An extended-reach-drilling case study
Britannia Operating Ltd.
Baroid Product Service Line, Halliburton Energy Services
An extended-reach well was planned and successfully drilled in the North Sea Central Graben with a water-based drilling fluid in an area where wells have histor-ically been drilled with mineral oil-based and ester-based drilling fluids.
Water-based drilling fluids were used because of the high costs of shipping cuttings drilled with inverts to shore for processing and disposal, coupled with the potential shutdown of drilling operations in times of bad weather when cuttings could not be transferred from the rig to supply vessels. These issues are especially important in the large-diameter hole section where high amounts of cuttings are generated.
Many water-based systems have been used previously with little success in the Central Graben. The conditions indicated a need to develop a stable, shale-inhibiting water-based mud system capable of performing well in extended-reach drilling (ERD) applications.
The drilling program for this development well called for the intermediate hole interval to be drilled from the 20-in. casing shoe at ± 3,500 ft MD (± 3,400 ft TVD) to a TD of ± 8,900 ft MD (± 7,600 ft TVD) with a final hole angle of 63°. Additionally, the program included a suitable water-based system for this ERD based on shale inhibition and formation analysis, mud weight optimization, and hole-cleaning efficiency.
Mineral oil-based drilling fluids have traditionally provided the most inhibitive drilling fluid in this area. The challenge was to design a water-based drilling fluid system with performance near that of the mineral oil-based fluids. Shales encountered in the Central Graben area of the North Sea are some of the most chemically reactive found in any drilling area. They contain high percentages of mixed layer illite/smectite clays, which are prone to swelling and dispersion and can lead to diminishing control over mud system rheological properties. Laboratory work on shale samples collected from a nearby well included x-ray analysis, cation exchange capacity, exchangeable cations characterization, and the use of the dielectric constant measurement (DCM) technique.
The DCM technique provides a quantitative determination of rock properties by measuring the specific surface area per unit weight, thus representing the total hydratable surface area of a sample. The DCM is dominated by the presence of smectite and is also strongly influenced by the other hydratable clays. The DCM measurements indicated increases from 250 m2/g initially to a maximum of 580 m2/g at 7,500 ft TVD in the Alba and Lothian formations. The DCM showed where the most reactive shales would be encountered and at what point increased fluid inhibition properties would be required.
Mud weight optimization
Previous wells had trajectories of 82° in the deeper 12.25-in. hole interval. The high angle of attack into problematic shale at 12,800-13,000 ft TVD has been a source of several hole instability occurrences. These instabilities were mainly in the form of hole pack-offs and collapse primarily due to weak rock strength and laminated structure of the shale.
As a consequence, a field wellbore stability analysis resulted in the following conclusions:
- The higher hole angles drilled perpendicular to the orientation of the maximum horizontal stress required higher mud weights compared to those drilled parallel to the maximum horizontal stress
- The higher deviation angles required higher mud weights, which served to increase mud pressure penetration into the weak shale laminations and hasten wellbore instability
- The laminations in the shale limited the angle of attack due to weak bedding plane-related rock strength anisotropy. Consequently, when the well trajectory was drilled parallel to the bedding planes, more hole-collapse problems were observed.
Accordingly, the well trajectory design for the ERD well was changed so that the problem shale encountered in the 12.25-in interval would be drilled at a lower angle. This change required a shallower kick-off depth for the large-diameter interval. A borehole stability analysis was performed for the large diameter interval. Based upon anticipated hole angles with depth, a mud weight schedule with an initial mud weight of 11.5 ppg and a final mud weight of 12.5 ppg was developed. The minimum mud weights necessary to initiate hole collapse, breakdown, and formation fracture pressures were analyzed. Caution was given that any increases in mud weight above 12.5 ppg level should be carefully considered in order to avoid differential sticking across any permeable zones encountered while drilling the section.
Once the mud weight schedule was identified, pre-well planning focused on hole cleaning in the large-diameter interval. The hole-cleaning methodology was developed from earlier steady-state hole cleaning modeling work. Key parameters included mud density, fluid rheological parameters, cuttings size and shape, pump rate, hole geometry and angle, drill pipe eccentricity, drill pipe rotation speed, and rate of penetration (ROP).
