IPTC-18113-MS

Unconventionals Meets Deepwater; Lower Completion Limited Entry Linerwith Retrievable Ball Drop Diversion System Applied in a Deepwater BrazilCarbonate Field

M.A. Fowler, R.D. Gdanski, P. Campbell, W. Bode, J.M. Baima, and S. Hensgens, Shell

Copyright 2014, International Petroleum Technology Conference

This paper was prepared for presentation at the International Petroleum Technology Conference held in Kuala Lumpur, Malaysia, 1012 December 2014.

This paper was selected for presentation by an IPTC Programme Committee following review of information contained in an abstract submitted by the author(s).Contents of the paper, as presented, have not been reviewed by the International Petroleum Technology Conference and are subject to correction by the author(s).The material, as presented, does not necessarily reflect any position of the International Petroleum Technology Conference, its officers, or members. Paperspresented at IPTC are subject to publication review by Sponsor Society Committees of IPTC. Electronic reproduction, distribution, or storage of any part of this paperfor commercial purposes without the written consent of the International Petroleum Technology Conference is prohibited. Permission to reproduce in print is restrictedto an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paperwas presented. Write Librarian, IPTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 1-972-952-9435

Abstract

In order to economically develop the Albian Carbonate Macae formation in the Campos Basin DeepwaterBrazil an extended reach horizontal well is required. Due to the tight nature (low permeability) of the rock,stimulation of the reservoir is required. Completion designs and techniques for treating extended reachhorizontal wells in deepwater are limited. Most applications for these type wells in Deepwater areineffective (high skin/low PI) and inefficient (trip/time intensive) affecting project economics. A com-pletion system was developed that allows for safe, effective and efficient stimulation and installation ofsuch an extended reach horizontal well in Deepwater.

This paper will describe the design, testing and execution of a unique Deepwater completion systemthat adapts a known multi-stage ball drop system used in onshore unconventional reservoirs, for examplethe US and Canada, to a known horizontal open hole sand control system to effectively matrix acidize a2000 m horizontal open hole thru a Limited Entry Liner with reservoir segmentation. The system uses aRetrievable Ball Drop Diversion System (RBDDS) consisting of multi-stage frac sleeves run on washpipe. The RBDDS is run inside a limited entry liner that is segmented into stages or intervals by openholeprogrammable interventionless packers (Mendes et al. 2014) all run in a single trip and with no pipemanipulation required while stimulating. It furthermore leaves behind a robust lower completion systemfor the life of the field.

This 2000 m lower completion system was successfully run in the Albian Carbonate reservoir and allstages treated with 15% HCl with minimal NPT, no harm to people or environment. At the time ofcompletion, this well was the longest horizontal reservoir section and longest step out well drilled andcompleted in Brazil. This Limited Entry Liner System together with the RBDDS proved to be a veryefficient, effective and deepwater friendly system.

IntroductionLow permeable deepwater carbonate reservoirs have created many challenges to produce economically.The wells typically require extensive reservoir contact that has to be stimulated in order to achieve an

acceptable Productivity Index, PI. Field A, Albian Carbonate in the Campos basin offshore Brazil fallsinto this challenging category. Effectively stimulating long horizontals has long been an industrychallenge onshore and even more so in a deepwater environment. With the advent of Unconventionals inthe US and Canada the ability to stimulate long horizontals has generated a great deal of focus and effortover the last few years. The high rate and multi-stage fracture stimulations done on these wells hasresulted in the development of new industry techniques, notably using frac sleeves operated by droppingballs, and pumping down plugs and perforating guns to isolate and perforate each stage.

This paper will review the concept selection process, detailed design work, testing, and implementationresults of an extended reach horizontal well that utilizes onshore Unconventional ball drop technologyadapted to a Deepwater setting to effectively stimulation a 2000 m long horizontal section. The well andstimulation treatment was considered an apparaisal effort such that if the production from the well issuitable it will open up the full Albian Carbonate for further development.

BackgroundThe field was discovered in the early 90s and lies in the Campos basin offshore Brazil, see Figure 1showing field location. The field is mature having been on production since the early 90s from variousEocene sandstone reservoirs. Initially the field was produced under depletion via an early productionsystem up to the late 90s. Then, in early 2000s, there was a field redevelopement done, that was lateracquired by the current operator, and production was re-started in 2003 to an FPSO, now with waterinjection support. The Quissama member of the Albian age Macae formation in the Campos basinunderlies the existing Eocene age reservoirs, but has not been produced to date. To help extend field life

Figure 1Field location Campos Basin, Rio de Janeiro State, Brazil

2 IPTC-18113-MS

and to evaluate the production potential of this formation a development / appraisal well was planned tobe drilled/completed and hooked up to the FPSO and subsea infra-structure. The Albian Carbonate belowField A and B lie in 769 m of water and is low permeability.

