In-Situ testing of soils and pavement systems using spectralanalysis of surface waves(SASW)

Abstract

The determination of shear wave velocity profile with depth for any ground or pavement site is essential for: (i) performing earthquake resistant analysis, (ii) carrying out soil liquefaction studies, and (iii) assessing the existing condition of roads and runways in order to evaluate their structural capacity. Spectral Analysis of Surface Waves (SASW) and Multichannel Analysis of Surface Waves (MASW) are nondestructive testing techniques frequently used to determine the variation of shear wave velocity with depth for layered media. These techniques are based on the measurement of the dispersive characteristics of Rayleigh (surface) waves in layered elastic media. In the SASW technique, typically two receivers are used, whereas MASW commonly employs more than 24 receivers. MASW uses a phase difference method and a two dimensional wave field transformation technique, resulting in more reliable and accurate results and avoiding the phase unwrapping problem often encountered in SASW, particularly in pavement testing. However, SASW remains comparatively cost effective, requiring fewer receivers and simpler acquisition systems. The present thesis uses the SASW technique but employs 8 receivers instead of 2 to address the phase unwrapping issue and to generate input data rapidly for multiple receiver spacing values. A complete SASW site investigation consists of three phases: (i) field testing, (ii) generating the field dispersion curve, and (iii) determining the stiffness profile through inverse (back) analysis. A disturbance is applied on the ground or pavement surface to generate Rayleigh waves that propagate at velocities corresponding to different frequencies. These waves are captured using geophones or accelerometers connected to a data acquisition system. The recorded input signal is analysed using Fast Fourier Transform (FFT) to construct the dispersion curve, showing variation of wavelength with frequency or phase velocity with wavelength. The dispersion curve is then used in inverse analysis to obtain shear wave velocity and layer thicknesses of the pavement or ground. From literature, it is noted that for roads and runways, higher frequency Rayleigh waves are generated using a sledgehammer, and measurements are made using shorter receiver spacing (0.5–5.0 m). For deep ground profiling, heavy impulse sources (bulldozers, rollers, or heavy weights dropped from a height) are used to generate low frequency waves, with receivers placed at wider spacing. Controlling the input energy from a sledgehammer manually is difficult, while dropping a known mass from a fixed height provides better control. The main objective of this research is to examine the effect of height of fall of the dropping mass on the zone of influence for which shear wave velocity can be successfully obtained. SASW tests in this thesis were performed on: (i) two ground sites, (ii) seven asphaltic concrete pavement sites, and (iii) two cement concrete pavement sites. Ground sites were in Bagalkot district (Karnataka). Pavement sites (asphaltic and cement concrete) were in Bangalore and Bagalkot districts and in the IISc Bangalore campus. At ground sites, a 65 kg cylindrical mass (SPT hammer) was dropped from heights of 1–4 m, either directly on the ground surface or on a steel base plate. Measurements were made using both accelerometers and geophones. At pavement sites, a 6.5 kg spherical mass was dropped from heights of 0.5–3.0 m, and signals were recorded using accelerometers. Various combinations of source distance (S) and receiver spacing (X) were used. Key findings include:

Increasing the height of fall increases the ratio of peak power amplitude to threshold power amplitude for the same S and X. For a given height of fall, this amplitude ratio is greater when S and X are smaller. Using a base plate increases the amplitude ratio at ground sites. Phase unwrapping can be resolved if dispersion data are obtained at different X values for the same frequency range so that _unwrap/ becomes nearly constant. Increasing the height of fall increases the maximum wavelength ( _max) up to which shear wave velocity can be reliably predicted. For all combinations of S, X and H, a unique combined dispersion curve is obtained if: (i) coherence > 0.90, and (ii) peak to threshold amplitude ratio > 1.1. For the same H, stiffer strata produce larger _max. Base plates enhance _max at ground sites. Cement concrete pavements produce larger _max than asphaltic pavements for the same H. For pavement testing, H = 0.5 m is adequate to assess top pavement layers and soil subgrade. Cement concrete pavements require placing the first receiver farther from the source compared to asphaltic pavements. Layer thickness and shear wave velocity values from inverse analysis are reasonable and consistent with literature. Without inverse analysis, total pavement thickness can be estimated as */3 to */2, where * is the minimum wavelength beyond which phase velocity starts increasing.

The thesis provides practical guidelines for determining appropriate energy input (drop height) for SASW testing on ground and pavement sites.