Straddle-pack selective stimulation via coiled tubing improves production from Ekofisk chalk reservoir

ROBERT MURPHY and STONE FAGLEY, Weatherford International

Many of the more recent wells at Ekofisk field in the Norwegian North Sea have been completed as monobores, with lengthy horizontal or high-angle sections and several distinct perforation clusters, with as many as 20 separate zones. Early methods of stimulating these chalk reservoirs involved high-volume acid jobs, requiring bullheading, and the use of frac balls to divert flow to less-permeable sections. However, after collection and evaluation of much data, it became clear that many of the perforated intervals were unproductive. To address this situation, it was decided to change to a selective stimulation method, using newly designed coiled tubing (CT)-deployed straddle pack-off tools.

Subsequent evaluation of this technique indicated that it was a more productive approach; however, its reliability was less than desirable, due to equipment and methodology failures. This led to a detailed study of these CT operations, carried out over a period of nine months and involving the evaluation of results of nine separate well stimulations, with perforation sets ranging from seven to 20. The study involved review of equipment design and function; material choice and methodology, such as pumping techniques; use of glycol trains; accuracy of depth measurements; reliability of pressure measurement equipment, etc., with the objective of optimizing them for future operations.

This article briefly reviews the history and results of the initial stimulation of these wells and the decision to convert to straddle pack-off methods. Described in detail are results of the study to evaluate and improve equipment and techniques, lessons learned, and changes that resulted and led to the more reliable, effective stimulation techniques now being used.

BACKGROUND

Ekofisk field in Block 2/4 of the Norwegian North Sea’s southern part was discovered in 1969 and was the first major oil discovery in the area. Production commenced in 1971 from four wells drilled from a jackup rig and produced to a nearby tanker and, since that time, the field has been under continuous development from freestanding platforms set in 250-ft water depths. The field is located in the Central Graben area, and production comes from the naturally fractured chalk of the Ekofisk and Tor formations of early Paleocene and late Cretaceous age. The reservoir has very high porosity (30% to 40%), but relatively low permeability, with an original oil column of about 1,000 ft at a vertical depth of approximately 9,500 to 10,500 ft. The producing formation has low mechanical strength, which results in compaction of the chalk reservoirs and can lead to healing of natural fractures and those produced by stimulation, especially with water-based fluids.1 This, in turn, leads to seabed subsidence, and eventually, to problems with platform structural integrity.

The field was originally produced by pressure depletion, but now, limited gas injection and an aggressive waterflood program started in 1987 have contributed to a substantial increase in projected oil-in-place recovery from an initial estimate of 17% to the current level approaching 50%.2 In recent years, development wells have been drilled and completed as monobores, with measured depths ranging from 13,000 to 21,500 ft, and often lengthy horizontal or highly deviated sections. These wells are generally perforated with multiple clusters of perforations (1 to10 ft in length up to 200 ft apart and as many as 20 or more per well) and are produced co-mingled, typically from a 6⅜-in., 65.8-ppf liner.

From the outset, wells were stimulated using high-rate/high-pressure bullheading of high-volume concentrated hydrochloric acid with frac balls being used to divert flow from the more permeable perforations into those sections needing the most stimulation. The problem with this approach was not recognized initially, but the wells had heterogeneous pay zones, and their efficient stimulation was difficult to realize because of differences in porosity, permeability and formation pressures. This practice was continued for several years, but as more production history was accumulated and running of production logs became the norm, it became easier to make detailed evaluations of well performance. It soon became evident that most of the production was coming from a small number of perforation clusters, while many others were not contributing to overall well performance. It was clear that varying zonal properties along the horizontal section length were making the diversion of fluids normally anticipated in bullheading operations less than desirable, leaving many zones effectively untreated and directing most of the treating fluids into the zones that were already productive.

As a result, a new approach was indicated to attempt improving the spread of stimulation fluids more evenly over the perforation sets and, in 2003, three wells that had been drilled in 2002 were subjected to detailed production logging.3 The data evaluation performed on these wells resulted in their selection for a pilot program to test a zonal isolation tool that would allow selective stimulation of different perforation sets by straddling them individually, using a straddle system run on coiled tubing.