The Albian shallow water shelf carbonates comprise of three main families of facies. The best qualityreservoir units (porosities up to 28% and permeabilities up to 1000 mD) are the oolitic facies depositedin a high-energy environment. The second family of facies is composed of the more porous (up to 25%)but less permeable (up to 100 mD) oncolitic-peloidal packstones and grainstones, deposited in a shallowmarine environment with moderate water agitation. The third family of facies seen in the southern Camposbasin during the Albian stage are fine-grained limestones (Mudstone/Wackestone) deposited in deeper andlower energy environments. These form reservoir quality rock with medium-high porosities (average16%), but relatively low permeabilities (average 0.5-3.0 mD). The Field A, Albian well has been assessedto belong to this third category.

Well Functional Specification/RequirementsAs part of the well functional specification the overall objective was defined as evaluation of theproduction potential of the Albian Carbonate via an appraisal / development well. In order to achieve boththe appraisal and development aspects of the well, a long horizontal lateral that transected the reservoirwas required. This was to expose as much reservoir for evaluation and production as possible. A keydesire was to find secondary permeability (fractures or vugs) even though none appeared on the availableseismic. As it is anticipated that further well stimulation may be required, the completion design must besuitable for treatments in the future. As part of well planning efforts the following key performanceindicators were established;

Achieve a drilled length of 2000 m in the Albian Carbonate Limit Dog Leg Severity to 2.0 degr/100ft in the reservoir section Successfully run the lower completion to TD Effectively stimulate the 2000 m horizontal section by matrix acidization Achieve maximum possible well performance, by utilizing a unique stimulation system thatincorporated a limited entry concept with a retrievable ball drop diversion system (RBDDS)

Demonstrate lower FLCV integrity prior to installing the upper completion

Figure 2Seismic cross section showing the Albian reservoir structure

IPTC-18113-MS 3

Install upper completion with functional and tested production packer, downhole pressure gaugesand SSSV

Achieve objectives within budget

The well materials and equipment should be suitable for a 20 year life. Since no fluid or core sampleswere available for evaluation, a review of nearby fields was made to help predict the reservoir propertiesand fluid composition of the Albian Carbonate. A geo-chemical study concluded that a good analog fluidwas from the Congro and the Field A sands. Thus the upper and as much of the lower as possiblecompletion equipment were designed to match existing wells. The wells comprised of GRE tubing witha SCSSSV, dual gas lift mandrels (with one unloader and one orifice valve installed), permanentdownhole gauges, and a production packer. The use of GRE tubing was required due to concerns over latelife souring and possible presence of CO2. All upper completion hardware was made of High Alloymaterials. As additional insurance downhole chemical injection was added to address any potential scalingissues.

Subsurface pressures at the top of the reservoir are around 5000 psi and bottom-hole temperatures are90 C. The lower completion equipment was all 13Cr material with the casing packer and Fluid LossControl Valve (FLCV), being the same design that was utilized on the previous wells in the Eocene sands.The lower completion design had to be suitable for a matrix acid stimulation treatment early and late life,thus a suitable lower and upper completion mechanical and dimensional configuration was required. Fluidloss control was a big concern should secondary permeability be encountered while drilling or stimulatingthe well. Thus the lower design had to control losses while stimulating each stage and after finalizing thetreatment and Pull out of the Hole, (POOH). If losses could not be controlled, damaging LCMs wouldbe required to control the well, thus negating the positive effects of the treatment.

Concepts ReviewedInitial reviews focused on how best to improve production potential in the well. Since stimulation wasrequired, the treatment concepts focused on two primary options; Acid Fracturing or Matrix Stimulation.Initial concepts sought to allow for both matrix and fracture acidizing treatments. This was to allow forflexibility, once the well was drilled. Completion concepts that allow for both matrix or fracture acidizingare either too costly (i.e. trip intensive) or are typically optimized towards one method over the other. Thusa system that would allow for both optimized treating options was dropped from consideration. Thiscoupled with a review of case histories in the Albian Carbonate in Brazil and the fact that the reservoiris in the grey area, for a fracture treatment and matrix treatment, it was decided to pursue only matrixtreatment options for the Albian Carbonate well. It was felt that an acid fracture treatment would not beas effective, because of the expectation that the rock would be too soft (analog data), thus the fractureswould close once the pressure is released (Azevedo et al. 2010). The matrix option would still allowevaluation of the production potential in a more economical and less risky scenario, thus acid fracturingwas no longer pursued for this well, see Figure 3 below showing selection criteria and scoring.

Several completion concepts were reviewed for Albian Carbonate Well. These concepts focused onprevious methods used in Brazil, the North Sea (Chalk), and Shale Unconventionals. All of the concepts

Figure 3Matrix versus Acid Fracturing

4 IPTC-18113-MS

had the objective of effectively completing and treating long horizontal low permeable wells. Theconcepts used in Brazil (Rodrigues et al. 2007) and in the North Sea utilized a coil tubing pin pointtreatment method that jetted the formation pumping down both the coil tubing and the annulus (Rodrigueset al. 2005). Other systems utilized in the North Sea were Plug and perforate methods; the zones wereperforated, isolated then treated in successive order. Optimized variations of this system are currentlybeing used onshore for Unconventionals. This is then replicated for each interval. While an effectivemethod for treating it is expensive in a deepwater setting. The pin point systems require Coil Tubing andpumping down the annulus with acid to be effective. Thus in a subsea environment a tie back system thatisolates the Riser, BOPs and the annulus is required. An alternate method would be to pump a geldiverting agent down the choke or kill lines. This, however, is ineffective in treating the interval. Thesystems reviewed are listed below;

