SPE-170789-MS

Deepwater Gas Zone Shutoffs by Wireline in Openhole Gravel Pack

Completions
Claudia Zettner, Hank Bensmiller, and Tom Lane, ExxonMobil Production Company, a division of Exxon Mobil
Corporation

Copyright 2014, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Annual Technical Conference and Exhibition held in Amsterdam, The Netherlands, 2729 October 2014.
This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
An operating company successfully executed two wireline through tubing gas shutoff (GSO) workovers
in high angle openhole gravel pack (OHGP) completions to isolate high gas oil ratio (HGOR) zones,
resulting in significantly increased oil production. This paper summarizes the GSO concept, production
logging results, completion modeling, job planning, and prejob surface testing of the completion
equipment, job execution, and production results.
The GSO workovers were performed on a tension leg platform (TLP) in a deepwater offshore field.
Isolation of HGOR zones was achieved by installing a wireline deployed stackable straddle system across
the heel of the OHGP interval. Although both wells have OHGP completions for sand control, effective
isolation of the gas zone was predicted using completion modeling and was validated by sustained
production performance. These wells do not have external casing packers (ECPs) to prevent flow through
the gravel pack interval. Isolation of the gas along the annulus is provided only by increased hydraulic
resistance in the gravel packed annulus.
The candidate wells had unfavorably HGORs attributed to gas breakthrough from ongoing gas
reinjection. To evaluate if production could be improved by reducing gas production, detailed mathematical modeling of an isolation straddles impact to inflow from the reservoir through the completion
was performed. Results suggested that internal isolation could provide effective GSO without ECPs.
Optimal straddle length was determined to help minimize production from the HGOR zone without
affecting low GOR zones. In preparation for the workovers, multiple surface tests were conducted to test
installation and retrieval of the completion equipment as well as integration and function of tools and
equipment from multiple service companies. Work on the first well included the first use of coiled tubing
(CT) by the operator on a TLP for a workover. After the CT milling of downhole restrictions, 16
components were sequentially installed by means of electric line (e-line) of a stackable straddle system
across the high gas zone. After straddle installation, both wells transitioned from being cycled to
continuous lower GOR production. The first well produced 5,000 BOPD at 6,000 GOR before the GSO,
and was increased to 10,000 BOPD at 1,000 GOR. The second well produced 6,000 BOPD at 6,000 GOR
before the GSO, and 10,000 BOPD at 500 GOR afterward.

SPE-170789-MS

Figure 1Conceptualization of gas cap moving into producer.

The keys to success involved strong teamwork and support from multiple service companies using CT
milling, completion equipment, wireline, slickline, and conveyance tools. Significant scalability was
achieved during planning of the second workover based on observations from the first.

Introduction
All producers in the field of study were drilled from a 36-slot TLP at a water depth of a little more than
1000 m. Reservoirs are deepwater channel depositional complexes that contain multiple sand systems with
some degree of pressure communication between these zones. Development wells typically encounter
multiple zones. OHGPs are the preferred completion option in this field to provide high rates. It was
recognized during the project design phase that controlling unwanted gas and water in the future would
be challenging. The wellwork solution proposed and implemented can address that challenge by providing
lower GOR oil and additional reserves recovery.
Although the long-term depletion plans for most reservoirs in the field rely on waterflooding, some
producing reservoirs have been used to store surplus gas until gas export facilities were operational. As
these injected gas caps expanded toward producers, GORs increased to the point where processing
facilities were gas constrained. Producer well seriatims based on GOR are used to optimize oil production.
The highest GOR wells are typically cycled to manage GORs and then shut in when cycling is no longer
effective.
Engineering principles and field experience have shown that gas breakthrough in any zone in a well can
significantly impact overall production. High gas mobility coupled with good injection support in the gas
producing interval can result in a HGOR, non-competitive producer.
Fig. 1 illustrates what can often happen during this scenario. As the gas cap expands toward the
producer, the gas leading edge breaks into the top of the completion. High gas mobility results in the
producing GOR quickly escalating. This can result in inefficient gas cycling in the reservoir between the
injector and producer.
The completion diagram in Fig. 2 illustrates the general design of the lower completion, noting that the
OHGP was not run with ECPs. External shunt tubes prevented ECPs on early wells. Later wells have
internal shunt tubes that allow external packers in the gravel packed section of the well. Completions
include GPs to minimize sand production from the unconsolidated reservoirs. Gravel pack screens are
shunted to help ensure gravel is transported and placed along the length of the screen. Gravel placement
is generally effective and sand production is rarely a problem in this field. The gravel pack annulus is
packed with 30/50-mesh ceramic proppant along the length of the 8.5-gauge screen with 5-in. 15-ppf base
pipe.

