Three drillstring fatigue failures occurred while drilling two deep wells below 16,500 ft true-vertical depth (TVD) in the U.S.A. midcontinent region. All the failures occurred across 2°/100 ft to 3°/100 ft dogleg severity intervals from 6,000 ft to 8,000 ft. The well conditions (i.e., pipe condition and directional plan) were not significantly different from other deep wells in the area, which had not failed. A deep-well drillstring failure study was conducted which included a review of drillstring inspection reports, daily drilling reports, digital data, technical literature, and engineering analysis for the two wells. A Cumulative Fatigue Analysis (CFA) modeling technique taking into account specific well conditions (i.e., wellbore geometry, rotary speed, rate of penetration, hook load, and drillstring configuration) was applied. The model indicated that drillstring failures would occur across shallow doglegs mainly because of high hang-down loads combined with slow rate of penetration (ROP). The results of the study led to the development of new deep-well design criteria and implementation of new drilling guidelines. The new guidelines included utilization of look-ahead CFA modeling when approaching drillstring endurance limits to minimize drillpipe fatigue failures. Look-ahead CFA modeling and the new drilling guidelines were used on two subsequent deep wells in the area, leading to successful drilling to total depth (TD) of 18,000 ft TVD without failure. One of the wells had a 1.4°/100 ft dogleg severity (calculated based on 100 ft survey spacing) at 1,500 ft and drillpipe shuffling was required to prevent drillstring failure in the deep hole section. The drillstring fatigue failure prevention guidelines apply to deep wells drilled worldwide. Introduction BP America Inc. experienced three high-load cyclic fatigue tube failures while drilling in the U.S.A. midcontinent region's deep Anadarko basin. All the failures occurred within a four-month period, on two wells drilled by different rigs. The three fatigue failures occurred in the highest shallow dogleg interval of the wells, and all three failures were in the inclination-angle dropping section of the well. In Well A, poor mud properties caused the hole to pack off, leading to a heat-related tensile failure of a crossover in the bottomhole assembly (BHA). An unsuccessful fishing job for the remaining BHA led to a sidetrack. The sidetrack around the fish created shallow doglegs and led to the two drillpipe fatigue failures in Well A. Shallow directional walk problems combined with a small directional target created shallow doglegs in Well B. The well scope was also changed at TD to drill 375 ft deeper. Mud circulation was lost, and the drillstring was rotated without circulation while building mud volume. Well B subsequently experienced a fatigue failure while pulling out of the hole at TD of 16,628 ft. After these failures occurred, a comprehensive study was performed to understand why these two wells experienced drillpipe failures while other area wells with similar conditions did not experience a failure. The study included a review of drillpipe inspection reports, daily drilling reports, digital drilling recorder data, and engineering analysis. Drillpipe Failures in Well A Well A had been drilled to 11,409 ft TD. The hole was losing mud returns and starting to pack off because of hole instability. The drillstring was backreamed (rotating the drillstring while pulling out of the hole) with 325,000 lbf pickup weight to pull out of the hole. The hole finally partially packed off (circulation returns were lost) with very limited drillstring movement up or down at the 9?-in. bit depth of 10,530 ft. The drillstring was rotated for 1.5 hours at 10,530 ft (without circulation) before the failure.

A real-time, bi-directional, drill string telemetry network has been proven reliable over the course of seventeen field trial wells, including seven wells within the U.S. The network has demonstrated simultaneous upward and downward data rates of up to 57,000 bits per second, with reliability comparable to current mud pulse telemetry technology. The network utilizes unique inductive coupling coils and armored coaxial data cables embedded within premium double-shouldered drilling tubulars to provide high bandwidth telemetry without impacting drilling operations. U.S. field trials have included multiple vertical and directional gas wells drilled to depths exceeding 14,000 feet in the Arkoma region of southeastern Oklahoma. The drilling environment has involved extremely harsh vibrational conditions, including air hammer drilling and multiple jarring events. During these trials, a leading oilfield service company has deployed a number of different downhole measurement-while -drilling (MWD) tools interfaced directly to the drill string telemetry network. This interface allows real-time surface control and interrogation of the downhole tools and transmission of high-density, low-latency drilling dynamics, formation evaluation and directional MWD data at previously impossible speeds. In a recent well, the operator elected to eliminate their usual mud pulse transmission tool, utilizing the telemetry drill string network and compatible MWD tools as the primary downhole data source. This paper builds on prior publications and provides details of the latest field trials, including, for the first time, information on the development and performance of network enabled MWD tools. A summary of the mechanical and electrical design considerations associated with tool conversion is offered. Observations regarding mud-pulse and drill string telemetry performance, including operational differences, rig time impact, added value and deployment issues are provided. Finally, this paper provides details, and value propositions, of downhole measurement and drilling applications that are enabled by the availability of a reliable telemetry drill string network. Introduction Seven years of engineering and development, funded in part by the U.S. Department of Energy, has produced the IntelliServ® network, a high-speed, bi-directional drill string telemetry system.[1] This network makes it possible to obtain large volumes of data from downhole tools and other measurement nodes along the drill string instantaneously - greatly expanding the quantity and quality of information available in ‘real-time’. The system's bi-directional architecture allows high-speed transmission of downhole data to the surface and commands from the surface to downhole devices simultaneously. Through a physical and electrical interface to the telemetry drill string, existing MWD/LWD/RSS tools can be made fully compatible with the network, allowing high band-width communication between all connected tools and a surface acquisition/control system. Drill String Network Technology Overview The drill string telemetry network comprises conventional drilling tubulars modified to incorporate a high speed, low loss data cable running the length of each joint. The cable terminates at unique, inductive coils that are installed in the pin nose and corresponding box shoulder of every connection and transmit data across each tool joint interface. Second-generation, double-shoulder connection configurations provide an ideal location for coil placement, with each coil installed in a protective groove in the secondary torque shoulder. Figure 1 illustrates a coil installed in the pin end of a drill pipe joint.