Ball Drop systems Work String Ball Drop system (selected concept) Acid Diversion System (polished rod isolation system) Conventional Coil Tubing Pump down systems

A system that combines the ability to treat long intervals with the flexibility to add a Limited Entryinjection profile will provide the best opportunity to stimulate the Carbonate. This technique is currentlythe leading edge design for stimulating long horizontal Carbonates. (Furui, et al, 2010, Hanson et al.,2002) Limited Entry is having defined numbers of small controlled diameter and strategically spacedholes in the liner to ensure a more even distribution of acid across the full interval, basically mechanicaldiversion. The number of holes and sizing is derived by a balance of production rates, required treatingrates, the interval length and completion sizing (i.e. allowable rates). The holes can be pre-drilled or donevia perforating as long as the perforation hole size can be controlled.

As with all subsea wells, fluid loss control after and during the stimulation treatment is important forwell control and to ensure the well productivity is not damaged by any fluid loss pill. It is thus critical tohave fluid loss control capability for any system that is used. The Fluid Lose Control Valves (FLCV) thatare typically used in deepwater sand control wells, can function as a barrier for upper completioninstallation and well suspension, plus control losses after treating. These valves are opened remotely fromthe host, thus intervention is not required. Because of these features it is highly desirable to have a designthat allows for the use of a conventional FLCV.

The Acid Diversion system and the workstring ball drop system (RBDDS) were the only two systemsthat could incorporate a FLCV and the Limited Entry concept, thus ultimately eliminating the otherconcepts. The Acid Diversion system required the pipe to be moved between each stage, thus having thepotential for acid exposure, losses between stages, and the potential troublesome issue of having tore-pressure test all of the treating temporary pipe work (Jouti et al. 2011). In addition to the HSSEexposure, the re-testing and pipe movement during the treatment made the system more costly than theworkstring ball drop system.

The conventional sleeve ball drop system was eliminated because it was not limited entry capable andthe removal of the balls is problematic in a deepwater well. Catching and leaving the balls in the lowercompletion was also considered, but in the end concerns over plugging and potential ball debris in thesubsea systems eliminated this concept.

A ranking matrix was established to allow for subjective analysis of the features and benefits of eachsystem as they relate to the well functional specifications. This ranking matrix was divided into categoriesthat had weighted factors, see Figure 4 below. Based on these factors the chosen system was the Workstring or Retrievable Ball Drop Diversion system in combination with Limited Entry liner subs. While

IPTC-18113-MS 5

novel in the combination of components, the system utilized previously proven equipment, thus greatlyreducing any required testing, risk or development.

By only using the ball drop sleeves and openhole packers, the reservoir can be effectively segmentedinto stages and allows for the independent treatment of each stage. This is a good technique for fracturing,but does not provide the formation coverage required for effective matrix acid stimulation needed for theAlbian Carbonate well. History and more recent developments in Limited Entry stimulation techniques(K. Furui et al. 2010) have shown greatly improved stimulated skin values from the traditional designswith -1 to -2 skins, and -3.5 to -4 using specialized Limited Entry. For this reason, the RBDDS uses thelimited entry joints to implement affective acid coverage across the formation at specific rates thusimproving the wormholing affect. This allows for lower skin values and better zonal coverage with thesame amount of acid. Thus incorporating the Limited Entry feature into the liner and inner string hassignificant value.

Detailed DesignThe design concept chosen employs a concentric string that is similar to an open hole gravel pack systembut modified for the purpose of limited entry matrix acidizing. The external string was comprised of a jetdown shoe, a liner with Limited Entry subs, Interventionless openhole packers, seal bores for inner stringisolation, a FLCV and a Gravel Pack (GP) packer to secure the liner in place. The inner string consistedof washpipe to space out the ball drop sleeves and seal assemblies, and the GP packer setting tool. Theseal assemblies were spaced out to land in the seal bores at each of the 12 openhole packers, thusproviding zonal isolation and creating an annulus between the liner ID and washpipe OD to generate thelimited entry affect between each openhole packer, (see Figure 5). The well is stimulated from the toe toheel using successively larger diameter balls for each stage as you come up the hole. The balls open thesleeves and provide isolation to the lower just treated zone, thus enabling independent treatment of eachstage. The inner washpipe string is run in the treating position; therefore once all the packers are set nopipe movement is required. This feature along with the use of dual ball droppers ensured that the pipewould not have to be manipulated or connections broken at any time during the treatment, which was akey HSE driver. This helped to ensure the fluids, specifically the acid, stayed in the pipe during thetreatment. The lower completion also included features that allowed wash-down while running in the hole,optional rotational release, and reverse out of any acid in the workstring should the system lock-up. Theliner and inner string are run in one trip.

