32 Oilfield Review

Extended-Reach Drilling: Breaking the 10-km Barrier

Geosteering, torque reduction and casing flotation have all contributed

to record-breaking extended-reach drilling achievements. The limits of

directional drilling continue to be pushed back as horizontal reservoir

sections greater than 2500 m are being drilled, cased, cemented and

completed to tap reserves at extreme distances from surface wellsites.

Frank AllenPaul ToomsBP Exploration Operating Co. Ltd.Poole, England

Greg ConranPoole, England

Bill LessoPatrick Van de SlijkeSugar Land, Texas, USA

For help in preparation of this article, thanks to LesleyBrown, Susan Caito and John Minge, BP ExplorationOperating Co. Ltd., Poole, England; John Gammage,Deutag Drilling, Poole, England; Dave Bergt, Jim Jares,Toni Marszalek, Charlie Pratten and David White,Anadrill, Sugar Land, Texas, USA; Mike Williams,Anadrill, Poole, England; and Ty Rivet, Anadrill, New Orleans, Louisiana, USA.ADN (Azimuthal Density Neutron), CDR(Compensated Dual Resistivity) and GeoSteering aremarks of Schlumberger.

■■Wytch farm extended-reach drilling.Logging-while-drilling and GeoSteeringtools provided real-time data for the geolo-gist to adjust the well path and guide thewell along an optimum route through thethin pay zones at Wytch Farm, southernEngland. In long extended-reach wells,geosteering techniques are instrumental inmaintaining a smooth, accurate wellborethrough the reservoir.

Extended-reach drilling technology recentlyachieved a new milestone with the drillingand completion of a 10-km [33,000-ft]stepout well, some 2 km [6600 ft] longerthan the previous world record. The maintechnical hurdles to the success of this wellwere reducing torque and drag, controllingfluid circulation and maintaining directionalcontrol. Solutions were not simple becausemethods to mitigate technical challenges inone area often had adverse consequences inothers. One factor, drillstring rotation,repeatedly surfaced as part of the solution toall the major problems in this extended-reach well.

Guiding a wellbore accurately through thepay zone at such extreme distances would bevirtually impossible without geosteering (pre-vious page). Geosteering involves takingpetrophysical measurements at or near the bitand relaying the information in real time tothe drilling team. The team can then adjust thebit direction to aim the wellbore optimallythrough the formation. The result is thatsmaller targets at greater distances can bedrilled successfully. Geosteering has been asuccess in terms of both the technical prac-ticalities and the productivity benefits inextended-reach wells.

This record extended-reach well, M-11, atthe BP Exploration Operating Co. Ltd. WytchFarm development in southern England, hasa horizontal displacement of 10,114 m[33,182 ft ] at a true vertical depth (TVD) ofonly 1605 m [5266 ft]. Well M-11 is the sec-ond extended-reach record well at WytchFarm (above). Both wells used Anadrill log-ging-while-drilling (LWD), measurements-while-drilling (MWD) and GeoSteering

Winter 1997 33

tools. The M-05 well had set a displacementrecord of 8035 m [26,361 ft] in 1995 (right).

The M-05 record was broken last year byPhillips China, Inc., which drilled the Xijiang24-3 in the South China Sea to a horizontaldisplacement of 8063 m [26,446 ft].1 Hori-zontal displacement, also called stepout ordeparture, is the farthest horizontal distancefrom a vertical line below the surface loca-tion to the tip of the well.

The total measured depth of Well M-11 is10,658 m [34,967 ft], a remarkable achieve-ment that puts this well second among thelongest wellbore paths drilled in the world.In comparison, the world’s deepest well at12,869 m [42,226 ft] measured depth is avertical well, SG-3, drilled by the Ministry ofGeology of the former Soviet Union for sci-entific exploration on the Kola Peninsula.2

The stepout ratio (horizontal displacementdivided by TVD at total depth) is 6.3 for WellM-11 (below). Generally, a well is defined asextended reach if it has a stepout ratio of 1 ormore. A horizontal well is loosely defined ashaving a final segment at an 85° to 90° incli-nation from true vertical. An extended-reachwell may also be a horizontal well, but thisis not a requirement.

34 Oilfield Review

* Directional drilling, MWD, LWD or GeoSteering services provided by Anadrill

Rank Horizontal Measured TVD, ft Operator Well Locationdisplacement, ft depth, ft

1* 33,182 34,967 5266 BP M-11 UK, Wytch Farm

2* 26,446 30,308 9554 Phillips Xijiang 24-3 South China Sea

3* 26,361 28,593 5285 BP M-05 UK, Wytch Farm

4 25,764 30,600 NA Norsk Hydro 30/6 C-26 North Sea

5* 25,108 27,241 NA BP M-09 UK, Wytch Farm

6* 23,917 28,743 9147 Statoil 33/9 C-2 North Sea

7* 22,369 24,442 NA BP M-03 UK, Wytch Farm

8* 22,180 24,680 5243 BP M-02 UK, Wytch Farm

9 21,490 26,509 NA Norsk Hydro 30/6 B-34 North Sea

10* 21,289 25,991 NA Norsk Hydro 30/6 C-17 North Sea

11 20,966 24,670 9076 Amoco SEER T-12 North Sea

12* 20,577 21,102 NA Norsk Hydro 31/4 A-8A North Sea

13* 20,577 25,010 NA Norsk Hydro 30/9 B-30 North Sea

14* 20,514 22,907 5528 Total Austral ARA S7/1 Tierra del Fuego, Arg.

15 20,289 25,164 NA Norsk Hydro 30/6 B-6 North Sea

16* 20,151 25,541 NA Norsk Hydro 30/6 C-24A North Sea

17* 19,667 22,559 5324 BP M-06 UK, Wytch Farm

18 19,209 23,786 8845 Statoil 33/9 C-3 North Sea

19* 19,030 23,835 NA Norsk Hydro 30/9 B-48 North Sea

20* 18,758 24,540 13,940 BP 2/1 A-13 North Sea

0 1000 2000 3000 4000 5000 6000 7000 8000 9000 11,00010,000

Measured depth, m

Wytch Farm wellsOther wellsDeparture/TVD = 1Departure/TVD = 3Departure/TVD = 5

TVD

, m

0

1000

2000

3000

4000

5000

6000

7000

Wytch FarmM-09

Standard technology Advanced technology

Wytch FarmM-05

Wytch FarmM-11

■■Industry comparison of extended-reach wells. What was once considered the envelope of extended-reach drilling now merely indicatesthe difference between standard and advanced technology. That envelope continuously enlarges as companies push technology to thelimit.

■■Top 20 extended-reach wells worldwide.