Summary Three drillstring fatigue failures occurred while drilling two deep wells below 16,500 ft true-vertical depth (TVD) in the US midcontinent region. All the failures occurred across 2°/100 ft- to 3°/100 ft-dogleg severity (DLS) intervals from 6,000 to 8,000 ft. The well conditions (i.e., pipe condition and directional plan) were not significantly different from other deep wells in the area, which had not failed. A deep-well drillstring failure study was conducted, which included a review of drillstring-inspection reports, daily drilling reports, digital data, technical literature, and engineering analysis for the two wells. A cumulative fatigue analysis (CFA) modeling technique taking into account specific well conditions [i.e., wellbore geometry, rotary speed, rate of penetration (ROP), hook load, and drillstring configuration] was applied. The model indicated that drillstring failures would occur across shallow doglegs mainly because of high hang-down loads combined with slow ROP. The results of the study led to the development of new deep-well design criteria and implementation of new drilling guidelines. The new guidelines included the use of look-ahead CFA modeling when approaching drillstring endurance limits to minimize drillpipe-fatigue failures. Look-ahead CFA modeling and the new drilling guidelines were used on two subsequent deep wells in the area, leading to successful drilling to total depth (TD) of 18,000 ft TVD without failure. One of the wells had a 1.4°/100-ft DLS (calculated based on 100-ft survey spacing) at 1,500 ft, and drillpipe shuffling was required to prevent drillstring failure in the deep-hole section. The drillstring-fatigue failure prevention guidelines apply to deep wells drilled worldwide. Introduction BP America, Inc. experienced three high-load cyclic fatigue tube failures while drilling in the US midcontinent region's deep Anadarko basin. All of the failures occurred within a four-month period on two wells drilled by different rigs. The three fatigue failures occurred in the highest shallow dogleg interval of the wells, and all three failures were in the inclination-angle dropping section of the well. In Well A, poor mud properties caused the hole to pack off, leading to a heat-related tensile failure of a crossover in the bottomhole assembly (BHA). An unsuccessful fishing job for the remaining BHA led to a sidetrack. The sidetrack around the fish created shallow doglegs and led to the two drillpipe-fatigue failures in Well A. Shallow directional walk problems combined with a small directional target created shallow doglegs in Well B. The well scope was also changed at TD to drill 375 ft deeper. Mud circulation was lost, and the drillstring was rotated without circulation while building mud volume. Well B subsequently experienced a fatigue failure while pulling out of the hole at TD of 16,628 ft. After these failures occurred, a comprehensive study was performed to understand why these two wells experienced drillpipe failures while other area wells with similar conditions did not experience a failure. The study included a review of drillpipe-inspection reports, daily drilling reports, digital drilling recorder data, and engineering analysis.