Figure 4Selction matrix with concept rankings

6 IPTC-18113-MS

The chosen concept yielded the following features or benefits:

Single trip, run, set and treat decreasing HSE exposure, well cost and risk No pipe manipulation required during the treatment reducing HSE exposure Ability to wash-down while tripping in Reservoir segmented into 12 stages allowing optimized treatment of each stage for the full 2000m well bore

Isolation of each stage during and after the treatment for leak-off control The liner in each stage had limited entry ports for effective mechanical diversion at each stage Post treatment Fluid Loss Control via a barrier valve The ability to isolate at each stage while POOH should losses be excessive Full liner ID post treatment Uses existing equipment, but configured in a unique way

For subsea completions, wellbore isolation and fluid loss control is a critical aspect of well design. TheAlbian Carbonate well was designed to incorporate a fluid loss control valve (FLCV) and provide arestriction to heavy losses. The FLCV is closed when a shifter, run at the end of the washpipe, is pulledacross it closing a ball isolating the lower zone and stopping any potential losses. As the Albian Carbonatewell is an extended reach horizontal, the ability to control losses while tripping out with the washpipe isimportant. To control and limit losses while tripping, a sealbore was added below the lower packer andheavy wall pipe was run above the last stage. The heavy wall pipe works to restrict / choke the losses andthe sealbore allows for stopping any losses should the well need to be topped off. Calculations showedthat with an over balance of 100 psi the loss rate between the washpipe and heavy wall pipe is limited to2.0 bpm.

The shifter used to activate the FLCV has to be pushed through each of the seal bores in the completionwhen being made-up and then pulled back through the same seal bores at the end of the job and still havesufficient strength to close the FLCV. The shifter also could not damage the seal bores jeopardizing theseal integrity and treatment effectiveness.

TestingAs part of the design the following tests were executed:

FLCV shifter test

In order to confirm that the FLCV shifter was suitable for this application a test was conducted thatsimulated actual operations expected on the completion. The shifter collet was subjected to 5 initial cyclespushing through a valve nubbing and 5 cycles pulling through the same nubbing to record the baselinerelease forces of the collet. Then it was cycled 37 times each way through a 4 ft long seal bore. The designof the collet is such that forces to push the collet through the nubbing are around of the loads than is

Figure 5Albian Carbonate Lower Completion Design

IPTC-18113-MS 7

required to pull back through the nubbing. The pre-test loads were within specification, however, after 37push and pull cycles the collet loads to shift the valve closed increased by over 70%. This is evidence thatthe collets had work hardened during the test. The changes in shift forces were not evident during the sealbore cycle testing, this could have been masked by the 30 degree lead-in angle of the seal bore. The colletwas inspected after the test and found to be within print tolerance and did not show any adverse wear ordamage, see Figure 6. Although not expected, the increase in shift value yielded a positive result, thatbeing able to effectively close the FLCV.

SIT of the surface ball drop system and diversion sleevesA system integration test was conducted to verify the surface ball drop setup in combination with thechosen manifold line up and downhole diversion sleeves. The setup consisted of the, Drill Pipe Landingstring complete with TIWs suspended vertically from a crane, the remote controlled ball drop system,temporary pipe work configured per the planned offshore line up and a set of diversion sleeves to simulatethe downhole activation of the same. As part of the test an accoustice device was used to witness the balldropping at surface. The test demonstrated that the ball drop system in combination with compositeballs and an 80 lb HEC viscousified brine pill could effectively transported the activation balls 6 mvertically into the landing string, and that the acoustic system allowed for detecting the dropping of theballs at surface (be it the smaller sizes were harder to observe). The test also confirmed the line up wouldallow for efficient offshore execution.

Openhole IsolationA design review was conducted after the concept select stage where it was agreed that the primary meansfor isolation was water Swellpackers. This would allow a single trip system and simplify the lowercompletion. The objective was to spot fresh water across the interval to allow the packers to swell onceon depth and while making up and testing the temporary pipe work. A cost and risk benefit analysis wasconducted showing that even with 2 days of wait time the water Swellpacker option was still the preferredoption. To ensure that we fully understand the Swell times of the packers and to ensure that they do notswell too early i.e. while tripping in, testing was conducted on the Water Swellpacker. The testing wasdone to simulate the time that the packers will be in brine while making up the liner in the riser, trippingin the hole and after circulating out the brine to freshwater. Testing was done with freshwater, 10% and25% NaCl solutions to simulate residual brine left in the openhole after displacing to fresh water. The 10%NaCl and freshwater solutions resulted a swell time of 8 days until a seal was formed. The other samples,all done at 100 C to represent well conditions, did not reach the seal expansion point and thus werestopped. The expansion to seal time did not allow for the Water Swellpackers to be used for this well, asthey were required to seal within 48 hrs after getting to depth.

