Drilling and Testing Hot, High-Pressure Wells- Disk 2/media/Files/resources/oilfield_review/ors93/... · Pressure High Bottom Hole Temperature Exploration and Appraisal Wells: Supplementary

Drilling and Testing Hot, High-Pressure Wells

DRILLING

As cars get more powerful, driving is made safer by widening roads and enforcing speed limits.

Similar tactics are used to safely drill and test high-temperature, high-pressure wells. To meet extreme well

conditions, higher capacity hardware is deployed—the road is widened. Then, to maintain a speed limit,

tight controls are implemented to ensure that safety margins remain unbreached.

Robert MacAndrewRanger Oil (UK) LtdAberdeen, Scotland

Nigel ParryPhillips Petroleum Company United Kingdom LtdAberdeen, Scotland

Jean-Marie PrieurConoco (UK) LtdAberdeen, Scotland

Jan WiggelmanShell UK Exploration and ProductionAberdeen, Scotland

Eric DigginsBrunei Shell PetroleumBrunei

Patrick GuicheneyMontrouge, France

Doug CameronAdrian StewartAberdeen, Scotland

nThe HTHP challenge. High-temperature, high-pressure, usually deep and often corro-sive wells impose the most severe constraints on drillers and well testing engineers.

For help in preparation of this article, thanks to WillyBrandt, Sedco Forex, Aberdeen, Scotland; Andy Vigor,Schlumberger Wireline and Testing, Aberdeen, Scotland;Robbie Rounsaville and Benoit Vidick, SchlumbergerDowell, Aberdeen, Scotland.KickAlert, MDS (Management Drilling System), IDEAL(Integrated Drilling Evaluation and Logging system),CemCADE, and CSI (Combinable Seismic Imager) aremarks of Schlumberger.SideKick is a mark of Borland International; Fann is amark of NL Industries Inc.1. “Applications for Consent to Drill or Reenter High

Pressure High Bottom Hole Temperature Explorationand Appraisal Wells: Supplementary Information to beSupplied in Addition to That Required by CSON 11,”Continental Shelf Operations Notice No. 59, Department of Energy, London, England, May 1990.In 1992, the UK Department of Energy (DOE) wassubsumed into the Department of Trade and Industry. At about the same time, some of the DOE’sduties were transferred to the UK Health and SafetyExecutive.

April/July 1993

High-temperature, high-pressure (HTHP)wells present special challenges to drill andtest (above). Predominantly gas producers,HTHP wells may yield significant reservesin some areas. But the wells stretch conven-tional equipment beyond normal opera-tional capacities. To safely meet theseextreme conditions, traditional procedureshave been modified and extra operationalcontrols devised.

What constitutes HTHP is debatable. Per-haps the best definition has been coined bythe UK Department of Energy:

“Wells where the undisturbed bottomholetemperature at prospective reservoir depthor total depth is greater than 300°F [150°C]and either the maximum anticipated porepressure of any porous formation to bedrilled exceeds a hydrostatic gradient of 0.8psi/ft or pressure control equipment with arated working pressure in excess of 10,000psi is required.”1

HTHP drilling is not new. In the late 1970sand early 1980s, many gas wells were

15

nUK CentralGraben. Reservesin this area may be a source of gasto replace thedeclining produc-tion from the UKSouthern Basin.The CentralGraben has theadditional attrac-tion of relativelyrich condensatedeposits—1 to 2barrels of fluid per10,000 standardcubic feet of gas.However, to fullyexploit this poten-tial, HTHP wells are required.

North Sea

Shetland

East Shetland Basin

BergenNorway

StavangerNorwegiansector

St. Fergus

Aberdeen22 7 8

2

Danishsector

Central Graben

1

3029

39

A B

Germansector

Dutchsector

The Netherlands

Amsterdam

RotterdamGermany

The Southern Basin

Great Yarmouth

London

UK

UKsector

Edinburgh

E S

drilled in the Tuscaloosa trend, Louisiana,USA, and other southern US states. Theseencountered temperatures above 350°F[177°C] and pressures of more than 16,000psi, not to mention highly corrosive environ-ments.2 When HTHP interest switched to theNorth Sea in the mid 80s, new hazards wereintroduced. The wells were drilled offshore,in extremely hostile conditions and some-times using floating semisubmersible rigsrather than fixed jackups.

Most of the North Sea HTHP wells are sit-uated in the Central Graben—a series ofdownthrown and upthrown blocks (above).3

16

The Central Graben contains several Juras-sic gas condensate prospects at 12,000 to20,000 ft [3660 to 6100 m], with pressuresof 18,000 psi or more and temperatures ofup to 400°F [205°C].

Water depth in the Central Graben variesbetween 250 to 350 ft [75 to 105 m]. Bothjackups and semisubmersibles have suc-cessfully drilled wells in the sector, harshenvironment jackups up to about 300 ft [90m] and semisubmersibles for deeper water.Jackups offer the advantage of contact withthe seabed, eliminating heave and simplify-ing many drilling and testing operations. Onthe downside, in an emergency, jackupscannot be moved off location quickly. Also,few deepwater jackups are available.4

This article looks at three key areas ofHTHP operations in the UK Central Graben:drilling safety, casing and cementing, andtesting. It also examines how North Seaexperience has been used to help convert ajackup to drill demanding wells off Brunei(see “Readying a Jackup For Brunei’s HTHPWells,” page 18).

Drilling SafetyPreventing and controlling influxes of reser-voir fluid into the well—called kicks—arealways central to drilling safety, but inHTHP wells the dangers from a kick areamplified. The volume of a HTHP gas kickremains virtually unchanged as it rises in theannulus from 14,000 to 10,000 ft [4265 to3050 m]. From 10,000 to 2000 ft [610 m]its volume triples. But from 2000 ft to thesurface, there is a hundred-fold expansion.

Put simply, a gas influx of 10 barrels at14,000 psi becomes 4000 barrels underatmospheric conditions. As reservoir fluidrapidly expands, it forces mud out of thewell—unloading—reducing mud in thewell, cutting hydrostatic pressure at the for-mation, allowing additional reservoir fluidsto enter, and ultimately causing a blowout.

Wells drilled in the Central Graben haveanother complication—an unpredictableand sharp increase in pore pressure over ashort vertical interval, sometimes less than100 ft [30 m]. And, while the pore pressuremay rise rapidly, the fracture pressure doesnot. In some cases, convergence of poreand fracture pressures means that a smalldecrease in the mud weight of 0.5 poundsper gallon [lbm/gal] or less changes thewell from losing circulation to taking a kick(next page).5

The difficulty of drilling in the CentralGraben was highlighted in September 1988when a blowout on the semisubmersibleOcean Odyssey resulted in fire and loss oflife.6 Consequently, the UK Department ofEnergy essentially banned the drilling andtesting of prospects with anticipated reser-voir pressures exceeding 10,000 psi.

As a result, the UK Offshore OperatorsAssociation (UKOOA) collated the experi-ences of those involved in HTHP wells anddrew up guidelines.7 In addition, in 1992,the Institute of Petroleum (IP), London, Eng-land, published a comprehensive set of rec-ommended practices to provide informationand guidance on HTHP well control activi-ties.8 With these two sets of guidelines,drilling and testing has resumed.

Before an HTHP well is spudded, contin-gency plans are made. The maximum vol-ume and flow rate of formation fluid, andassociated peak temperatures and pressuresare anticipated for a number of worst-casescenarios. Well-control hardware may then

Oilfield Review

2. Huntoon GG: “Completion Practices in Deep SourTuscaloosa Wells,” Journal of Petroleum Technology36, no. 1 (January 1984): 79-88.Schultz RR, Stehle DE and Murali J: “Completion ofa Deep, Hot, and Corrosive East Texas Gas Well,”paper SPE 14983, presented at the SPE DeepDrilling and Production Symposium, Amarillo,Texas, USA, April 6-8, 1986.

3. JM Peden: “The Central Graben—The Next NorthSea Challenge,” presented at the Petroleum Scienceand Technology Institute Seminar, The CentralGraben, The Next North Sea Challenge, The Needfor Joint Industry Research, Edinburgh, Scotland,September 10, 1990.

