Abstract This paper reviews Mobil Oil Indonesia's experience in the completion of high temperature, high rate gas wells, gained during ten years as operator for the Arun Field. Over 50 production and injection wells have been completed in the Arun Field, and some of these wells have been worked over once or even twice since their initial completion. problems which have been encountered during the operation or workover of the wells have led to changes in completion practices and equipment design. These problems are described in detail as are the subsequent completion changes and the results of these changes. Introduction The Arun Field is a giant natural gas condensate field located in Aceh Province at the northern tip of the island of Sumatra (figure 1). It was discovered in 1971. Mobil Oil Indonesia (MOI) is the operator for the field, acting as a production sharing contractor to the Indonesian national oil company, Pertamina. The field is a major source of LNG for Japan and Korea. Abnormally high producing temperatures and pressures, the corrosive constituents of the well stream fluid, and high field production rates, have combined to present many challenging engineering and operating problems. Individual Arun wells are capable of production rates in excess of 150 MMSCF/D with wellhead temperatures above 300F. The initial reservoir pressure was 7100 psia and wellhead producing pressures were 5500 psig or greater. As a consequence of these extreme conditions, the original Arun completions developed wellhead seal and tubing connection leaks. A comprehensive design review, including laboratory testing, identified the mechanism by which leakage occurred. An alternative completion design was then developed based on metal to metal wellhead seals on the final cemented casing, and premium shouldered connection tubings with multiple seals. New Arun wells were completed in this way and a workover program was undertaken to recomplete all wells which had the original completion. The Arun Field workover program indicated that the weighted invert emulsion packer fluid resulted in solids settlement and emulsion breakdown. Extensive investigation eventually led to the specification of an unweighted refined oil as a suitable replacement packer fluid. The workover program further showed that the seal assembly in every polished bore receptacle (PBR)/seal assembly type of completion was firmly stuck in the PBR. No seal assembly was pulled in any of the 29 workovers. A testing program was initiated to determine if a packer type completion with a rigid tubing to packer connection would be suitable for Arun usage. Subsequently an hydraulic set packer was specified. II. THE EARLY ARUN COMPLETION DESIGN The first Arun well to be drilled and completed as a development well was CIII-2 (A-15), in January 1977. The well was completed as shown in figure 2, and included many of the features typical of the early Arun completion design. Features which should be specifically noted are as follows: ○ Resilient elastomer seals were used on all wellhead casing packoff sealing systems with only the tubing; hanger neck seal incorporating a metal to metal seal. ○ A premium thread tubing connection with limited back-out resistance was specified. ○ The production tubing downhole isolation assembly was the PBR/seal assembly type with the PBR being run as part of the 9 5/8 inch cemented casing. ○ A weighted invert diesel water emulsion was used as the packer fluid in the 7 inch by 9 5/8 inch annulus. It became evident after some operating experience had been gained that this type of completion was not suitable for Arun producing conditions. The problems encountered will be discussed in the following sections. P. 218^

Abstract This paper presents a study on effects of perforation flow geometry on the evaluation of the flow performance of perforations. From data obtained in the laboratory serious deficiency and poor accuracy were identified in exploring the linear perforation targets recommended by API RP 43 Section II for the evaluation of the flow performance of perforations. This discredits the validity of the procedure and poses doubt to the correctness of applying API RP 43 results to the downhole conditions. The work suggested that perforations with radial flow configuration should be utilised in such an evaluation. Introduction Perforating is one of the most important elements in the well completion process. The ultimate test of perforation effectiveness has usually been the well productivity. As a result, much attention has been devoted to the laboratory test of perforating equipment and perforations so generated as a means of predicting and improving the well performance. Over the past 30 years, laboratory procedures for evaluating gun performance have been evolved. The first comprehensive laboratory