SPE-170789-MS

Figure 2Completion sketch for Wells 1 and 2.

Figure 3Well path for Well 1.

Oil producers are directionally drilled because most reservoir objectives are not directly underneath the
TLP. Reservoir penetrations are often at a high angle from vertical to provide the permeability thickness
necessary to deliver target rates. A typical well path is provided in Fig. 3.

GSO Concept and Supportive Modeling

Concept
Options to shut off gas production in the heel of an OHGP are limited. GSO polymers are used in some
applications, but generally are not a permanent solution. Squeezing fine cement through the screen and
then drilling out was considered, but rejected, because of execution complexities and concerns regarding
seal effectiveness. The recommended gas shutoff option was installation of a straddle across a high GOR
zone to shut off production from this zone using a straddle long enough to increase the hydraulic
resistance, as illustrated in Fig. 4. This reduced flow from the high GOR zone allows additional oil
production from other lower GOR zones and additional production from other wells in a gas constrained
environment. This smaller diameter straddle would create a restriction, but modeling indicated that it
would have a minor effect.

SPE-170789-MS

Figure 4 Gas shutoff concept in openhole gravel pack. The completion without the straddle (left) and a straddle across the high GOR zone in the
OHGP completion limits gas production by forcing gas around the straddle (right).

The straddle would be installed across and extend below the zone with the excessive free gas
production. The straddle seals the screen with packers and helps prevent production through that section
of the screen. Gas could still flow from the zone by traveling through the reservoir, downhole through the
gravel packed annulus, or through any gap between the straddle and the inside diameter (ID) of the gravel
pack base pipe and around the bottom of the straddle into the well. However, its ability to flow is reduced
by the tortuous path through the gravel pack and around the straddle, as shown in Fig. 4.
Modeling
Detailed mathematical well flow modeling was conducted to determine whether the gas could be
adequately restricted by a straddle in an openhole gravel pack completion without zonal isolation (swell
packers) behind casing, without requiring a prohibitively long straddle or adversely affecting rate.
Scientists affiliated with the operating company completed this proof of concept analysis. The multiphase
flow was modeled around the straddle and packers, through the gravel packed annulus, and through the
annular space between the straddle and the inner diameter of the gravel pack base pipe, as shown in Fig.
5. The software package used research versions of well performance analysis software as well as
computational fluid dynamics to supplement the analysis. The reservoir model for the inflow calculation
was matched to the production log and well tests. Analysis included straddle length and placement of
packers. Modeling results indicated that the gas could be sufficiently restricted, even without zonal
isolation in the OHGP completion. Despite the modeling results, concerns still existed that gas would
quickly bypass the straddle.

Reservoir Background
Numerous channel complexes exist in the field, and each has a development plan to maximize recovery
given its specific reservoir and fluid properties. The two GSO wells are located in a channel complex that
is divided by two major cross-channel faults, as shown in Fig. 6. Updip gas injection (GI) is used in both
reservoirs. The larger, western reservoir area also includes two downdip water injectors (WIs), although

SPE-170789-MS

Figure 5Sketch of flow path modeling for GSO proof of concept.