4. Low E and Seymour KP: “The Drilling and Testing ofHigh-Pressure Gas Condensate Wells in the NorthSea,” paper IADC/SPE 17224, presented at the I988IADC/SPE Drilling Conference, Dallas, Texas, USA,February 28-March 2, 1988.Seymour KP and Mackay A: “The Design, Drillingand Testing of a Deviated High-Temperature, High-Pressure Exploration Well in the North Sea,” paperOTC 7338, presented at the 25th Annual OffshoreTechnology Conference, Houston, Texas, USA, May3-6, 1993.

5. Ross I, Seymour K and Whyte B: “The Role of Jack-ups and Permanent Completions in Central GrabenExploration Drilling,” presented at the PetroleumScience and Technology Institute Seminar, The Cen-tral Graben, The Next North Sea Challenge, TheNeed for Joint Industry Research, Edinburgh, Scot-

nLithology versuspressure in theCentral Graben. Inthe highlightedarea, an increasein pore pressurefrom 13.5 to 17.5lbm/gal can occurover an interval ofless than 100 ft, yetthe fracture gradi-ent of the perme-able formationsremains below18.5 lbm/gal. The increase mayoccur in the LowerCretaceous, Kim-meridge clay orthe top of the reser-voir sands, but thelocation and mag-nitude are difficultto predict.

KimmeridgeClay

Tertiary

Jurassic

Paleocene

Cretaceous

Hod

Bryne

Und

iffer

entia

ted

UpperTriassic

10 12 14 16 18 20

Pressure gradients, lbm/gal

Pore pressure

Frac pressure

Mud weight

App

roxi

mat

e de

pth,

ft

Lower Cret.

500

2500

4500

6500

8500

10,500

12,500

14,500

7. “Procedures for High-Pressure Drilling,” UK OffshoreOperators Association Drilling Practices Committee,no. 2, London, England, January 1990.Hoopingarner JB, Greif CV, Neme EE, Rodt GM andBates TR: “Rig Modifications Meet New UK High-Pressure Requirements,” paper IADC/SPE 19976,presented at the 1990 IADC/SPE Drilling Confer-ence, Houston, Texas, USA, February 12-March 2,1990.

8. “Well Control During the Drilling and Testing ofHigh Pressure Offshore Wells,” Model Code of SafePractice, Part 17. London, England: Institute ofPetroleum, 1992.

9. Prieur JM: ”Control Aspects of Drilling High PressureWells,” paper SPE 19245, presented at OffshoreEurope, Aberdeen, Scotland, September 5-8, 1989.

10. The majority of wells have been drilled using low-toxicity OBM, which is relatively easy to design forHTHP duty. Today, more environmentally safewater-base muds (WBM) are being used. These tendto be more difficult to use in HTHP wells since theymay gel or thin. WBM sometimes proves problem-atic because of difficulty in maintaining fluid properties. “Well Control When Drilling With Oil-Based Mud, Recent Experience in Deep Wells,” UKDepartment of Energy, Offshore Technology Report,OTH 86 260, London, England: HMSO 1990.

be designed to cope with these worst casesfor at least an hour, the minimum timeneeded to evacuate a rig.

Because the consequences of failure inHTHP wells are so great, worst-case scenar-ios tend to be more conservative than fornormal wells. Usually, the maximum antici-pated size of a kick is set at the limit ofdetection—often 10 to 20 barrels. In HTHPwells, many contingency plans are based onthe worst case of an influx completely fillingthe well at reservoir pressure.9

When drilling with oil-base mud (OBM),there is a likelihood that gas entering thewellbore will dissolve into the mud’s oilphase.10 This affects the how the kick movesup the annulus and may mask detection.Since 1986, researchers at SchlumbergerCambridge Research (SCR), Cambridge,England, have been studying the behavior ofgas kicks, particularly in OBM. This workhas resulted in Anadrill’s SideKick software

(continued on page 20)

land, September 10, 1990.6.”Ocean Odyssey Blowout Fuels UK Sector Safety

Debate,” Offshore Engineer (October 1988): 11.

17April/July 1993

Readying a Jackup forBrunei’s HTHP Wells

Brunei Shell Petroleum has a number of offshore

drilling prospects with formation pressures

exceeding 10,000 psi. The first well in an HTHP

campaign was spudded in August 1992 and com-

pleted in December. These wells are expected to

be about 13,125 ft [4000 m] deep with bottomhole

pressures greater than 14,000 psi and bot-

tomhole static temperature peaking at approxi-

mately 300°F.

At the planning stage, one of the first consider-

ations was choice of rig. With no suitable unit in

the sector, Brunei Shell could have mobilized a

North Sea rig. But the milder weather offshore

Brunei does not merit such ruggedized units with

high mobilization costs. A more cost-effective

option was to upgrade a rig already in the sector.

Attention turned to the jackup Trident XII, oper-

ated by Sedco Forex and contracted by Brunei

Shell since February 1990 (next page). Prior to

modification, the rig could drill in water up to 300

ft deep, had topdrive and was fitted with the MDS

computerized drilling monitoring system. But it

could handle only a maximum wellhead pressure

of 10,000 psi. By early January 1992, the scope of

18

work was defined to bring the rig up to 15,000-psi

status. The plans were based largely on North

Sea experience, and included:

• Replacement of the 20 3/4-in., 3000-psi BOP by

a 21 1/4-in., 5000-psi BOP—necessary because

the large diameter casings were to go deeper;

replacement of the 13 5/8-in., 10,000-psi BOP

by a 13 5/8-in., 15,000-psi BOP; and an upgrade

of the hydraulic control unit and the

handling systems to accommodate the larger

BOPs.

• Upgrading of the choke, kill and cement lines to

handle 15,000 psi. Also replacement of choke,

kill, and safety valves by equipment with

15,000-psi working pressure.

• Installation of a 15,000-psi choke pressure

manifold and addition of a glycol injection unit.

• Upgrade of the MDS system to monitor HTHP

parameters. To achieve this, temperature and

pressure sensors were installed in strategic

positions.

From agreeing on this plan to spudding the first

HTHP well took only six months—including

drilling a normally-pressured “shakedown” well.

Rig work took three months and included an

extensive program of inspection and maintenance

on key rig equipment not strictly part of the high-

pressure upgrade—like the topdrive, drawworks

and blocks.

To meet the schedule, Sedco Forex determined

equipment needs while engineering the modifica-

tions. Although time was vital—especially con-

sidering the increased delivery time on some

high-pressure components—care was taken to

ensure equipment was delivered with a specific

quality file including traceability, interim inspec-

tion, test reports and certificates of conformance.

Some critical items were inspected and pres-

sure tested by certification authorities. Designs

for the structural modifications to the rig—like

the upgrade of the BOP handling system—were

also reviewed and approved by an authority,

which also surveyed the work when completed.

Because this was an upgrade and not a new

build, available space could rarely be increased,

and never by much. For example, the 15,000-psi

choke manifold weighs twice its 10,000-psi coun-

terpart, yet had to be located at the same place

and offer increased circulating options—both

lines from the BOP to the choke manifold can be

either a choke or a kill line.

The MDS system, which had been installed two

years previously, was upgraded to monitor more

parameters in real time with screens on the rig

floor, and in the operator representative and tool-

pusher offices. New measurements included:

• Pressure and temperature upstream of the

chokes, on two flow paths as choke and kill

lines were dual purpose, to ensure BOP and

valve elastomer ratings are not exceeded.

• Pressure and temperature downstream of the

choke. Temperature monitoring enables detec-

tion of possible hydrate formation beyond the

choke, indicating when glycol injection is

required. Pressure monitoring ensures the

10,000-psi rating downstream of the choke

manifold is not exceeded.

• Mud-gas separator (MGS) pressure. This

ensures that low-pressure vessel capabilities

are not exceeded. If the pressure approaches

the limit, a hydraulic valve is opened to divert

mud overboard.

Oilfield Review

But it did more than just ready the fabric of the

rig. The group also prepared personnel who

would be involved in the drilling. First, it drew up

a set of HTHP drilling and well control procedures

to fit local conditions and equipment. Particular

attention was focused on stripping drillpipe into

the well (see page 24). Second, it coordinated

HTHP-awareness sessions for project personnel.

Sedco Forex rig manager. Additional representa-

tives—like geologists, equipment engineers and

production engineers—attended when needed.

The team’s work was judged a key element in the

smooth startup and safety of subsequent drilling.