STRADDLE PACK-OFF STIMULATION

The treatment proposed presented several major challenges that had to be addressed before the first such treatment could be undertaken:

• The 5½-in. Jet Frac straddle packer system was in the design and testing process, and its specifications were modified to fit the requirements of the proposed operation. It was designed without slips or the need for any manipulation; rather, it was a hydraulic tool in which the packing elements were activated by pressure applied from surface against a spring-loaded piston. It needed to be set at a low flow/pressure rate and still be able to achieve high injection rates into the formation with minimal pressure drop across the tool. This need was fulfilled by means of a new design of the injection valve with a large flow area, and a spring-loaded sleeve actuated when differential pressure exceeds 1,500 psi, at which point, the packing elements would be set. The design was such that it could be set and unset several times on the same run in the well to enable multiple zones to be treated on a single CT trip.4
• Evaluation of well data indicated that to achieve acid fracturing in the carbonate formation, surface pressures of about 7,400 psi would be required, which was outside the scope of available CT units. This necessitated the design of a 2⅞-in. CT string, the weight of which was beyond the capabilities of the platform cranes, and so a split-reel design was utilized to overcome the problem. 
• Because of platform limitations, accommodation of required pumping equipment, tankage and stimulation fluids onboard was not possible, so the acid/frac fluid had to be pumped from a stimulation vessel tied up alongside by way of a 3-in. Coflexip hose. This, in turn, presented challenges, not only with sea conditions and weather fluctuations, but in communication between the pumping operators on the vessel and the CT/downhole tool operators on the platform.
• To monitor downhole pressures and temperatures, and provide for detailed operational data for subsequent analysis, the straddle packer system design was modified so that it could be equipped with memory gauges.

All of these concerns were addressed and overcome, and a total of six non-productive zones in the first three wells were treated using 28% hydrochloric acid. This resulted in a production increase from the three wells of close to 6,600 bopd, with increases in wellhead pressure averaging 60 psi. This post-treatment production increase was not sustained for more than a few months, but the downhole flow profile was altered and became more evenly distributed across the perforation clusters, leading to ultimate recovery increases in the long term.

Several lessons were learned from this initial campaign and were subsequently applied in later operations, including:

• Lower acid strength gives more control on its extended reach; 10% to 15% concentration is equally effective, both in results, and especially, in overall cost
• Additional memory gauges to monitor pressure and temperature above and below packing elements would make it easier to evaluate the straddle performance
• Increased gauge ring size would improve packing element performance and reduce extrusion
• High injection rates were often difficult to obtain, due to the low number of perforations exposed; consideration should be given to re-perforating before treatment to address this situation
• Low-pressure zones can go on a vacuum, causing problems in moving the CT with a full column of fluid
• Larger campaigns should be initiated to reduce overall cost per job by spreading the inherent expense of equipment mobilization and demobilization.

SUBSEQUENT TREATMENT CAMPAIGNS

The initial testing of the straddle pack-off was employed as a basis for continued selective stimulation of these wells in 2006, and later, when a major campaign was undertaken in 2009 and 2010 to attempt stimulating nine wells on the 2/4 M platform. This program included a total of 96 perforation clusters with a newly designed system known as Surge Frac, Fig. 1. This system had several important features and improvements that were derived from experience with the first-generation straddle pack-off system employed earlier:

• Pack-off can be set and unset multiple times in one trip
• The pack-off sets at low differential pressure created by fluid velocity; no pipe movement is needed
• Pack-off incorporates an internal pressure-limiting valve (PLV) that controls overpressure on the packing elements, prolonging their life
• The pack-offs are released by stopping the pumps and activating an integral set down un-loader, which does not entirely eliminate the need for pipe movement, but does negate the axial force on the elements while they retract to the neutral position.

Fig. 1. Surge Frac system

A total of 69 zones was individually stimulated, and efforts to treat the other 27 zones were abandoned for various reasons related to known downhole restrictions, proximity of perforation clusters to the oil/water contact, weather conditions, pack-off problems, etc.

During the latter part of this campaign, pack-off failure rate was increasing, and a contractor’s engineer was dispatched to the location to review the tool’s performance and make further recommendations for improvements in design and operating procedures. The balance of the campaign was completed with positive results by taking into account some of the design findings, and then a detailed study of its progress and effectiveness was undertaken by the operator.