they are water alternating gas (WAG) capable, and

one has recently been used to also inject gas. The oil
producers (OPs) and crestal GI wells are drilled
from the TLP. Reservoir pressure data indicates
water and gas injection has been successful for
supporting reservoir production. Current reservoir
pressure is close to the initial reservoir pressure.
The average field producing GOR has steadily
increased as a result of expanding gas caps and
cycled reinjected gas. Before the GSOs, both of the
candidate wells were producing 6,000 GOR, above
Figure 6 Generalized reservoir map.
the field average solution GOR of 700. Well performance was generally consistent with cumulative
reservoir production, injection, reservoir volume,
and structural location of each producer. Production was choked at the wellhead to reduce flow so that
total field production could be managed within the gas compression facility limits. These wells were also
cycled to optimize field production by (1) reducing the production contribution from the highest GOR
wells while (2) resting them to allow gas cusps in the reservoir to recede. Nonetheless, well GORs
continued to increase because of gas cap expansion.
GSOs were performed in Wells 1 and 2 on the map. Well 1 is located in the eastern reservoir area in
a downdip location relative to the GI well. This reservoir area is bound by the two north-south faults with
limited connectivity to other fault blocks. Well 2 is located in the larger western reservoir area and is also
in a relative updip location in a position to be directly influenced by updip GI. Although located within
different fault blocks, the wells are along the same depositional channel system and have a broadly similar
vertical sand distribution based on openhole logs.

SPE-170789-MS

Figure 7Well 1 daily production before workover.

Figure 8 Well 1 openhole log and completion interval.

Figure 9 Well 1 production log.

Workover Candidate SelectionWell 1

Production was initiated from Well 1 in 2004. Offset GI was initiated in August of 2005. As shown in Fig.
7, Well 1 initially produced up to 20,000 BOPD at low GOR. In 2006, the well began a steep oil
production decline as it was choked back to manage GOR, which, by then, had increased to more than
2,000. Gas breakthrough was evident within months of the initiation of gas injection in the offset well. By
late 2007, well cycling was initiated and production was below 5,000 BOPD for the days the well was
online (Note: because the production plot uses daily values, the cycling period appears to be an area of
solid green on the plot). Oil production and GOR improved somewhat in late 2009 as a result of an
extended shut-in period that year. By 2012, the GOR climbed to 6,000 and the well was only on
production approximately 30% of the time. Because this GOR was higher than the field marginal GOR,

SPE-170789-MS

Figure 10 Views of the 16-row insert step mill.

Figure 11The straddle assembly and the relative placement across the high gas zone in the completion.

Well 1 did not add substantially to field incremental daily production when it was produced. Water
production is minimal in this fault block because of the lack of a natural aquifer.
Well 1 was completed across multiple sands, as shown in Fig. 8. Three primary sands are present within
the 500-m completion, notionally represented as the Upper, Middle, and Lower sands. The three sands are
also present in the updip GI well. All were considered to have good permeability and connectivity to the
GI. Based on well and reservoir performance, it was expected that free gas had broken into only one or
two of the zones, creating the opportunity to restore Well 1 to a lower GOR by isolating the zone(s)
producing excess gas.
Based on the simulation work performed to investigate GSO effectiveness in the downhole completion
assembly, the team was increasingly confident that a straddle assembly could substantially reduce gas
influx in the OHGP. The next step of the workover process was to run a production logging tool (PLT)
to identify the source of the free gas production. Data provided by the PLT allowed discerning oil, water,
and gas fractions along the completion while the well was flowing and shut in. As shown on the PLT GR
track in Fig. 9, Well 1 was determined to have a high gas influx in the Upper sand. No free gas was
observed from the Middle and Lower sands. Results also indicated that approximately two-thirds of the

SPE-170789-MS

oil was from the Middle and Lower sands. Shut-in passes showed crossflow into the Lower sands
indicative of lower pressure, possibly attributed to uneven GI support. If the gas in the Upper sand could
be shut off and drawdown increased on the Middle and Lower sands, it was reasonable to expect that low
GOR production could be restored. One key preworkover risk was production sustainability from the
remaining zones. Because of the wells high GOR history, uncertainty regarding reservoir vertical
connectivity between zones, and proximity to the gas cap, early gas produced from the Middle sand after
the PLT run was a possibility, but not significant enough to prevent the project moving forward.