The dielectric constant measurements indicated increases from 250 m2/g initially to a maximum of 580 m2/g at 7,500 ft TVD in the Alba and Lothian formations.
The goal of the extensive pre-well simulation process was to determine the ranges of various parameters that would provide good hole cleaning while drilling with water-based mud (WBM) and to keep equivalent circulating density (ECD) below the fracture gradient at the 20-in casing shoe. Key conclusions included:
- While cleaning in a 16-in. diameter hole was not very good, especially at the higher angles planned for the interval, it was better than with a 17.5-in diameter hole. Cleaning problems in a 16-in. interval were expected
- Larger-sized drill pipe would improve cleaning
- Control over the size of cuttings generated while drilling was critical
- A WBM with intermediate viscosity would perform better than a high-viscosity fluid.
With the results of the hole cleaning simulations in hand, predictions of ECD with cuttings in the annulus were made using the following input parameters:
- Hole inside diameter: 16-in.
- Drill pipe outside diameter: 6.625-in.
- Pump output: 800-1,200 US gal/min
- ROP: 50-100 ft/hr
- Average cuttings diameter: 0.25-in.
Hydraulic simulations showed a baseline ECD of 12.13 ppg with zero ROP. With the ROP levels simulated, predicted ECDs rose to 12.35-12.45 ppg. As expected, the higher the ROP, the higher the predicted ECD. These predicted ranges were well within the anticipated formation fracture gradient of 14 ppg plus.
A review of various WBMs to determine the optimum choice for drilling the reactive shales resulted in the selection of a KCl/polymer system. The formulation for the whole mud dilution volume was divided into two sections based on the DCM results. Down to the top of the Alba formation (± 6,500 ft TVD), the system would be formulated with a KCl brine (8.6-14.3 wt%) containing polyglycol (3-4 vol%). With the rapid increase in DCM results below this point, the KCl concentration was to be increased to 14.3-17.2 wt%, with additions of a shale-stabilizing surfactant.
The 16-in. hole was successfully drilled to a TD of 8,843 ft. The 13 3/8-in. casing was run and landed without problems. Total drilling-related time was within the planned limit of 10 days. No significant hole instability problems were observed while drilling.
The mud weight schedule was altered just prior to spud for an initial mud weight of 11 ppg and a minimum mud weight of 11.5 ppg prior to drilling into the Alba formation. To help minimize mud pressure penetration into the shale zones, the mud weights used while drilling were maintained slightly above the predicted minimum level required for shale stability.
The frequent wiper trips and extended circulation accounted for a significant portion of the total time in the interval. Because this was the first attempt to drill a large-diameter interval with a water-based mud, a very conservative approach was taken with respect to the overall drilling operation. Incorporating the lessons learned from drilling this large-diameter interval can reduce this cost on future wells. Several notable drilling achievements were identified:
- Two critical sections were drilled without any sliding problems
- Estimated near-gauge hole with an average hole diameter of 16.3-in.
- Hole conditions and cuttings integrity were excellent throughout the entire interval
- Shaker efficiency was excellent throughout the entire interval
- No over-pull of greater than 50,000 ft-lb on trips
- No wiper trip required at TD prior to running the 13 3/8-in. casing
- No down time to mud-related incidents (bit balling, stabilizer balling, mud rings)
- Maximum ECDs near interval TD of 12.25-12.3 ppg were well below predicted fracture pressures.
Several conclusions can be drawn from this study:
- Thorough pre-well planning is necessary to ensure the success of non-routine drilling projects
- Extensive wellbore stability and hydraulic modeling can be handled together in pre-well planning. They are often complementary, especially in challenging well construction projects
- Predictions from a comprehensive, appropriately designed wellbore stability model can be confidently used to produce mud weight schedules in ERD wells that keep the hole stable
- ECD measurements near the bottom of the 16-in. interval agreed closely with those generated in the pre-well planning process, even when the actual ROP levels, drill pipe rotation speeds, drilling fluid rheology, and other parameters are not exactly known. With a very good hydraulic modeling, ECDs in pre-well planning can be considered better than "ballpark" estimates.
The authors wish to thank Britannia Operating Ltd., ChevronTexaco Petroleum Technology Co., and Halliburton Energy Services Inc. for permission to publish and present this paper.