This paper discusses case histories of more than seventy completions requiring sand control. Approximately half of these wells were gravel packed (both slurry and water packs) between 1990 and 1995 while the remaining completions were fracture stimulated (frac-packs) from 1992 through 1995. The case histories include: geopressured oil reservoirs of moderate permeability, normal pressured high perm oil and gas formations, partially depleted high perm gas sands, and shallow dry gas formations. Results of bottom hole pressure transient analyses are included that compare skin values and completion efficiencies of both gravel packed and frac-packed completions. Production plots and decline curves are presented depicting accelerated as well as improved reserve recovery with the frac-packed completions. The associated costs of frac-packing is discussed along with a net present value analysis justifying these costs. Introduction The popularity of fracture stimulation combined with sand control as a completion technique has intensified in recent years. Due to the pronounced success of frac packing, this technique has become the preferred completion method in sand control environments with several operating companies. However, there has been some debate within the industry concerning the feasibility of this technique in moderate to high permeability formations. This paper reviews the results of one operator's experiences in a variety of applications, including both oil and gas reservoirs with permeability ranges of 3 to 4000 md and bottom hole pressure gradients of 0.17 to 0.84 psi/ft. The skin values of 35 frac-pack completions are compared to those observed in 29 gravel packs employing similar completion techniques with the exception of the sand control method. The relationship of skin damage to flow efficiency is also discussed. Several case histories confirm the theoretical calculations, comparing original gravel pack completions that sanded up to their subsequent frac-pack workovers in the same perforated interval. The rate acceleration and improved recovery of these direct comparisons is presented as well as the rig time and costs of the two different techniques. A comparison of productivity index (PI) and the improvement of PI observed over time in frac-packs is summarized. Finally, the economic impact of these improvements is evaluated relative to the increased cost of the frac-pack technique. Skin Damage The most obvious economic justification for frac-pack completions is improved completion efficiency (lower skin) and the resultant higher flow rates. P. 201

This paper describes the "frac-and-pack" completion technique currently being used in the Gulf of Mexico, and elsewhere, for stimulation and sand control. The paper describes process applications and concerns that arise during implementation of the technique and discusses the completion procedure, treatment design, and execution.An innovative well-completion technique, frac and pack, recently has been introduced into Gulf of Mexico well-completion operations. The technique combines the stimulation advantages of hydraulic fracturing with the sand-control technique of gravel packing. The method involves a one-step fracturing and gravel-packing operation performed through conventional downhole gravel-packing hardware.The technique is being used successfully to reduce skin damage effects and to improve the productivity of unconsolidated sandstone completions significantly by creating better communication between the wellbore and the reservoir. An example of the productivity improvement related to frac and pack is provided in the Conclusions and References. Successful fracturing of soft formations poses challenges different from fracturing of typical hard rock. The primary difference is the use of tip-screenout designs and very high sand concentrations to ensure adequate fracture conductivity between the reservoir and the wellbore. Many authors have described tip-screenout fracturing techniques.

The paper presents the findings of 20 completions in both oil and gas wells. These completions were performed from the period of 1973 through 1994. The early completions employed sand control technology of the 1970's and 1980's with most requiring acid stimulation immediately after sand placement to assure the removal of completion damage. The 1994 completions Involved fracturing combined with sand control (frac-packing) in the same reservoirs which had been gravel packed in the '70's and '80's.The unique aspect of this review is that these fracturing treatments were carried out using a solidsfree viscoelastic surfactant fluid. The fluid system is discussed along with the reasons why this particular fluid was selected. The implications that this fluid has on the fracture placement as well as productivity are discussed.Laboratory testing indicates that the solids based polymer fluids actually exhibit "deep bed" filtration and plug the formation pore throats, the fracture packs, and perforations. This damage can require acid stimulation for optimized production.The completion results are compared by use of a productivity index (J), normalized Jl\ and bottom hole pressure build-up analysis. 5%VESS

