Tucker Lake SAGD liner design, fabrication and installation evaluation

Sand-control device prevents production of plugging, erosive sands while allowing smooth production without excessive pressure drop.

J. J. Forsyth, Husky Energy and B. Fermaniuk, Regent Energy Group Ltd.

In recent years, slotted liners have become the dominant means for providing sand control in SAGD applications due largely to the technological advancements in slotted liner design and, the simple but robust, low cost producability of slotted liners. Broad adoption of this completion technique has intensified the need to develop rigorous design bases for both flow and structural characteristics of intermediate casing, liners and slot sizing of liners. This article discusses the necessary steps the operator has taken to design, fabricate and install the Tucker Lake horizontal sand control slotted liner program. It also discusses the making of a successful slotted liner program, with a discussion of the unique issues that should be addressed for slotted liner success in thermal service.

INTRODUCTION

Located 30 km northwest of Cold Lake, Alberta in Township 64, Ranges 4 and 5, W4M, is the operator’s Tucker Lake Oil Sands project, Fig. 1. Completed on schedule and under budget, first oil was achieved in November 2006 with production expected to peak at 30,000 barrels of bitumen per day.¹

Fig. 1. Husky’s Tucker Lake project in relation to its other locations.

Fig. 2. Slot geometry. Blue area represents liner strut. Red area represents liner sections.

Fig. 3. Stress/strain graph: Post-yielding stiffness.

The operator’s experience with thermal oil recovery methods started at the Bolney/Celtic heavy oil fields in Saskatchewan. They are being used at both the Tucker Lake and Sunrise holdings. Tucker Lake, the operator’s first oil sands project, takes advantage of existing pipelines, and infrastructure in the Cold Lake area and Upgrader at Lloydminster. The operator’s 100% working interest at Tucker Lake is estimated to contain a discovered resource of 1.27 billion barrels with a project recovery estimation of approximately 352 million barrels of oil over a 35-year project life. In total, the operator has an interest in 510,890 acres, which are estimated to contain a discovered resource of 40.9 billion barrels.

SAGD, a thermal in situ recovery process using horizontal well pairs of which a horizontal well, equipped as the bitumen producer, is located near the bottom of the reservoir. Steam is injected into a second horizontal well placed approximately five meters above and running parallel to the producer. Steam is continuously injected into the reservoir through the upper well forming a steam chamber, heating the bitumen, thus lowering its viscosity and enabling it to flow.

The bitumen and condensed steam drain to the lower horizontal well, under the influence of gravity. These fluids are produced through the wellbore to the surface. As steam injection commences, the steam chamber grows larger as a result of heat and mass transfer. When the steam chamber has reached the desired size, determined by reservoir modeling, the production of the bitumen/water mixture begins. Natural gas or pumps may be used in assisting the produced fluid in the wellbore to the surface. The horizontal well pairs are typically 700 to 800 m in length with a subsurface lateral spacing of approximately 100 m.

CLEARWATER FORMATION-INTERSPERSED/SURFACE SHALES

Due to the overwhelming costs associated with SAGD exploitation technology, thorough design considerations are critical to the success of the Tucker Lake SAGD project. The use of the SAGD recovery process provides a number of advantages over other technologies. Although SAGD has a slower recovery rate, the ultimate production rate is greater, allowing for fewer wells to be drilled. The SAGD recovery method is both a continuous and a low-pressure thermal process, which generates much less severe tubular loading conditions than older thermal techniques such as Cyclic Steam Stimulation (CSS). Capturing a large percentage of over 1.2 billion barrels of oil from Tucker Lake will rely on pertinent knowledge of the Clearwater reservoir. The pay zone in the Clearwater Formation consists of a layer of sand 30-50 m thick is ideally suited for SAGD. The operator has identified the SAGD process as being ideal for the Tucker Lake project and Clearwater formation due most in part to the reservoir’s relatively consistent homogeneity. The Clearwater has often been described as “a large sand box” and therefore has little heat damming characteristics. The homogenous nature of the Clearwater Formation is ideal for the required heat and mass transfer necessary for the success of SAGD. From the pads, wells are directionally drilled to approximately 480-m TVD, then they are directionally drilled horizontally, penetrating into the oil bearing sand for approximately 700 m.

