Batch drilling in the North Sea expedites extended-reach drilling

Brian G. Teggart, Charles L. Dunne, Richard D. Wilkinson
British Gas Exploration & Production Ltd.

Batch setting of surface casing strings produced substantial efficiencies in extended reach drilling for British Gas Exploration & Production in the Armada field by reducing down time.

In addition, effective hole cleaning techniques and reductions of torque and drag effects substantially improved drilling efficiencies.

Lessons learned on the project will be beneficial to future operations.

Armada complex

The Armada complex is located approximately 250 km northeast of Aberdeen in the North Sea.

The complex extends into five separate exploration areas and produces from three reservoirs: Fleming, Hawkins, and Drake (Fig. 1 [44757 bytes]).

The Paleocene Fleming field is approximately 24 km long by 3 km wide and is the largest reservoir.

The Jurassic Drake and Hawkins reservoirs are smaller accumulations lying directly below the Fleming field.

All three fields together have estimated reserves of 1.2 tcf of gas and 70 million bbl of condensate. The planned maximum depletion rate is 450 MMscfd.

The Armada complex is expected to be operational for 20-25 years.

Development plan

The original development concept involved drainage of the fields using two platforms, primarily due to the length of the Fleming field.

Unfortunately, the economics did not justify this option and it was finally determined that a single, centrally positioned platform, using extended-reach drilling, was the most feasible option.

Drilling targets

The predrilling program included targets in all three reservoirs (Fig. 1).

Five targets were identified in the Paleocene Maureen formation (Fleming field), two targets in the late Jurassic Ula formation (Drake field), and one target in the late Jurassic Ula formation (Hawkins field).

Composite stratigraphic columns and casing schematics are illustrated in Fig. 2 [39336 bytes] and Fig. 3 [41632 bytes].

Three of the Armada wells are classified as extended-reach wells including the A4, A5, and A8 (Table 1 [28117 bytes]).

The industry definition for an extended reach well is when measured depth (MD) exceeds true vertical depth (TVD) by a factor of two.

Well A8 may have achieved a semisubmersible drilling record, reaching a total of 23,905 ft MD at a TVD of 9,147 ft.

Borehole stability

Obtaining a pore-pressure profile is critical for well-planning purposes, particularly for extended-reach, high-angle wells where borehole stability is a concern.

Derivation of the pore-pressure profile was based on two areas of study:

1. Examination of existing offset data, final well reports, mudloggers' reports, and repeat formation test (RFT) data.

2. Petrophysical pore-pressure and fracture-gradient research using existing data.

Borehole stability factors were predicted with a mechanical stability program using data from a previous study.

Unfortunately, application was difficult because mud densities were unknown.

Most of the input parameters used in the simulations were based on assumptions and published information rather than actual field evidence.

The derived pore-pressure regime exhibited north-south variations across the Armada complex. This information was used in combination with various operating scenarios to define the mechanical design of the casing strings.

Picking the 95/8-in. casing seat was critical in controlling the pore-pressure regression of the Paleocene shales.

In addition, anticipated fluid losses at the base of the Balder formation was a cause for concern.

The Jurassic design incorporated a contingency 75/8-in. drilling liner with a 63/4-in. hole to total depth (TD).

Equipment specification

For wells over 20,000 ft, the rig capability must satisfy minimum specifications for torque, drag, hydraulics, and logistical requirements.

There were eight major equipment requirements:

1. A topdrive drilling system with a minimum of 45,000 ft-lb of continuous torque

2. A 65/8-in. pipe handler

3. Three mud pumps

4. A derrick board capable of handling 65/8-in. and 51/2-in. drillstrings

5. Sufficient storage for casing, drill pipe, mud, and bulk

6. A high-performance solids control system for use with synthetic mud

7. Four linear-motion shakers

8. Two high-volume centrifuges.

Equipment on this drilling unit dictated the equipment requirements for the template, subsea well head system, and subsea release equipment.

Drillstring design

Drillstring selection depends on a combination of anticipated torque levels and the required circulation rates needed for effective hole cleaning.