The Red Oak Field is located in the Arkoma Basin in Southeastern Oklahoma. The Field recently exceeded peak production levels in what was previously deemed a fully developed reservoir, through an ongoing successful infill drilling program based on the use of three-dimensional (3-D) seismic. Moreover, the redevelopment program has surpassed the 100 well mark and has delivered some high rate gas wells. The field may be characterized by its dry gas and multiple pay horizons in what has long been known to be "crooked-hole country". In situ compressive strengths range from 10,000-psia through the Pennsylvanian to 55,000-psia through the Ordovician. Much work has been done in the past to optimize air drilling operations for shallow wells; however, as deeper horizons are exploited new technologies have been implemented in order to deliver continuous improvement. Drilling improvements in recent years have included the introduction of a state-of-the-art drilling rig, further optimization of air drilling operations and the introduction of Polycrystalline Diamond Compact (PDC) bits. A down-hole vibration mitigation effort was also initiated which yielded improved bit runs and the identification of micro-tortuosity and weight transfer issues. Vibration mitigation resulted in the redesign of bottom hole assemblies (BHA) and the optimization of bent-housing steerable-motor angle settings. Rig rates have increased over time, along with the need for designer wells. Rotary Steerable Systems (RSS) were therefore implemented to address these issues. Initial attempts indicated that RSS were not able to overcome the formation tendencies in 3-D space. A detailed investigation resulted in the hypothesis whereby speeding up the RSS with a motor would provide enough side force with a push-the-bit system to overcome formation tendencies. Real-time Mechanical Specific Energy (MSE) measurements were used in conjunction with the implementation of a Powered Rotary Steerable Systems (PRSS) and revealed the fact that the full benefit of the RSS was not consistently realized (mostly due to the extreme nature of the drilling environment). As such, prototype extended-gauge PDC bits were designed in order to further reduce down-hole vibration, improve well bore quality and bit performance. The result has been sustained top quartile performance, a state drilling record and the continued growth of a mature field. Introduction Flournoy1 elaborated on the benefit of air hammer drilling in the shallow surface hole sections of wells in the Arkoma Basin. At the time of his publication optimization efforts resulted in rate of penetration (ROP) of 100-ft./hour versus the average of 25-ft./hour. Since then ROP's in excess of 200-ft./hour are not uncommon for air hammer systems. Although the surface hole sections are air drilled to this day, the primary reason is not penetration rate but rather the likelihood of mud losses with conventional mud systems. It should be noted that in recent years PDC bits run with mud systems just below surface casing routinely deliver similar performance. However, drilling with a mud system in the shallow surface section of the well has a distinct disadvantage in that severe losses of mud are likely. Unfortunately, gyroscopic surveys from surface hole sections which have been air drilled with hammers at high penetration rates have shown excessive dog legs and inclinations as high as 7-degrees in some wells. The result has at times been unpredictable drift of the well path which may not be an issue for shallow Red Oak wells but is certainly an issue for wells drilling to deeper depths. Most of the wells drilled within the last two years in the Red Oak Field required drilling from less than ideal locations atop mountains or drilling multiple target wells with tails below typical pay horizons. Improvements in seismic and geologic interpretation by the Sub-surface Technical Team resulted in designer wells (see Figure 1) with tight tolerances. The thrusted, faulted and folded nature of the formations in the basin made directional operations onerous and costly. Directional constraints coupled with the fact that air hammer drilling operations for the surface section of the wells provides very little directional control often results in directional operations beginning from an even less-desirable location. As a result ROP's in the air drilled surface section are often constrained in order to alleviate potential shallow dog legs for deeper wells.

A jackup-drilling rig drilled seven exploration wells in the U.K. and Norwegian sectors of the North Sea during a two-year drilling program. By using sublet well commitments and assignments, the rig was secured in a very tight rig market. The drilling rig was shared between two Amoco (now a part of BP) business units for four wells and another major operator for three wells. The application of rig and personnel sharing produced over 10% (U.S. $12 million) cost savings to Amoco through shared lessons learned, continuous improvement, and reduced permitting time. An additional savings of over 10% (U.S. $12 million) was achieved through detailed planning and risk assessment that significantly reduced unscheduled-event (trouble) time. Key factors for success of the program include continuous use of the same rig, crews, and offshore team members; the vision and support of management; adequate well design and pre-planning time; and the commitment of the right people with the right talents to the projects. The same Amoco wellsite supervisors were utilized throughout the program (even as the rig moved from operator to operator). This paper summarizes how the sharing of a rig, people, and best practices can improve exploration drilling performance, even across different teams, business units, operators, and countries. Examples of sharing knowledge between project teams are presented to quantify the value of a long-term drilling rig contract for one-off exploration wells. The paper concludes by providing lessons learned from rig-sharing agreements and future applications of rig sharing. Introduction Amoco Norway signed a long-term rig contract for a harsh-environment, heavy-duty jackup-drilling rig (Transocean Nordic) in February 1996. Prior to commencing the contract term in October 1996, the rig was mobilized from the U.K. sector to Rotterdam, The Netherlands for modifications (pipe-handling equipment installation and other upgrades) in order to comply with Norwegian regulations. Amoco Norway drilled the second and fifth wells in the program, including a challenging high-pressure, high-temperature (HPHT) well and a 58 (geological sidetrack. Amoco U.K. drilled the third and sixth wells in the program (both HPHT). Another major operator (Eni-Agip) drilled the first, fourth, and seventh wells. Neither Amoco nor Eni-Agip had any financial interest in the other company's respective exploration project areas. The rig was released in August 1998. When the rig was contracted, there was a tight market for rigs, especially HPHT, 300-ft-water depth, harsh-environment rated jackup rigs capable of working in the Norwegian sector. Although the contracted rig met the minimum project requirements (after the pre-contract upgrades), it was less than optimal in regards to spud-can design, variable load, derrick load, living-quarters capacity, pipe-handling capability, and water-depth rating. In general, the rig was older and smaller than most competing (and unavailable) rigs in the class. Availability of the most versatile rig possible to meet the project timing requirements was the key driver in rig selection. In a tight market, it is often necessary for an operator to secure rig slots by executing rig contracts well in advance of the anticipated contract commencement (eight months in this instance). Amoco obtained sublet well commitments from several operators prior to signing the contract. The rig contract contained an assignment clause, which allowed Amoco to assign the rig to other operators. The rig contract, sublet well commitments, extension options, assignment rights, and rig-sharing agreements allowed Amoco to optimize rig utilization during tight market conditions. Amoco was able to keep the rig working throughout the contract with only minimal standby time.