Keywords: expansive soil, swell strain, instrumentation, telemetry, sensor, calibration, pressure response.

A laboratory research study conducted at The University of Texas at Arlington proposed the use of four novel stabilizers to stabilize soft expansive soils of Southeast Arlington [1]. A field monitoring study is needed to validate the effectiveness of these stabilizers in the field conditions. Such field studies will provide better understanding of stabilizers when they are exposed to variations in temperature, humidity and various external disturbances. In order to compare the stabilizing effects and hence forth to develop specifications for their usage, an instrumentation system was designed and procured.

A large-scale laboratory study, with semi-infinite boundary conditioned soil samples, was designed and conducted to measure the swelling and loading potentials of the soil in terms of pressures and strains. The main objective of this experiment was to compare the results obtained from the sensors used in the experiment with results interpolated from standard soil test results. This comparison will provide an explanation on the suitability of the present instrumentation setup (including data acquisition and sensors) to provide data that realistically represent the swell and strength characteristics of soil.

Selection of sensors in this research was based on laboratory study results, previous documentation in the literature, and financial considerations. Measured swell strains in soils obtained from laboratory experimental evaluations were close to 12%. However, these strain values are expected to be low in real pavement site conditions. Also, the pressures experienced by the site soils depend on location of soil from ground surface, traffic loads and their magnitudes.

The laboratory small scale experiment was conducted in a container large enough to accommodate the pressure cells, which were 0.45 m long and 22.6 cm in diameter. The container was first surrounded inside with a geotextile and fastened with screws to ensure proper wrapping. Geotextiles were used to facilitate moisture movements during soaking to ensure wetting of the soil. The first four inches of the container from the bottom were filled and compacted with a coarse sand layer. This sand layer was then saturated with water. The sand layer served two purposes, one was to anchor the strain gages and the other was to facilitate drainage boundary condition to allow water to reach clay layer compacted above the sand layer A total of four strain gages, with three to measure vertical movements and one to measure horizontal movements, were used. Two pressure cells, one each to measure vertical and horizontal pressures in the soils were used. One dummy gage (a 350-ohm resistor sealed in a watertight container) was used which provides readings that reflect any undue temperature variations during testing.

The setup was subjected to moisture hydration, static loading and a simulated dynamic loading. The data was recorded for five consecutive days at a sampling rate of 0.08 Hz. Five-point running average method was used for filtering the noise from the data. Data was acquired for all loading conditions and was used to calculate the mean and standard deviation of the measured maximum pressure values.

The strain gages measured the tensile strains on the soils subjected to swelling. The magnitudes of the measured values are quite low compared to the laboratory swell test values. From the swell strain versus time plots, it can be mentioned that the soil in the large-scale setup did not reach the full swell potential when compared to the results obtained from the laboratory tests, which reached full saturation due to small size of the samples. The horizontal stress, measured was higher than the theoretical determined value and this could be attributed to the rebounding of the stress from the walls of the container.

Both static and dynamic loading conditions provided pressure measurements that matched with theoretical predictions. Static load induced pressures are close to those predicted using simple linear elasticity based pressure measurements. Dynamic pressure increase due to simulated cylindrical wheel load is repeatable and matches well with the linear elasticity model predictions. Overall, these results suggest that the present instrumentation will be able to capture soil swell response under moisture hydrated and loading conditions. In conclusion, a few recommendations for sensor installation in field conditions are provided.

1

A.J. Puppala., E. Wattanasanticharoen, and K. Punthutaecha, "Experimental Evaluations of Stabilization Methods for Sulphate-Rich Expansive Soils", ISSMFE, Ground Improvement Journal, Thomas Telford Publication, Vol 7, No 1, pp 25-35, 2003. doi:10.1680/grim.7.1.25.37796

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £135 +P&P)