Based on Santa Fe 135's racking capacity and a limitation to 5,000 psi of pump pressure, a 65/8-in. drillstring was selected for the 16-in. hole section.

Combination strings of 51/2 in. and 65/8 in., or 5 in. and 65/8-in. drill pipe were selected for the 121/4-in. section.

For the 81/2-in. hole section, 5-in. drill pipe was chosen.

Casing wear was a serious concern. As a result, all drill pipe required hard banding.

In addition, plans were made for the use of nonrotating drill pipe subs spaced every 60 ft across the build section. This would reduce casing wear as well as reduce torque levels.

Planning and inspecting drillstring components used the DS-1 standard as a basis.2

Drilling fluid

Based on requirements for rheology, chemical inhibition, lubricity, equipment capacities, and problems associated with the highly reactive Tertiary claystones, a synthetic mud system was selected for drilling the Armada wells.

Cementation

Cementation guidelines depended on the semisubmersible's cementing equipment.

Before cementing 133/8-in. casing, a water-based mud scavenger was pumped ahead of the cement in order to recover the annular volume. This reduced the amount of synthetic mud left down hole.

For wells with casing shoes deeper than 20,000 ft MD, a 95/8-in. liner was hung inside the 133/8-in. casing. This was due to impracticalities in cementing at such depths.

Subsequently, the 95/8-in. casing strings were tied back to the subsea wellhead either before or after drilling the 81/2-in. hole section.

Inside the 95/8-in. casing, 7-in. liners were hung. The liners included an integral liner top-packer which could be used for rotating to bottom when required.

Torque and drag simulations have demonstrated the capability of rotating long liner strings up to 7,000 ft with suitable premium casing connections.

Directional drilling and surveying

The planning phase focused on nonproductive time due to stuck pipe, down-hole tool failure, and casing wear.

The primary objectives of the directional planning process were to:

• Prevent excessive doglegs in the build section, not to exceed build rates of 2.5°. This would reduce torque, drag, and drill pipe wear and fatigue.

• Minimize tortuosity of the well path. This would also reduce torque, drag, and reduce the possibility of mechanically stuck pipe.3

• Maximize the rate of penetration (ROP).

• Avoid tangential inclinations in the range of 45-55°. This would reduce the possibility of stuck pipe due to mechanical pack off. Hole angles in this range are the most difficult to clean due to buildup of cuttings.

Wireline-conveyed, pump-down gyros were used for wells with inclinations greater than 60°.

Batch setting

The 30-in. and 20-in. casing strings were batch set on all eight Armada wells.

By batch setting, the following benefits were achieved:

• Allowed subsea blow out preventers (BOPs) to be moved from slot to slot at the end of each well, rather than pull the BOP to surface. This saved approximately 1 day per well.

• Improved efficiencies due to task repetition.

• Simplified logistical operations.

A great deal of importance was placed on remotely operated vehicle (ROV) operations. Two ROVs were utilized, a work class Trojan and a surveillance Seaeye.

36-in. hole section

The 36-in. hole section began with the ROV connecting two guidelines to the template miniposts for the required slot.

Next, a 26-in./36-in. hole opener assembly was run into the template receptacle. A utility guide frame (UGF) was clamped around the bottom hole assembly (BHA), allowing the assembly to be easily stabbed into the template receptacle.

A centralized bushing was bolted onto the UGF. In order to maintain hole verticality, a 95/8-in. mud motor was utilized.

Hole verticality in predrilled wells is critical because of the constrained nature of the template receptacles. If the hole deviates more than 0.25°, the casing will not land correctly.

Next, the 36-in. hole section was drilled to 620 ft MD (seabed at 380 ft) before 30-in. casing was run.

The casing was then cemented. A retrievable guide base assembly was used to land the casing in the receptacle.

Problems with unconsolidated gravels and sands were solved by using a 12 ppg viscous displacement mud at total depth (TD). Viscous pills were used for trouble spots.