Wytch Farm Development

Drilling the “10-km challenge,” as Well M-11 had been nicknamed throughout plan-ning and operations, was driven by environ-mental and economic as well as technicalobjectives. Simply put, the well was drilledbecause it economically tapped additionalreserves more than 10 km from the surfacewellsite. These reserves lay in a section of theSherwood reservoir which extends offshorebeneath Poole Bay on the southern coast ofEngland. Wytch Farm is western Europe’slargest onshore oil field, but roughly one-halfof its 467 million bbl [74 million m3]of oilextends offshore. The Wytch Farm oil fieldcomprises three major reservoirs, the shal-lower Frome Limestone at 800 m [2625 ft],the Bridport reservoir at 900 m [2900 ft] andthe larger, more productive Sherwood reser-voir at 1600 m [5200 ft]. The Bridport hasbeen on production since 1979 and was thefirst stage of Wytch Farm development. Thesecond stage consisted of the onshoreSherwood reserves, and third stage the off-shore Sherwood (above left).

In 1990, BP began analyzing methods ofproducing offshore reserves from theSherwood reservoir, including evaluation ofsetting a platform or constructing an artifi-cial island (left). The field sits near a naturepreserve and is in an area of outstandingnatural beauty frequented by tourists. Thus,any development plan had to be aestheti-cally pleasing with minimal effect on thearea. The development plan also had toadhere to stringent environmental regula-tions. The initial plan called for constructionof an artificial island with conventionaldirectional wells at a cost of £180 million[$330 million].3 In contrast, development ofthe offshore reserves with extended-reachwells would cost less than half as much atan estimated £80 million [$150 million] andwould better protect the environment.Furthermore, the use of extended-reachwells accelerated production by three years.

Winter 1997 35

Dai

ly p

rodu

ctio

n, th

ousa

nd B

OP

D

Bridport

OnshoreSherwood

OffshoreSherwood

0

110

100

90

80

70

60

50

40

30

20

10

1975 1980 1985 1990 1995 2000 2005 2010 2015

■■Daily production forecast. The third-stage development of Wytch Farm comprisesextended-reach wells drilled from two onshore sites to produce reserves that sit offshore.More than 80% of the current field production comes from extended-reach wells. Theextended-reach program increased reserves from 300 million bbl before the program to467 million bbl now.

1. Talkington K: “Remote South China Sea ReservoirPrompts Extended Reach Record,” Oil & Gas Journal95, no. 45 (November 10, 1997): 67-71.

2. Petzet GA: “Unreal ‘Depth’ at Wytch Farm,” Oil &Gas Journal 96, no. 7 (February 16, 1998): 17.

3. “BP Taps Wytch Farm with Extended Reach Wells,”Oil & Gas Journal92, no. 1 (January 3, 1994): 30-31.

0 1000 2000 3000 50004000 6000 7000 8000 9000

0 1000 2000 3000 50004000 6000 7000 8000 9000

Island concept

Poole Harbor Hook IslandSea level

Extended-reach well

Sherwood reservoir

Sherwood reservoir

Departure, m

■■Artificial island development concept. The original proposal for development of the off-shore reserves called for construction of an artificial island and simple directional wells.Instead, extended-reach wells were chosen because there would be less impact on theenvironment and development would cost less than half and occur three years earlier.

Well M-11 is the fourteenth extended-reachwell drilled in the third stage of Wytch Farmdevelopment (below).4

A significant factor in the decision to useextended-reach drilling (ERD) was the suc-cess of other companies, particularlyUnocal Corp. in the Point Pedernales field insouthern California, USA.5 In the late 1980s,collapsing oil prices prompted Unocal todesign and drill extended-reach wells froman existing platform rather than set a secondplatform. For Unocal, extended-reach wellsachieved several objectives: they eliminatedthe high capital cost of a second platform,intersected more of the formation with near-horizontal wellbores and demonstrated con-clusively that such difficult wells could bedrilled and completed economically. These

36 Oilfield Review

BournemouthPoole

Poole Harbor5 km

6 km7 km

8 km9 km

M-01

M-08F-20 F-21

M-02

M-05

0 1 2 3 4 5

N

F-19M-10 M-07

F-18M-03

M-06M-09

Purbec

B1 B2A F

M-11

10 km

Gathering station

Sherwood sandstone reservoir1585 m TVD subseaWellsites

Sherwood existing entry andTD locationsRadius of development of offshorereserves by extended-reach drilling

PooleLondon

■■Wytch Farm extended-reach drillingradius. Extended-reach wells from F andM sites on a peninsula in Poole Harbor tapreserves as far as 10 km away. The off-shore reserves are developed in a step-by-step manner, generally with each wellslightly farther than the previous wells.

Arne fault

Northern fault

Oxford clay kellaways bedsCornbrash/Forest marbleFullers earthBridport sandstoneLiae

Mercia mudstone

Sherwood sandstone

Aylesbere group

133/8-in. casingat 1329 m MD

185/8-in. casingat 232 m MD

West

500

1000

1500

2000

TVD

sub

sea,

m

1 Sea level 2 3 4 5

Wellsite M

■■Geological cross section. The world’s longest extended-reach well has a stepout of 10,114 m and a total measured depth of 10,658 m.

wells were relevant to Wytch Farm becauseof the shallow TVD of 1350 m [4420 ft].Many other wells had been drilled with longstepouts, but none at such shallow depths.The base requirement for further field devel-opment was the capability to drill and com-plete wells at up to 6000-m [20,000-ft]departure from onshore wellsites, which BPfelt to be a reasonable extension of existingtechnology.6

The use of extended-reach wells results inless surface disturbance because fewer wellsare needed and surface sites have a smallerfootprint. In developing Wytch Farm field,BP sought to maximize oil recovery as eco-nomically as possible with the least distur-bance to the environment and thesurrounding community. Drilling from small

surface sites instead of an offshore locationhelped accomplish this goal. BP establisheda partnership with local communities andregulatory authorities to ensure that the nat-ural beauty of the Poole area remainedunspoiled. More than 300 formal meetingsand countless informal discussions were heldwith local authorities, government depart-ments, environmental and conservationorganizations and the general public toensure that the views of area residents beconsidered in the development of the field.

The area around the oil field forms part ofthe Dorset Heritage Coastline and includesareas of special scientific interest, a wildlifespecial protection area, a wetland birds siteand a national nature reserve. The surfacesite is designed to blend into the

environment, and equipment is painted inearth-tone colors to minimize visual impact.All lighting is judiciously placed and pointeddownward. Strict noise regulations areimposed and ensure minimal disturbance.An extensive impermeable containmentditch surrounding the site can hold any flu-ids from potential accidents.

Teamwork and PlanningAn extended-reach well requires extensiveplanning and involves the commitment anddirect participation of the operator, rig con-tractor and all service providers. The evalua-tion, design and planning of Well M-11lasted more than a year, from March 1996until drilling began in May 1997.