SAND CONTROL AND SLOTTED LINER DESIGN

Heavy oil operations in unconsolidated clastic formations can be plagued with sand production problems, therefore it is important to make well informed selections of sand control liners. Sand production must be avoided, since it can be detrimental to the SAGD surface processing equipment. To eliminate sand invasion, the operator required a sand control device that would prevent production of erosive sand, reduced pore throat plugging and slot plugging, yet allowing smooth production of reservoir fluids without introducing excessive pressure drop. The operator selected slotted liners as the major sand control application for both the producers and injectors.

Often sand control can be achieved, but at a high price with respect to liner permeability, through the selection of too small of an aperture opening or slot width.¹ This usually leads to a higher than necessary pressure drop, lost production and/or costly wellbore intervention. Therefore, proper identification and selection of the slot specifications such as slot width, slot lengths and slot density, as well as liner weight and material slotting properties, will improve the probability of success. Proper slot specifications permit continuous fluids production, continuous non-detrimental fines production and wellbore performance, while ultimately maintaining optimum sand control of potentially damaging Clearwater sands.

Slotted liners have become the dominant sand control means for SAGD applications due largely to the technological design advancements and low-cost producability. Broad adoption of this simple completion technique has intensified the need to develop rigorous design bases for both flow and structural characteristics of slotted liners and for proper slot sizing in correlation to the geological properties.

Regent Energy Group Ltd. of Nisku, Alberta was contracted to provide all slotting and seaming for the operator’s Tucker Lake project. Custom design of slotted liner for sand or particle control while optimizing fluid flow, through maximization of open area and the manufacturing of tight slot openings was the service company’s main engineering focus for the Tucker Lake project. Proper knowledge of the reservoir parameters such as unconfined compressive strength, fluid viscosity, particle size distribution, etc., provide a necessary basis of information for preliminarily evaluation of the optimal sand control slotted liner design. The final criterion for selecting slot -width specifications is fluid-flow testing of representative reservoir sand.

For the Tucker Lake project, representative reservoir sand was flow tested using selected slotted liner design specifications provided by the service company. All fluid flow testing of representative sand for the operator’s Tucker Lake project was conducted at Hycal Energy Research Laboratories Inc. of Calgary, Alberta. The method of extensively testing the Tucker Lake representative reservoir sand was to pack the sand into a test cell above a single slot, in a cold flow loop using water, water/oil and water/oil/gas mixtures.1

This is to determine the optimum sand particle control width or specification. Initial concerns regarding high pressure drops associated with low open-area slotted liners proved to be irrelevant through research/modeling, testing and the operator and the service company’s previous experience.2 When comparing grain size analysis conducted by Hycal for the A, B and C pad, and sieve analysis from wells in section 28 - 064 - 04 W4, the results can be summarized as follows:

• Particle size from well 00/15-28 is coarser (average 145-135 m SS)
• Particle size from well 02/14-28 is coarser (average147-138 m SS)
• Particle size from well AB/05-28 is finer (average 150-135 m SS)
• Particle size from well 00/14-28 is finer (140 m SS).

Based on Hycal’s slot flow testing program and utilizing the above sieve analysis, the following liner specifications for the Tucker Lake SAGD wells were determined:

1. Injector wells: All slotted liner: slotted at 0.020 in. and seamed to 0.016 with 525 slots per meter (40 slots per column)

2. Producer wells: 2 wells using stainless steel wire wrapped liner: 0.014 in.

3. Producer wells: 26 wells using slotted liner: slotted at 0.020 in. and seamed to 0.014 in. with 682 slots per meter (52 slots per column)

4. Producer wells: 12 wells using slotted liner: slotted at 0.020 in. and seamed to 0.018-in. with 682 slots per meter (52 slots per column).

Typical slotted liner density is designed based on the installation loading limits, thermal loading limits and the optimum radial flow calculated from the reservoir. With technologically advanced slotting and seaming production equipment, the service company provided tight tolerances on the slot specifications; essential for utilization in the Clearwater reservoir. The equipment used to slot the Tucker Lake liner was manufactured by plunging and cutting radially over the entire length of the pipe, on 6-in. centers with the second column being a 3-in. staggered offset. A total count of approximately 6,000-8,000 slots was produced per pipe (dependant on configuration). The staggered longitudinal slot orientation provides an engineered factor of pipe and torsional strength necessary for installing the liner to TD.