A new hole opener was required for each slot due to the abrasive nature of the gravel formation.

Gyro single-shots were found to be ineffective on the first slot because of rig heave. MWD surveys were successfully used on subsequent slots.

Inclination was less than 1° for all slots.

Hole and casing

The 26-in. hole assembly incorporated a steerable motor allowing build angles of up to 22°.

The 20-in. casing strings were set at depths up to 2,425 ft MD, 2,385 ft TVD.

Sea water was used as the drilling fluid.

Casing TDs were selected in order to take advantage of the higher leak-off values for deeper formations. Thus, heavier-than-normal mud weights could be used in the 16-in. hole sections.

The only problems experienced during cementing were casing pressure tests on three of the wells. The 20-in. wiper plugs could only take 1,000 psi.

After cementing 20-in. casing, it was necessary to remove all loads from the template by retrieving the guide base.

Next, the retrieving tool was run on drill pipe. The guide base was retrieved without problems on every occasion.

Finally, an ROV placed a light-weight trash lid over the wellhead for protection, and then moved on to re-establish guidelines for the next slot.

Moonpool gyro surveying

Rather than run wireline gyros from the catwalk and through the V-door, survey operations for the 20-in. casing were taken from the moonpool area.

This method saved 36 hr of rig time. Total time for all 20-in. casing surveys was 4 hr.

Surveys in the top section of the 26-in. hole used a wireline fed up through the mousehole.

This avoided resighting the unit at the end of the catwalk and avoided interfering with crane operations.

After the casing surveys were finished, a corrosion cap was installed on the wellhead with the aid of an ROV.

An ROV then inserted a hot-stab line into the corrosion cap so that a corrosion inhibitor could be pumped into the wellhead.

Operations on the slot were completed once the hot-stab line was released.

Because a tugger line was used instead of drill pipe during these operations, no rig time was lost.

Performance

All batch setting operations were completed in 33 days, 18 days ahead of schedule. This was due to:

• The experience curve-Trouble time was reduced from 0.98 days on the first slot, to zero on the last well.

• Mild weather-Total weather downtime was 0.68 day as opposed to the 7.2 days in the drilling plan.

• Using proven time-saving techniques perfected by other operators.

• 16-in. hole section

The BOP was run with a capture funnel. Guidelines were not used. This allowed for easy skidding between well slots.

Two ROVs were oriented at right angles to each other for visual monitoring.

In general, the BOP remained subsea for the duration of two wells, then it was retrieved to surface for inspection and maintenance.

The 16-in. hole section was drilled with a steerable assembly and a tri-cone bit.

For the longer-reach wells with sections extending to 8,000 ft (Wells A4 and A8), a polycrystalline diamond compact (PDC) bit was utilized instead of tri-cone bits due to limited bearing life.

The PDC bits experienced no problems with directional control. However, penetration rates were so high that there were hole cleaning problems.

A torque and drag monitoring program assisted the rig crew in determining when a period of circulation was needed to clean the hole.

By time drilling at 250 ft per hour, the hole was effectively cleaned. The same drill rates were used in the 121/4-in. hole section.

The majority of directional work was carried out in the 16-in. hole section with angle build rates of up 75°.

Minimal steering was required through this section since it was predefined in the 26-in. hole.

Dog leg severity was limited to less than 3.5°. This reduced torque and drag, and minimized drillstring failures.

While drilling, each stand was reamed and back-reamed at 150 rpm to ensure a clean hole.

Flow rates were increased to a maximum of 1,250 gpm once the mud was sufficiently conditioned. The use of 65/8-in. drill pipe allowed these kinds of flow rates.

Hole conditions remained stable with mud weights of 12.0-12.5 ppg. The higher angle wells needing 12.5 ppg to reduce cavings.

The earlier wells in the program tended to suffer from the production of fines within the first 1,500 ft of the section.

However, this problem was reduced by maintaining the mud rheologic properties to one and 11/2 times the hole size and using a setting of 6 rpm on a rotating viscometer.