Close cooperation and a smooth workingrelationship were essential for such a chal-lenging, high-profile project as Wytch Farm.Contractors were paid on a day-rate basiswith total well performance incentives to beshared among all participants, furtherencouraging teamwork and ensuring align-ment of goals. Anadrill supplied directionalwell planning and drilling engineering sup-port, directional drillers and equipment, sur-veying, MWD/LWD, mud logging andcoring services. Baroid provided mud sup-plies, mud engineering, and drill-cuttingsand waste fluids disposal. BJ Services pro-vided cementing services. Deutag Drillingsupplied the drilling rig, crews, drillstringtubulars, tubular running services and fishingtools. Lasalle designed, installed and com-missioned completion and electrical sub-mersible pump equipment. Schlumberger

Winter 1997 37

Oxford clay kellaways beds

Cornbrash/Forest marble

Tertiary group

Fullers earth/Inferior coliteBridport sandstone

Mercia mudstone

Gault and Greensand formations

Sherwood sandstone

Aylesbere group

Chalk group

East6 7 8 9 10 km

7-in. casingat 8881 m MD

M-11zTD 9688 m 1584.8 m TVDSS

M-11yTD 10658 m 1585.2 m TVDSS

51/2-in. liner 10655 m

Fault to south / 6-8 region

Wireline & Testing supplied openhole log-ging services, cased-hole logging services,drillstem test and tubing-conveyed perforat-ing and well testing; and Dowell suppliedcoiled tubing services.

The focus during the early part of the thirdstage of development was to build multidis-ciplinary teams, with contractors working inclose proximity to the operator’s staff.Contractor senior representatives haveoffices close to one another within the oper-ator’s office complex in Poole. This setup fos-ters a close but informal “roundtable”arrangement. Communication barriers havedisappeared, and everyone on the projecthas the same goals—to drill the wells effi-ciently, correctly and economically.Decisions are made and involve both theoperator and contractors. For such a systemto work effectively, a high degree of trust andopenness is necessary among personnel ofall rank. This working environment has pro-vided the opportunity to obtain excellentbenchmark data for drilling subsequentextended-reach wells, leading to the record-setting 10-km target.

At the onset of third-stage development,two of the least known factors were theincreased time and cost for extended-reachcompared to conventional wells. Typically,the first one or two wells of any project incurthe greatest time and cost as the learningcurve begins. Efficiency and performanceimprove rapidly on subsequent wells as teammembers work together more efficiently, andtechnology is applied more effectively.Generally, each well at Wytch Farm has been

drilled farther than the previous wells, allow-ing for incremental learning through experi-ence. Had such an incremental learningprocess not been used, many problemswould undoubtedly have occurred on the M-11 well.

Every aspect of the M-11 well plan under-went a rigorous peer review to identify poten-tial pitfalls and develop contingency plans.Experts from BP and its partners thoroughlyanalyzed key risks and processes in reservoirissues, well placement, drilling mechanics,hydraulics and safety. Having outsiders ana-lyze the well plan provided a useful checkagainst existing contingency plans.

Profile Design The reservoir section targeted by Well M-11lies between 8- and 10-km [26,000- to33,000-ft] departure from the M site (previ-ous page, bottom and below). For this lengthdeparture, the relatively shallow depth of theSherwood reservoir posed some specialdrilling challenges not present at greaterdepths. There was limited scope in the tra-jectory design to allow the target to be

4. Knott D: “BP Completes Record Extended-ReachWell,” Oil & Gas Journal96, no. 3 (January 19, 1998): 24-26.

5. Mueller MD, Quintana JM and Bunyak MJ:“Extended-Reach Drilling from Platform Irene,” paperSPE 20818, presented at the Offshore TechnologyConference, Houston, Texas, USA, May 7-10, 1990.

6. Banks SM, Hogg TW and Thorogood JL: “IncreasingExtended-Reach Capabilities Through Wellbore ProfileOptimization,” paper IADC/SPE 23850, presented atthe IADC/SPE Drilling Conference, New Orleans,Louisiana, USA, February 18-21, 1992.

reached optimally.7 The majority of wells atWytch Farm have been drilled with 80° to82° tangent angles and were controlled bychoice of kickoff point and build rate in theupper section.

This well was planned to have a shallowkickoff point to allow inclination angle to bebuilt to 82° in the 171⁄2-in. hole. The 133⁄8-in.casing was to have been run to about 1400m [4600 ft] measured depth, and then the121⁄4-in. hole was to be drilled as a 7500-m[24,600-ft] tangent section into the top of theSherwood reservoir. The length of the 81⁄2-in.hole section was kept to a minimum becausereservoir drilling is three to four times moreexpensive than in the 121⁄4-in. section due toextra bottomhole assembly equipment, mudlosses and slower penetration rates. In addi-tion, the mudstone above the reservoirwould be exposed to a low-weight mud,leading to concerns about borehole stability.

A tangent angle of 82° was used to keeptorque levels manageable, maximize thelikelihood of being able to run casing to a10-km departure and permit oriented drillingin the 81/2-in. section (above and next page,top). The final double build and hold trajec-tory was similar to that on other wells,except that steering would be further limitedin the 81/2-in. section. This trajectory wouldallow the well to build angle to horizontalstarting at the 95⁄8-in. casing shoe. Previouswells used an instrumented positive dis-placement motor, LWD tools and anadjustable stabilizer to geosteer in the reservoir.

Because steering capability would be lim-ited, Well M-11 was designed to be drilled asgeometrically as possible through the reser-voir, with some geosteering performed byrotary steering drilling tools.

The key to success would be performingevery phase of the well plan flawlessly.Directional control, hole cleaning, torqueand drag, and casing flotation each played arole. The rest of this article describes howthey come together in the drilling of thisrecord-setting 10-km well.

Directional ControlSteering by slide drilling is impossible atextreme horizontal distances. Experience onWells M-05 and M-09 indicated that slidedrilling would be practically impossiblebeyond 8 km. Drilling in the sliding moderesults in several inefficiencies that are com-pounded by extreme distances. The motormust be oriented and maintained in a partic-ular direction while drilling to follow thedesired path. This orientation is achievedthrough a combination of rotating the drill-string several revolutions and working thepipe to turn it to the desired direction. At 8 kmor more, the pipe may need 15 to 20 turns atsurface just to turn the tool once downhole,because the drillstring can absorb the torqueover such a long distance. For the directionaldriller, this technique is as much art as it is sci-ence. After the tool is positioned, drillstringtorque is required to hold the motor in properorientation against reverse torque created bythe motor as the bit drills.8

This situation is beneficial in wells with lowfrictional drag because adjusting weight onbit changes reactive torque, changing tool-face direction. Thus, small changes in orien-tation can be made by varying weight on bit,giving the directional driller better control. Inhigh-drag situations like Well M-11, however,it is difficult to keep torque constant in thelower part of the drillstring, causing difficultyin maintaining toolface orientation. Anotherproblem with slide drilling in high-anglewells is that cuttings removal suffers from thelack of drillstring rotation. In wells with highdrag, the drillstring cannot be loweredsmoothly and continuously, which preventsthe motor from operating at optimal condi-tions. In combination, these factors result in alower penetration rate compared to that dur-ing rotary drilling (next page, bottom). Forextended-reach wells, not only does penetra-tion rate suffer, but there is a point at whichslide drilling is no longer possible.