The Troika Subsea development project has produced the top four hydrocarbon producing wells in the Gulf of Mexico - three wells each delivering greater than 40,000 STBOE/D sustained rate. This paper will give the reader an overview of the completion designs and installations, but equally important, what lessons we learned to apply on our next deep water high rate producers. The Troika field is located in 2700' of water and was a fast paced development project which was delivered (from discovery to first oil production) in only 39 months - a deep water industry record! This subsea development consists of five subsea wells clustered around a production manifold and tied-back 14 miles to the host platform. Many well design aspects, various technologies, and timing issues had to converge quickly for the success of this project. Completion practices will be reviewed and resulting productivity performances will be charted. The following areas will be discussed:Well performance historyPerforating practicesFluid loss control practicesSand control operationsTubing installationFlowback operationsSubsea operationsBatch operationsPersonnel and training Introduction The Troika development is located about 100 miles offshore Louisiana in the Gulf of Mexico in Green Canyon blocks 200, 201, 244, and 245 blocks (Figure 1). The subsea development consists of an eight slot manifold, an electrical umbilical, an hydraulic/chemical injection umbilical and two 10 inch flowlines. The facilities are operated remotely from the Bullwinkle platform located 14 miles northwest of the Troika manifold location (Figure 2). The initial development of the field consisted of batch completing three wells in the S-10 reservoir followed by "drill-then-complete" for each of the final two wells, also in the S-10 reservoir. The S-10 is an unconsolidated Pliocene sand, described in some detail below. First production from the field occurred in November of 1997 with production facilities designed to handle 80,000 STBO/D, 146 MMCF/D and 40,000 BWPD; but based on the results of the initial well completion, the facility was upgraded to increase the Troika processing capacity to 100,000 STBO/D and 200 MMCF/D. The S-10 reservoir thickness is between 200' (crestal site) to over 600' (OWC site) with sidewall core permeabilities ranging from 1,000 md to over 5,000 md (Figure 3). The oil gravity is 37 degree API with a GOR of about 1975 SCF/STBO. The bottom hole pore pressure gradient ranged from 14. 2 at the crest to 13. 7 ppg lower on structure with a bottom hole temperature range of 175 to 185 degrees Fahrenheit. Seismic and well data suggested a homogeneous reservoir both laterally and vertically. The combination of the above factors led to the belief that the S-10 production wells would have a high deliverability. Forecasted rates were between 20,000 and 25,000 STBO/D/well, a target rate which no well in the Gulf of Mexico had ever consistently produced at the time of the planning of the Troika project. Recoverable reserve estimates were between 200 and 300 million barrels of oil in the S-10 sand. A generic wellbore sketch is shown as Figure 4.

Recently, advances in sand control completion practices have lead to the evolution and popularity of two different completion techniques: the high rate water pack (HRWP) and the frac-pack. A wealth of information has been published regarding each separate completion technique, but little information has been published on the direct comparisons of the two techniques. This paper describes three cases, from three separate fields, in which the HRWP and frac-pack techniques were employed. In each case, a completion was performed in formations with similar reservoir characteristics utilizing both the HRWP and frac-pack techniques. Completion steps for both the HRWP and frac-pack wells are discussed at the end of this paper. Production performance from each example is compared, as well as results from bottomhole pressure transient tests. System analysis is used in each case to compare completion effectiveness. The results are summarized with a brief economic evaluation of the two completion techniques for each case history. All three of the example sets presented in this paper consist of high permeability gas-bearing sands located in the Gulf of Mexico. Each example completion is from an area known to produce sand in large quantities if a successful sand control treatment is not performed. Introduction The examples that follow contain a brief description of the wells to be compared, along with a general description of the wellbore configuration and initial flow rates. A more detailed description of the frac-pack and HRWP completion techniques can be found at the back of this paper (Attachments 1 and 2). Table 1 contains the wellbore geometry and reservoir flow parameters for each case discussed in this paper. A comparison of well performance based on production rates, as well as a comparison of bottomhole pressure build-up data, is included for each well. In each gas well used in the following comparisons, a multi-rate test was performed in order to determine the turbulent coefficient associated with each completion. The analysis of the turbulent coefficient for each case history yielded interesting results. However, more conclusive results were drawn when the turbulent coefficients were compared by completion type. A brief discussion on the -system analysis results and potential productivity gains associated with each completion technique is included. Finally, a net present value (NPV) calculation was performed on each completion based on the cost associated with the completion, the production rates seen by each well and the forecasted production for each well. Capital expenses and first production were assumed to occur in June 1997. Expenses were forecasted based on each field's historical operating costs. Product prices were assumed to average $1.81/Mcf for 1997, $1.88/Mcf for 1998 and $1.92/Mcf for 1999. All net present values discussed in this paper will be discounted at 15%. A summary section will provide an overview of the data presented in an attempt to draw guidelines for selecting a completion technique of choice in sand control situations, based solely on deliverability. Other design considerations that are not discussed in this paper are the relative location of nearby fluid contacts (oil/gas, water/gas or water/oil), down hole equipment limitations and the dependability of the completion to control sand production. P. 269^