Fracture Width-Design vs. Measurement
- Michael B. Smith (Amoco Production Co.) | Randi J. Rosenberg (Amoco Production Co.) | Jerry F. Bowen (Amoco Production Co.)
- Document ID
- Society of Petroleum Engineers
- SPE Annual Technical Conference and Exhibition, 26-29 September, New Orleans, Louisiana
- Publication Date
- Document Type
- Conference Paper
- 1982. Society of Petroleum Engineers
- 2.5.2 Fracturing Materials (Fluids, Proppant), 4.1.5 Processing Equipment, 5.5 Reservoir Simulation, 3.2.3 Hydraulic Fracturing Design, Implementation and Optimisation, 2.4.3 Sand/Solids Control, 4.4 Measurement and Control, 4.1.2 Separation and Treating, 3 Production and Well Operations, 2.5.1 Fracture design and containment, 4.3.4 Scale, 2.2.2 Perforating, 5.2.1 Phase Behavior and PVT Measurements, 5.6.1 Open hole/cased hole log analysis
- 7 in the last 30 days
- 572 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 8.50|
|SPE Non-Member Price:||USD 25.00|
Current fracture designs are based on choosing fracture height and using a 2-D approximation to calculate fracture width. There are however, two approximations which can be used, leading to greatly different designs in terms of fluid volume, pad volume, etc. Given this discrepancy, accurate in-situ measurements of fracture width are needed to help define fracture geometry.
The purpose of this paper is to present measurements made with a downhole closed-circuit television camera during a fracture stimulation (without proppant) of an oil bearing sandstone formation in proppant) of an oil bearing sandstone formation in an effort to investigate the applicability of the two fracturing models. Data is presented showing fracture width as a function of height and pressure. The measured widths are briefly compared to different design theories.
Since its inception 35 years ago, the field implementation of hydraulic fracturing has seen tremendous evolution of both materials and equipment. Unfortunately, theoretical understanding of the fracturing process and design procedures have not generally matched this improvement. The current state of the art in stimulation design is to guess a fracture height (often tempered by experience) and then apply either of two 2-D assumptions concerning fracture width. Coupled with calculations for fluid loss, fluid volume requirements can then be determined for a desired fracture penetration. Differences between these two basic assumptions have been adequately discussed in the literature and these will not be reiterated here. These differences can, however, lead to significant differences in total volume requirements, percent pad volume, etc.
Although fracture design is based on width assumptions, there has been no data available as to actual fracture widths. Only recently have comprehensive lab tests been published studying fracture width as a function of length and pressure. These tests have the laboratory advantages of precise measurement and control of variables. However, laboratory studies of hydraulic fracturing suffer from two difficult problems: geometry and rock strength.
Since hydraulic fractures, and in particular Massive Hydraulic Fracturing (MHF) stimulations, are typically designed for length/height ratios from 2 to 20, the experiments need to match this. However, since confining pressure such as used in Reference 2 is also required, the sample size becomes prohibitive. The second problem is rock strength. For a prohibitive. The second problem is rock strength. For a fracture with a length (or diameter) of a few inches, rock strength or fracture toughness is very important; however on a field scale, rock strength is a minor variable, since only minimal pressure is required to propagate a fracture several feet in length.
The alternative to lab tests is field tests, where precise knowledge and control of the variables are impossible. Generally only indirect data are available with bottomhole treating pressure being the only easily quantified variable. This paper discusses a field experiment where an additional variable is measured; direct measurement of fracture width using a closed circuit downhole television logging tool. The test was conducted in a relatively shallow 4500 ft (1371.6 m) oil bearing sandstone and while yielding significant data, the test is clearly limited in two respects: obviously only wellbore data is measured and, as discussed below, the width measurements made from the TV logs are slightly subjective.
Fig. 1 shows a gamma ray log of the test well along with laboratory measurements of Young's modulus. Modulus was measured in triaxial compression tests on horizontal, oil saturated core plugs. These tests were conducted at a confining pressure of 3000 psi (20.7 MPa). For the upper pay, the average modulus is 3.2 million psi (22,000 MPa), and using a Poisson's ratio of .21 measured at 4574 ft (1394 m) gives E' of 3.3 million psi (22.752 MPa). This value is used below for fracture width calculations.
|File Size||756 KB||Number of Pages||9|