The M-11 trajectory was designed to mini-mize the amount of sliding directional workin the 81/2-in. reservoir section. In some ofthe earlier Wytch Farm wells, a new tech-nique was pioneered to overcome these

38 Oilfield Review

Sherwood zone thickness

Zone

5.5 mAbsent2.8 m

5.5 to 8 m7.1 m5.5 m

102030405060

Mercia mudstone

Zone 10Zone 30Zone 40

Zone 50

Zone 60

Zone 70

Estimated present oil-water contactat 1621 m TVDSS

20 m Standoff

Original oil-water contact at 1623 m TVDSS

Stratigraphic dip4° west

Stratigraphic dip0.4° west

3D seismic survey1550

1560

1570

1580

1590

1600

1610

1620

1630

164090008900880087008600850084008300820081008000

TVD

sub

sea,

m

M-11ykickoffpoint

Casing shoe

7. Modi S, Mason CJ, Tooms PJ and Conran G:“Meeting the 10km Drilling Challenge,” paper SPE38583, presented at the SPE Annual TechnicalConference and Exhibition, San Antonio, Texas, USA,October 5-7, 1997.

8. Warren TM: “Trends Toward Rotary SteerableDirectional Systems,” World Oil 218, no. 5 (May1997): 43-47.

Winter 1997 39

M-11y Sidetrack drilled

Stratigraphic dip1° east

2D seismic survey

M-11z 9688 m total measured depth

10,20010,10010,000990098009700960095009400930092009100

Horizontal departure, m

M-11z Trajectory

M-11y 10,658 m total measured depth 1585.7 m TVDSS

Fault throw 7.6 m

■■Sidetrack cross section. When the original borehole (M-11z) reached 9 km, the well was sidetracked to access a better part of the reser-voir. The sidetracked lower bore (M-11y) was cased and completed, and the upper borehole left open. The tip of the well veers upwardto stay away from the oil-water contact and to penetrate additional formation layers for added geological information.

Measured depth, m

RotarySliding

Pen

etra

tion

rate

, m/h

r

20

18

16

14

12

10

8

6

4

2

06000 6500 7000 7500 8000 8500 9000

■■Rotary and sliding penetration rates. In offset Well M-05, rotary drilling penetrationrates were several times greater than slide-drilling penetration rates. Conventional slidedrilling techniques were ineffective at extreme stepout distances because of difficultycontrolling downhole torque and weight on bit.

problems with conventional sliding fordirectional control.9 The combination of theGeoSteering tool near-bit inclination mea-surement and a downhole variable-gaugestabilizer positioned above the motorenabled drilling the wells almost entirely inrotary mode.

Standard steerable-motor directional drillingequipment is generally based on the tilt-angle principle. A bend between 0.5° and 3°in the motor provides the bit offset necessaryto initiate and maintain changes in coursedirection. Three geometric contact points(bit, near-bit stabilizer on the motor and astabilizer above the motor) approximate anarc that the well path will follow, and thusthe curve rate or dogleg severity for the sys-tem. This curvature is formed and built byholding the entire drillstring still so that thebend can work in a preferential direction orin sliding mode.10

Conventional practice is to drill in rotarymode, rotating the drillstring from the surfaceto drill a straight path. If a change in directionis needed, the drillstring is stopped with thebent housing or tilt on the steerable systemoriented in the desired direction. This orien-tation is called the toolface angle and is mea-sured downhole by MWD systems. Whendrilling in this oriented mode, the entire drill-string has to slide. Drillstring drag problemsbecome acute in extended-reach wells andcause problems in setting the toolface angleand applying weight to the bit. Rate of pene-tration suffers. Techniques are needed to pro-vide greater directional flexibility with rotarydrilling in extended-reach wells.

40 Oilfield Review

■■The art and science of direc-tional drilling. The directionaldriller closely monitors surfaceand downhole measurments oftorque, drag and direction. Ittakes a skillful directionaldriller to interpret these dataand then steer the bit accord-ingly by tweaking the down-hole tools from surface. Theresult is a smooth wellbore righton target.

The primary components of the GeoSteeringtool are a steerable motor with an instru-mented section and a fast wireless telemetrysystem that passes data to the MWD systemhigher up in the BHA. The instrumented subis built into the motor near the bent housingwhich is typically about 1.5 m [5 ft] abovethe bit. Packaged in the sub are directionaland petrophysical sensors, electronics forcontrol and telemetry and batteries forpower. Three inclinometers provide inclina-tion data at the bit in both a survey and con-tinuous mode. Above the GeoSteering tool, astabilizer with an adjustable gauge allowsthe directional driller to change the direc-tional characteristics of the BHA in rotarymode.11 Unfortunately, rotary directional ten-dencies are not as predictable as those insteering mode. The inclination-at-bit mea-surement from the GeoSteering tool is criti-cal to judge the cause-effect relationshipfrom gauge changes to well path curve rate.Minor directional changes can be made reli-ably with rotary drilling, reserving the steer-ing mode for larger or more difficult changes(previous page).

The first three extended-reach wells atWytch Farm, Wells F-18, F-19 and F-20,were designed based on a reservoir charac-terization developed from the onshore reser-voir, 2D seismic surveying and more recent3D seismic acquisition, outcrop studies andoffshore appraisal wells. The basic design ineach case was to traverse the reservoir sec-tion at a constant angle of 86°. A wellencountering a 50-m [160-ft] oil columncould then be drilled 20 m [65 ft] above theoil-water contact.12 These wells were suc-cessful and had departures of about 5 km[16,000 ft]. With the additional reservoirinformation from these wells, subsequent

wells were designed with an even largerstandoff from the oil-water contact. Wellsfrom the M site accessed poorer-qualitysands in the Sherwood reservoir and neededeven longer reservoir sections for sufficientwell productivity. BP decided to drill the nextwells with 7- to 8-km [23,000- to 26,000-ft]ERD stepouts and vertical depth control tol-erance of 1 m [3 ft].

At such distances, orienting the pipe in slid-ing mode would be difficult and time-con-suming. Penetration rates would sufferbecause of the frictional losses reducing theamount of weight reaching the bit.Minimizing the amount of sliding was critical.The nature of the Sherwood reservoir compli-cated the picture, since it has many distinctzones separated by near-horizontal shales.13

Previous directional drilling had shown theformation response to be erratic, with rotarymode performance of a BHA varying sud-denly and unpredictably from holding angle,to building angle, to dropping angle.