The group fine-tuned technical matters of the

rig upgrade—for instance, new requirements for

the MGS system—and monitored the timing of

project landmarks and their impact on the startup

date. It established a rig commissioning list to

ensure the upgrade met objectives of the initial

plan and that the rig was ready to safely start the

HTHP campaign.

• MGS seal height. This ensures that the operat-

ing capabilities of the separator are not

exceeded. If this happens, mud flow is reduced

or the line overboard opened.

From the beginning, a joint Shell-Sedco Forex

HTHP team was established and met regularly

during the six-month project. The team included a

core of representatives from Brunei Shell, partic-

ularly from the drilling department, and the

19April/July 1993

n The jackup Trident XII

tine is usually adopted to check whetherswabbing will cause an influx.

Before the assembly is pulled out ofhole, the mud at the bit is circulated tosurface—a procedure called circulatingbottoms up. If this is free from gas, tenstands of drillpipe are pulled. The string isthen run back to total depth (TD), andbottoms up are circulated. Gas in the mudis measured again, with an increase indi-cating swabbing.

If swabbing does cause an influx, themud weight may be raised slightly andthe string pulled out of hole more slowly.Also, circulating mud while pulling out ofhole helps stop swabbing—a processmade easier by topdrive. If the well isbeing drilled from a semisubmersible,severe vessel heave may swab the well. Ifheave becomes too great in stormyweather, drilling may have to stop untilconditions improve.

•The combination of relatively high-viscos-ity mud, deep wells and small annular

model which simulates gas kicks and maybe used to plan methods of controllingHTHP wells (below).11

Planning requires realistic data: well tem-perature profile, nature of the anticipatedreservoir fluids, expected maximum bottom-hole pressure and pressure gradient, androck strength and permeability. These aremost often estimated using offset data—rela-tively plentiful in the North Sea. But whereoffset data are sketchy, predictive modelingmay be employed. Koninklijke/Shell Explo-ratie en Produktie Laboratorium (KSEPL),Rijswijk, The Netherlands, has developed amodel to predict rock strength and porepressure in many areas of the North Sea.KSEPL has also modified a model designedto predict wellhead temperatures in offshoreproduction wells to estimate surface equip-ment temperature when controlling a kick.

Worst-case scenarios are used not only tospecify equipment but also to draw up spe-cific operational procedures, for exampledetailing what to do if the well takes a kick

(next page). Training is then used to com-municate these procedures to drilling per-sonnel. Long before drilling starts, specificHTHP training courses may be run. Onceon the rig, there are prespud meetings, crewsafety meetings before starting key sectionsof the well, preshift meetings to discuss thecurrent situation, and regular drills to prac-tice important techniques like closingblowout preventers (BOP).12

The three issues at the heart of HTHPdrilling safety are kick prevention, kickdetection and well control.

Kick Prevention: The best way of avoidingwell-control problems is to anticipate situa-tions known to precipitate kicks and takepreventive action. Here are four examples:•When high-pressure formations are drilled,

kicks commonly occur when the drillingassembly is being pulled out of hole. Themovement of the assembly creates a pis-ton effect reducing pressure below the bit,called swabbing. A time-consuming rou-

14,0

0010

,000

6000

2000

Time

Free gas

Dissolved gas

Free gas

Dissolved gas

Free gas

Dissolved gas

10%0 10%010%0 10%0 10%0 10%0

Mea

sure

d de

pth,

ft

nInflux simulationusing SideKick soft-ware showing thedevelopment of agas kick of about10 barrels at14,000 ft as it is cir-culated out of thewell. Mud is blue;gas in solution,red; and free gasis white.

20 Oilfield Review

12. Lindsay G and White D: “The Use of a Gas KickSimulator to Produce an Oil Based Mud TrainingPackage,” presented at the 1993 IADC EuropeanWell Control Conference, Paris, France, June 2-4, 1993.Hornung MR: “Kick Prevention, Detection, andControl: Planning and Training Guidelines forDrilling Deep High-Pressure Gas Wells,” paperIADC/SPE 19990, presented at the I990 IADC/SPEDrilling Conference, Houston, Texas, USA,February 28-March 2, 1990.

11. Leach CP and Wand PA: “Use of a Kick Simulator asa Well Planning Tool,” paper 24577, presented atthe 67th SPE Annual Technical Conference andExhibition, Washington, DC, USA, October 4-7,1992.

21April/July 1993

nGeneralized decision tree for controlling kicks.

clearances leads to higher than normalfriction pressure during mud circulation.At the formation, mud hydrostatic pres-sure and friction pressure then combine togive the equivalent circulating density(ECD). This may be designed to balanceformation fluid pressure. But during aconnection, mud flow stops and frictionpressure is zero. With reduced ECD, smallquantities of gas, called connection gas,may permeate from the formation. If bot-toms-up circulation time exceeds the timeto drill to the next connection point, gasthat entered during the previous connec-tion may remain undetected. Additionalgas may then enter as another connectionis made, significantly increasing risk of aserious kick. The safe procedure is toensure that bottoms up has been circu-lated before making the next connection.

•Kicks don’t occur just during drilling. Cor-ing also causes problems. The relativelysmall clearance between core barrel andopen hole increases the possibility ofswabbing when pulling out of hole. Thismay be combated by limiting the amountof core cut at any one time—usually to 30ft [10 m] or less—and pulling out of holevery slowly, checking for flow and moni-toring gas in the mud.

•Tight margins between pore pressure androck strength, as in the Central Graben,make lost circulation common, compli-cating well control. The normal practiceon encountering losses is to pump lost cir-culation material (LCM) in the mud. IfLCM fails to block the formation, the strat-egy is to pull out of hole, run back in withopen-ended pipe and spot cement acrossthe loss zone. Some slurry is squeezedinto the formation and, once set, theremainder drilled out. However, in HTHPwells, the swabbing effect of pulling thebottomhole assembly out of hole prior tospotting the plug may induce a kick else-where in the wellbore. In this case, theonly solution is to spot the cement plug

Well influx

Operation inprogress

Pulled out of hole,no pipe in BOP

Closeshear rams

Openchoke line

Check surfacepressures

Tripping,bit off bottom

Install openkelly cock

Closekelly cock

Close annularpreventer

Openchoke line

Install killassembly & test

Checkspace out

Close 5-in.pipe rams

Land string,close Posilocks

Openkelly cock

Yes

Yes

No

No

Observe well

Muster all crewsfor information

Prepare to kill well

Withdraw allwork permits

Advisestandby boat

Drop string,wait then close

shear rams

Drillingbit on bottom

Raise kelly

Stop pump

Close annularpreventer

Openchoke line

Closekelly cock

Install killassembly & test

Checkspace out

Close 5-in.pipe rams

Land string,close Posilocks

Pressure up to shutin drillpipe pressure

Open kelly cock

Collarsin BOP?

Upwardforce acting on

collars greater thanstring weight?

through the bit (below). To make this eas-ier, rotary drilling is favored, rather thanusing a downhole motor that may clog upwith cement.

Kick Detection: Because no technique canguarantee kick-free drilling, influx detectionremains vitally important. Traditional influxdetection relies on observing mud levelincreases in the mud pits, or performing flowchecks—stopping drilling to see if the well isflowing. Comparisons of mud flow rates intoand out of the well are also used. To makedetection more reliable, transfer of mud intothe active system is tightly controlled andusually not allowed while drilling.

Recently, Anadrill has introduced theKickAlert early gas detection service basedon the principle that acoustic pulses createdby the normal action of the mud pumpstravel more slowly through mud containinggas they do through pure mud. The pulsesare measured as pressure variations at thestandpipe as the mud enters the well and atthe annulus as it comes out. If the well is

stable and no gas is entering, the phase rela-tionship between the pressure pulses in thestandpipe and annulus is constant, orchanges gradually as the well is drilleddeeper. When gas enters, the pulses travelmuch more rapidly up the annulus, dramati-cally changing the phase and setting off analarm on the drillfloor.

The presence of high-pressure gas mayalso be indicated by changes in drilling con-ditions. Increases in rate of penetration,torque or mud temperature in the mudreturn flowline on surface may all signifythe onset of a kick. Computerized monitors,like Sedco Forex’s MDS rig information sys-tem and Anadrill’s IDEAL Integrated DrillingEvaluation and Logging system, help drillingpersonnel keep track of trends and spotabnormal situations using quick-look inter-pretations on a drillfloor screen.13

Well Control: As soon as a kick is detected,drilling is stopped and the well is shut in.The influx must then be circulated outwhile keeping the pressure under control(next page).

