Fracturing/pressure pumping

Evolution of Multistage-Fracture Completions in a North Sea Environment

This paper reviews two newly developed novel completion systems that significantly reduce time spent performing multistage stimulation in environments where cost and consequence of failure are high.

jpt-2015-04-fig1evolution.jpg
Fig. 1—Setup for erosion-flow testing.

This paper reviews two newly developed novel completion systems that significantly reduce time spent performing multistage stimulation in environments where cost and consequence of failure are high. Both coiled-tubing and wireline-manipulated sliding-sleeve/valve systems and ball-drop-actuated systems have been developed and deployed, depending on the various completion and stimulation challenges faced. Since their first installation in 2009, these systems have been proven and refined in multiple wells for two major operators.

Background

For many of the fields requiring stimulation in the North Sea, cemented plug-and-perforation (plug-and-perf) completions have been used historically. It is a flexible solution in terms of the various types of stimulation designs that can be accommodated. However, because of the large number of perforating runs required, the bridge-plug setting, the retrieving or milling involved, and the cleanout runs required, it is also a very time-consuming option. The turnaround times per stage have been widely documented and often range from 3 to 7 days.

Prototype Development, Testing, and Qualification. Sliding sleeves have been used for flow-control purposes in oil and gas wells for decades. Although sliding sleeves are simple in design, many users have questioned their reliability, and they did not see much use for fracturing operations until the mid-2000s, when ball-drop-actuated sliding sleeves entered the market. The preferred method for stimulating a reservoir with a cemented liner—stage by stage—has been with plug-and-perf methods. Plug-and-perf provides a proven method of initial stimulation, compartment for compartment, but provides no solution for efficient production management over the life of the well.

The entry into the stimulation-sleeve market for one operator was triggered by a need for a robust sliding sleeve that could be used for multiple applications in a field offshore Norway. This initial new-product-development project materialized into the 5.5-in. coiled-tubing (CT) -actuated valve, a high-performance, mechanically operated stimulation/production valve for 8.5-in. open hole. The key parameters in the development of this sleeve were mechanical robustness to cope with the subsiding reservoir, erosion resistance to cope with planned proppant placement, scale tolerance to cope with known scaling issues in the field, and sealing integrity for the life of the well to allow future production management and water shutoff. The sleeve would be operated by pipe or CT and a flow-actuated shifting tool.

The field for which the equipment was designed is a mature chalk field, and the reservoir is undergoing considerable subsidence. To cope with geomechanical forces, these wells are typically designed with heavy-walled liners. After agreement was reached on final design criteria, the project was kicked off in April 2008. The planned activities included final detailed design; a third-party design verification; and prototyping followed by erosion testing, function testing with CT tools, and testing to Organization for Standardization (ISO) 14310 V3 (Fig. 1 above).

The first test was a third-party qualification-test program performed by the International Research Institute of Stavanger (IRIS) in accordance with ISO 14310 V3 criteria. The next test was to simulate a flow rate of 6000 L/min at 4 lbm of ceramic proppant added per gallon through the 16 sleeve ports. To protect the sleeve, all the ports had been fitted with tungsten carbide nozzles and the inner surfaces were coated with antiabrasion coating. The test was performed at the facilities of a major service company in Stavanger (please see the complete paper for details of the testing).

After the testing was concluded and accepted by the client, a series of 14 sleeves was manufactured for the first installation in an 8.5-in. open hole. Because of other delays, however, the first installation was not performed until July 2009 (please see the complete paper for details of this installation), allowing time to test the sleeve in accordance with ISO 14310 V0 criteria, thus allowing the 5.5‑in. CT-actuated valve to be considered as a barrier during the installation. The sleeve was gas tight through the entire test program and was V0 qualified at 15,000 psi and 350°F.

Ball-Drop Actuation and Multientry Sleeve Design. Client interest had been growing for a ball-drop-actuated sleeve system allowing multiple sleeves to be opened by a single ball. A system such as this would allow multiple stimulation exit points to be opened within each stimulation stage by one ball; this would eliminate the need to cluster perforate to achieve the same result. The initial ball-drop sleeve system was designed for use on 4.5-in. tubing for open hole in the Bakken shale. The initial system design was built around replicating a four-cluster perforation job by running four sleeves for each stage. In order to achieve this, a “flex seat” had to be developed that would allow the ball to open and pass several sleeves. In addition, a regular fixed-seat sleeve to stop the ball in the final sleeve of each stage and isolate the lower stage had to be developed. All four sleeve seats within one stage would have to rely on the same ball size, and this would allow one ball to open each of the flex sleeves before opening the last fixed sleeve and isolating the preceding stage.

A prototype was designed in Norway and tested at the IRIS facility in Stavanger. The remaining prototypes were manufactured and assembled in the US. After a successful land trial, the equipment was shipped to North Dakota and two sets were installed in Stages 15 and 16 in the trial well in the Bakken shale together with third-party openhole isolation packers. Three sleeves were run in each stage. On both stages, the ball landed with a rate of 12–13 bbl/min, and a good pressure signature indicated that all three sleeves in the stage were opened by the ball. During this field trial, a mixture of sands was pumped, but the total amount of proppant pumped into the well was 2,573,000 lbm.