investigation on the flow properties of perforation was carried out in 1953 ! This study led to the introduction of the laboratory flow test procedure which was devised in 1956. In the procedure, perforation effectiveness was described in terms of a Flow Index (FI). The Fl compare, the flow rate through a perforated target after backflowing (qp) to a flow rate previously obtained through the same target with both the faces open to flow (qi) i.e.: FI = qp/qi Having recognised the importance of the flow properties of perforations, the industry adopted in 1962 a standard test procedure, namely API Recommended Practice No. 43 (API RP 43). The 1962 edition of the RP 43 describes perforation effectiveness in similar terms as defines for the Flow Index above. It employs the Well Flow Index (WFI) instead of the Flow index. The WFI's defined as the ratio of the apparent permeability of the perforated core (Kp) made from cylindrical Berea sandstone to the permeability of the unperforated core with both ends open (Ko) i.e. WFI = Kp/Ko A perforation made into a cylindrical core target influences the flow properties of the core target in such a way that the core target is effectively shortened. The apparent permeability, Kp, determined for the perforated core target, based on the original length and cross-sectional area, hence Kp will increase with core target perforation. Thus, the WFI can be misleading because the effects of perforation damage masked by the effects of perforation length. The term, Well Flow Index, is also a misnomer because it has no relation to the downhole productivity in a radial system. In an effort to provide more meaningful data, the API procedure was revised in 1971 and once more revised in 1974. The new procedure, the apparent permeability measured in a core with a real damaged perforation (Kp) is compared with the permeability of a core with an ideal, clean perforation of the same perforation depth (Ki). This new measure is called Core Flow Efficiency (CFE). The CFE is defined as: where Ki is the permeability of a perforated core with ends open. Thus the API RP 43 now provides a measure of perforation quality. The API RP 43 test procedure includes two types of tests which are referred to as Section I - Surface Test and Section II - Laboratory Flow Tests. Section I defines multishot perforation penetration performance in a large concrete drum type target using actual field guns. P. 322^

Abstract Completion operations, for oil and gas wells, are usually performed in casing sizes not exceeding 13-3/8" (34.0 cm) with the majority of completions being performed in 9-5/8" (24.5 cm) and smaller. As such, the availability of completion equipment, for large diameter casings, becomes extremely limited or nonexistent, Occasionally, conditions may exist which require the completion to be designed in a casing size larger than 13-3/8" (34.0 cm). In order to complete in these casing sizes, special equipment must be designed. This paper discusses the design criteria, development, and subsequent use of perforating and gravel pack equipment generated for use in 20" (50.8 cm) casing. The theoretical approach to the evolution of this equipment is applicable to the development of additional systems. Introduction While completion equipment is largely designed for 9-5/8" (24.5 cm) casing sizes and smaller, with some equipment available up to 13-3/8" (34.0 cm), perforating and gravel pack equipment is virtually nonexistent in sizes greater than 9-5/8" (24.5 cm). Due to specific well completion requirements it may be advantageous to be able to complete in the larger diameter casing sizes. With equipment being unavailable for these larger sizes, special equipment must be developed. The herein discussed completion system has been specifically developed for 20" (50.8 cm), 133 lb/ft (198 kg/m) casing. In this application a very shallow, low pressure, gas zone exists beneath a platform. It is the operator's intent to set 20" (50.8 cm) casing across the zone and attempt to blow down the gas. Following depletion of the zone, the 20" (50.8 cm) wellbores can be cleaned out and deepened to the primary objective zones. Further development drilling could then commence with less risk and lover costs. DESIGN PARAMETERS With the primary objective of the project being the depletion of the reservoir, the resultant completion had to be installed in the most efficient manner possible. Owing to the less than 500 psi (3447.5 kPa) static reservoir pressure and the desire to deplete as quickly as possible, the perforating system had to deliver the highest shot density possible with the greatest amount of charge performance. This would provide for the maximum flow rate by minimizing pressure loss through the completion. A systems analysis plot, Fig. 1, supports this and, thus, fostered the need of