EFFECTS OF DISTURBANCE, STRAIN RATE, AND PARTIAL DRAINAGE ON PRESSUREMETER TEST RESULTS IN CLAY

SINNADURAI PRAPAHARAN, Purdue University

Abstract

The pressuremeter is a promising in situ device for the determination of strength and modulus parameters. However, the past experience in clays indicate that the pressuremeter yields unreasonably high undrained strength values. It is believed that the effects of disturbance caused by the insertion of the device, high strain rate used in the pressuremeter test compared to the ones used in the laboratory test, and partial drainage that takes place during the test are the main reasons for the error in the pressuremeter test results. The effect of disturbance is analysed by considering a remolded annulus around the pressuremeter and by initially unloading the cavity. The pressuremeter expansion curve for a disturbed soil is predicted using an anisotropic constitutive model, and the strength parameters are derived using the conventional interpretation procedures. A parametric study is conducted by varying the amounts of disturbance. An attempt has been made to evaluate the use of pore size distribution analysis to determine the thickness of the disturbed annulus. A curve fitting equation is developed to describe the pressuremeter expansion curve. A methodology is developed to analyse the effect of strain rate on the stress-strain curve derived from the pressuremeter test. The effects of partial drainage on the pore pressure distribution around the pressuremeter probe and the pressuremeter expansion curve are studied by using a linear-elastic model. The results of this study show that disturbance is the most important factor affecting the pressuremeter test results. The effect of unloading of the borehole during the drilling overwhelms the effect of the presence of a remolded annulus when the initial unloading is large. The undrained strength can be over-estimated by as much as 100% and the modulus can be underestimated by 40%, depending on the amount of initial unloading. Highly anisotropic and strain softening soils are affected most by disturbance compared to relatively insensitive soils. The proposed curve fitting equation describe the experimental and theoretical pressuremeter curves very well. The higher strain rate used in the pressuremeter test and the variation of strain rate across the soil mass are shown to significantly affect the stress-strain curve derived from the pressuremeter expansion curve. Partial drainage significantly affects the pore pressure generated in the soil mass during the pressuremeter test; however, the pressuremeter expansion curve is not significantly affected.

Degree

Ph.D.

Subject Area

Civil engineering

COinS