After completing the mechanical slotting process, it was necessary to remove burrs and/or “wickers” caused by the mechanical nature of the slot cutting process. The service company’s patented process called Hot Pig remediates this challenge of removing the burrs and/or “wickers.” This process was used for the operator’s Tucker Lake slotted liner, thereby reducing the slot plugging potential along with reducing the pressure drop associated with a “wicker” plugged slot as experienced with mechanical cleaning processes.

The last value-added process to the Tucker Lake slotted liner was seaming. The service company’s patented Transverse Rolled Slotted Liner (circumferential seaming) process is a localized plastic deformation process, which plastically deforms each edge of the slot into a taper, thereby reducing the aperture of the straight cut slot width on the OD of the pipe. This process reduces the likelihood of plugging within the wall of the liner, important for SAGD applications. Standard quality control and quality assurance programs were conducted for all of the operator’s Tucker Lake liner.

INSTALLATION LIMITS

Two major considerations for a SAGD thermal liner design are installation and thermal issues. Although thermal loading considerations and thermal structural performance are extremely important for operational consideration at temperatures exceeding 200° C, thermal structural performance first depends on successful liner installation, where permanent liner deformation does not occur.

Torque capabilities-pipe body considerations and slot deformation. Torque and drag analysis is one of the most effective ways to evaluate the loads that the liner will experience during installation. The torque and drag models run for the operator’s Tucker Lake project were performed using Landmark software. The Landmark software requires several input parameters, with some of the most significant variables being liner size, liner weight, well trajectory and the formation-liner friction factor. By inputting these variables, the Torque and Drag software/model provides a maximum expected torque value that the liner body will experience during installation. Slotted liner structural loading limits can be considerably lower than unslotted pipe of the same material properties.3 These predicted values are quite accurate for a complete pipe body; however, for a slotted liner, extra modeling must be preformed.

The Torque and Drag evaluation does not factor slotted liner slot geometry, therefore, slotted liner specifications are further evaluated using an analytical model, which has been calibrated using physical testing and detailed Finite Element Analysis (FEA) results. This aids in the evaluation of the installation loading limits to resultant slot deformation. The design basis for installation load limits is in maintaining global stresses to a value below the elastic limit (with a safety factor). This also aids in predicting the installation torque limits that the liner struts and liner sections can tolerate without exceeding the plastic slot deformation for a given configuration (sand control slot specification must be predicted to retain its manufacturing tolerance during installation).

FEA inputs such as liner size, (assigned by initial well design), liner weight, slot dimensions (determined from the sieve analysis), allowable slot deformation and the maximum expected installation torque, determined by the torque and drag analysis, can be used to calculate the maximum slot density of the liner. This calculation must be performed through an iterative process. For example, to increase the slot density value (i.e., open area/reduce flow convergence), while maintaining the allowable slot deformation criteria, it may be logical to increase the pipe weight. This however will give rise to new slot geometry values for slot dimensions (due to the slot manufacturing process) and would therefore require new installation torque modeling and evaluation.

Once a compromise of all the liner design variables can be reached, such as operational thermal loading considerations, flow/open area requirements, installation loading conditions, original liner design and cost, a suitable liner design and configuration is achieved. It is important to note that both installation and thermal service considerations were fundamental in reaching a final design for Tucker Lake’s liner.

From the Torque and Drag modeling, required open area, experience, and FEA theory the following specs were determined for the slotted liners at the operator’s Tucker Lake Thermal Project:

• Size: 8 5/8 in.
• Weight: 32 lb/ft
• Grade: K-55
• Slots: injectors: 0.020 in. seamed to 0.016 in., 525 slots/m
• Producers: 0.020 in./0.018-in. and 0.020-in./0.014-in. 682 slots/m
• Allowable slot spec variance: +0.001 in. and -0.002 in.

Torque capabilities-liner connection. When deciding on a thermal liner connection, several specifications must be considered, the most important being maximum operational torque value (maximum installation torque), connection efficiency and connection dimensions. Due to the maximum predicted torque of 11,000 ft-lbs, a connection was chosen to be able to handle this torque maximum during installation. It is also important to ensure that the connection is not below the maximum allowable torque of the slotted pipe body.

