This paper presents a drilling solution through application of an automated managed-pressure-drilling (MPD) technique proved to identify and react to actual wellbore pressures and detect and control gains and losses while still having the ability to maintain a constant bottomhole pressure (BHP) while drilling through tight windows. The paper demonstrates the successful application of advanced automated MPD technologies on the Dover well close to Fort McMurray, Alberta, Canada.
Introduction
A well in the Dover field had multiple failures in the liner that resulted in excessive sand production, causing the well to be shut in. After reviewing the options of well repair or redrilling the horizontal section to install a new slotted liner, it was determined that redrilling was the best option.
After the well-schematic analysis and in collaboration with the operator, the combination of a proprietary control system and MPD techniques was recommended along with a water-based mud (WBM) weight to drill the well and still be able to maintain the BHP required to overbalance the formation. The capability to detect microinfluxes/-losses while drilling, combined with an automated control system covering the drilling parameters [including the surface backpressure (SBP) and constant BHP], allowed the well to be drilled while maintaining the downhole pressure values as required.
This approach additionally mitigated the challenges with weighted-WBM systems. For this particular well, reservoir temperature had cooled down to 145°C. If the well was not circulated for a period of time during drilling operation, mud temperature would increase, which would lead to the breakdown of the polymers and the loss of suspension of weighting material, effectively reducing the density to values near water density, thus creating an underbalanced situation.
Redrilling a producer well in a shallow, overpressured, hot reservoir was a first for the operator in this area. Typically, during any completion workover, full fluid/mud losses occur, so the potential of these to occur during drilling operations was considered high.
Sequence of Events. The drilling process included the following:
- MPD package was rigged in and pressure tested.
- Well was displaced to a 1070-kg/m3 polymer WBM while setting the bottomhole equivalent circulating density (BH ECD) to 1250 kg/m3.
- In-casing/fingerprinting test was performed.
- Window was milled out of intermediate casing with a weighted brine.
- Brine was displaced out of the well to 1070-kg/m3 polymer WBM while pump rates and choke were adjusted to remain above a BH ECD of 1250 kg/m3.
- Drilling commenced from 256-m measured depth (MD).
- Total depth was reached at 985-m MD.
- Well was displaced to a weighted brine, and pump rates and choke were used to avoid exceeding fracture pressure.
- Drillstring was pulled out of the hole, and the slotted liner was run.
By managing the BHP, MPD restricted any steam influxes from injector wells or potential commingling from entering into the wellbore while drilling by applying SBP, which also compensated for the annular frictional pressure loss during pumpoff, maintaining a constant BHP on connections also. Formation pore pressure was estimated to be approximately 1371 kPa, and fracture pressure was estimated at approximately 2380 kPa on the basis of offset wells and known pressure information in the area. Because of a tight drilling window of approximately 1000 kPa, there was no room for misinterpretation of volumes, ECDs, or BHPs. Detailed work went into modeling various mud weights, pump rates, and surface friction pressure on the basis of the MPD surface equipment, as well as required SBP while drilling or during connections.
After analyzing the modeled data, the well was drilled maintaining a bottomhole circulating pressure (BHCP) of approximately 1650 kPa with a 1070-kg/m3 drilling-fluid density, which resulted in an approximately 300-kPa overbalance of the estimated formation pressure. Shutting the pumps off at any point resulted in the control system automatically calculating the resulting friction loss and thereby in the chokes closing in order to hold the desired SBP, maintaining the BHCP at all times.
The Challenge
Drilling with a narrow pore-pressure/fracture-pressure window (approximately 1000 kPa) in a mature steam chamber while having adjacent steam injectors and producers within the field operating may create difficulty in controlling the BHP because the adjacent steam chambers may have commingled and there may be communication with adjacent steam-assisted-gravity-drainage (SAGD) wells.
Caprock integrity is always a concern in SAGD wells because compromised caprock can lead to major environmental and safety issues. Major considerations when drilling SAGD wells usually include stress changes caused by erosion of quaternary channels providing stress relief, stress release caused by erosion of modern river valleys, tectonic stresses, and basement movements. In-situ stresses can be affected by the steam-injection pressure, which can cause minifracturing of the formation.
The fluctuations in BHP must be minimized; these can lead to the weakening of caprock strength/wellbore stability. The BHP must be maintained dynamically and statically within the drilling window at all times.
Possible hole-stability issues can be caused by the exposure to high temperatures from steam injection over time. The temperature of the formation increases the temperature of the circulating fluid, which could soften the bitumen, resulting in hole-stability issues, washout, and an increasing cuttings volume.
When the BHP is not controlled precisely, drilling projects can become environmentally unsafe (e.g., the Joslyn Creek thermal oil-sands project in Alberta in 2006, which resulted in a catastrophic release of steam influx).
Additionally, the Dover field is shallow, which requires all wells to be drilled with a 45° slant rig. This would be the first time for the rotating-control device to be installed on top of the blowout preventer and to be required to hold pressure at a 45° angle. Only the pipe and blind rams are rated above temperatures expected to reach surface in the case of a blowout or substantial well influx.