a specialized perforating gun. Existing perforating equipment would have required multiple trips to achieve the desired perforation shot density, and would not have been able to offer positive, shot orientation. Performance would also suffer as clearance between the gun outside diameter and the casing inside diameter would have been difficult to control. Consequently, a perforating system was needed that would offer a very high shot density with maximum performance, while affording positive shot orientation. Due to the zone being gas, positive well fluid isolation and pressure containment became a criteria. This precluded the use of existing large diameter equipment, which typically have lead seals that are not capable of effecting a suitable gas seal. Additionally, the original design parameters called for the use of a retrievable seal bore packer as the perforating packer. With the expressed desire to underbalance perforate, the requirement for positive gas pressure containment, and the need for secure casing anchorage, a special packer and associated gravel pack assembly was essential. P. 250^

Abstract This paper discusses the experience of and the solution to problems which have arisen concerning tubular products used by Mobil Oil Indonesia, Inc in Arun Field gas well completions. Mobil Oil Indonesia has developed and implemented an intensive Tubular Quality Control Program to maximize the quality and integrity of the tubular products which are subjected to the extreme conditions of this prolific gas field. prolific gas field. The Tubular Quality Control Program which monitors the specified tubular quality from the steel mill to the rig floor has successfully maintained safe and dependable well operations with higher tubing integrity. Introduction Mobil Oil Indonesia, Inc's operating experience in the Arun gas Field has shown downhole failures result from the unique field producing conditions. These conditions demanded tubular products with greater integrity. Producing hot, sour gas from the Arun limestone, at high rates and Producing hot, sour gas from the Arun limestone, at high rates and pressures, subjects tubing to compressive and tensile forces which can pressures, subjects tubing to compressive and tensile forces which can exceed the strength of tubular connections. This paper discusses the experience of and the solution to problems which have arisen concerning tubulars used in the Arun well completions. High annular pressures in the Arun wells suggested that tubing leaks were one of the problems. During workovers, the recovered tubing exhibited tubular defects such as connection leaks, internal and/or external pits, cracks and overlaps. This experience demonstrated the need for an improved system of tubular testing and quality control. This paper focuses on the intensive testing and quality control program developed by Mobil Oil Indonesia, Inc to prevent similar future tubular failures. The Tubular Quality Control Program comprises manufacturing quality control, third party inspections, and improved transportation and handling procedures. Since this program was implemented in 1981, the overall tubular rejection rate has been less than 5.0 percent. Most of the rejections occur as a result of thread damage inflicted during tubing installation. Approximately 0.6 percent ore rejected as a result of damage in transit, and third party inspection losses (mill defects) account for less than 1.6 percent. percent. p. 10–87 p. 10–87

Abstract This paper addresses the basic configuration of simple subsea completions. In an attempt to present an overview of the options open to an operator, examples are taken from various systems which have been designed and/or installed. Features from more advanced designs have become available to upgrade the state of the art in this field. It should be emphasized that like land and platform wellheads and trees, a vast array of options and customized designs are available to meet particular requirements in this environment, and it is difficult to propose particular requirements in this environment, and it is difficult to propose a standard completion. To enable a concise view of a subsea completion all major areas have been touched, from the drilling equipment required, the interface equipment and finally the tree and control systems typically available. WELLHEADS Wellheads for offshore wells fall into three main categories. Marine Wellheads for Floating drilling units. Mudline Suspension/Surface Wellhead for 'fixed bottom' Jackup units. A Hybrid System of 1 and 2 generally a special purpose design for specific applications. MARINE WELLHEADS (See Figure 1) This type of wellhead is located on the ocean floor level. Its primary functions are: To provide a termination of the casing strings with a means to connect the primary well pressure control equipment. Provide a means of guidance for the Blowout Preventor stack, and guidance of any additional equipment such as a subsea tree (if required). Provide a means of suspending the various casing strings and sealing the annuli. The footing rig used to drill the well is equipped with a subsea Blowout Preventor stack (BOP). Preventor stack (BOP). p. 10–1 p. 10–1