LESSONS LEARNED

Straggle design and operation. Following several days of monitoring tool performance on location and direct participation in its running, disassembly and redress, the following recommendations were made and are being acted upon:

• The first major issue identified was the presence of acid pitting on ealing areas and surfaces. The condition of parts of the tool was such that leakage was occurring during post-assembly pressure testing, and though minor leaks can occur without compromising a good stimulation, some critical seals and parts in the pressure limiting valve (PLV) were exhibiting severe pitting, Fig. 2. Resultant leaks were rendering the packer’s overpressure prevention feature inoperable. This undoubtedly contributed to failures. Significant work was done to replace critical seals with acid-resistant sealing systems, also resulting in improved blowout and extrusion resistance over the original design. Materials engineering studies were made to review the situation, and these indicated that the acid inhibitor used was ineffective in protecting the tool. After discussions with the operator, a request was made to manufacture these critical parts from Inconel material rather than change the additives formulation for the acid.
• Many of the packing elements were not recovered when pulling out, and the few that were, had indications of erosion wear, durometer loss, chemical attack, “nibbling” and gas impregnation/decompression bubbles, Fig. 3. On one occasion, the tool was run in the well to target depth, but due to problems with the stimulation vessel, no pumping was possible for almost 48 hr, and when it was, the tool failed immediately. The elastomeric elements were recovered, having been exposed to wellbore chemicals for the whole time, and they were found to have lost all of their properties. An elastomer analysis and chemical compatibility tests were initiated, and a review of the material used for the packing elements against the well conditions was started. As a result, new elastomers were developed that can still hold high differential pressures after prolonged exposure to the specific well conditions. Testing continues to develop a multi-element configuration and improve the geometry to reduce stress concentrations in the rubber. 
• In earlier operations, pump rates were increased to offset the pressure loss at surface, as the heavy acid entered the coil. Even though surface pressures remained similar, the additional hydrostatic and increased back pressure from flow were responsible for numerous overpressure element failures. To remedy this, the procedure was modified to maintain a constant pump rate until the acid reached the zone, and then was chased with a post-flush. 
• A CT dump valve design was initiated, to be run below the straddle, which would close for stimulation but reopen afterwards to dump fluid below the straddle to alleviate CT overbalance when tripping (thought to be a major cause of packing element loss after stimulations and while tripping out of hole).
• Analysis of gauge data showed that elements were being moved to the next zone before packers were unset. After stimulations, the zones were feeding fluid back into the wellbore with enough pressure to keep the elements energized and set. This zone feedback has to return into the coil through orifices in order to drain. In some cases, it requires hours to drain off. Procedures were modified to increase waiting times before moving.
• It was found that utilizing an orifice or having a leak below the lower packing element, when there is no permeability below the straddle, can lead to pressure build-up underneath and ultimately cause the bottomhole assembly (BHA) to slip, or the lower packer to unset. For this reason, an effective stimulation of the lowermost zone is considered critical to increasing the success rate of the later stimulations. There are a number of solutions available to help prevent this from occurring, including a programmable dump valve that will also reduce CT overbalance and zone feedback.
• Consideration should be given to developing a hydraulic anchor to eliminate tubing movement, keep the elements centered, and reduce eccentric rubber loss. This, too, is being pursued.
• Memory gauge reliability has been an issue, with a failure rate exceeding 40%. Review of these failures indicates a need for better operator training and the use of higher-range gauges to handle any overpressure situation.

Fig. 2. Acid pitting of seal surfaces.

Fig. 3. Damage to packing elements

Operating procedures. When the campaigns of 2004 and 2006 were completed, little or no post-operative review and evaluation was performed, so that when the 2009 campaign commenced, many of the lessons that could have been learned from them were not available. A detailed design review, in common with a Failure Mode and Effect Analysis (FMEA), would surely have identified some cause-and-effect data. To address the shortfalls, the operator undertook a comprehensive review of all the operating procedures and practices used, and the following recommendations were made:

• A real-time telemetry system capable of providing downhole pressure and temperature data during the operation was noted to be useful for interpreting BHA behavior.
• It became clear that many of the dynamic BHA components were not manufactured from acid-resistant materials. This is perhaps a function of lack of communication between operator and contractor, in not fully defining operating conditions, a situation that is now being vigorously addressed, so that everyone will be on the same page in the future.
• It was noted that pump trip events (sudden changes in pumping rate or pressure) coincided with sudden packing element failures, and recommendations were made to review pumping practices accordingly.
• The method of activating the straddle pack-off is being achieved by a combination of applied pump rate/pressure and controlled size orifices; a ball and seat; dual flappers below the lower packer holding pressure from above; or a combination of these elements. A new dump valve design is being pursued that may be preferable, but at present, this situation needs further evaluation to develop the best solution.
• It is clear that major fluctuations in pump rate directly affect the pack-off seal, and a review of the pumping data clearly suggests that smooth rate adjustments are critical to achieving optimum performance. This is difficult to achieve with manual pump operation and makes the job of the CT operator in preventing sudden rapid changes in CT tension all but impossible. Lacking automated pump control, all rate adjustments should be done in small, 1.0–1.5-bpm increments, and communication between pump and CT operators is critical to avoid loss of pack-off. When a pumping vessel with automated pump control was available for use, this situation became much easier to handle.
• The CT reel is full of fluid with the BHA at surface, but on bottom, the fluid level is only that which balances well pressure. When pumping begins to fill the reel, bottomhole pressure in the BHA does not change. Once the reel is full, fluid is pumped to the BHA, but it arrives as a mist rather than a fluid plug, and the BHA starts to set. In this scenario, the frac valve could open before the fluid level reaches the surface, and in any situation where pump rate drops below leak-off rate at the BHA, the pressure across the straddle drops. Stable pump rates and controlled rate changes are vital.
• The spacing between the straddle elements is theoretically sufficient to provide a 10-ft space between the elements and the closest perforation. This should be adequate, but field experience shows that, in many cases, packing elements have been set across perforations, due to depth uncertainty; this is particularly prevalent in wells completed with tubing-conveyed perforating (TCP). Historically, depth concerns have been an issue in almost 50% of straddle runs in some cases, leading to premature pack-off failure. The application of CT-conveyed memory casing collar locator (CCL) or the use of real-time downhole telemetry to obtain reliable CCL data is recommended to address this issue. 
• Each monobore liner, which in many cases, is heavy-wall, non-API pipe, may have a specific internal diameter (ID) substantially different from nominal. Detailed pipe tallies should be made available to enable straddle designs to be equipped with the exact gauge ring required to enable optimum packing element performance.
• Detailed analysis of packing element damage, which typically occurs on the low-pressure side of the element, indicates that the major cause is the result of movement, due to inappropriate CT manipulation or pipe elongation/contraction as a result of temperature changes, pump trips, etc. To eliminate such movement, consideration should be given to the design of a BHA anchoring device.
• In general, operational procedures can be enhanced by detailed evaluation of historic data, better up-front planning and subsequent job preparation, and closer cooperation between operator and contractor(s). Performance of pre-job simulations and detailed evaluation and documentation of BHA status, both pre- and post-job, are essential to improve planning and execution of the straddle pack-off stimulation process. 

ACKNOWLEDGMENTS
The authors thank the operator and Weatherford International Ltd. for permission to publish this article and for their encouragement to do so. This article is based upon paper 153990-PP, “Selective stimulation of multiple sets of perforations with straddle tools: Lessons learned led to equipment and methodology optimization,” presented by the authors at the SPE/ICoTA Coiled Tubing & Well Intervention Conference & Exhibition, The Woodlands, Texas, March 27–28, 2012.

REFERENCES
1. Van Den Bark, Edwin, and Owen D. Thomas, “Ekofisk: First of the giant oil fields in Western Europe,” AAPG Bulletin, Vol. 65, Issue 11, November 1981.
2. Snow, S. E., and E. V. Hough, “Field and laboratory experience in stimulating Ekofisk area North Sea chalk reservoirs,” paper 18225 presented at the SPE Technical Conference and Exhibition, Houston, Texas, Oct. 2-5, 1988.
3. Halvorsen, Helge; Kjetil Ormark and Alistair McLeod, “High-rate stimulation using 2⅞-in. coiled tubing zonal isolation straddle packer,” paper 89660, presented at the SPE/ICoTA Coiled Tubing Conference and Exhibition, Houston, Texas, March 23-24, 2004.
4. “Unique approach to selective stimulation of carbonate reservoirs with multiple sets of perforations and comingled production pays dividends,” World Oil, February 2008.

The authors
	ROBERT MURPHY is global product line manager, Thru-Tubing Packers, with Weatherford International. He has 33 years of experience running packers and bridge plugs in completion and remedial applications. Mr. Murphy has spent the last 18 years working exclusively with coiled tubing-conveyed packers, including thru-tubing inflatables.
	STONE FAGLEY is a research and development Engineer with Weatherford International’s Thru-Tubing Packers & Fishing group in Houston. Mr. Fagley graduated from Texas Tech University with a BS degree in mechanical engineering.