Workover Planning
The subject GSO workovers were the most complex wellwork conducted by the operator to date on a TLP
because of the required CT milling of tubing ID restrictions and the setting of a 150-m straddle section
requiring e-line with a tractor to set 16 separate components. Because of the complexity of the workover,
significant effort was made with respect to planning and is described in the following four sections: vendor
engagement, new equipment operation, procedure development, and contingency planning.

Vendor Engagement
Services were provided by four suppliers during the first workover. The service companies gave this
workover the same level of priority and focus as the operator. Equipment reliability and crew competency
were very important factors to the success of these workovers because of the complexity and the
cumulative risk from the sequential e-line runs. Any small improvement to equipment reliability would
dramatically improve the opportunity for a successful workover. The service companies participated in
planning, numerous equipment tests, procedure reviews, and well-on-paper exercises. Global subject
matter experts (SMEs) for the tools and equipment were made available to the local team and operator.
The local crews made efforts to understand operation of the new tools and potential troubleshooting before
traveling offshore. Service providers supplied experienced personnel and largely maintained the same
crews for both workovers. Normal crew rotations were changed to provide two crews in-country at the
same time for one month of 24-hour operations, followed by a month off. They also used quality plans
for these workover operations. Quality plans for work execution specified items, such as pre-job
checklists, detailed vendor procedures, modeling, and competency assurance.

New Equipment Operation

This workover was the first use of CT equipment on a TLP by the operator. A motion compensation
system was used because of concerns regarding the relative motion of the TLP (heave) during milling
operations. The straddle equipment for this workover was also new to the operator. A series of system
integration tests (SITs) were conducted on the CT milling, motion compensation system, and straddle
equipment to test the equipment, develop procedures, and confirm tool functionality and communication
between service providers and crews.
CT Milling SIT
The objective of CT milling was to enlarge a nipple restriction in the lower completion from an ID of
3.895 to 4.393 in., over a length of 5 in., at an approximate 2500-m working depth to allow the 4.3-in.
outside diameter (OD) straddle packers to pass through. The nipple was 13Cr stainless steel, which can
be challenging to mill because of its material toughness. A SIT was conducted in Texas to develop a CT
milling procedure for the proposed work and to test the effectiveness of several mills and milling
techniques. Key operational learnings from this testing were incorporated into the procedure. The final
mill design was a 16-row insert step mill with 3.3-in. length transitions. Fig. 10 illustrates this mill.
Milling performance of several motors were also compared.

SPE-170789-MS

CT Motion Compensation System SIT

Concern existed regarding the relative motion from
the TLP during the milling operation, so a motion
compensation tower was used. The tower moved to
offset heave or vertical motion of the TLP relative
to the fixed well. The riser contained a special
titanium component to help absorb the horizontal
motion of the TLP. A loading analysis on the CT
stack-up to verify that the stresses and strains were
within acceptable ranges and a deck loading analysis were also conducted.
A SIT was conducted in the United States where
the system was stored and included several members of the in-country crew. The objectives of this
SIT were to develop a rig-up procedure including
equipment spacing and to train local personnel from
the service provider. Improvements were made for
equipment handing and reducing time working at
height. A plan for rescue at height was developed.
After the equipment was shipped for this specific
job, a second SIT was conducted onshore to familiarize the two CT crews with rig-up and rig-down
operations of the motion compensation system.
Straddle Equipment and Installation
The straddle assembly used was designed for wireline installation and retrieval. It is a modular system
consisting of retrievable packers, straddle (3.5-in.
OD, 2.935-in. ID tubing) sections with special anFigure 12CT motion compensation system.
chors that snap together. Packers and straddles can
be retrieved by shearing connection pins with upward pull. The connections are designed to shear one component at a time from the top.
The maximum length straddle that the 90-ft mast and lubricator could accommodate was 10 m. The
planned 150-m straddle length for the first workover would require 16 sequential wireline runs to install
the four packer sections and 12 straddle sections. Fig. 11 illustrates the straddle assembly and relative
placement across the high gas zone.
Several SITs were conducted for the straddle equipment, including running the straddle and setting and
retrieving packers. Additional testing was conducted onshore for tool string operation with the straddle
equipment and the wireline and conveyance service providers.
Straddle Equipment SITs
SITs both in a lab setting and in a test well were conducted to test deployment and retrieval of the straddle
assembly using e-line, tractor, stroker, and downhole power unit (DPU) packer setting tools. The
objectives of the tests were to determine whether tractor, running tools, and assemblies could fit through
the tight tolerances; testing of the mechanical aspects of the straddle assembly; calibrating the tractor
surface data with packer and anchor latch movement and setting force; verifying quality assurance/quality
control (QA/QC) plans; helping ensure the accuracy and completeness of assembly procedures, operating
procedures, and contingency plans; and improving procedures to increase efficiency, reliability, or
functionality during the field application.