The BOPs are the primary means of wellclosure. Once a kick is suspected, the annu-lar blowout preventer is first closed. A flexi-ble rubber element is inflated usinghydraulic pressure, and is sufficiently flexi-ble to seal around any downhole equip-ment. When it has been established that notool joints are in the way, the pipe rams arethen shut, sealing around the drillpipe. Nowmud can no longer return through the flow-line to the shale shakers and mud pits.Instead it must travel through the chokelineto the choke manifold, which is used torelieve mud pressure at surface.

Most operators favor what is called a“hard” shut-in—closing BOPs with thechoke already closed. Sometimes “soft” clo-sure is used—the choke is closed only afterthe BOPs have sealed. Some believe thatthis reduces the hydrostatic shock to the for-mation, but it has the severe disadvantage ofdelaying closure and allowing additionalformation fluid to enter the wellbore. It isalso a more complex procedure, increasingthe likelihood of errors.14

The capacity of a BOP to resist pressuredepends on the elastomeric seals inside therams and their likelihood of not beingextruded. As temperature increases, extru-sion becomes more likely. Seals may haveto withstand prolonged temperatures thattop 400°F [205°C]—beyond the limits ofordinary components. Finite-element analy-sis has been used to identify which areas ofthe BOPs are most affected by heat andwhich seals need special elastomers rated to350°F.15 Sometimes, special BOP tempera-ture monitors are used to ensure theseextended limits are not breached. However,high-temperature elastomers are harder thantheir low-temperature counterparts and maynot seal at ambient temperature, makingsurface pressure tests difficult.

Once the BOPs and choke are closed,pressure builds in the annulus and drillpipe.The maximum drillpipe pressure is used tocalculate bottomhole pressure, which isused to plan the kill strategy. Well-kill strat-egy also takes into consideration the drillingoperation underway during the kick.16

22 Oilfield Review

Cement Lost circulation zone

Mud

200 ft

nSpotting a plug without cementing the pipe in the hole. With the bit 100 to 200 ft above the loss zone, preferably insidethe previous casing shoe, 5 barrels of mud-base oil, 20 barrels ofspacer, 50 to 100 barrels of cement and a further 20 barrels ofspacer may typically be pumped until the base oil reaches the bit. The annular BOP is closed and the plug squeezed untilthe top of cement reaches 50 ft above the top of the loss zone.The last 10 barrels are displaced down the annulus to ensurethe bit is free of cement. While the cement is setting, the pipe isworked up and down with the annular preventer closed. Toavoid plugging the bottomhole assembly due to gelation, sedi-mentation or premature setting, extensive tests are needed toensure that the slurry has uniform, predictable properties atdownhole conditions.

If the kick occurs during drilling, weightedmud—either from a premixed or speciallyprepared supply—may immediately bepumped down the drillpipe. The formationfluid influx is gradually displaced up theannulus, expanding as hydrostatic pressuredecreases. At surface, the mud-influx mix-ture travels to the choke manifold via thechokeline and has its pressure reduced bythe choke. The well is slowly brought undercontrol by carefully selecting mud weightand choke opening.

It is vital that the surface drilling andpumping equipment withstands the pres-sures during the kill—for example, the kellyor topdrive are usually rated to only 5000psi. The surface pressure during the kill isestimated by adding the shut-in drillpipepressure to the friction pressure of the fluid

as it is pumped into the well. Friction pres-sure is routinely measured by the drillingcrew at the start of every 12-hour shift. Ifcalculations show surface limitations will bebreached, a kill assembly rated to 15,000psi must be temporarily installed after thewell has been shut in. Two valves in thedrillstring just below the kelly or top-drive—called surface valves or kellycocks—are closed to seal inside the drill-string allowing removal of the topdrive orkelly and installation of the kill assembly.

Controlling a kick while the drillstring isbeing pulled out of hole is less straightfor-ward. Reservoir fluid enters below the bot-tom of the drillstring and cannot be circu-lated out of the well until drillpipe is runback in hole below the kick. Running indrillpipe through a closed BOP is calledstripping and requires careful coordinationof several critical operations (next page).Once stripped back in hole, a conventional

13. Weishaupt MA, Omsberg NP, Jardine SI and Patter-son DA: “Rig Computer System Improves Safety forDeep HP/HT Wells by Kick Detection and WellControl Monitoring,” paper SPE 23053, presented atthe Offshore Europe Conference, Aberdeen, Scot-land, September 3-6, 1991.

14. Jardine SI, Johnson AB, White DB and Stibbs W:“Hard or Soft Shut-in: Which is the Best Approach?,”paper SPE/ IADC 25712, presented at the 1993SPE/IADC Drilling Conference, Amsterdam, TheNetherlands, February 23-25, 1993.

15. McWhorter DJ: “High-Temperature Variable BoreRam Blowout Preventer Sealing,” paper OTC 7336,presented at the 25th Annual Offshore TechnologyConference, Houston, Texas, USA, May 3-6, 1993.

16. Jardine SI, White DB and Billingham J: “Computer-Aided Real-Time Kick Analysis and Control,” paperSPE/ IADC 25711, presented at the 1993 SPE/IADCDrilling Conference, Amsterdam, The Netherlands,February 23-25, 1993.

23April/July 1993

nThe well-control layout. Normally, drilling mud returns to the pits via the flowline and the shale shakers. If the well takes a kick,the BOPs shut off flow, directing it through the chokeline to the choke manifold and then to the mud-gas separator (MGS).

Inside the MGS, gas is safely vented and the mud returned to the mud system. Fluid in the MGS acts as the only barrier preventinggas from forcing its way into the mud room. Separator seal pressure is reduced as heavy mud is diluted by the relatively light hydrocarbons from the influx. In some MGS systems, fresh mud may be circulated down the seal leg, displacing the hydrocarbonsand ensuring a tight seal. BOPs have a combination of annular preventer and rams—the exact choice of elements varies. At hightemperature, normal elastomers that seal inside the rams may soften and be extruded. For HTHP drilling, harder high-temperature elastomers are usually used.

Vacuum degasser

Divert flow overboard

Inlet for fresh mud injection

Choke inlet line

Secondary vent

Overboard lines

Mud flow when shut-inChokeline

Choke

Trip tank

Blind rams

Shear rams

Slip rams

Pipe rams

Kill line

Annular preventer

Stripper

Rotary table

Flowline

Kill pump

Shaker room tanks

Vent to top of derrick

Top sealPacker

BOP ram elastomersMud flow during normal drilling

Kill well

Mud MixGas

Mud gas separator

Active and reserve mud pits

nStrippingdrillpipe into thewell. After theannular preventerin the BOP stack isclosed around thedrillstring, newjoints of pipe areadded to the stringand forced throughthe preventer intothe well. While thisis happening, theinternal pressure ofthe drillpipe mustbe controlled. Twokelly cocks or sur-face valves arealways included inthe drillstring, justbelow the topdriveor kelly. For strip-ping, they areclosed, sealinginside the drillpipe,then the elevatorsare removed (A). A set of internalBOPs is fitted insidethe pipe. This actsas a one-way valveholding pressurefrom below whileallowing flow topass from above.In this way, thekelly valves maybe reopened andjoints of pipeadded withoutexposing thedrillfloor to a flow-ing well (B). Suffi-cient pipe is addeduntil the bit isbelow the influx,which may then becirculated out (C).

This processmust take intoaccount theincreased volumeof the string asjoints are added,and the increasedvolume of thereservoir fluid as ittravels up the well.Careful control ofthe choke isneeded to bleed offcalculated vol-umes of mud andmaintain the rightbackpressure.

24 Oilfield Review

A B C

Kelly valves closed

Internal BOP closed

Kelly valves open

New joint of drillpipe

17. Low E and Jansen C: “A Method for Handling GasKicks Safely in High-Pressure Wells,” paperSPE/IADC 21964, presented at the 1991 SPE/IADCDrilling Conference, Amsterdam, The Netherlands,

nIn search of the ideal casing point. Insome cases, the preferred place for the 9 5/8-in. casing shoe is in the upper sec-tion of the Kimmeridge clay, which issometimes as little as 50 ft [15 m] thick intotal. But there are potential high-pressuresand lenses within the clay. This maymean setting the shoe in the fracturedHod chalk. In this case, the aim is to havethe shoe at the bottom of the zone, wherefractures are least prevalent.

