains of 80% and 64% relative density sample are 13.3% and 7.2% lower than 48% relative density sample respectively at normalized elevation 0.5 for Kobe earthquake test. So, the changes of strain have been decreased at higher relative density. Dynamic characteristics observed from this research is helpful for the implementation of wrap-faced sand retaining wall at a site.