During drilling of the early wells, theMWD survey measure point was typically 20 to 25 m [66 to 82 ft] behind the bitface. The typical BHA consisted of the bit,motor, stabilizer, CDR Compensated DualResistivity tool, MWD, and ADNAzimuthal Density Neutron tool (above). Asimple geometrical calculation showedthat if the build rate were to change by4°/30 m [4°/100 ft], the true vertical depthof the well would be altered by 0.75 m[2.5 ft] before the problem could bedetected. With this BHA design, the devia-tion from plan could be quite large, andonly a rapid response with sliding modedrilling would allow the well to stay withinthe tight tolerances required. Even with

Winter 1997 41

ADNAzimuthal Density

Neutron toolMWD

Inclination,azimuth

and shocks GeoSteering tool,motor, inclination,

resistivity and gamma ray

0.75°bend

CDRCompensated

Dual Resistivity tool

Resistivity andgamma ray

Variable-gauge stabilizer: 71 /4. to 81

/2 in

Density andporosity

■■GeoSteering assembly. The Anadrill GeoSteering tool was used to drill the 81/2-in. reservoir sections of early Wytch Farm wells and todrill out past the 95/8-in. casing shoe of Well M-11. This BHA configuration included a variable-gauge stabilizer to steer the well. Mudpulses adjust the stabilizer blades in or out to alter the inclination of the assembly.

9. Bruce S, Bezant P and Pinnock S: “A Review ofThree Years’ Work in Europe and Africa with anInstrumented Motor,” paper IADC/SPE 35053, pre-sented at the IADC/SPE Drilling Conference, NewOrleans, Louisiana, USA, March 12-15, 1996.Peach SR and Kloss PJC: “A New Generation ofInstrumented Steerable Motors Improves Geosteeringin North Sea Horizontal Wells,” paper IADC/SPE27482, presented at the IADC/SPE Drilling Confer-ence, Dallas, Texas, USA, February 15-18, 1994.Lesso WG and Kashikar SV: “The Principles andProcedures of Geosteering,” paper IADC/SPE 35051,presented at the IADC/SPE Drilling Conference, NewOrleans, Louisiana, USA, March 12-15, 1996.

10. Poli S, Donati F, Oppelt J and Ragnitz D: “AdvancedTools for Advanced Wells: Rotary Closed LoopDrilling System—Results of Prototype Field Testing,”paper SPE 36884, presented at the SPE EuropeanPetroleum Conference, Milan, Italy, October 22-24, 1996.

11. Bonner S, Burgess T, Clark B, Decker D, Orban J,Prevedel B, Lüling M and White J: “Measurements atthe Bit: A New Generation of MWD Tools,” OilfieldReview 5, no. 2 (April/July 1993): 44-54.

12. Harrison PF and Mitchell AW: “ContinuousImprovement in Well Design OptimisesDevelopment,” paper SPE 30536, presented at theSPE Annual Technical Conference and Exhibition,Dallas, Texas, USA, October 22-25, 1995.

13. Cocking DA, Bezant PN and Tooms PJ: “Pushing theERD Envelope at Wytch Farm,” paper SPE/IADC37618, presented at the SPE/IADC DrillingConference, Amsterdam, The Netherlands, March 4-7, 1997.

sliding, it would be difficult to keep thewell within the 1-m vertical tolerance.14

The solution was to use the GeoSteeringtool to provide near-bit inclination dataand give immediate warning of anychange in BHA response. Gamma ray andresistivity data could also be checked forany unexpected changes in geology.Another drilling equipment change forthese long wells was the inclusion of adownhole-adjustable, variable-gauge sta-bilizer.15 If the GeoSteering tool identifieddeviations from the required well pathearly, then the trajectory could be alteredin the rotary mode by adjusting the stabi-lizer. This BHA design drilled the reservoirsections successfully in Wells M-02, M-03and M-05.

In Well M-02, with a stepout of 6760 m[22,180 ft] and a reservoir vertical toleranceof 2 m [7 ft], 15% of the reservoir sectionwas drilled in sliding mode, and the stabi-lizer gauge setting was changed 16 times indrilling 891 m [2923 ft]. The sliding pene-tration rate was low—only one-third therate achieved during rotary drilling. BHAperformance in Well M-03 was similar, buteven better in the M-05 Well. In each case,without the flexibility provided by this setup, considerably more time would havebeen spent in sliding mode to adjust thewell path, with a corresponding decrease inpenetration rate.

42 Oilfield Review

ADNAzimuthal Density

Neutron tool

Density andporosity

CDRCompensatedDual Resistivity

tool

Dual resistivity,gamma ray andannular pressure

MWD

Downholeweight on bit,

downhole torque,multi-axis shocks anddirectional information

StabilizerStabilizer

Steerable rotarydrilling tool

The highly variable gauge stabilizer has sixpossible settings between 71⁄4 in. and 81⁄2 in.The stabilizer blade position is controlledfrom the surface by applying mud-flowsequences that are interpreted by a micro-processor in the tool. This adjustable stabi-lizer allowed some degree of build and dropduring rotary drilling and worked adequatelyin previous wells at Wytch Farm, but it didnot provide azimuth control.16

To drill out 8 km and then on to 10 km,rotary steerable directional systems weretried for even greater control of inclinationand azimuth. It was clear that a new type ofdirectional drilling tool would be needed tocontrol direction and azimuth adequatelywhile rotating in the Sherwood at greatdepartures. Several designs were nearingcommercialization, each with a differentapproach to achieving vertical and lateralsteering while the drillstring is rotated.

A prototype from Camco Drilling GroupLtd. had been tried on Wells M-08, M-09and M-10 at Wytch Farm to encourage andaccelerate their development and test thefeasibility of using these tools to drill the 10-km well.17 These tools further reduced theamount of time spent on slide drilling. Thisrotary steerable system synchronously mod-ulates the stabilizer blade extension andcontact pressure as a particular blade passes

a certain orientation in the wellbore. Byapplying hydraulic pressure each time arotating blade passes a specific vertical or lat-eral orientation, the near-bit stabilizer forcesdrilling away from that direction. Continuousrotation of the drillstring results in reducedtorque and drag and improved hole cleaning.

These improvements led to more efficientweight transfer to the bit and higher penetra-tion rates. Hole direction and dogleg werecontrolled from the surface during drillingoperations by sending information to thedownhole tool using a sequence of mudpulses. This coded sequence specified steer-ing commands from the multiple commandset preprogrammed into the tool on surface.The BHA at total depth included the bit, steer-able rotary system, stabilizer, MWD and LWD(above). The steerable rotary drilling systemwas instrumental in cutting sliding time to lessthan 5% on the M-11 well and was critical todrilling the well beyond 10 km.