Tertiary

Jurassic

Paleocene

CretaceousHod

Bryne

Und

iffer

entia

ted

UpperTriassic

KimmeridgeClay

App

roxi

mat

e de

pth,

ft

500

2500

4500

6500

8500

10,500

12,500

14,500

30 in.

20 in.

133/8 in.

95/8 in.

7 in.

Lower Cret.

kill operation may be started and the influxcirculated to the choke manifold.

For HTHP drilling, the choke manifold istypically rated to 15,000 psi, while equip-ment downstream of the choke is rated to5000 psi, sometimes 10,000 psi. Redun-dancy demands no fewer than two flowpaths, so the manifold comprises at leasttwo chokes. A large pressure drop across thechoke may result in adiabatic cooling of thefluid on the low-pressure side, and coolingleads to development of hydrates andplugged lines. To prevent this, glycol ormethanol must be injected upstream of thechoke. Remote sensing of upstream anddownstream choke temperature and pres-sure enables monitoring of potential hydrateformation.

Once fluid pressure has been reduced bythe choke, returns are routed in one of twodirections: diverted overboard or to themud-gas separator—also known as thepoor-boy separator. Flow is normallydiverted overboard only in an emergency—for example, a blocked line.

The MGS is traditionally employed tovent small volumes of free gas in the mud,preventing frothing and reducing the loadon the subsequent vacuum degasser. Insidethe MGS, internal weirs allow the gas toseparate from the mud and be vented (page23). A small hydrostatic head of less than 15psi prevents gas from forcing its waythrough the MGS into the mud tanks. ForHTHP duty, pressure sensors warn when thegas pressure threatens to exceed the hydro-static head so that the well may be shut inuntil the liquid seal is reestablished.17

In some cases, the influx may be so large,the anticipated pressure so high or the con-centration of hydrogen sulfide (H2S) sogreat, that circulating the reservoir fluid tosurface is not considered safe. In these rarecases, the only alternative is to squeeze thefluid back into the formation—a procedurecalled bullheading.

Bullheading presents additional chal-lenges. It may involve considerably highersurface pressures and pump rates than cir-culation and may pressure up or fracturethe formation. Also, the fractured forma-tion may not be the source of the kick, cre-ating the conditions for an underground

April/July 1993

blowout—reservoir fluid flowing intoanother formation. Once bullheading starts,it may be difficult to keep track of what flu-ids are in the well and where they are. Fur-thermore, the operation may burst casingthat has been weakened during drilling.

Casing and CementingIn Central Graben wells, choosing the loca-tion of the intermediate—usually 95/8-in.—casing shoe is crucial. Ideally, casingmust be set above the high-pressure reser-voir and just below a zone of weak Hodchalk (right). If it is set too high, the weakformation will be exposed to subsequenthigh pressure, and the only solution is to seta short, perhaps less than 100 ft, 7-in.drilling liner. This has the undesirable effectof reducing the diameter of further drilling.

Consequently, the shoe is usually set atthe bottom of the Hod chalk, in the LowerCretaceous clays or in the Kimmeridge clayjust above the reservoir. But finding the cas-ing point is not easy—vertical seismic pro-file surveys may be employed.18 As drillingapproaches the likely casing point, it isintermittently halted, bottoms up circulatedand cuttings examined by a geologist ormicropaleogeologist.

Pressure is a key consideration whendesigning the casing string. The 95/8-in. cas-ing is often designed to withstand completeevacuation to atmospheric pressure withreservoir pressure in the annulus betweenopen hole and casing. Given the high pres-sures, heavy-weight casing is usuallyrequired throughout the string.

H2S is another consideration. At high tem-peratures, the corrosive gas does not affectsteel. But when temperatures fall belowabout 185°F [85°C] higher up the well, cas-ing may be prone to attack. To combat this,special-grade steel is required. This mayreduce casing inside diameter preventingpassage of an 81/2-in. bit. The solutionfavored by many North Sea operators is touse thick-walled, special-grade steel 97/8-in.or even 103/4-in. casing higher up the welltapering to 95/8 in. as the temperature

March 11-14, 1991.18. Meehan R, Miller D, Haldorsen J, Kamata M and

Underhill B: “Rekindling Interest in Seismic WhileDrilling,” Oilfield Review 5, no. 1 (January 1993): 4-13.

25

nEffects of mud contamination on slurrystrength after 8- and 16-hour setting times.

Neat Class H Cement (16.5 lbm/gal)

Mud Compressive strength

contamination (psi at 170° F)

(% by volume) 8 hr 16 hr

0 4547 5862

5 3512 5300

10 2519 4538

20 2378 2331

50 245 471

increases and H2S attack ceases to be a dan-ger.19 However, if the design is too conser-vative, the casing string may become soheavy that the rig cannot bear the hook loadto run the casing.

Once the casing string has been run, theshoe must be cemented to resist the highreservoir pressure that will be encounteredalmost as soon as the next section of drillingstarts. Location of the top of cement (TOC)of the 95/8-in. casing is sometimes an issue.In normal wells the TOC is usually abovethe previous casing shoe, with fluid trappedin the annulus above the TOC. When HTHPwells are drilled, hot mud passing up thedrillpipe-casing annulus heats fluid in thecasing-casing annulus, causing it to expand.For a subsea-HTHP well, the pressure hasno escape and it can burst or collapse thecasing. For this reason, the TOC for 95/8-in.casing in a HTHP well is sometimes keptbelow the 133/8-in. shoe to allow annularpressure to dissipate into the formation.

In most HTHP wells a 7-in. liner is run,although in some cases it may be possibleto cement a 7-in. casing to surface. In eithercase, the cement job must isolate the high-pressure zones to facilitate well testing. Thisrequires good cementing practices and acarefully designed slurry.

Mud removal is vital in achieving strongcement bonding to the formation and cas-ing, and sealing against high pressure. Evensmall quantities of contaminant in thecement slurry compromise the final settingstrength (above, right). Spacers reduce con-tamination, but high temperatures may thinor destroy spacer polymers causing weight-ing agents to settle. An emulsified Dowellspacer or XC polymer, both weighted byhematite, have been successfully employed.

26

The tight pressure constraints found inHTHP wells mean that the traditional den-sity hierarchy—cement heavier than themud with an intermediate spacer—is diffi-cult to achieve without exceeding formationfracture pressure. A viscosity hierarchy isalso desirable, but when cement is thickerthan mud, the friction pressure may increasebeyond the limit. This emphasizes theimportance of other good drilling andcementing practices: drilling an even well-bore, circulating and conditioning the mudcorrectly, and centralizing the casing.20

Cement slurry rheology is usuallydesigned to give the best flow regime formud removal using allowable pump rates.Although turbulent flow is normally pre-ferred, pump rate restrictions often result inthe use of effective laminar flow. A majorconstraint on pump rate in HTHP wells isthe high friction pressure in the 7-in. linerannulus. To predict friction pressures andkeep ECDs below the formation fracturegradient, a Fann Model 70, HTHP viscome-ter is used to measure slurry rheologies atup to 500°F [260°C] and 20,000 psi. Thesedata may then be fed into design softwarelike Dowell’s CemCADE service.21

Besides rheology, HTHP cement slurrieshave many other design considerations: •Some cement slurries in HTHP wells

require long setting times. Depths of morethan 15,000 ft [4575 m], high frictionpressures and low fracture gradients meanlong pumping times—in some cases 7hours or more. Bottomhole static tempera-tures (BHST) are high and accurate esti-mation of the bottomhole circulating tem-perature (BHCT) is vital to ensure that theslurry sets at the right time.

BHST—usually defined as the tempera-ture 24 hours after the last circulation—is

measured during logging and convertedinto BHCT using American PetroleumInstitute (API) conversions.22 Software tem-perature models are also available. Dow-ell’s CemCADE temperature simulator cal-culates BHCT versus time and depth,taking into account mud properties, wellgeometry and deviation, borehole geology,mud temperature in and out of the well,and the pump rate (next page).