As a general rule, the torque design limit should be reached by the slotted pipe body and not the connection; hence, the slotted liner-pipe body, torque-installation load limit and resultant slotting specification can be calculated. A connection with 100% tensile and compressive efficiency or greater is also of importance. By having a connection with 100% efficiency or greater, the connection is as strong as the pipe body and, hence, the connection is not the weakest point in the liner. This is especially important during thermal operation of the well when strain localization (commonly referred to as shear bands-a precursor to catastrophic failure) occurs.

Dimensions are also important as the connection OD must not be too large; as this could greatly affect the formation-to-pipe friction loads (drag) on the liner during installation. Ideally, a flush connection would be the best choice from a “dimensions point of view” (i.e., the OD of the connection is equal to the pipe body). However, flush connections have efficiencies much less than a 100%.

In some cases connection manufactures have made special adjustments to certain dimensional components, such as Tenaris’s special clearance (reduced coupling OD) and special bevel, designed to reduce the installation drag of the coupling. Often the question of connection sealablity arises, however this is not a feature of a SAGD liner connection, since the liner string is filled with slots. Therefore, there is little concern for the connection not maintaining a complete seal as is the case for the intermediate build section casing.

After taking all the design features into consideration, the decision was made to use the Tenaris SAGD Blue Special Bevel (previously Tenaris ER) connection. This connection meets all installation requirements and provides a 100% tensile and compressive efficiency.

Horizontal-section hole size vs. liner size. The hole size is an important decision, as too small a hole increases torque and drag, making liner installation difficult. In the case of drilling too big a hole, running through the intermediate casing with the BHA can become difficult, not to mention the need for costly special drift bits.

After running the Torque and Drag models and using past experience with hole size vs. liner size, the decision was made to drill a 10 5/8-in. hole for the 8 5/8-in. liner. Moreover, the horizontal section was drilled 20 m passed liner TD, which allows sufficient area for the liner to expand when in thermal operation. No installation issues were encountered while running the liners at Tucker Lake. All liners were successfully run to TD without major installation issues, which qualified the slotted liner specification based on the installation load modeling limits.

THERMAL LINER DESIGN CONSIDERATIONS

Liners for wells produced using SAGD must balance sand control requirements and inflow characteristics against the structural demands of thermally induced loads.4 Recently, much consideration has gone into the operator’s thermal casing and liner design with Tucker Lake being one of the main focus areas. Focus has been on intermediate casing and liners as they are the sections of pipe most affected by thermal loading due to being frictionally constrained. The internal tubing strings are also in direct contact with the steam, but aren’t loaded to the same degree due to the fact that they are (more) free to expand than the cemented intermediate casing or the reservoir contacted liner.

Thermal loads and material selection. The design basis of the slotted liner has been derived from two major loading paths. The first being the installation loads and the second being the thermal loads that the liner will experience during operation (i.e., thermal expansion/contraction).

In any SAGD operation, installation and thermal service are important design basis considerations. Depending on whether it is an injector or producer well, the thermal loads can vary, which include loads experienced during either startup or production operations. This creates liner growth due to the coefficient-of-expansion and the subsequent contraction, while the wellbore cools down during the production phase or while servicing the well. The operator modeled the thermal loads by combining engineering techniques such as Finite Element Analysis (FEA), physical testing and empirical evidence. FEA modeling provides material property response for various liner configurations. These techniques were also performed for other oil sands operations with the operator, with several inferences being made toward the Tucker Lake liner design basis.

It was identified that the thermal operating conditions at Tucker Lake would yield the liner pipe body; therefore, the “post-yielding” properties of the liner material had to be carefully considered. Extensive FEA modeling was performed by Noetic Engineering Inc., of Edmonton, Alberta for the operator’s Sunrise project reaffirming that the liner pipe body would yield under the thermal loading conditions. The expected loading during thermal operation for the candidate liner configurations posed a risk to structural integrity of the liner, and more specifically sand control.³ This led to the careful selection of string material with adequate post-yielding stiffness. Post-yield stiffness, the most important post-yielding property, is the ability of the pipe body to resist further deformation after its yield point has been exceeded. In Fig. 4, the stress-strain graph, post-yield stiffness is the instantaneous slope of the stress-strain curve.

Fig. 4. Ovalization and slot deformation: opening and closing of slots.