Abstract A new generation of fire-resistant wellhead equipment has been developed for offshore platforms. This cooper presents the design concepts and principles of operation of this new equipment, and it describes the principles of operation of this new equipment, and it describes the engineering approach taken during its development. These new wellheads and Christmas trees have survived numerous laboratory fire tests which simulated offshore platform fire conditions. No leakage was permissible during the test burns nor during the subsequent cooldown periods. The laboratory fire test results confirmed the performance periods. The laboratory fire test results confirmed the performance predicted by computer Finite Element Analysis and verified the design predicted by computer Finite Element Analysis and verified the design procedures employed. procedures employed. This new fire resistant wellhead equipment promises to significantly reduce the fire hazard associated with the production of oil and gas from offshore platform wells. platform wells. Introduction The "state-of-the-ort" for fire-resistant petroleum equipment has advanced rapidly during the last few years. This advancement has been driven by the petroleum, industry in response to several catastrophic offshore platform petroleum, industry in response to several catastrophic offshore platform fires. On many offshore platforms, production wells are closely spaced to minimize deck area requirements. Thus, should a fire occur on one well (typically during workover operations), adjacent wells con quickly become overheated to the point of failure. Leakage from these adjacent wells can cause the fire to increase in intensity and spread to other wells in "chain reaction" fashion. Such fires con result in the loss of the entire production platform. The dire consequences of such disasters emphasize the importance platform. The dire consequences of such disasters emphasize the importance of fire-resistant (F-R) wellhead equipment. The petroleum industry responded by developing a recommended fire test specification for valves, API RP6F, in 1978. RP6F is now in its third edition, reflecting the continuing evolution and escalation of fire- resistance standards. During the past, three years, even more demanding "improved" fire test specifications and acceptance criteria have been developed by a major oil company. Table 1 reflects the basic differences between the original API RP6F and the new "improved" F-R specifications. Note "that the "improved" specifications require significantly higher test temperatures and allow no leakage during or after the fire test The "Improved" specifications are already being voluntarily adopted by other major oil companies and may become the "standard" fire-resistance criteria for wellhead equipment in the future.

Discovery and delineation wells which proved the Udang Field did not yield sufficient data to indicate the complexity of the reservoir. Development drilling showed the Udang pay section to be a series of interbedded shales and sands having individuals gas caps and water/oil contacts. The initial plan, to complete the wells by perforating a 40-foot section and installing a screen with slurry pack, was quickly changed in favour of multi-zone single selective completions. This paper describes the various types of completions and equipment used, and the success achieved by utilizing single selective completions to control excessive gas and water production. The field commenced production in January 1979. Since then to the present, about 100 trips have been made with wireline tools to oen or close sliding sleeves with only one failure. In 3 wells the same sliding sleeve was successfully actuated six times. This has been achieved even though some wells have 65 degree inclinations and most all wells produce measurable amounts of sand. It is concluded that these type completions have substantially increased present and projected oil recovery from the Udang Field. Introduction Conoco, as operator for itself, Java Gulf Limited, Getty Oil International (Indonesia) Ltd. and Inpex Natuna Limited, has completed development of the Udang offshore field in the Natuna Sea, Indonesia. It is located approximately 1000 Km north of Jakarta as shown in Figure 1 . Water depth at this location is 300 feet. Production commenced January 18, 1979 with completion of well A-1 and reached 40,000 BOPD after completion of development drilling in the â??Bâ?