Temperature probes have also beenused. Heat-sensitive paper is encased insmall spheres and pumped with the mud.Recovery varies from 10 to 50%, but onceat surface, a color change in the paper isused to determine the maximum tempera-ture the sphere encountered.

Retardation is complicated by the needto design a slurry that may encounter ashoe up to 70°F [39°C] hotter than theliner top. Yet good cement is required atboth locations. In general, if the statictemperature at the top of the liner is lessthan the circulating temperature at theshoe, the slurry will set satisfactorily inboth locations. However, if the situation isreversed, there can be difficulties formu-lating a slurry to set at both temperatures.This requires a retarder that is not tootemperature-sensitive.

•In gas zones with a low overbalance, thereis the risk that high-pressure gas will enterthe cement during hydration and createlarge channels. Elsewhere, loss of fluidinto the formation reduces the slurry liq-uid-to-solids ratio, changing rheology,density and setting time. Resistance to gasmigration and fluid-loss control are oftenhandled using latex additives. This is alsothe case in HTHP wells. In some cases,gas-tight, high-temperature slurries havebeen designed using latex additive andClass H cement rather than the more usualClass G.23 At these extreme conditions, theClass H slurries show longer natural thick-ening times, advantageous rheologies anda right-angle set—a rapid transition to set-ting rather than a gradual one.

•Set cement must exhibit good compressivestrength at the shoe and liner hanger.Strength may deteriorate with time, some-thing that becomes more likely once theexpected BHST exceeds 225°F [107°C].Hottest North Sea wells drilled so far have

Oilfield Review

Max

imum

tem

pera

ture

, °F

Time, min

Dep

th, f

t

390

310

370

350

330

0 70 140 210 280 350 420 490

Circulating Cementing Waiting on cement

CemCADE-predicted BHCT

API-derived BHCT

19,000

19,400

19,800

20,200

nPredicting maximum bottomhole circulating temperature. This plot shows the magnitude and depth of the maximum BHCT atany given time. Traditionally, API conversions have been used totransform bottomhole static temperature to bottomhole circulatingtemperature. This gives a single value. Increasingly, software systemslike the CemCADE temperature model are used to simulate cementjobs and show how temperature varies with time and depth.

exhibited BHSTs of around 400°F, socement recipes always include silica flour,which prevents a loss of strength andincrease in permeability that may other-wise occur in set cement over time.

•Narrow margins between pore pressureand fracture pressure mean that thehydrostatic pressure during cementation iscrucial. To achieve the correct density, aweighting agent—usually hematite—isrequired. In the field, a 100-barrel batchmixer is often used to produce a homoge-neous slurry with a very accurate density.

•Solids in the slurry must remain in suspen-sion. This is sometimes difficult toachieve. For example, weighting agentsare obviously heavy and need to be sup-ported to avoid settling. However, toachieve the desired flow properties theslurry may need to be thinned. But if it istoo thin, settling may occur, threateningthe quality of the set cement.

To balance design factors like these, exten-sive laboratory testing is required to ensurethat the slurry exhibits the right propertiesat downhole conditions. Tests should alsoreflect field mixing technique. Batch-mixingis usually simulated by stirring the slurry ina consistometer for the appropriate periodto impart an appropriate mixing energy.

Once the 7-in. liner is cemented, casingpressure tests simulate losing control duringwell testing and exposing the entire string toformation pressure. Tests are generally car-ried out using a retrievable packer set abovethe theoretical top of cement in the annulus.This avoids creating microannuli, tiny con-duits between cement and casing that formwhen the casing expands under pressure andcompacts the annular cement. But the testsuse mud, creating a different fluid gradient inthe well than would be found with gas. Tocombat this, the packer is progressivelymoved up the well in a series of tests toreproduce the worst case of a gas-filled well.

April/July 1993

19. Krus H and Prieur JM: “High-Pressure Well Design,”SPE Drilling Engineering 6 (December 1991): 240-244.

20. Bittleston S and Guillot D: “Mud Removal: ResearchImproves Traditional Cementing Guidelines,” Oil-field Review 3, no. 2 (April 1991): 44-54.

21. Vidick B and Acock A: “Minimizing Risks in HighTemperature/High Pressure Cementing: The QualityAssurance/Quality Control Approach,” paper SPE23074, presented at the Offshore Europe Confer-ence, Aberdeen, Scotland, September 3-6, 1991.

22. API Specification for Materials and Testing for WellCements, 5th ed. Dallas, Texas, USA: AmericanPetroleum Institute, 1990.

TestingCores are taken and logs run where possibleand used to decide whether and where totest. Coring may be limited by its propensityto swab the well and cause kicks. For log-ging, the tolerance of all standard wirelinelogging tools to high temperature may beboosted by thermal insulation.24 Wirelinemust also be protected. Choice of insulationmaterial and armor is influenced by depth,wellhead pressure, bottomhole temperatureand pressure, the possible presence of H2Sand job duration. A dedicated suite of toolsand wireline reel is often prepared, testedand certified for HTHP use.

Once cores and logs have indicated thepresence of hydrocarbons, a well test isneeded to determine parameters like reser-voir extent and permeability, and to samplereservoir fluid. In almost all cases, a cased-hole drillstem test is used.25 The well is nor-mally shut in using a downhole valve andflow is controlled at surface using a choke

23. Cement is traditionally divided into classes definedby the API that broadly describe the proportions ofthe different chemicals making up the compoundand their particle size distribution.

24. A universal Dewar flask protects logging tool elec-tronic cartridges. For example, if a CSI CombinableSeismic Imager tool is put in an oven at 415°F[212°C] for four hours, its internal temperature risesto about 350°F. However, if the tool has been pro-tected by UDF equipment, its internal temperatureincreases only to about 200°F [93°C], comfortablywithin its operating specifications. Insulationincreases tool outside diameter from 33/8 to 35/8 in.

25. Vella M, Veneruso T, Le Foll P, McEvoy and Reiss A:“The Nuts and Bolts of Well Testing,” OilfieldReview 4, no. 2 (April 1992): 14-27.

manifold. Periods of flow and shut-in allowcollection of data like flow rate and pres-sure changes. In HTHP tests, fluid samplesare usually gathered at surface during aflow period (see “An HTHP Test in Depth,”page 30).

The rates and pressures experienced dur-ing testing HTHP wells are prodigious. Onetest by Ranger Oil Ltd. in the Central Grabenusing a jackup rig resulted in 44 MMscf/D ofgas and 4400 B/D of condensate. The maxi-mum recorded tubing-head pressure was12,500 psi, the bottomhole temperature was386°F [197°C] and the surface temperaturereached 300°F—even though surface equip-ment was cooled with water.26 Equipmenthas been designed and built to test 20,000psi and 400°F formations.

Well-control equipment used duringdrilling is designed to handle reservoir fluidsfor relatively short periods. During a test, thesurface equipment must cope with longflow periods. Where possible, elastomers

27

26. Ford J, Peden J and Rae G: “A Review of CompletionDesign for Deep, Hostile, High Pressure, High Tem-perature Wells,” presented at the Petroleum Scienceand Technology Institute Seminar, The CentralGraben, The Next North Sea Challenge, The Needfor Joint Industry Research, Edinburgh, Scotland,September 10, 1990.

P T

T

SC

SC

Surgetank

Transferpump

ShutdownPanel

Subseavalve

controlpanel

Remotemanual

emergency shutdown stations

To starboardflare

To port flare

To port flare

To starboardflare

To port flare

To starboardflare

Relief line diverter

Gas manifold

SeparatorChoke

manifold

Steamexchanger

Steamsupply

Condensateto overboard

Relief lineoverboard

Water lineto overboard

Methanol injection unit

Oil manifold

Lubricator valve

Openventoverboard

Surface test tree

High Pressure Low pressure

T

T P

Stand-aloneshut-in valve

Hydrocarbon process lineRelief lineSeparator oil lineSeparator gas linePneumatic signalHydraulic signalTemperature monitoringPressure monitoringP

T

P

nSimplified process and instrumentation diagram of the surface testing system. Well fluidflows through the surface test tree to the choke manifold where pressure is reduced by thechoke from 15,000 psi to about 1000 psi. If limits are breached, the flow may be shut in atthe subsea valve, surface test tree or stand-alone isolation valve. To prevent hydrate for-mation, methanol or glycol may be injected upstream of the choke and the fluid heatedby the steam exchanger downstream.

are replaced by metal-to-metal seals,removing the temperature limitation of testequipment. Surface and subsea equipmentare monitored using temperature and pres-sure sensors that report back to a real-timemonitoring system, which initiates the emer-gency shutdown (ESD) system if limits arebreached. In addition, the number of down-hole test tools and the number of operationsthey perform are kept to a minimum.