Figure 6FLCV shifter after cycle testing

8 IPTC-18113-MS

As a result of the water Swellpacker testing, an extensive review of available openhole isolationsystems that met the requirements for application in the Albian Carbonate well was conducted. To keepthe single trip feature an interventionless packer was required. An openhole packer that contained anelectronic trigger activation method was chosen. The trigger is designed to activate once a pre-definedpressure and temperature is seen by the packer sensors. This activates a timer, that once expired activatesthe e-trigger, releasing the pressure in an atmospheric chamber and setting the seal elements. Theinterventionless features of the packer makes it uniquely suitable for this application (Mendes et al. 2014).Figure 7 is a sketch of the packer with the trigger mechanism shown in the photo.

Prior to running the completion the setting parameters of the openhole packers were changed to havethem activated by pressuring up the well and holding the pressure for a pre-determined amount of time.The early parameters focused on using the well hydrostatic pressure and temperature to set the packers,while using the timer to control any safety factors while running in the hole (RIH). The ECDs seen whilecementing the 9 5/8 casing were sufficient to allow the change in packer setting parameters. The packertimer starts once the pre-programmed pressures and temperatures are seen. Once the time expires thepackers set using well hydrostatic pressure. The settings were changed on the rig when connecting thebattery and turning on the packers. This was originally planned to save battery life. The temperaturesettings were set to ensure that the temperature deviation was always active, thus eliminating temperatureas a setting parameter once on depth. See Figure 8 below for the packer setting parameters. The settingpressures took into account the slight change in TVD from heel to toe. Sub-Assembly, SA-2 was the firstin the hole and placed near the toe of the well and SA-15 was at the heel.

The liner assembly contains seal bores at each OH packer and Limited Entry Liner subs in betweeneach OH packer. The inner string seals were spaced-out to align with liner sealbores to provide zonalisolation in the liner to washpipe annulus for each stage. By having both the liner to washpipe and the openhole to liner annulus isolated at each stage, acid can be effectively injected into an individual zone. Thisalong with the limited entry subs spaced along the liner provides the mechanical diversion via the holesfor efficient reservoir coverage. For a given rate, quantity of holes, and hole size the back-pressure and

Figure 7Openhole Interventionless Packer and tigger mechanism

Figure 8Openhole packer final setting parameters

IPTC-18113-MS 9

rate per hole can be estimated using a newly developed Diversion Model which takes into account thefriction driven pressure drop along washpipe to liner annulus. The Diversion model was used to optimizethe limited entry design while adhering to the stimulation principles as defined by Gdanski et al. 1999.These principles were used to determine hole spacing and placement and volumes required per stage whilemaintaining the 25/75 rule for carbonate stimulation. The 25/75 rule states that treating 25% of the zonereally well achieves the same production as treating 75% of the full zone.

The division of the wellbore into 12 zones with limited entry joint clusters in each zone allows for goodmechanical diversion of the acid to help ensure adequate coverage over the full wellbore length, see Figure8.

Another advantage of the RBDDS is that the number of zones to be stimulated can be tailored to theavailable volume and pumping horse power of the selected stimulation vessel. For the Albian wellsstimulation vessel which had a max treating rate of 20 BPM the appropriate number of stages was 12stages allowing 1/8th inch incremental size balls), and the number and size of holes in the limited entryliner were optimized accordingly. If a larger vessel capable of double the rate (40 BPM) had beenavailable it is conceivable that the number of stages could have been reduced in half.

As part of the acid stimulation pump schedule, a brine preflush was required. Due to the low perm ofthe rock and effective displacement it was expected that the birne would have to be injected under fractureconditions until the acid hits and the pressure drops allowing matrix stimulation and wormholing. Agraphical representation is shown below in Figure 10 and is based upon transient spherical flow as definedby K. Furui, et al. 2010.

ExecutionIn order to ensure accurate space out of the Limited Entry Liner with isolation packers and in the innerRBDDS, the system and components was laser tallied by two independent companies. In comparing thetwo tallies, three liner joints were found to be different in length and re-measured to ensure accuracy.

In addition, prior to mobilizing the equipment offshore, the inner liner washpipe, seal assemblies, andsleeves were spaced out against the liner in the yard. This confirmed the various tally spreadsheets madeby the service companies and Shell and enabled confidence in the final space out of each stage offshore.

The liner and washpipe joints were then loaded in sequence such that they could be run straight fromthe baskets or tubing racks. This helped to ensure that the items were run in the correct order andcontributed to the reduced amount of time needed to run the liner and washpipe.

While being made-up vertically at the rotary offshore in a 9.7 ppg NaCl Brine, the washpipe space-outwas checked against the liner by tagging the float shoe with the FLCV shifter, the space-out was foundto be exactly as measured onshore. This was also confirmed by high seal drag when stabbing all the sealassemblies at once with 25 klbs stab-in and 40 klbs pick-up force.

Figure 9Well segmentation for mechanical acid diversion

10 IPTC-18113-MS

Prior to running the completion the openhole section was logged with a very stiff and long BHA. TheBHA had an 8.5 bit on bottom with 8.375 centralizers across the logging BHA to a length of 52 m. Itis believed that this assembly cleared any ledges and reduced the severity of any micro doglegs. Asignificant amount of Stick-slip was seen while working the logging BHA to TD. This affected the qualityof the logging data and resulted in a hole out of gauge.