Hydraulics and Hole CleaningSelection of a drilling fluid must balance anumber of critical factors. The fluid must pro-vide a stable wellbore for drilling long open-hole intervals at high angles, maximizelubricity to reduce torque and drag, developproper rheology for effective cuttings trans-port, minimize the potential for problemssuch as differential sticking and lost circula-tion, minimize formation damage of produc-tive intervals, and limit environmental

■■ Steerable rotary assembly. A BHA with a steerable rotary drilling tool was the primary directional system for most of the 12 1/4-in.and 81/2-in. sections of Well M-11. The CDR tool was incorporated into the BHA design during the latter part of the 12 1/4-in. section tohelp identify the top of the reservoir and determine the 95/8-in. casing seat.

exposure through a wellsite waste-minimiza-tion program.18 A growing industry trend ofdesigning wells to extend casing seats tolonger intervals requires the use of oil-baseor synthetic-base mud to provide lubricity tohelp control torque and drag. This trendmeans larger hole diameters deeper in theERD trajectory.

With 8- to 10-km extended-reach sections,the circulation rates necessary to ensure ade-quate hole cleaning in these larger holes aredifficult to attain because of increases inannular pressure losses. These pressurelosses increase the equivalent circulatingdensity (ECD) at the end of the well. Higherequivalent circulating densities may causelost circulation, especially in fractured ordepleted upper zones. Higher pump ratesensure a clean hole but may lead to lost cir-culation from higher ECD.

In Well M-11, annular pressure whiledrilling was monitored in real time as part ofthe LWD package. On several occasions, thetool detected increases in ECD and warnedthe driller before a lost-circulation problemoccurred. When the tool detected increasesin annular pressure at the BHA, the drillerwould stop drilling to remove cuttingsbuildup, thus lowering ECD.

Efficient cuttings transport is one of the pri-mary design considerations for the drillingfluid in an extended-reach well. The factorsthat affect cuttings transport include drilling-fluid density, low shear-rate rheology, flowrate, cuttings size and concentration in theannulus, drillpipe size, rotary speed anddrillstring eccentricity in the wellbore.

Hole cleaning is of critical importancewhen drilling the high-angle tangent sectionsof extended-reach wells because of the ten-dency for cuttings to fall to the low side ofthe well. To obtain the necessary annularvelocities for hole cleaning in high-angleand horizontal sections, high flow rates areneeded, resulting in greater demands on themud pumps. Experience at Wytch Farm hasshown that above a tangent angle of 80° in a121⁄4-in. hole, flow rates of at least 1100gal/min [4200 L/min] are necessary to keepthe hole clean as it is drilled.

To meet such rates, an existing rig mayneed to increase the number of mud pumpsfrom two to three, increase the power ratingof the pumps from 1600 hp to 2000 hp ormore, and increase the pressure rating of thepumps and surface system from 5000 to7500 psi. A downside to these changes is theincrease in capital cost for additional pumpsand a doubling of maintenance costs for thehigher-pressure equipment.19 At higher sur-face pressures, parts replacement will occurmore frequently, but having an extra pumpresults in less disruption to the drilling pro-gram during pump repairs.

To maintain sufficient flow rates on theWytch Farm wells, Deutag’s T-47 rig used a5000-psi surface system and three 1600-hptriplex pumps electronically controlled toincrease efficiency. If the three pumpsworked independently, there would bepotential for synchronization of the cylindersin each pump. In such a case, the pumpscould produce pressure surges up to 750 psi,much more than could be handled ade-quately by the pumps’ pulsation dampeners.In effect, this electronic control systemworked like a camshaft, keeping the cylinders40° out of phase with each other. The threepumps worked as one nine-cylinder pump.

In addition to reducing pressure surges, thissystem reduced surface vibrations, decreas-ing the wear and tear on the pumping system,and detected impending pump failure toreduce downtime. A further benefit was animprovement in the quality of the MWD sig-nal. With a smooth, regular pump signal, theMWD pulses were much clearer and easierto read at the surface, improving data quality.

Increasing the pumping capacity of the rigrequired sufficient capacity in the solids-control equipment, particularly the shaleshakers, to handle the increased flow rates.Removing the maximum amount of solids atthe shakers reduces the need for furthersolids-control equipment. The T-47 rig hadfour state-of-the-art, linear-motion shale

shakers. One of these shale shakers was out-fitted with larger-mesh screens to salvageand reuse lost-circulation material. Two cen-trifuges, in line after the shakers, removedfine low-gravity solids.

Pipe rotation is another critical factor inhole cleaning. The objective of the hole-cleaning program is to improve drilling per-formance by avoiding stuck pipe, avoidingtight hole on connections and trips, maxi-mizing the footage drilled between wipertrips, eliminating backreaming trips prior toreaching the casing point and maximizingdaily drilling progress.20 The more anextended-reach well can be drilled in therotary mode instead of sliding, the better thehole cleaning. Pipe rotation helps preventcuttings from accumulating around stabiliz-ers, drillpipe protectors and tool joints.Rotating pipe helps disturb any cuttings thatmay settle to the low side of the wellbore,keeping the cuttings suspended in and trans-ported by the mud. Faster rotational speedsof the pipe improve hole cleaning, but thereare some drawbacks to very high rotationalspeeds. Although good for hole cleaning,excessive rotational speed can increase theseverity of downhole vibration and shocks,putting directional drilling and LWD equip-ment at electronic and mechanical risk.Furthermore, excessive rotary speeds mayincrease drillpipe and casing wear.

Winter 1997 43

14. Bruce et al, reference 9.15. Odell AC, Payne ML and Cocking DA: “Application

of a Highly Variable Gauge Stabilizer at Wytch Farmto Extend the ERD Envelope,” paper SPE 30462, pre-sented at the SPE Annual Technical Conference andExhibition, Dallas, Texas, USA, October 22-25,1995.

16. Poli et al, reference 10.17. Barr JD, Clegg JM and Russell MK: “Steerable Rotary

Drilling with an Experimental System,” paperSPE/IADC 29382, presented at the SPE/IADC DrillingConference, Amsterdam, The Netherlands, February28-March 2, 1995.

18. Payne ML, Wilton BS and Ramos GG: “RecentAdvances and Emerging Technologies for ExtendedReach Drilling,” paper SPE 29920, presented at theInternational Meeting on Petroleum Engineering,Beijing, China, November 14-17, 1995.

19. Gammage JH, Modi S and Klop GW: “Beyond 8kmDeparture Wells: The Necessary Rig & Equipment,”paper SPE/IADC 37600, presented at the SPE/IADCDrilling Conference, Amsterdam, The Netherlands,March 4-6, 1997.

20. Guild GJ, Wallace IM and Wassenborg MJ: “HoleCleaning Program for Extended Reach Wells,” paperSPE/IADC 29381, presented at the SPE/IADC DrillingConference, Amsterdam, The Netherlands, February28-March 2, 1995.Lockett TJ, Richardson SM and Worraker WJ: “TheImportance of Rotation Effects for Efficient CuttingsRemoval During Drilling,” paper SPE/IADC 25768,presented at the SPE/IADC Drilling Conference,Amsterdam, The Netherlands, February 23-25, 1993.