Because of the extreme conditions, HTHPtest planning and equipment selection haveto be meticulous, and the personnel per-forming the tests highly trained. With infor-mation from offsets, the first task is to antici-pate likely maximum values for several keyparameters like shut-in tubing-head pressure

28

and wellhead temperature, downhole tem-perature and pressure, and flow rate. Thesemaxima are used to select equipment withthe necessary operating capabilities. If thesecapabilities are exceeded, the test must stopor the test objectives be reviewed. In estab-lishing the maxima, attention must be paidto data collection. For example, to acquirethe correct data, the test will have a mini-mum flow period, and the length of thisperiod will then affect temperature ofseabed equipment.

Next, the individual safety requirementsof each component are determined—forexample, pressure relief valves and temper-ature monitors. Then the components areconsidered as part of the whole test system,allowing elimination of any redundantsafety devices.

When the equipment package is deter-mined, a piping and instrumentation dia-gram may be prepared, which specifies allthe equipment, piping, safety devices, andtheir operating parameters (above). A riglayout diagram highlights the positions ofkey well test equipment making sure thatthey interface with existing rig ESD systemsand fit into limited space. Safety checks andanalyses are carried out according to APIrecommendations.27 Procedures are estab-lished for key operations like perforating thewell, changing chokes or pressure testing allequipment. Contingency plans are made tocope with a range of possible incidents:downhole leaks or failures, surface leaks, adeterioration in the seastate or weather, orthe formation of hydrates at surface.28

Oilfield Review

This information is submitted to an inde-pendent certifying authority that mustapprove the plans before the test can pro-ceed. In addition, inspection certificates arechecked before each piece of equipment isdispatched offshore. Finally, the certifyingauthority has to approve the rig up.

Test equipment and operations may bedivided into three sections: downhole, sub-sea and surface.

Downhole Equipment: Sealing off thecandidate formation requires a packer. Dur-ing an HTHP test, differential pressuresacross the packer may exceed 10,000 psi.For this reason, permanent packers are usu-ally chosen, rather than the retrievablepackers used in lower pressure tests. Withwireline (or very occasionally drillpipe), thepacker is installed complete with a sealbore,and a seal assembly is then run with the teststring to seal into the packer. The sealassembly is usually about 40 ft long to allowfor the thermal expansion of the test stringas hot reservoir fluid flows.

Perforating with wireline guns is generallyavoided during HTHP tests, so tubing-con-veyed perforating (TCP) is preferred. Unlikewireline perforating, TCP allows the reser-voir to be perforated underbalanced andimmediately flowed through the test string.29

Because the guns will spend hours in thewell prior to firing, high-temperature explo-sive is used. In most cases, the TCP guns arerun as part of the test string, rather thanhung off below the packer. This reduces thetime that the explosives spend downholeand allows the guns to be retrieved in caseof total failure.

In most HTHP wells, TCP guns are firedusing a time-delay, tubing-pressure firingmechanism. Tubing pressure initiates the fir-ing process, but the pressure is then bleddown to underbalance pressure. The gunsfire after a preset delay, long enough toachieve underbalanced conditions. A sec-ondary firing system is usually included incase the primary system fails.

nDownhole test string. • Upper and lower single-shot reversing

valves—opened using annular pres-sure to communicate between theannulus and the tubing and allowreverse circulation. Once open, thesemay not be closed.

• Pump-through safety valve—closeswhen the annulus is pressured to a predetermined value (usually signifi-cantly higher than the pressures usedto cycle the tester valve). It is a one-shot tool for emergencies only.

• Tubing-test flapper valve—holds pres-sure from above to enable tubing andtools to be pressure tested but allowsfullbore flow from below. The flapper is permanently locked open duringtesting using annular pressure.

• Multicycle reverse-circulationvalve—in some cases, a recloseablerecirculating valve is included in thestring. It is opened by cycling tubingpressure a preset number of times andclosed by circulating through it at apreset rate.

• Tester valve—annular pressure opensand closes the ball valve, allowingflow and buildup periods during thetest. The valve also acts as an addi-tional emergency barrier to flow.

• Pressure gauges in carriersubs—gauge reliability at high temper-atures is major bugbear. At 350°F,most electronic gauges are at the limitof endurance. Combinations of high-accuracy electronic and less accuratemechanical gauges, with a limit of500°F, are often run.

Although the number of downhole tools isreduced to a minimum, HTHP tests stillrequire a number of components to allowdownhole shut-in, pressure testing of thestring, reverse circulation to remove hydro-carbons from the string prior to pulling outof hole, and downhole measurement ofpressure changes (left ). Sometimes to sim-plify the test procedure surface shut-in issubstituted for downhole shut-in. However,this introduces wellbore storage—the springeffect of the column of fluid in the wellbelow the surface valve that must beaccounted for by data analysis and usuallynecessitates longer shut-in periods.

In most cases, test tools are operatedusing annular pressure. The condition of thefluid in the annulus, usually drilling mud,plays a critical factor. High-density, high-solids drilling fluid may plug pressure portsand reduce tool reliability. Solids may alsosettle, potentially sticking the test string. Theeffects on heavy, water-base mud of beingstatic in a hot well have been thoroughly

29April/July 1993

27. Recommended Practice for Analysis, Design, Instal-lation and Testing of Basic Surface Safety Systemsfor Offshore Production Platforms, API Recom-mended Practice 14C, 4th ed. Washington, DC,USA: American Petroleum Institute, 1986.

28. Davidson AR, Prise G and French C: “SuccessfulHigh-Temperature/High-Pressure Well Testing Froma Semisubmersible Drilling Rig,” SPE Drilling &Completion 8, no.1 (March 1993): 7-13.

29. Cosad C: “Choosing a Perforation Strategy,” OilfieldReview 4, no. 4 (October 1992): 54-69.

Upper single-shot reversing valve

Tubing-test flapper valve

Tester valve

Lower single-shot reversing valve

Gauge carrier subs

Gauge carrier subs

Locator and seal assembly

Permanent packer

Sealbore extension

Hydraulic firing system

Safety spacer

TCP guns

Shock absorber

Gun release sub

Pump-through safety valve

Multicycle reverse- circulation valve

Upper circulation sub

Seal assembly

Planning an HTHP test may take three to six

months. During this time a well test supervisor

liaises with the oil company to ensure that a test

program is prepared and approved. The following

example is a simplified program to test one HTHP

zone from a semisubmersible rig. Even so, it

shows that a test requires a plethora of coordi-

nated operations involving most of the people

on the rig.

First, having ensured that the well is debris

free and the mud is in good condition, run in hole

(RIH) with the packer on an electric-wireline set-

ting tool; a casing collar correlation log confirms

setting depth.

Hold a safety meeting and clear all nonessen-

tial personnel from the work area. Pick up the

tubing-conveyed perforating guns. Make up the

test assembly (previous page) and pressure test

the tool string against the tubing-test

flapper valve.

RIH with the test string but do not stab into the

production packer with the seal assembly. Pres-

sure test the string, then confirm the packer depth

and the space out needed by running a wireline

depth correlation log—gamma ray and casing

collar correlation tools.

Make up the subsea test tree to the tubing and

then the landing string. Rig up the surface test

tree and surface lines and RIH landing the fluted

hanger at the bottom of the subsea test tree onto

the wear bushing at the top of the casing on the

seabed (next page). At the same time, the seal

assembly enters the packer sealbore. Perform the

final pressure test against the tubing-test flapper

valve—a bypass ensures that a leak in the flap-

per valve does not accidentally fire the TCP guns.

Commence testing operations. Close the mid-

dle pipe rams and pressure the annulus to 1500

psi, locking open the tubing-test flapper valve

and establishing a reference pressure in the

tester valve. Bleed the annulus and surface pres-

sure to zero, open the middle pipe rams and

unsting the seal assembly from the packer. Circu-

late cushion fluid down the tubing—the cushion

controls the underbalance during perforation.

