Case History: Single-String Horizontal Well Design for Shallow, Unconsolidated Sand

Abstract

SPE/IADC Members Abstract Most "conventional" horizontal well designs employ an intermediate casing string through the build section to stabilize the wellbore isolate the production interval, and provide a mechanical transition for change over to "drill-in" fluid. On shallow wells, capital savings of up to 25%–35% may be obtained if the intermediate string is omitted and a single-string of pipe serves the functions of production casing and slotted or pre-drilled completion. This technique raises special concerns, however, in unconsolidated sands where poor hole stability is a factor. The following paper describes the successful planning and implementation of such a design for eight wells in the unconsolidated Tulare heavy oil sand in the San Joaquin Valley, CA. An understanding of the stress regime and its impact on wellbore stability, combined with an assessment of the economic risk, made this design breakthrough possible. Lessons learned in this pilot program will have application to broader issues surrounding shallow horizontal drilling and extended reach wells. Time and cost analyses are included with comparisons to the benchmark conventional horizontal well. The wells were drilled as a single hole size from surface to TD. A clay-based, fresh water mud system was used for the build section, with an open-hole change over to a potassium chloride/biopolymer drill-in fluid made prior to drilling the lateral completion interval. A single-string of tubulars consisting of a slotted liner tied back to production casing was run to TD. The upper section of the string was isolated from the completion interval with an external casing packer, then cemented in place via a port collar or stage tool. Field results showed the design to be a success. By the end of the program, the wells achieved the targeted 25% cost reduction relative to the benchmark conventional horizontal well. Based on projected activity levels, this will result in a annual savings of at least $1.0 MM for the field development. Background Hundreds of horizontal wells have been drilled and completed in the San Joaquin Basin in recent years for the recovery of heavy oil. The area is distinguished from other drilling environments by the extremely shallow producing intervals the unconsolidated and highly permeable sands, the cyclic thermal loading due to steam injection, and the economic constraints imposed by the low margins associated with heavy oil production. Reservoir Data and Design Considerations. The Tulare formation in California's San Joaquin Basin is a Pliocene-Pleistocene Age unconsolidated sand formation ranging in depth at the South Belridge field from 400' to 1200' (122 m to 367 m). The sand is highly permeable (average in air: 3,000 md) and completely uncemented the only binding mechanism is the heavy oil (10–13 API, 1800 cp. @ 90 F) which it bears. The formation is broken into multiple zones bounded by clay intervals. Each zone is further divided into 3–5 sands by discontinuous clay stringers. Oil is produced by means of continuous and cyclic steam injection. Average reservoir pressure during flooding is roughly 120 psi (0.8 MPa). The Tulare sands contain 1–8% clay, which historically has caused the sand to degrade severely when exposed to typical drilling and completion fluids for more than 3–4 days. Likewise, sands that have been swept by the steamflood have a tendency to break down as a result of fluid invasion and subsequent degradation in the absence of the binding oil. This tendency is normally controlled by maintaining low fluid loss in drilling mud and with by adding of potassium chloride (KCl) for inhibition to clear completion fluids. Despite this control, hole collapse and running sands are not unusual during extended open-hole operations in vertical wells. Overlying the Tulare oil-producing horizon are air-bearing sands, which also have a high potential for fluid losses. A loss of hydrostatic pressure below water gradient will lead to hole collapse and running sands. P. 333^