Laboratory modelling of grouting in poorly-graded sand

Abstract

This paper presents a newly designed laboratory apparatus for grouting in poorly-graded sand. Sandy soil specimen was prepared by a wet-raining method. Bleeding tests were carried out to select a cement-bentonite mixture with minimal bleeding. Both water injection tests and grout injection tests were conducted to investigate the threshold seepage velocity, the fracture initiation pressure, and factors affecting the fracture initiation pressure using a plot of injection pressure (p) against injection rate (q). Under low q value, the development of p-q curve was not relevant to the overburden pressure applied and the p value was generally in a proportional linear relationship with the q value. It is also noted that for the same q value, the fracturing pressure required was increased with an increase in effective overburden pressure. Vertical oriented fractures were formed and then turned to a horizontal orientation. The subsequent turning of the fractures might be related either to the minimum compressive stress somehow being increased during the test or to the inhomogeneous nature of the sand specimen.