The lower completion was run to TD without encountering any issues or significant drag. The liner slidto TD much smoother than the logging BHA. The actual friction factor (FF) recorded was 0.25, see Figure11. The discontinuity in the chart is a result of re-calibration of the Hookload sensor. At the time of thewriting of this paper, this well was the longest step-out and completed reservoir section in Brazil, a 1995m horizontal.

Figure 10Treatment injection flow model

Figure 11Actual Torque and Drag chart

IPTC-18113-MS 11

The design base case friction factors were 0.25 in the casing and 0.40 in the openhole from analog data.The actual FF recorded while RIH matches fairly consistently with the modeling done prior to running thecompletion when 0.25 FFs are used, see Figure 12. The hole was enlarged from a gauge hole of around8.5 to an average around 9.1 to 9.6 ID with some larger sections in stage 8 of 12.6 ID. The enlargedhole most likely resulted in lower actual FF. It is surmised that the heavy stick slip and vibration seenwhile logging and drilling enlarged the hole.

After getting the completion to depth the Solids Free Drill in Fluid (DIF), which was in the OH toreduce sliding friction and control fluid loss while RIH was circulated out. The maximum circulation ratewas limited to 3.0 bpm to prevent pre-setting of the GP & isolation packers. The pressure at the cementunit was limited to 300 psi until sufficient volume was pumped to have the DIF out of the openholesection. After this the rate was increased with a back pressure limited to 500 psi and below 3.0 bpm. Bycontrolling the pump pressure we ensured that the GP & openhole packers did not accidently set duringthe displacement. The mud was completely circulated out of the hole, because if the well went on lossesafter the treatment the solids free mud could have been pulled into the reservoir causing potentialimpairment.

The GP packer setting ball was pumped at the end of the mud displacement and was pumped down ina 10 bbl 80 lb HEC pill. The GP packer setting ball was the smallest of all the balls to be dropped at 1.125OD. The GP packer was set and pressure tested. The well was pressured up to 1200 psi over hydrostaticto set the openhole packers starting from the heel working down towards the toe. The pressure was heldto 800 psi for an hour and 20 minutes. The set time for the packers was 70 mins. The Chart below, Figure13, shows the openhole packers setting recorded on the downhole gauges ran with the RBDDS. This wasunexpected but probably caused by the tight nature of the rock and volume displacement in the packerhydrostatic setting chamber and packer element.

After rigging up and testing the iron on the rig and setting the pop-offs on the stimulation vessel,stimulation operations were ready to commence. Per the program the sleeve activation balls were droppedvia the ball dropper made in line with the treating iron per the P&ID diagram shown in Figure 14. Theballs were dropped in an HEC gel pill and displaced with the cement unit. Once the ball was dropped andthe gel pill spotted, the treatment commenced via the stimulation vessel. The purpose behind dropping the

Figure 12Modeled Torque and Drag Chart

12 IPTC-18113-MS

ball in a gel pill was to ensure that the balls landed on seat when required. It should be noted that someballs landed as designed, but most did not and there were no surface indication of sleeve shearing seenfor stages 3, 5 and 12. Most balls landed early. Prior to commencing operations, reviews of previousoperations in the Eagleford were done to help gain from learnings in dropping the balls and opening thesleeves. Spotting the balls in a gel pill was a learning used from these reviews, however, the timing ofopening the sleeves with the balls was and still is an unknown element. Further review or operationalconsiderations needs to be made to ensure proper stimulation on future wells. Some observations of theball dropping done on this well are noted below for consideration in subsequent operations;

1. The system is dynamic and with each stage the balls are getting larger and the distance the ball

Figure 13Openhole packer setting via the bottomhole gauges

Figure 14P&ID of temporary pipe work on the rig

IPTC-18113-MS 13

travels shorter.2. The ball drop method of shutting down to spot the ball with gell then switching over to the pumpsfrom the stimulation vessel, could have allowed the balls to drop out of the gell pill.

3. Upon surface inspection it was found that all sleeves had sheared. The fact that the packer ball(1.125 OD) landed 100 bbls early leads one to believe that the other sleeves sheared early. The1.125 ball was the smallest OD ball dropped. The ball sizes increased by 0.125 for each stagethereafter.

4. Workstring design, specifically the IDs that the balls must pass can impact ball speed. It istheorized that the balls once entering a restriction will speed up and continue at a higher transportrate after passing a restriction, thus adding to the early landing of the balls.

5. It is recommended to adjust the pad stages between the acid, ball drop, and post flushes to allowfor sleeve shear inaccuracies. Based on item 4 above one could assume a much higher transportrate once the ball is in the WP, thus the displacement rate should be slowed prior to the ballentering the WP. The rate for each ball size and workstring configuration is not known.

6. On stage 3 the rate was increased to land the ball from 5 bpm to 10 bpm, thus it is surmised thatthe sleeve sheared too fast to be seen with the gauges.

7. Options to pin the sleeves to a higher value should be considered. All sleeves were set to shear at2500 psi with stage 6 set at 4500 psi. During the system onshore SIT we did not see the large2.625 ball sleeve shear even at a low rate of 2 bpm. Water hammer effects could have caused thesleeve to shift without any pressure indications via the gauges. Visual indications where observedduring the onshore test with the pipe jumping.