Reenter the seal assembly, close the middle pipe

rams and pressure the annulus to 1500 psi.

Prior to perforating, hold a safety meeting to

ensure that everyone understands what is

expected during the test. To fire the TCP guns,

pressure the tubing to shear the pins in the

hydraulic delay firing head. Bleed back the pres-

sure to the required underbalance. The guns will

fire after 15 minutes.

Once an increase in wellhead pressure shows

that the well has been perforated, open the choke

valve and allow the well to flow through an

adjustable choke for about 15 minutes, so that the

volume below the tester valve is filled with reser-

voir fluid. Bleed off the annular pressure to close

the tester valve for a pressure buildup of about an

hour.

Reopen the tester valve for the main flow,

which at first passes through the adjustable

choke to the burner. After several hours, when

the pressure and flow rate are constant, divert

flow though a fixed choke and into the test sepa-

rator. Throughout the main flow, continually mon-

itor hydrogen sulfide/carbon dioxide [H2S/CO2]

concentrations, base-solids and water volumes,

surface pressure and temperature. Closely

observe the surface temperature to ensure that

the temperature ratings of the BOP and flowline

are not exceeded.

After flowing the well, close the tester valve by

bleeding off the annular pressure for a shut-in

period that is usually one-and-a-half times longer

than the main flow. Additional flow and buildup

periods may be required depending on the test

objectives. Once these have been met, the well

must be killed and the test string must be pulled

out of hole.

To kill the well, lock open the tester valve using

sufficient annular pressure cycles and bullhead

the reservoir fluid and about 10 barrels of kill-

weight mud into the formation.

If bullheading creates too much pressure in the

tubing, close the tester valve and pressure up

inside the tubing so there is a 1000-psi differen-

tial between the tubing and the annulus to open

the multicycle reverse-circulation valve. Reverse

out the fluid inside the test string to the burner

using kill-weight mud. Once mud is recovered at

surface, start pumping down the tubing directing

mud via the rig choke and mud-gas separator to

the mud pits. Condition the mud to a constant

weight. Stop pumping, close the reversing valve,

open the tester valve and bullhead the volume

below the reversing valve to the perforations plus

10 barrels into the formation.

Bleed down the test string pressure to zero and

for 15 minutes check that the well is not flowing

or losing fluid. Open the middle pipe rams, pick

up the test string and unsting the seal assembly

from the packer. The test string may be further

circulated. If the well is stable, rig down surface

equipment and pull out of hole.

30 Oilfield Review

An HTHP Test in Depth

investigated in the laboratory and the per-formance of test tools has been improved toreduce downhole failures. In some cases,the annular fluid is changed to high-densitybrine, which is solids free but increases theexpense of the test.

Subsea Systems: Like drilling, testing isgenerally simpler on a jackup than on asemisubmersible. On a jackup, the piping tosurface is fixed and the control valves areon deck. For a semisubmersible, a subsea

test tree is located in the BOPs on theseabed to allow quick and safe disconnec-tion of the test tubing during testing. Abovethe tree, there is a conventional riser discon-nect mechanism and a riser running to therig’s deck. The choke and kill lines are flexi-ble to compensate for vessel heave (above).

Surface Equipment: At any time duringthe test it must be possible to shut in thewell. Conventionally, this is carried outusing the choke manifold valve. In HTHPwell tests, a hydraulic actuator is fitted tothe flowline valve of the flowhead, or christ-

mas tree, and a hydraulic isolation valve isinstalled between the flowhead and thechoke manifold (page 28). Furthermore, ashut-in valve within the subsea safety tree islinked to the ESD panel.

At the heart of the pressure control equip-ment is the choke manifold. Although sepa-rate from the drilling choke, the test mani-fold has the same purpose, to reduce fluid

31April/July 1993

Seabed

Rig’s Texas deck

Single master valvehydraulically operated

Y-spool with double flow wing valves and double kill wing valve

Manual master valve

Lower manual valve

Production riser

Upper valve hydraulically actuated

To choke manifold

Riser disconnect

Subsea test tree

Fluted hanger

Shear rams

Pipe rams

Pipe rams

Pipe rams

Jackup Semisubmersible

Lubricator valve

Tubing

3-in. Coflexip flowline

Check valve

Hydraulic-actuated kill wing valve

Manual master valve

2-in. Coflexip kill line

To choke manifold

Hydraulic-actuated flow wing valve

Swivel

Retainer valve

Spacer

Rig floor

Riser

Rig floor

Wear bushing

Landing String

nComparing semisubmersible and jackuprig surface and subsea systems. For jack-ups, there is no subsea system; all casingis cemented to surface and a straightfor-ward test tree, or wellhead, is used atdeck level. Rather than a temporary testhead, testing on jackups often utilizes apermanent-type tree. The BOP stack isusually removed for jackup testing. How-ever, it is now possible to retain the BOPstack and run a safety valve below it. Therams seal around the safety valve, whichincorporates a nitrogen chamber to assistclosure and enable it to cut wireline.

For semisubmersibles, a subsea test tree(SSTT) in the BOP on the seabed allowssafe disconnection during testing. Abovethe tree is a conventional riser-disconnectmechanism and a riser running to the rig’sdeck. In the SSTT are two subsea controlvalves below the BOP rams—a flappervalve and a ball valve that operatetogether. If the rig has to move off station,the valves may be closed using ahydraulic control on surface, and ahydraulic latch above the valve opens todisconnect the string above. The weight ofthe string left in the hole is borne by afluted hanger below the control valve. Theriser may then be disconnected with theupper part of the test string still inside it.

A retainer valve inside the string isalways run in HTHP tests to prevent hydro-carbons in the upper test string from con-taminating seawater. The upper test stringalso contains a lubricator valve—some-times two for redundancy—to allow wire-line assemblies to be used without a lubri-cator. At deck level is the flowhead orsurface test tree. This incorporates theshutoff, kill and choke valves, and allowswireline entry.

nChoke manifold configurations. To givedouble isolation when changing a chokeduring a test, either the eight-valve ordouble-choke manifold is required.

32

Double-choke manifold

Four-valve choke manifold

Eight-valve, double- block choke manifold

pressure, usually to less than 1000 psi. Themanifold contains adjustable and fixedchokes. To change one of these—eitherbecause a different size is required orbecause of choke erosion—the path throughthe choke must be isolated by closing valveson either side of it. When a choke is beingchanged, conventional four-valve manifoldsdo not offer the double isolation requiredfor HTHP tests. For this reason, eight-valvemanifolds that are nearly twice the size ofthe four-valve version are often used. Inother cases, two four-valve manifolds sepa-rated by isolation valves are specified (left).

Hydrate formation is a serious problem,especially early in the test when the wellhas not been warmed by extended flow. Toavoid plugging the line with hydrate, gly-col or methanol may be injected into thefluid before it reaches the choke. Addition-ally, a heat exchanger warms fluid down-stream of the choke. Peculiar to HTHPtests, an extra 15,000-psi choke is some-times incorporated in the heat exchanger.Therefore, early in the test when hydratescould form in the line—which may be tensof meters long—between the choke andthe heater, pressure is initially reduced bythe heater choke.

Heating the reservoir fluid also aids sepa-ration. For HTHP wells, conventional sepa-ration and sampling techniques are suffi-cient. Fluid volumes are then metered anddisposed of, usually by burning at a flare.

Although North Sea HTHP wells presentformidable challenges, about 50 have beensuccessfully drilled and many of themtested. And there have been spin-offs. Thelessons learned in coming to terms withHTHP wells are now being applied in lessextreme conditions.

To date, most of the HTHP wells havepredominantly been around 15,000 psi and320°F [160°C]. But there is a new chal-lenge waiting around the corner: ultra-HTHP wells topping 20,000 psi and400°F.30 These will be even more taxing andensure fresh challenges for well planners,drillers and testing engineers. —CF

Oilfield Review

30. Seymour KP, Stuart C, Simpson B, Lorenson P andMackay A: “The Drilling of a High-Pressure, High-Temperature Well in the North Sea Using 20,000-psiWell Control Equipment,” paper OTC 7337, pre-sented at the 25th Annual Offshore Technology Con-ference, Houston, Texas, USA, May 3-6, 1993.