Experimental methods and results are presented for two-dimensional (2D) hydraulic fracturing experiments investigating the interaction and crossing of a hydraulic fracture with a pre-existing frictional discontinuity. The 2D experimental results include direct viewing and measurement of fracture path along with fracture and interface displacement. These data are compared directly to results from a 2D numerical hydraulic fracture model. The 2D experimental method involves propagating a hydraulic fracture from the centre of a 350 by 350 by 50 mm rock sample. Transparent shims are used to transmit the normal stress to the face of the sample and allow continuous digital video records to be made of the fracture growth and fracture interaction with the orthogonal discontinuity. Displacement of a grid drawn onto the sample face is measured by optical analysis methods, allowing the fracture and interface opening to be measured to better than 10-micron resolution. The data provide information about fracture crossing/arresting, path, and fracture and interface displacement with time. The numerical model matched the experiment in the sense of correctly predicting crossing in one case and blunting in the other. The model was unable to match the pressure from the experiments in siltstone, primarily because of not including effects from the pump and injection system compliance.
The mechanics involved in a hydraulic fracture interacting with a natural fracture have been extensively studied over the past years because this process produces a number of first order effects on fracture growth. The crossing interaction can result in branching, offsetting, and even blunting of the hydraulic fracture tip, significantly changing the fracture growth, excess pressure, and width. A verified method to predict the outcome of a crossing interaction is therefore necessary in any fracture stimulation model intended for design and post-treatment analysis of hydraulic fractures placed in naturally fractured rock. This paper is primarily concerned with describing and illustrating the use of a new experimental method that allows two dimensional hydraulic fractures to be grown and studied as they interact with a frictional interface. Therefore, only a brief review of previous work on the topic of hydraulic fractures interacting with frictional discontinuities is provided.