8. It is critical that all sleeves open and that the system does not lock-up, so sleeve pinning valuesneed to be reviewed and considered with a great deal of care. It is recommended to performadditional onshore testing of pinning configuration; this could also be used to learn more about balltransport speeds and sleeve shearing.

The treatment was pumped with all volumes displaced. One of the two primary pumps on thestimulation vessel went down while pumping the first stage of the treatment. To compensate for this lossin hydraulic horsepower a back up pump on the stimulation vessel and the rig cement unit was broughtonline once the acid had passed below the landing string to aid in achieving target rates, see Figure 15.The cement unit was able to add 2.0-2.5 bpm to the treatment, with total treating rates ranging from 15-18bpm throughout the job versus the design rate of 19 bpm. Limited entry diversion models were reviewedbased upon the achievable rate of 12 - 15 bpm with the pumps remaining on the stimulation vessel andwith the addition of the cement unit. The rates checked were 10 bpm and 16 bpm against the design of19 bpm see Figure 15 that summarizing this analysis.

The purpose of the diversion model is to calculate the rate, pressure, hole size, qty of holes and intervalspacing for the washpipe to liner annulus to ensure that sufficient volumes are displaced across the fullinterval to be treated. The desired range was checked against what was achievable to ensure adequate acidcoverage with the limited entry system. At 16 bpm we could achieve the required pressures and diversionrates. Due to concerns about over heating the cement unit it was only used while pumping the acid intothe formation.

Figure 15Diversion model calculated results

14 IPTC-18113-MS

Post job Stimulation AnalysisThe well was stimulated from the toe to the heelwith stage 1 at the toe and stage 12 near the heel, bydropping successively larger balls for each of theappropriate sleeves. Figure 16 shows the InjectivityIndex for each stage. This shows the affects of theacid on the formation. From this data one can seethe breakdown of the formation as the acid hits.Once the Brine post flush enters the formation theInjectivity index decreases to a value greater thanthe pre-flush brine stage indicating etching orwormholing of the reservoir.

The Injectivity Index was calculated by sub-stracting out washpipe and limited entry frictionpressures for the brine and the acid blend. Thetreating pressures used to determine the InjectivityIndex were from the downhole pressure gauges runjust above the GP packer setting tool. The pipefriction from the washpipe is a function of thedistance from the gauges, pipe roughness, ID, fluidtype and rate, see Figure 17.

The limited entry friction was determined bycalculating the pressure loss in the liner ID to wash-pipe OD annulus and across the limited entry holesat a give rate. The objective of the design is toensure that one gets adequate acid coverage over the full interval or stage. Thus the design is a balanceof hole sizing and available Hydrualic Horsepower (HHP) to achieve effective rates at each of the LimitedEntry holes. Once the hole sizing and HHP where balanced then we can determine pressure loss at eachstage for a given rate, see Figure 18.

An example of the treatment at Stage 8 is shown in Figure 19. The table shows the actual calculationsat given points during the treatment. These points are defined by the dotted lines as indicated in thepumping chart on the bottom right, with the red line indicating pump pressure and the blue pump rate. TheBottom Hole Treating Pressures (BHTP) at each of these points where plotted with the Acid Stage shown

Figure 16Treatment Injectivity Index per stage

Figure 17Washpipe friction as a function of distance from the Down-hole gauges

Figure 18Limited Entry Liner pressure drop for a given hole size

IPTC-18113-MS 15

in the shaded section. We can see from the pressures and Injectivity Index that the formation did breakback a bit, but not as was expected. It is theorized that the Limited Entry design may have masked someof the effects of the acid desolving the formation. Once the pumps where shut down we could see that thereservoir trapped around 1700 psi between the well and the check in the treating lines. This was evidencethat no secondary permeability was located by the treatment.

Prior to pumping the first stage an injectivity test was conducted. During this test the ISIP wasdetermined to be 7183 psi. Based on this value we can see that the bulk of the treatment for this stage wasdone at or above fracture pressures. Also if we assume a homogenous formation then we can expect thatwe did not have any communication between stages. The Injectivity Index is used to normalize treatingpressures at various rates during the job. Figure 16 above shows the Injectivity Index that varies fromstage to stage. For some of the stages it was difficult to determine the treating pressures and rates. For allstages we saw the pressure decline when the acid hit the formation and increase as the brine began todisplace the acid from the treatment. It is believed that based on these results that no secondary perm wasfound by the treatment and that once the acid was spent the brine was injecting into low permeable rock.

ConclusionsThe Limited Entry Liner with Retrievable Ball Drop Diversion System (RBDDS) proved to be anoperationally efficient system with significant potential to effectively matrix treat and develop deepwatercarbonate reservoirs. The 2000 meter system was run, set, stimulated and secured in just under 9 days. Thetime to complete and stimulate the lower completion was 3.5 days less than a similar concept (Jouti et al.2011). In addition, by designing safety into the system, significant risk related to pumping acid, pressuretesting and potential exposure to hazardous chemicals was greatly reduced. The Albian Carbonate wellwas drilled and completed as the longest horizontal reservoir section and longest step-out of any well inBrazil to date.