The Solution
MPD control-system equipment was rigged in. To minimize reservoir damage and to control the BHCP within the tight drilling window, the well was displaced to an unweighted WBM system of approximately 1070 kg/m3. While drilling, the hydrostatic pressure from the WBM and the wellhead pressure from the surface equipment resulted in an overbalance of approximately 300 kPa. Drilling this well as slightly overbalanced allowed for minimizing the possibility of formation issues. Because the mud weight statically would result in a lower BHP, SBP was held during connections to ensure that the BHP remained constant at approximately 1650 kPa. The stability in the BHP reduced any pressure fluctuations within the wellbore, thereby minimizing any induced steam influx during static conditions and any possible wellbore-instability issues.
Risks of minifracturing the caprock were eliminated by holding constant BHP, with the system run in an auto-control mode with the ability to detect micro-influx or losses while drilling. A virtual pressure-relief valve built into the system ensured that the pressures were maintained below the expected fracture pressure at all times while drilling. Should the pressure have reached the set high limit within the system, the chokes would have failed to open to minimize any risk to the formation.
Low-volume losses were detected by the mass-balance flowmeter. The total volume of losses calculated was 7.7 m3 of WBM. Because the mass-balance flowmeter measures flow in and flow out accurately (within liters), it is assumed that the volume of fluid lost was to the formation and not on the surface.
The mass flowmeter also measured the density and temperature of the fluid returns. This was valuable real-time information to have while drilling. The temperature data allowed ensuring that all surface equipment was within working specifications. A mud cooler was used to cool the drilling fluid before injection down the hole. To ensure steam was not being induced into the wellbore, the density-out data allowed for a direct comparison with the density-in data. Throughout the well, the density in/out was observed to be relatively constant.
The control system continually calculated and adjusted accurate BH ECD values throughout the drilling profile of the horizontal section of the SAGD well. Constant-BHP operations were automatically adjusted by the system to maintain the desired pressure set point of approximately 1650 kPa. To aid in ensuring constant SBP during connections, an auxiliary pump was used to pump approximately 700 L/min of fluid across the control-system manifold, thereby ensuring that accurate SBP was being held for the duration of the connection. Once the connection was made, the auxiliary pump was staged down as the rig pumps came back on for drilling to continue.
There were no trips made while drilling. Once total depth was reached, the well was displaced back to 1430‑kg/m3 brine. The displacement began by pumping the kill mud down at 1 m3/min. Once the fluid hit the bit, the pump rate was slowed to 0.8 m3/min and, throughout the displacement, BH ECD was maintained at 1250 kg/m3 by use of the hydraulics model of the control system. Once the circulating pressure increased, both chokes were opened and flow was directed to the gut line to reduce friction pressure caused by the surface-line restrictions. The BHCP remained fairly constant (near 1650 kPa) during the entire process. For a description of the MPD equipment used in the project, please see the complete paper.
Conclusions
Drilling SAGD wells with an MPD technique ensures constant-BHP control precisely, with detection and reaction for mud losses or steam influxes. This further helps to maintain constant strength for the formation rocks and avoids formation damage. MPD provides additional value for SAGD wells by increasing the safety of the drilling process and helping to protect the environment by maintaining bottomhole properties as required and as constantly as possible. It allows for formations to be drilled with light mud weight, which results in higher rate of penetration and eliminates static/dynamic chip-hold-down effects.
Most importantly, MPD also minimizes any environmental release of fluid or steam while drilling caused by the use of specialized surface equipment, real-time monitoring and automated control of surface backpressure, or exceeding fracture pressure, which would exceed the caprock integrity and cause a steam and oil release to the surface.
A total of 717 m were drilled for the re-entry production section of this SAGD well in 19.25 bit hours with a 1070-kg/m3 polymer WBM, resulting in an average ROP of 37.25 m/hr.
Well Objectives Achieved. The following objectives were achieved in this project:
- The horizontal section of the well was drilled safely with MPD services in less than 2 days (19.25 bit hours).
- The control system maintained a constant BHP while drilling and while making connections and during MPD ranging. Friction pressure was replaced with SBP when rig pumps were turned off; the well did not become underbalanced.
- Caprock fracture pressure was not reached or exceeded at any time during operations.
- Only small amounts of loss were seen throughout the well.
- There was no more damage to the formation by drilling with the WBM/polymer system through the control system than there would have been if drilling had been conducted conventionally with heavy mud.
- The slotted liner was installed successfully.
- The well has since been put on production and is producing at a higher-than-expected rate.
This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 170152, “The Successful Application of Automated Managed-Pressure Drilling To Protect Caprock Integrity by Narrow-Margin Drilling in SAGD Wells,” by Nadine Osayande, Elvin Mammadov, and Sheldon Sephton, Weatherford Canada, and Vincent Boucher, Suncor Energy, prepared for the 2014 SPE Heavy Oil Conference—Canada, Calgary, 10–12 June. The paper has not been peer reviewed.