Moving to Cementable Multistage-Fracturing Sleeved Completions.Until 2011, all installations had been in openhole completions. However, many fields require liners to be cemented for a variety of reasons, and therefore there was an interest in fracture sleeves that could be cemented.

The next step in the stimulation-sleeve evolution was adapting the system to enable it to be run in cemented-liner applications. The object was to be able to cement directly through the sleeves without the use of inner strings. The sleeves would be wiped clean on the inside by the cement dart. The basis for the cementable ball-drop-actuated-valve system was the sleeve for open hole.

The cementable ball-drop-fracture-sleeve system was tested and qualified in much the same way as the original openhole ball-drop-actuated-sleeve system. First, a string of five ball seats in decreasing sizes was rigged up in series. A batch of cement was pumped through, followed by a wiper dart (Fig. 2), followed by clean water. This test was performed to ensure that a “floppy dart” would wipe the ball seats properly clean and to ensure that the pressure needed to pump such a dart through the sleeves would be far less than the pressure needed to actuate the opening of the sleeve. This test has since been repeated many times to qualify the cementable ball-drop-actuated-valve system with various liner-hanger/wiper-dart combinations, and wiper darts have been pumped through as many as 80 sleeves without any problems.

jpt-2015-04-fig2evolution.jpg
Fig. 2—Liner wiper dart.

Second, a system test was performed in which cementable ball-drop-actuated sleeves were installed inside a 9⅝-in. casing. Cement was pumped down on the inside of the sleeves and up though the sleeve annulus inside the 9⅝‑in. casing. A cementing dart was pumped down the tubing to wipe clean the inner diameter of the cemented sleeves; then, the ports to the annulus were closed off, and the cement was left for 2 days to set up. A cement sample was taken before attempting to open the sleeves, confirming that the compressive strength of the cement had reached 3,000 psi.

The two cementable ball-drop-­actuated sleeves cemented in the casing had two different seat sizes, so balls of two different sizes were used to actuate them individually. Both sleeves opened as expected, within the predicted opening shear pressures. After the cement test, the open/close (OC) function of the test cell was tested. This was achieved by pulling the sleeves closed with a shifting tool on a hydraulic ram multiple times. After this, the test cell was cut in half in order to ensure that no cement had ingressed into critical areas. No cement residue was found in critical areas of the cementable ball-drop-actuated sleeve.

Following a successful land test, the first cementable ball-drop-actuated-sleeve system was installed in the field offshore Norway in July 2011 when 73 cementable ball-drop-actuated sleeves were run to depth and cemented in place as a part of a 5-in. reservoir liner in 6.5‑in. hole (please see the complete paper for details of this installation).

A 2100-m multistage fracture job was planned together with a major operator on the Norwegian continental shelf with the cementable ball-drop-actuated sleeves in a 4-in. reservoir liner in 6.5‑in. hole. The well is a horizontal producer, and the OC functionality enables the operator to close sleeves at a later stage with a shifting tool in order to shut off water or crossflow. The possibility of adjusting the fracture-port size proved useful for the design engineers when designing the fractures for a reservoir with varying layer thickness. The initial option was to run an openhole completion with swell packers for zonal isolation, with a contingency to run the reservoir completion as a cemented liner if required as a result of challenging hole conditions.

Drilling proved to be a challenge, and the decision was made to cement the liner, with a reduction from eight to five fracture stages.

The liner was cemented and the wiper dart bumped after having passed all sleeves without any problems. Cleanout was performed before stimulation. The first ball-drop-actuated-valve stage was opened by running CT with a ball-nose tool and an impact hammer. This was done to provide an injection point for displacement of fracture balls for later stages. During liner run in, total depth had to be set high because of collapsing hole, resulting in incorrect space out of the fracture-sleeve completion. A decision was made not to open Stage 5 because it was too close to the preceding liner shoe. Stages 1 through 4 were fractured successfully with a built-for-purpose fracturing vessel.

The last step in the evolution of these completions was to merge all of the knowledge and features now gained with the development of the cementable ball-drop-actuated sleeve and the original CT-actuated-valve product line. This resulted in two new product functionalities:

  • The cementable CT-actuated valve, a mechanical operated sliding sleeve that can be cemented in place as a part of a cemented liner
  • The cementable ball-drop-actuated multi-OC valve, a ball-drop-actuated sleeve that incorporates shifting profiles at both ends, giving production-management, restimulation, and water-shutoff capabilities for the life of the well

Both have been designed and tested in Stavanger. The cementable ball-drop-actuated multi-OC valve was designed as a full hybrid between the ball-drop-actuated valve and the CT-actuated valve. Initial opening can be performed by ball, and one ball can open multiple sleeves if desired by use of flex seats and fixed seats. After the initial ball drop, the sleeves can be closed and reopened as many times as needed, either with CT or with a wireline tractor. If the ball-seat sizes are smaller than the intervention tool string, the seats must be milled out first. All of these features have been tested on land at various testing facilities in Norway. For a discussion of testing results for both of these new product functionalities, please see the complete paper.
This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 170641, “The Evolution, Optimization, and Experience of Multistage-Fracturing Completions in a North Sea Environment,” by T.R. Koløy, K. Brække, T. Sørheim, and P. Lønning, Trican, prepared for the 2014 SPE Annual Technical Conference and Exhibition, Amsterdam, 27–29 October. The paper has not been peer reviewed.