Rojas Vidovic, Juan Carlos (2008) LOADING RATE EFFECT ON THE MECHANICAL BEHAVIOUR OF A PYROCLASTIC SOIL. [Tesi di dottorato] (Unpublished)

Preview

PDF (Thesis) RojasVidovic_JuanCarlos.pdf Download (6MB) | Preview

Collection description

This thesis represents a laboratory-based experimental study into the rate of loading effect on the mechanical behaviour of an unsaturated pyroclastic soil. The study includes equipment design and laboratory testing which are addressed to analyze the rate of loading influence by means of equip-ment modification and suction controlled triaxial tests. The studied material, characterized by its very low density, is typical of the area around Naples where natural slopes are covered by loose silty sands of pyroclastic erupted by Vesuvius and by other volcanoes. The analyzed soil is a volcanic ash constituted by non plastic silty sand with pres-ence of pumices. The equipment design component of this research consists of the development of specialized de-vices: triaxial apparatusses and suctions probes.Two similar triaxial apparatuses capable of testing unsaturated samples under suction-controlled condition (USPv2) were developed. The main new feature of the USPv2 is the inclusion of a double drainage system to shorten the testing time. As two are the devices constructed, preliminary evaluation tests were performed on similar samples under similar stress conditions to assess the repeatability of tests independently of the apparatus used. On the other hand, two different types of high capacity tensiometers were developed. One of them is an integral strain-gauged tensiometer and the other one a commercial-transducer based ten-siometer. Both of them include solutions allowing the easy substitution of their parts. For each ten-siometer, a saturation process and a saturation system were designed; the main condition for these tasks was the application of relative low pre-pressurization pressure (800 kPa). To evaluate the per-formance of the tensiometers free evaporation tests, prolonged high suction measurements and pres-sure reversals tests were carried out. Based on the tests results, the commercial-transducer based probe was selected to be used with the USPv2 triaxial apparatus in order to monitor pore-water pressure during suction-controlled isotropic compression tests. The laboratory-testing program of this study includes three series of tests. Isotropic compression (first serie) and deviator stage (second serie) at different stress-rate and strain-rate, respectively, were performed using the triaxial apparatus USPv2. During isotropic compression stages the load-ing rates of 2, 8, 32, 128 kPa/h have been applied under several constant suction values (i.e. 0, 15, 45 and 300 kPa). The drained deviator stages were carried out at the same suctions levels under strain rates of 0.25 and 2.50 %/h. In order to study the effect of loading rate on pore-water pressure variations, a third serie of test was conducted. Suction-controlled isotropic compression tests with pore-water pressure measurement were carried out by means the USPv2 triaxial apparatus used in combination with the commercial-transducer based tensiometer. The capacity to apply both fluid pressures to the top and the bottom of the sample significantly shortens the equalization stage. During isotropic compression stage, as the loading rate is increased the rigidity of the soil also increase, however, a loading rate value was identified from which no dif-ference was observed. The study also suggests that the excess pore-water pressure generated during isotropic loading process, that may reduce the suction value, is less significant than the creep phe-nomena for the pyroclastic soil studied. For the deformation velocities studied on deviator stage, non effect was observed on the soil behaviour. Finally, the control of pore-air pressure and pore-water pressure by both sample sides, accompanied by a short testing program addressed to deter-mine rational testing rates, reduce a lot the testing time.

Downloads

Actions (login required)

View Item