Liner ovality and slot deformation. Ovality was another mechanical property requiring careful consideration for liner design at Tucker Lake. Although ovality is a pipe characteristic more often seen in intermediate casing design, it still requires discussion for sand control liners. There are two different types of ovality, manufacturing ovality and deformation ovality.

Manufacturing ovality is the degree of deviation from perfect circularity (i.e., how oval is the pipe due to manufacturing processes?). It is important to ensure that the liner body is within American Petroleum Institute (API) ovality specification (API 5CT). API 5CT ovality specification doesn’t explicitly identify an exact value; instead there is a range of allowable outside diameters (+1% or -0.5% from nominal). However, the closer the pipe body is to circularity the better, especially from a structural integrity and ease of slotting point of view. The probability of liner strain localization is reduced with increased circularity.

Deformation ovality is the measure of deformation from circularity due to the loading conditions experienced by the pipe during thermal operation, Fig. 4. This is particularly important when considering slot deformation. Although FEA modeling has identified installation loading as well as thermal loading limits, FEA modeling has not been able to determine combined loading paths. Combined loading can be defined as the integrity response of two different dynamic or static loads on an object.

In our case it can refer to slots that may have been subjected to large installation loads, which could possibly carry residual torque, tension, or compression, and are then subjected to thermal operational loads. Another example would be the combined loading of constrained thermal expansion (formation-to-pipe friction constraint and external load) and formation stress (external pressure). Although many of the FEA modeling assumes no residual stress/strain buildup, there is a possibility that this could occur, at which point it is imperative to have a liner design robust enough to resist deformation to these combined loads.

The best way to ensure this thermal deformation resistance is to select a favorable combination of material, wall thickness, and slotting configuration. Although favorable material properties, such as post-yielding stiffness, are essential, it is important to remember that liner geometry and structure are also critical with respect to slot deformation resistance.

From the FEA analysis in combination with material tests, it was determined to use the Tenaris K-55 material instead of Tenaris L-80 material, as the K-55 material was identified as having preferential post-yielding properties and hence a greater resistance to slot deformation at high operating temperatures. It is important to note that pipe material selection should be based on the post-yield thermo-mechanical properties and not the grade designation. The operator also specified that the manufactured ovality of the liner body be less than the allowable API standard of 1.5%, thereby reducing the amount of induced deformation potential directly caused by ovality issues during the pipe manufacturing process. By specifying this as a pipe criterion to meet, the operator was effective in eliminating all suspect liner from the selection process thereby increasing all the credible liners deformation resistance to thermal loads during operation.

Liner hangers. An important design feature is the ability to handle installation loads (i.e. torque and compression/tension). It is imperative that the liner hanger be able to handle slightly higher loads than the rest of the liner string, so that a weak point is not introduced into the string.

Placement of the liner is also an important consideration. At Tucker Lake the liner hanger was placed on top of two blank joints, which allowed for approximately 26 m of overlap into the intermediate casing. This overlap allowance was necessary to ensure that the liner hanger would not travel past the intermediate casing section during operation, due to thermal expansion or contraction of the liner and the intermediate casing.

Another important feature of the liner hanger is the liner hanger sealing element. Second to the slots, the liner hanger sealing element is the main protection against sand production. The hanger must be able to withstand large pressures (up to 7 MPa at Tucker Lake) at high temperatures (up to 275°C) while providing a 360° debris barrier. Three liner hangers were used at Tucker Lake, the Hammer Seal hanger from Import Tool Corp. Ltd., the Metal Form Packer (MFP) and the CLP liner hanger from Weatherford.

PERFORMANCE-DRILLING COST AND TIME AND LINER PERFORMANCE.

Construction of the Tucker Lake project began in the fall of 2004 and at its peak used approximately 700 on-site workers from several contractors. Total project capital costs came in approximately 5% below the $500 million budget. The first phase of drilling and completions came in under budget by 4.8% and a month ahead of schedule. Drilling of phase two (C-Pad expansion project) came in under budget by 7% and on schedule.

Sand production from [Encana-Senlac] Phase C is negligible with the rolled-top slot design [seamed slot] and there are no indications of slot plugging in any of the Phase C wellbores.5 To date there have been no sand production issues at the operator’s Tucker Lake project, which implies successful quality slot design as well as success in maintaining a radial seal between the liner hanger and intermediate casing string.