¿ platform early October 1981. Exploration and Discovery On October 16, 1968 a Production Sharing Contract was signed between Pertamina and conoco Indonesia to explore for and develop hydrocarbons in the Block â??Bâ?¿ contract in the Indonesia South China Sea. The Udang anticline in Figure 2 was primarily delineated by data from seismic surveys run between 1968 and 1975, with most detailed data being acquired in the period 1973 – 1975. In 1977 additional detailed seismic data were obtained in the northeast and north areas of the structure, and in 1979 in the southwest portion of the anticline. Four exploratory wells, Udang 1, 2, 3, and 4, were drilled on the Udang anticline between July 1974 and May 1975. All wells tested hydrocarbons; however, Udang 1 and 3 had the most significant oil flows. Development The results of the first four exploratory wells justified the placing of a production platform (Udang â??Aâ?¿) in the central part of the anticline and the drilling of 13 development wells in the period December 1978 to November 1979. An additional platform (Udang â??Bâ?¿) was set in September 1980 from which an additional 13 wells were drilled with the last well completed on September 23, 1981. (See Figure 2 ). The producing reservoir in the Udang Field in the Udang Field is the Gabus Sandstone of Oligocene age. The sand was deposited from a very large deltaic fan identified to be 1500 feet thick and extending 30-40 miles in length with a width of 3 to 5 miles. Udang production occurs in the top 300 feet of this sand/shale sequence, hence the entire field is underlain with water. Several individual sand members have been identified in the immediate field area and given nomenclature as shown on Figure 3 . An example log is shown on Figure 4 . Most producing zones have average permeabilities exceeding one Darcy. The complexity of the reservoir was not known when initial development plans were formulated based on information obtained from exploratory wells, 1, 2, 3, and 4. The reservoir was described as two separate accumulations as shown by Figure 5 , the Upper Gabus as a thin sand having a large gas cap and a small oil ring; the Main Gabus as a massive sand, with interbedded shales, containing undersaturated oil. The development plan called for all Main Gabus wells to be completed by perforating a 40-foot section in the 7" production string and doing a single inside gravel pack completion using 40–60 mesh Ottawa sand resieved to meet specifications of less than 5% coarse or 2% fine particles. The screen and tell-tale was 8 gauge wrapped all-welded type 316 stainless steel double annealed wire having a controlled Rockwell â??Câ?¿ hardness of less than 22. Development well A-1, completed on 18 January 1979, did not substantiate the original concept of two main hydrocarbon bearing units, (Upper and Massive Gabus Sands), within the Gabus reservoir. It was temporarily completed as a single well without gravel pack in order to evaluate the sensitivity of sand production and check the performance of the reservoir under this producing condition. Well A-2 found an additional and distinct oil bearing sand not seen in other Udang wells; it was completed as dual single selective with the lower zone gravel packed as per original plans ( Figure 6 ).

Producing hot, high rate, high pressure wells can subject tubing threads to Producing hot, high rate, high pressure wells can subject tubing threads to compression and tensile forces that exceed the scope of current published data and specifications. Mobil Oil Indonesia, contractor to Pertamina, Indonesia's National Oil Company initiated the development of test apparatus and test procedures to simulate actual downhole stresses on tubulars. Mobil's intention was to fill the gap in the published data and to determine the cause of thread leaks in Arun wells. It was postulated that the leaks were the result of extreme compression and tensile forces acting on the tubing threads. Wells completed in the Arun Limestone are capable of producing 250 MMSCFD at well-head pressures and temperatures of 4000 psi and 347 degrees F. Forces on the tubing string varying from 350,000 lbs compression to nearly 500,000 lbs tension during pressure and temperature reversals between conditions of shut-in, producing and stimulating. The test apparatus and procedure can detect thread leaks while subjecting test specimens to: 7500 psi internal pressure Compression and tensile loads up to 500,000 lbs. Temperatures up to 350 degrees F Quenching of internal pin surface such that the temperature is reduced from 350 degrees F to 150 degrees in Application of the equipment and test procedures confirmed the failure mechanism and helped evaluate other thread designs for use in the harsh Arun wellbore environment. Introduction Tubulars, worldwide, are being