Once hole conditions improved, it was possible to dispense with wiper trips.

After circulating at TD, the assembly was back-reamed out of the hole in order to run casing.

The 133/8-in. casing jobs experienced no problems, although cement losses were experienced on several wells, probably into the shallow Utsira Sandstone.

For the longer-reach wells, A4 and A8, for which the 95/8-in. liners were to be run, 135/8-in. casing was incorporated into the build section.

This allowed for casing wear resulting from the increased rotational time.

121/4-in. hole

For the majority of the wells, the 121/4-in. hole encompassed a long tangential section. However, a drop-and-turn section was incorporated into one well.

A steerable drilling assembly was used for all wells, and no more than 5% down time was required for orientation purposes. PDC bits were used wherever possible.

The early Paleocene wells, A1 and A3, were terminated within the Balder and Lista claystones.

However, for subsequent wells (A4, A5, and A8), borehole stability problems from the Balder and Lista claystones required that the 95/8-in. casing point be deepened into the base of the Maureen claystone.

Flow rates of up to 1,100 gpm declined to 750 gpm for the longer reach wells due to hydraulic limitations.

High flow rates were achieved with a combination drillstring of 5 in. and 65/8 in. The 5-in. drill pipe was substituted with 51/2 in. when available.

In addition, the use of specially modified bits with large flow-through areas allowed maximum flow rates to be achieved.

Ultrasonic logs confirmed that it was unnecessary to use nonrotating casing protectors to minimize casing wear. Ultrasonic logs confirmed that new-generation hardbanding negated the need for these protectors.

In addition, the combination of synthetic-base mud and hard banding resulted in friction factors of only 0.13 in casing, and 0.16 in open.

The stability studies carried out in the design phase predicted that a maximum mud weight of 13.5 ppg would be needed to control the over-pressured Eocene formation.

However, this mud weight was insufficient, and 14.0-14.5 ppg mud was required to control the formation pressure, depending on hole angle.

Combination casing

Excluding the two longest reach wells, a combination 103/4-in. and 95/8-in. casing string was run on all wells.

A 95/8-in. liner was run in the two longest-reach wells to allow for a more-effective cement job.

The liner was tied back to the subsea wellhead using a combination 103/4-in. and 95/8-in. tieback string.

This operation was successful on the A4 well, but failed on the A8, resulting in the collapse of the 103/4-in. casing.

It was suspected that the tie-back seals were bypassed during pressure testing, trapping the pressure within the annulus.

While drilling the 81/2-in. hole section, the annular pressure built up as the temperature increased, causing the 103/4-in. casing to collapse.

81/2-in. hole

Hole instability in the Balder, Lista, and Maureen formations was caused by a chemical inbalance between the formation and brine-phase salinities of the mud.

An understanding of the pore-pressure below the Balder formation was gained on the first well, the A1, at the expense of a sidetrack.

It was obvious from the borehole collapse that the Balder and Sele formations required a mud weight of at least 13.0 ppg to ensure hole stability.

This led to the deepening of the 95/8-in. casing section described above.

The hole problems remained even after drilling into the Lista formation, until a link between hole instability and the chemcial balance was discovered.

Subsequent to the A4 sidetrack, maintaining the brine-phase salinity over 200,000 mg/l while drilling the Lista and Maureen formations resulted in a stable hole environment.

As an additional safety precaution, the 95/8-in. casing was set deeper in the Paleocene wells to ensure trouble-free operations.

Directional control and monitoring was less of an issue for Paleocene targets. Normally, these targets had already been penetrated and the section could be drilled with a packed rotary assembly.

However, for the Jurassic wells (A2, A6, and A7), a long section of Cretaceous limestone had to be drilled, and a steerable assembly was required.

Well A2 was drilled with a steerable motor and an adjustable stabilizer assembly.

The other wells were drilled without a steerable motor, using a rotary assembly and adjustable stabilizers.

Generally, the formations below the 95/8-in. casing shoe were drillable with a PDC bit.