44 Oilfield Review

Torque and Drag

Torque levels have been closely monitoredthroughout the extended-reach developmentprogram. In extended-reach wells, torquelevels are generally more dependent onwellbore length than on tangent angle.Higher-angle wells do, however, tend toreduce overall torque levels, as more of thedrillstring will be in compression, and con-sequently, tension and contact forces aroundthe top build section are reduced.

In Wytch Farm extended-reach drilling,three sizes of grade S135 drillpipe were used.In the Well M-11 121⁄4-in. hole, the drillstringconfiguration consisted of 8000 m [26,000 ft]of 65⁄8-in. drillpipe—a length dictated by theracking capacity within the derrick—on topand 51⁄2-in. pipe below. The larger pipe at thetop provided strength to resist torque loads.The main factors in drillstring design in thispart of the hole included pump pressure lim-itations, torque capacity and hole cleaning.In the 81⁄2-in. hole, the drillstring configura-tion consisted of 4500 m [14,700 ft] of 5-in.S135 drillpipe on bottom and then 51⁄2-in.S135 pipe to surface.

The considerations in drillstring design forthe 81⁄2-in. hole included fishing capability,equivalent circulating density and torquecapacity. The 51⁄2-in. drillpipe had double-shoulder, high-torque tool joints, and the 5-in.drillpipe joints underwent stress balancingand used high-friction-factor pipe dope toincrease the torque capacity to match that ofthe topdrive. Nonrotating drillpipe protectorswere run in trials on earlier wells and helpedreduce torque somewhat, but there was atrade-off because they caused an increase inannular pressure drop and therefore in equiv-alent circulating density. The drillpipe protec-tors also suppressed drillstring buckling.

0 1000 2000 3000 4000 5000 6000 7000 8000 9000

Measured depth, m

Su

rfa

ce

to

rqu

e,

tho

usa

nd

ft-

lbf

Bit

to

rqu

e,

tho

usa

nd

ft-

lbf

0

5

10

15

20

25

30

35

M-09 surface torqueM-11 surface torqueM-11z surface torqueM-11z bit torque

■■Hook load during slide drilling the 121/4-in. section of Well M-11. The hook load duringslide drilling followed the expected trend (parallel black lines) down to 6000 m mea-sured depth. After this depth, there was considerable gain in hook load. Beyond 8000 m,applying weight on bit during slide drilling was difficult, if not impossible. Drilling pastthis measured depth therefore required rotation to overcome friction and allow weight tobe transferred to the bit.

0 1000 2000 3000 4000 5000 6000 7000 8000 9000

Measured depth, m

Hook loadWeight on bit

Hoo

k lo

ad, t

hous

and

lbf

Wei

ght o

n bi

t, th

ousa

nd lb

f

140

0

20

120

100

40

60

80

■■Torque during rotary drilling of the 121/4-in. section of Well M-11. The parallel linesindicate the torque trend predicted for the M-11 121/4-in. section. The recorded surfacetorque matched the prediction quite accurately.

In drilling extended-reach wells, drillpipe isrotated not by the rotary table but instead bythe topdrive, which travels the length of thederrick and permits drilling with an entirestand of pipe. The topdrive also providesbackreaming capacity and the capability topush casing down the well when high drag isencountered. Maximum output from the top-drive system is closely related to the maxi-mum torque capacity of the drillpipe used.The Deutag T-47 rig has a continuous top-drive output of 45,000 ft-lbf and a maximumintermittent rating of 51,000 ft-lbf. Inevitablythere will be some torsional variation, so thetop-drive torque rating needs to be sufficientlyhigh to accommodate these peak values.

The use of MWD tools that measure down-hole torque, rotational speed and downholeweight on bit in real time helps identify con-ditions that lead to stick-slip, which can pro-duce detrimental torque variations in thedrillstring. During stick-slip, the bit willalternately stop rotating (stick) and then accel-erate (slip) while the drillstring rotates at a

Winter 1997 45

10-km wellsection attotal depth

F-19F-21M-03M-05M-08M-09

3000 4000 5000 6000 7000 8000 9000 10,000 11,000

Measured depth, m

Sur

face

torq

ue, t

hous

and

ft-lb

f

0

5

10

15

20

25

30

35

40

45

■■Surface torque during reservoir drilling. Modeling the actual surface torque in the 81/2-in.sections of offset wells produced reasonable estimates of the expected torque at 10-km.The predicted torque range at 10 km reaches the upper limit of the top-drive capacity.

constant speed. The long drillstring can windand unwind when this occurs, leading toexcessive torque on connections, drillstringfailures, premature bit wear, BHA failuresand inefficient drilling. Fluctuations indrillpipe torque from downhole stick-slipcan be minimized by adjusting drilling para-meters—weight on bit, pump rate or rota-tional speed—provided fluctuations indownhole torque are detected early.

Changing design parameters to control sur-face torque is not necessarily beneficial inminimizing drag. Overcoming axial drag inthese high-angle wells is a significant chal-lenge (previous page, top). The critical opera-tions, in particular, are running the 95⁄8-in.casing inside the 121⁄4-in. hole and drillingthe 81⁄2-in. hole in an oriented mode.Experience to date has shown that runningthe liner and completion tubing is less diffi-cult yet still requires close monitoring of drag.

Extensive reviews of previous wells helpedpredict surface torque for the 121⁄4-in. sectionof Well M-11. Two independent methodswere used to analyze these data. The firstmethod, composite forecasting, is a plot ofsurface torque from all previous Wytch Farmextended-reach wells with upper and lowertrend lines used to predict torque ranges forthe section. The second technique uses adrillstring simulator program to examinetorque values from the most recent wells andcalculate upper and lower bounds for thesection. This determination of friction factorhas proven remarkably consistent for the121⁄4-in. section (previous page, bottom).These analyses uncovered a strong correla-tion between wellbore length and torque inthe 121⁄4-in. hole.

Similar analyses were performed on the 81⁄2-in. hole to predict likely torque values.Composite plots from previous wells showeda trend of increasing torque with measureddepth (above left). Close assessment of eachwell generated some unexpected results:torque levels tended to flatten after a certainlength of openhole section was drilled,resulting in low openhole friction factors.

These low-friction factors reflect goodhole-cleaning practices and controlled addi-tion of fibrous lost-circulation material.Crushed almond hulls helped control fluidloss in the reservoir section, and this lost-cir-culation material had a beneficial side effectin that it reduced drillstring torque in open-hole sections during rotary drilling. Thesematerials apparently form a low-frictionlayer between the drillstring and formation.Once this relationship was discovered, arecovery system was added to the rig shaleshakers to collect and reuse the lost circula-tion material continuously as part of thetorque-reduction program. The effect of holecleaning on torque is significant in the 81⁄2-in.

hole. Low-torque troughs form after periodscirculating the well clean following a BHAchange or after a wiper trip. Higher torquelevels occur after slide drilling or rotarydrilling a long section (left, top and bottom).