subjected to a more severe environment as the economics of oil and gas production permit the industry to produce from formations that are deep, hot, sour and high pressure. Producing from these horizons introduce tubing connections to compression and tensile forces that exceed the range of currently published data. Mobil found it necessary to fill the gap between the conditions of published tubular performance data and actual producing conditions. Mobil published tubular performance data and actual producing conditions. Mobil Oil Indonesia operates the Arun Field in Northern Sumatra. High annular pressures in Arun wells suggested that tubing leaks were compromising an pressures in Arun wells suggested that tubing leaks were compromising an otherwise competent completion system. It was postulated that abnormally high wellbore temperatures and pressures were exerting forces on the tubing that exceeded the strength of the connections. Test apparatus and procedures were developed to stimulate actual downhole conditions. Use of procedures were developed to stimulate actual downhole conditions. Use of the equipment and techniques permits detection of thread leaks while subjecting test specimens to: 7500 psig internal pressure Axial loads from 500.00 lb tension to 500,000 lb compression. Temperatures up to 350 degrees F. Quenching of the internal pin surface such that the temperature is reduced from 350 degrees F to 150 degrees F in 30 seconds. Test Apparatus Figure 1 is a schematic illustrating the test equipment. A 10-¾ inch I>D. environmental test chamber (1) is mounted via an adapter flange (2) to a load cell (3). The chamber is rated at 10,000 psig burst pressure. The load cell is in the form of a hydraulic press capable of exerting 500,000 lb compression or tension. Two side outlets are fitted with blind hubs (40 and tapped for ½ inch NPT for recording chamber pressure and providing a relief valve. The test specimen is made up of two 7-inch, 35 pounds per foot nipples or pup joints (5) joined by a coupling (6). The specimen has had the threads pup joints (5) joined by a coupling (6). The specimen has had the threads cut that are to be evaluated. A load cell crossover (7) and a 10-¾ inch × 7 inch casing hanger (8) has been attached to the test specimen. The specially fabricated hanger has been tapped to provide: * Water injection to two spray nozzles (9) that direct a cold water sprayon to the seal area of the connection at each end of the coupling. * Nitrogen injection (10) to pressures up to 7500 psig. The pressure inside the test specimen is also recorded at this point. * A vent pipe to extract water and nitrogen and provide a means of stabilizing the internal pressure while injecting water. * Temperature recording (11) from a thermocouple attached to the inside of the top nipple near the lower seal area. * Two high pressure monitoring lines to record pressure from holes (12)tapped in each nipple opposite the sealing area of the connection. This monitors leakage within the connections. The specimen is attached to the load cell piston (13) inside the chamber. The annular volume between the specimen and the chamber is filled with oil. The purpose of the oil is to effect homogeneous heat transfer from the chamber exterior to the test specimen. The specimen assembly is held in position by a seal clamp (14) and hold-down ring (15). Heat is provided by position by a seal clamp (14) and hold-down ring (15). Heat is provided by twelve electrical heater strips that are wrapped around the exterior length of the environmental chamber. Figure 2 illustrates the peripheral attachments to the environmental chamber. Mobil used nitrogen to test the connections. A high pressure gas pump (A) and nitrogen gas supply (B) was used to pressure the test fixture. pump (A) and nitrogen gas supply (B) was used to pressure the test fixture. The spray nozzles are fed by a high pressure pump (C) capable of delivering 12 to 14 gpm against the internal test pressure. The water pump was manifolded (C) to circulate water through a 1/3 inch choke or to pump water to the fixture. A vent line (D) was provided to control the bleed-off of water and excess nitrogen pressure during the water injection cycle. The temperature control panel (E) regulates the heater strips. A strip recorder (F) monitors temperature and pressure exerted on the specimen. A relief valve (G) on the chamber upper port regulates the chamber annular pressure. pressure. Arun Test Apparatus The test facility is designed to simulate wellbore conditions. Table 1 shows the Arun wellbore parameters. Flow rates of up to 250 MMSCFOD through 7-inch tubing cause flowing wellhead temperatures to approach the 350 degrees F bottom hole temperature. Wellhead temperatures have been recorded at 342 degrees F at flowing pressure of 3500 to 4500 psig.