Pre-job testing and extensive onshore preparations had a large impact on job execution success. Thissystem is believed to become the system of choice for deepwater carbonate wells requiring matrixstimulation.

Figure 19Stage 8 Analysis summary

16 IPTC-18113-MS

AcknowledgementsMaking step changes in technology and operations is challenging, so the valued input of many individualsgreatly enhanced the design and operational effectiveness. The authors would like to thank the manyindividuals that contributed to the design, development and execution of this system. We would also liketo thank Shell for permission to publish this paper.

References1. Furui, K., Burton, R.C., Burkhead, D.W., Abdelmalek, N.A., Hill, A.D., Zhu, D., and Nozaki, M.,

2010, A Comprehensive Model of High-Rate Matrix Acid Stimulation for Long Horizontal Wellsin Carbonate Reservoirs, Paper SPE 134265

2. Hansen, J.H., Nederveen, N., 2002, Controlled Acid Jet (CAJ) Technique for effective singleoperation Stimulation of 14,000ft long reservoir sections, Paper SPE 78318

3. Stimulation Field Guidelines, Part III Carbonate Stimulation, Shell International Exploration andProduction, Dec 1999,

4. Gdanski, R., A Fundamentally New Model of Acid Wormholing in Carbonates, 1999, Paper SPE54719

5. Lechner, M., Ernst, S.D., Pitts, M.J., Lopdrup, T.P., Jaafar, M.R., Case Study: Improved ReservoirManagement From a Surface Controlled Two-Zone Open hole Packer Completion in a HorizontalWell in Al Shaheen Field, 2009, Offshore Qatar, Paper IPTC 13671

6. Rodrigues, V.F., Neumann, L.F., Torres, D., Guimaraes, C., Torres, R.S., 2007, Horizontal WellCompletion and Stimulation Techniques A review with emphasis on Low-Permeability Carbon-ates, Paper SPE 108075

7. Neumann, L.F., Fernandez, P.D., Rosolen, M.A., Rodrigues, V.F., Neto Silva, J.A., Redroso,C.A., Mendez, A., Torres, D., 2010, Case Study of Multiple Hydraulic Fracture Completion in aSubsea Horizontal well, Campos Basin, Paper, SPE 98277

8. Azevedo, C.T., Rosolen, M.A., Neumann, L.F., Melo, L.F., 2010, Challenges Faced to ExecuteHydraulic Fracturing in Brazilian Pre-Salt Wells., Paper ARMA 10-212

9. Damgaard, A.P., Bangert, D.S., Murray, D.J., Rubbo, R.P., Stout, G.W., 1992, A Unique Methodfor Perforating, Fracturing and Completing Horizontal Wells, Paper, SPE 19282

10. Surjaatmadja, J.B., McDaniel, B.W., Cheng, A., Rispler, K., Rees M.J., Khallad, A., 2002,Successful Acid Treatments in Horizontal Openholes using Dynamic Diversion and instantresponse downhole mixing An In-Depth Post job Evaluation, Paper, SPE 75522

11. Soliman, M.Y., East, L., Adams, D., 2008, Geomechanics Aspects of Multiple Fracturing ofHorizontal and Vertical Wells, Paper SPE 86992

12. Rodrigues, V.F., Fernandez, P.D., Rosolen, M.A., Franco de Almeida, M.L., Neumann, L.F.,Lima, C.B.C., Surjaatmadja, J.B., Miranda, C.G., Carneiro, F., 2005 First Application of aMultiple Fracturing Method in Noncemented Horizontal Offshore Wells, Paper SPE 94583

13. Jouti, I., Rafainer, G., Ferreira, A., Vidal, J., Villanueva, G.J., Canas, J., Rinto, A., Cancio, D.,Gigena, D., Scarcelli, D., Landinez, G., Barreto, W., 2011, Challenging Horizontal Ophen HoleCompletion in Carbonates: A Case History on Mechanical Isolation and Selective Stimulation inCampos Basin, Brazil, Paper, OTC-22417-PP

14. Al-Naimi, K.M., Lee, B.O., Bartko, K.M., Kelkar, S.K., Shaheen, M., Al-Jalal, Z., Johnston, B.,2008, Application of a Novel Open-hole Horizontal Well Completion in Saudi Arabia, Paper SPE113553

15. Mendes, G.D., Fowler, M.A., Hensgens, S.K., 2014 Electronic-Set Openhole Packer Installationin Campos Basin, Offshore Brazil: A Case History, Paper, SPE-170264-MS

IPTC-18113-MS 17

Unconventionals Meets Deepwater; Lower Completion Limited Entry Liner with Retrievable Ball Drop ...IntroductionBackgroundWell Functional Specification/RequirementsConcepts ReviewedDetailed DesignTestingFLCV shifter testSIT of the surface ball drop system and diversion sleeves

Openhole IsolationExecutionPost job Stimulation AnalysisConclusions

AcknowledgementsReferences