Casing FlotationOne of the major hurdles to overcome indrilling and completing a 10-km well was run-ning 95⁄8-in. casing to a departure beyond8000 m. Experience on other Wytch Farmwells showed that running the 95⁄8-in. casingbecame increasingly difficult with greaterdepartures. Casing design analyses using fric-tion factors from offset wells indicated thatdrag would be too high to run the planned8800 m of casing conventionally in Well M-11,even with full weight from the travelling block.

Various options, such as a tapered casingstring, altered fluid properties and casingflotation, were tried on intermediate-lengthwells to identify the best approach for WellM-11. Of the options tried, casing flotationproved to be the only method with sufficientpotential to get the casing to total depth. Inprinciple, casing flotation is a simple tech-nique. Essentially, casing is not filled as eachjoint is run into the wellbore, as is done intypical casing operations. The goal is to havethe casing close to neutrally buoyant, so itbecomes virtually weightless in the mud, anddrag is minimal.

On Well M-03, the entire 95⁄8-in. casingstring was floated into the well to observe theactual running weight compared to predic-tions. This exercise was important because itindicated that the mud weight as measuredat the surface needed to be slightly lowerthan calculated to reduce positive buoyancyand allow the casing to sink. The mud rheol-ogy had to be reduced as much as possibleprior to running the casing because of thehigh surge pressure caused as the casing waspushed into the well.21

46 Oilfield Review

200

0

20

40

60

80

100

120

140

160

180W

eigh

t on

bit,

thou

sand

ft-lb

f

Hoo

k lo

ad, t

hous

and

ft-lb

f

9000 9200 9400 9600 9800 10,000 10,200 10,400 10,600 10,800

Measured depth, m

Weight on bit

Hook load

50

0

5

10

15

20

25

30

35

40

45

Bit

Torq

ue, t

hous

and

ft-lb

f

Surface

9000 9200 9400 9600 9800 10,000 10,200 10,400 10,600 10,800

Measured depth, m

■■Hook load and weight on bit in the 81/2-in. section of Well M-11. During rotary drillingthe 81/2-in. section, the hook load decreased slightly while still maintaining effectiveweight on bit. Rotating the pipe lowered friction factors and helped keep the hole cleanto allow successful drilling past 10 km.

21. Cocking et al, reference 13.

■■Torque in the 81/2-in. section of Well M-11. Bit torque remained relatively stablethroughout the 81/2-in. reservoir section. The surface torque showed some fluctuationbut remained within the range predicted (parallel black lines) by torque and drag sim-ulators.

Winter 1997 47

In a long extended-reach section, an entireair-filled casing string can become positivelybuoyant and resist being pushed farther intothe well. In such cases, the technique isaltered by partial casing flotation, duringwhich the casing string is divided into twosections with the lower portion filled withair and the upper section filled with mud.The section filled with mud is in the near-vertical section of the well and providesweight to help push the lower, buoyant cas-ing into the well. A shear-out plug separatesthe air-filled and mud-filled sections of cas-ing. The plug holds the mud in the uppersection but can be opened with appliedpump pressure to circulate fluid through theentire casing string.

In a further prelude to Well M-11, the 95⁄8-in. casing was partially floated on WellM-08. The mud weight and rheology weredecreased prior to running the casing. Thefirst 2000 m [6600 ft] of casing were run withair. A shear-out plug was run in the casing atthat point, and the remaining 1500 m [4900ft] to surface was run and filled convention-ally. In addition to flotation, the casing was

133/8-in. casing at 1425 m

95/8-in. casing

Flotation collar

Double float collar

Side-jetted float shoe

121/4-in. openhole to 8890 m

Mud

Air

■■Partially floated 95⁄ 8-in. casing. Casing flotation experiments on Wells M-03, M-08 andM-09 proved the concept for the extreme test in Well M-11. The bottom section of casing,containing air, remained neutrally buoyant to allow the casing to slide more easilyalong the well path. The mud-filled upper section provided the weight necessary to pushthe entire string to bottom. Once the casing was landed, pump pressure was applied toshear the plugs in the flotation collars and allow circulation.

applied to the string. A set of bidirectional,hydraulic, flush-mounted slips held thebuoyed casing in the hole. The slips wereanchored to the rotary table and provided thenecessary hold-down force on the casing.

A Look at the FutureWell M-11 was completed with an electricalsubmersible pump and brought into produc-tion on January 12, 1998, at a rate of 20,000BOPD [3200 m3/d]. Wytch Farm productioncurrently averages more than 100,000 BOPD[16,000m3/d], 80% of which comes fromextended-reach wells.

The design of Well M-11 took more than ayear, and its completion provided an excel-lent test of the industry’s capabilities. The suc-cess of this well has opened up even moretargets and the potential to access reservesthat would have remained out of reach orrequired huge capital outlays just a few yearsago. The current focus at Wytch Farm is todrill 5-km stepout infill wells faster andcheaper than previous wells. Anotherextreme well, with a stepout of some 11 km[36,000 ft], is in the initial planning stages.Also under consideration are several 6- to 8-km multilateral wells.

All aspects of extended-reach technologyhave to move forward. The next step will bein completions and interventions, wherewellbore workovers and maintenance willbe critical.

The future for GeoSteering technology androtary steerable tools is bright. Currently,these steerable systems are used primarily onrelatively expensive extended-reach wellswhere they can provide a technical capabil-ity beyond the limit of standard motor-drivensystems. Here, these systems can be run eco-nomically even if their cost is high. Furtherwork will focus on increasing the reliability,upward telemetry systems and operating timeof these tools while cutting costs.

—KR

also rotated at various times during the oper-ation. Rotating the casing on Well M-08 wasmerely a test of the procedures before theyhad to be used on Well M-11. Once the cas-ing begins to float in the well, the actualtorque required for rotation is small. In fact,the actual running weight and torque closelymatched predicted values, proving that thesetechniques could be used effectively on theextreme-departure well.

This flotation method was used again onWell M-09 to run 95⁄8-in. casing to 6580 m[21,589 ft]. This flotation method was even-tually deployed on Well M-11 to run 95⁄8-in.casing to 8890 m [29,162 ft] measured depth(below). The casing was floated to 7080 m[23,228 ft] prior to installation of the flotationcollar. The casing was run into the openholefor another 20 stands prior to filling the uppersection of casing to achieve the additionalrunning weight.

To handle positive casing buoyancy safely, apush-tool was installed below the topdrive onthe rig. This tool engaged over the box con-nection of the casing and allowed the fullweight of the topdrive and blocks to be