The Attaka Unit, operated by Union Oil Company of Indonesia on behalf of Pertamina and Japex Indonesia Limited, is located 13 miles off the East Pertamina and Japex Indonesia Limited, is located 13 miles off the East Coast of Kalimantan, Indonesia, 90 miles north of the coastal city of Balikpapan (Fig. 1). The structure is a domal anticline complicated by transverse faults. The stratigraphy is predominantly a sequence of sand and shales reflecting in broad aspect, a transition from marine environments of deposition to shoreline and deltaic environments as the basin filled. The field was discovered in August. 1970 with the Attaka Well Number 1-A. Seven additional wells were drilled to define the limits of the field. Construction of offshore platforms, onshore process and oil storage facilities, and pipeline platforms, onshore process and oil storage facilities, and pipeline installation commenced in December, 1971. Initial oil production began in November 1972, 27 months after the production began in November 1972, 27 months after the discovery of the field. The initial producing rate from the nine wells on Platform A was 20,000 BOPD. By November, 1973, the original projected producing rate of 100,000 BOPD was attained from the field when the sixth Attaka-Unit platform was completed and placed on production. platform was completed and placed on production. As of November, 1973 there were 52 development wells in the Attaka Field with 37 dual and 15 single completions for a total of 89 production strings. Within the 52 wells, a total of 173 separate reservoir zones were perforated and available for production as needed. A cross-section of the Attaka structure (Fig. 3), reveals the lenticularity of the sands and the large gas caps which complicate the development of this field. Reservoir sand thicknesses range from 5 to 100 feet and many of the thinner reservoir zones are discontinuous through the field. Since there are over 50 sands of varying areal accumulation it would have required a prohibitive number of wells to properly develop each individual sand without the concept of multiple packer completions. As in many new fields, particularly where there are a large number of stray sands as at Attaka, there are questionable zones that are not defined by sidewall cores or downhole logs. These zones are usually left for secondary completions during the latter life of the field or, if the interval is large enough, a drill stem test or formation interval test is performed. At Attaka, having committed to multiple interval completions, the questionable zones were perforated and isolated from major reservoir zones by packers. Consequently, when the individual strings were put on production, there were 33 zones that each produced less than 10,000 barrels of oil before having to be shut-in because of gas, water, or sand production. Yet inspire of this, field production has been maintained at an average daily rate of 100.000 BOPD for the past two years. The success in maintaining the maximum rate can clearly be attributed to the completion techniques. As of December 1975, five and one-quarter years after the discovery of the field. one hundred million barrels of oil had been recovered from Attaka. Reservoir Engineering Concept and Development Criteria As the initial eight well exploratory program continued to establish the limits of the field, drill stem tests were performed to determine rock and fluid properties, productivity indices, skin damage, fluid interfaces and fault locations. Pressure build-up and drawdown data obtained from the major reservoir zones indicated that these zones could sustain daily producing rate of 1000 to 3000 BOPD and would adequately producing rate of 1000 to 3000 BOPD and would adequately drain a radial area of 1250 feet. This information was then utilized for determination of the optimum well spacing and locations. DST results also provided data for projecting eventual production rates. It was noted during several DST's that a production rates. It was noted during several DST's that a producing rate in excess of 100 BOPD per foot of perforations producing rate in excess of 100 BOPD per foot of perforations resulted in some sand production. Since it was felt that sand control completions would reduce the productivity of the wells, a value of 75 BOPD per foot of perforations was established as the maximum producing rate. This has proved to be satisfactory except when the wells began cutting water. In certain instances when water production was first detected, the producing rate was restricted to 50 BOPD or less per foot of perforations and sand production frequently decreased or was completely eliminated. However, some zones produced sand even during the period of waterfree production produced sand even during the period of waterfree production and in such cases the sand producing zones have been shut-in and other zones opened to maintain an active tubing string. The horizontal and vertical permeability of the loosely consolidated sands is virtually equivalent. In almost every case where gas-oil contacts are present in the wellbore or within a few feet of it, coning/fingering of gas has occurred when the producing rate has exceeded 50 BOPD per foot or perforated interval. However, by selectively opening and closing perforated interval. However, by selectively opening and closing sleeves, the field producing gas-oil ratio has actually decreased with time (Fig. 4). Producing intervals that increased to high gas-oil ratios have been shut-in and successfully allowed to re-saturate to avoid energy wastage while the tubing string was kept on production by opening alternate zones with wireline shifting tools. This flexibility has not only provided the opportunity to maintain maximum production. but has also enabled the reservoir engineers to closely monitor production and pressure decline so that proper reservoir production and pressure decline so that proper reservoir management could be practised, resulting in the maximum recover), of oil and gas.