Unfortunately, in the northwest quadrant of the field, the Lista and Maureen claystone formations were difficult to drill, and severe damage occurred to the PDC and insert bits.

In this quadrant, the only solution was to drill with an insert bit reinforced with additional gauge protection.

A combination of mud weights was used to drill the 81/2-in. hole section.

It may be possible to drill Paleocene wells with mud weights as low as 11.0-11.5 ppg mud as long as the majority of the Lista formation is cased off. And even lower mud weights may be possible, as long as the brine-phase salinity is maintained above 200,000 mg/l.

A lower mud weight would reduce the overbalance and the risk of differential sticking.

Mud weight in the Ekofisk formation was limited to 11.0 ppg due to the risk of differential sticking in the upper chalk.

As in the 121/4-in. hole section, the 81/2-in. hole did not require nonrotating drill pipe to minimize casing wear and reduce torque levels.

Formation evaluation

Logging while drilling (LWD) techniques were used extensively for formation evaluation purposes within the 81/2-in. hole sections.

In the earlier wells, a compensated dual-resistivity and compensated density-neutron LWD combination was used. Later, the compensated density-neutron was replaced with an advanced density-neutron tool.

Where formation pressures and fluid sampling data were required, it was necessary to make use of drill pipe-conveyed logging techniques. Wireline logging was not possible due to gravity constraints.

In the Paleocene reservoir, where a high mud overbalance existed, differential sticking was a real concern.

To reduce the risk of sticking the drillstring, the following techniques were used:

• Use of 77/8-in., low-torque stand-off subs, spaced every 60 ft to reduce contact with the reservoir

• Rotation and movement of drillstring as much as possible

• Placement of two sets of jars in the string, with one set inside the casing at all times

• Introduction of calcium carbonate bridging material into the mud system.

Three wells were successfully cored as shown in Table 2 [7045 bytes].

Coring on Well A1 was abandoned due to a combination of hole instability problems and lithologic cementation.

Initial runs employed a conventional core head and a 90 ft barrel. Subsequent runs used a face-discharge core head and barrel lengths of 180-270 ft.

Coring the Paleocene reservoir was generally more difficult, with the core barrel more prone to jamming.

Fractures were noted in some of the cores, probably from down hole stresses or through handling at the surface.

Liner

The 51/2-in. liner was run with a 7-in. rotating-liner hanger. Rotation was possible for all wells.

The Paleocene wells used relatively short liners for the high angle sections of the hole.

In the beginning, it was difficult, if not impossible to confirm whether these liners were released.

The system was changed to a liner hanger setting that incorporated a hold-down sub. This allowed for over-pull to be taken up against the liner in order to confirm release.

Use of the new system proved to be a success, with the hold-down sub providing a positive indication of release.

Liner cement operations were successful due to a combination of:

• Positioning of a fully centralized 51/2-in. liner within the 81/2-in. hole

• Ability to rotate the liner

• Attention to details on mud removal.

Solids control

The primary solids control system on the Santa Fe 135 consisted of five Thule VSM100 shakers and two Swaco variable speed centrifuges.

For secondary recovery, a Swaco drying screen was used downstream of the primary shakers, fed by a screw conveyor.

This resulted in a closed system. The mud recovered from the Drying Screen was routed back to the active system after centrifuge treatment.

The equipment proved invaluable in minimizing mud losses, especially during initial circulation when cold mud was lost over the primary shakers.

References

1. Kwakwa, K.A., Batchelor, A.S., and Clark, R.C., "Assessment of the Mechanical Stability of High-Angle Wells in Block 22/11, Nelson Field Discovery," SPE Drilling Engineering, March 1991.

2. "Standard DS-1: Drill Stem Design and Inspection," T.H. Hill Associates Inc., December 1992.

3. Banks, S.M., Hogg, T.W., and Thorogood, J.L., "Increasing Extended Reach Capabilities Through Wellbore Profile Optimisation," SPE paper 23850 presented at the 1992 SPE/IADC Drilling Conference, New Orleans, Feb. 18-21, 1992.

The Authors