The paper was presented at the SPE/DOE Unconventional Gas Recovery Symposium of the Society of Petroleum Engineers held in Pittsburgh, PA, May 16–18, 1982. The material is subject to correction by the author. Permission to copy is restricted to an abstract of not more than 300 words. Write: 6200 N. Central Expwy., Dallas, TX 75206.

Abstract

Hydraulic fracturing experiments have been performed in the laboratory in prefractured material under triaxial states of stress. Tests have been run on naturally fractured blocks of Devonian shale as well as blocks of hydrostone in which the angle of approach of the hydraulic fracture to the prefracture was varied in a systematic way. It was found that hydraulic prefracture was varied in a systematic way. It was found that hydraulic fractures tend to cross preexisting fractures only under high differential stresses and high angles of approach. In most cases the hydraulic fractures were either diverted or arrested by the pre-existing fractures.

Introduction

In fractured reservoirs constructive interaction with the natural fracture system is critical to the success of any stimulation treatment. To be most effective hydraulic fractures should cross and connect the natural fracture system, but it is possible that arrest, diversion, or offset could occur thus inhibiting fracture growth and proppant placement. In designing a treatment one would like to know under what conditions to expect these various types of behavior so that precautions could be taken against detrimental effects. The purpose of this project has been to provide this information by performing hydraulic fracturing experiments in provide this information by performing hydraulic fracturing experiments in the laboratory where the results can be examined directly. Two previous laboratory studies suggest that hydraulic fractures will cross closed preexisting fractures under triaxial stress. In 1963 Lamont and Jesseni performed 70 successful hydraulic fracturing experiments in six different rock types. These tests were run under triaxial compression up to 1142 psi and with angles of approach between hydraulic and pre-existing fractures from 30 to 90. Hydraulic fractures were found to cross closed pre-existing fractures at all of the angles of approach and all pre-existing fractures at all of the angles of approach and all combinations of stresses. Lamont and Jessen noted that the rates of fracture extension in their laboratory models were "considerably greater" than in field tests raising the possibility that fracture extension may have been unstable. Daneshy reports the results of three experiments in granite in a paper delivered in 1974. The pre-existing fractures were irregular natural fractures whose direction varied about 150 from right angles to the direction of propagation of the hydraulic fracture. The principal stresses for all tests were = 1000 psi, = 800 psi, and principal stresses for all tests were = 1000 psi, = 800 psi, and = 500 psi. The hydraulic fractures appeared to be arrested when the natural fractures were open at the point of intersection and appeared to cross the natural fractures when they were closed. Recent work by Andersons has shown the importance of friction on hydraulic fracture growth near unbonded interfaces in rock. In these tests hydraulic fractures were propagated at right angles to saw cuts in Indiana Limestone and Nugget Sandstone under uniaxial stress. The results of the tests are reported in terms of a threshold normal stress below which hydraulic fracture growth was arrested at the interface. This threshold normal stress was found to be inversely proportional to friction on the interface.

EXPERIMENTAL EQUIPMENT AND PROCEDURE

The apparatus used in this study was a 500,000-lb triaxial load frame capable of subjecting a 12 × 12 × 15-in. block to magnitudes of stress up to 3,000 psi. Cross-sectional diagrams of the apparatus are shown in Figures 1 and 2. Figure 3 shows the apparatus being assembled for a test. The two hydraulic systems, one for the flatjacks and one for the fracturing pressure used in running a test, are illustrated schematically in Figure 4. Each pair of flatjacks was pressurized independently with a hydraulic pump. The fracturing pressure was generated by a pressure intensifier actuated by a closed-loop servo-control system. This system was used to displace fracturing fluid at a constant flow rate of 0.05 cu in./sec. Fracturing pressures were recorded in analog and digital form for subsequent analysis and plotting with an HP 9835 desktop computer. An example of a pressure-time plot is shown in Figure 5.

p. 559

This content is only available via PDF.
You can access this article if you purchase or spend a download.