This paper presents a summary description of three types of sub-sea completions. Included in the considered systems are sub-sea trees of various degrees of sophistication which are landed on standard sub-sea wellhead systems. Running procedures, method of operation, and control systems used will be explained. The first tree to be discussed has a single control line which operates the hydraulic wing valve only. This tree is set with diver assistance, with manually operated master, swab, and annulus valves. The second tree has a single hydraulic control line that operates all the valves on the tree using pressure sensitive pilots. The system provides vertical rig control of valves and the hydraulic tree connector through a manifold spool and horizontal control from the platform after completion. TFL (through the flow line) loops and wireline set diverters are used for TFL work. A diver makes the flow line connections. The third tree has the characteristics of the second tree discussed. in addition, it has a hydraulic flow line connector, an electro-hydraulic multiplex control system, and is set and operated completely without diver assistance. Simple Tree Fig. 1 illustrates a single control line, single function. diver assisted tree. The major components to be discussed are: Tubing hanger orienting bushing. Tubing hanger and running tool. Manually operated tree connector. Valves. Control system. 1. Tubing Hanger Orienting Bushing Prior to running the tubing string and hanger, a tubing hanger Prior to running the tubing string and hanger, a tubing hanger orienting bushing is run. (Fig. 2). The orienting bushing is landed in the connector below the blowout preventers. After it is landed, it is rotated until a spring loaded dog locks into a vertical slot in the connector and is oriented with respect to the guide posts. As the tubing hanger is lowered into the bushing, a lug on the hanger lands on a camming surface in the bushing. As the hanger is lowered through the bushing, the lug slides down and around the camming surface and the hanger is oriented with respect to the guide posts. 2. Tubing Hanger and Running Tool The tubing hanger performs the same general functions as a land type hanger: suspending the tubing and isolating the tubing run or runs from each other and the annulus. There is usually one more vertical bore through the underwater hanger than there are tubing strings. This provides an access to the annulus with a vertical run through the hanger rather than horizontal access in the casinghead housing below the hanger as provided with land type equipment. The underwater tubing hanger is lowered and landed with a running tool made up on drill pipe. The running tool is locked into the hanger at the surface with hydraulically actuated dogs on the running tool. These fit into a recess at the top of the hanger. (Fig. 3). After landing the hanger, rotation of the drill pipe and running tool will mechanically wedge dogs out from the hanger into a machined recess in the casing seal assembly and lock the hanger securely into place. The running toot is released from the hanger by dropping an unlock dart down the drill pipe and applying fluid pressure down the drill pipe onto the top of the dart. This pressure moves a sleeve from behind the dogs on the running tool which frees it from the tubing hanger. In the top of the underwater hanger, just as with a land type hanger, there are seal subs or receptacles in each of the bores through the hanger. These receive the seals on the tree when it is landed. In the production bore through the hanger, there is a sealing nipple preparation to receive ' a standard wireline plug. Normally, this plug is installed before the BOP's are removed and is retrieved with standard wireline retrieving tools after the tree has been tested. 3. Tree Connector The connector (Fig. 4) used in this system is diver actuated. A series of locking dogs are run in by the diver to connect the tree to the wellhead. The tree is run on the guidelines used for the preventer stack. Guide funnels attached to the tree connector give initial orientation to the tree and final orientation is achieved by the lug on the hanger neck engaging the vertical slot in the connector.