On one basis, the operator has not experienced sand production issues and it could be said that to date the Tucker Lake Liner design has been successful. However, this opens the discussion to Tucker Lake liner optimization. Another basis is due to the operator implementation of a successful sand control design, there may be an opportunity to review slot dimensions based on the installation and thermal loading criterion, which could further minimize flow restrictions. Of course, careful consideration of liner installation loading limits and thermal loading should be re-evaluated for each revised slot design. This will more than likely be a topic that the operator will discuss for future development phases of Tucker Lake.

CONCLUSION

For SAGD wells, a rigorous structural and thermal design basis must be considered for slotted liner and intermediate casing for installation and thermal service loading. Sand control is paramount to the longevity of thermal wells with tight tolerance slotted liners providing prevention of producing erosive sand, reducing pore throat plugging and slot plugging, yet allowing smooth production of reservoir fluids without introducing excessive pressure drops. Therefore, slotted liner slot widths should be manufactured to preclude sand ingress and prevent sand plugging inside the pipe wall of the slot - of sand sizes not retained.

If the slotted liner can be installed to TD without rotation, then there is a great reduction in residual torque during thermal operation. No installation issues were encountered while running the liners at Tucker Lake, therefore all liners were successfully run to TD, which qualifies the operator’s slotted liner evaluation program-which was based on installation load modeling limits. The expected loading limits during thermal operation for the candidate liner configurations pose a risk to structural integrity of the liner, and more specifically sand control.³ This leads to the requirement of carefully selecting post-yield pipe material properties that would encompass adequate post-yielding stiffness. This is required due to combined loading issues of frictional restraint and external stresses during the thermal operation, which causes the pipe material to yield. Therefore, It is important to note that successful pipe material selection should be based on the post-yield thermo-mechanical properties and not the grade designation.³ To date, all due diligence in the design of the Tucker Lake SAGD project has been successful.

ACKNOWLEDGEMENT

The authors thank Dan Dall’Acqua of Noetic Engineering Inc. and Laurie Venning of Regent Energy Group Ltd., for their technical support towards this article. The authors also wish to thank HYCAL Energy Research Laboratories Inc., Tenaris Algoma Tubes, Import Tool Corporation Ltd., Weatherford International Oilfield Services Inc., and the author’s respective companies for supporting this article. This article was modified from WHOC 2008-385 presented at the World Heavy Oil Congress, Edmonton, Alberta, March 10-12, 2008.

LITERATURE CITED

¹ Bennion, B., “Optimizing production from heavy oil and bitumen reservoirs” presented at the HYCAL Technology Seminar Series, January, 2003.
² Kaiser, T. M. V., Wilson, S., L. A. Venning, “Inflow analysis and optimization of slotted liners” SPE 65517, 2000.
³ Dall’Acqua, D., Smith, D. T., Kaiser, T. M. V., “Post-yield thermal design basis for slotted liner” SPE 97777, 2005.
⁴ Slack, M.W., Roggensack, W.D., Wilson, G., Lemieux, R.O., “Thermal-Deformation-Resistant Slotted-Liner Design for Horizontal Wells” SPE 65523, 2000
5 Boyle T.B., Gittins, S.D., Chakrabarty, C., “The Evolution of SAGD Technology at East Senlac” 2002-300, 2002.

THE AUTHORS
	J. J. Forsyth is an E.I.T. at Husky Energy in the Oilsands Drilling Department. Forsyth graduated from the University of Calgary with a degree in chemical engineering. Forsyth played an integral role in the drilling and engineering of several SAGD and CSS projects within Husky, including Sunrise, Tucker Lake and Caribou. Forsyth’s focus is on thermal casing and liner design for Husky’s thermal projects.
	Brent Fermaniuk is the Technical Sales/Product Development Representative for Regent Energy Group Ltd. Brent graduated from the University of Alberta in 2000 with a Bachelor of Science Degree. Fermaniuk has worked at Weyerhaeuser Canada Inc., Celanese Canada Inc., Regent Control Systems Ltd. Fermaniuk established and implemented an optimization strategy for slotted-liner manufacturing efficiencies. His current role envelops slotted-liner slot size selection, sand-control completion products and systems and special projects management. He is a member of CHOA and SPE and is undertaking studies for his petroleum engineering degree.