10

SPE-170789-MS

Tools String SIT

The straddle installation involved three different
service providers one provided completion equipment (packers and straddles), another provided the
tractoring and stroker, and a third provided wireline.
Again, testing was performed in advance with the
personnel from the various service companies to
ensure compatibility of tools, control systems, and
equipment. Operation of two basic tools strings was
testedpacker and straddle settings.

Procedure Development
A series of wellwork-on-paper exercises was
conducted by the service companies and the operator to plan the workover. These wellwork-on-paper
meetings are a best practice and are expected
when performing non-routine wellwork globally.
The objective was to develop a single plan and
procedure supported, understood, and executed by
the all four service companies and operator. With
the number of participants, communication and coordination were critical. When the work was first
proposed, management participated in preliminary
Figure 13CT rig-up.
meetings. Technical and operational staff (engineers, technicians, workover supervisors, and
crews) participated in these meetings to help develop detailed procedures. A final wellwork on-paper meeting was conducted with the crews immediately before the job to review the procedure. Assembling wellwork contractors to obtain their input is
always valuable and results in a superior procedure.
Wellwork-on-Paper Deliverables

Communicate expectations: This workover would not be successful unless it was completed
safely and the well was left in no worse condition.
Procedure: The workover procedure combined multiple detailed vendor procedures. Each of the
vendor procedures had to be consistent.
Work schedule: Efficient 24-hour operations were required; this necessitated modifying the
rotations of crews so that two crews would be in-country at the same time, instead of the normal
staffing for 12-hour operations.
Technical details: Technical details that impacted multiple service providers deck layout, tool
string lengths, break points, etc.were agreed upon.
Stop points: Critical decision points during the workover were clarified and the responsible party
was identified (e.g., who was responsible for depth correlation and who was responsible to set the
first packer).
Equipment readiness: Equipment readiness and status were also important during the well-onpaper exercises.

SPE-170789-MS

11

Contingency planning: Key operational

risks were identified during the planning
processes. These risks were either mitigated
by additional testing or analysis to identify
lower risk alternatives or procedural
changes.

Workover 1 Execution
Workover 1 was completed in three phases to manage equipment, crew schedules, and uncertainty
during operations. All phases of the operation were
conducted safely without incident or injury to personnel.

Slickline Diagnostics
The operation began with a series of slickline diagnostics to confirm clearance to the working depth
for the straddle installation. This was conducted to
allow sufficient time for planning if diagnostics
were unsuccessful.
Phase 1 Operational SummaryConfirming
Clearance and Completion IDs
1. Handover well from production operations.
2. Mobilize, spot, and rig up 0.125-in. slickline
equipment.
3. Pressure test and preform prejob checks.
4. Perform drift runs/lead impression blocks
(LIBs) to nipple profile and total depth
(TD).
5. Demobilize slickline personnel.
6. Handover well to production operations.

Figure 14 RIH with lower packer.

CT Milling
CT milling was the second phase of the operation. The objective was to mill a tubing nipple restriction
to allow the straddle packers to pass through. Figs. 12 and 13 illustrate the CT motion compensation
system and the CT rig-up on the deck of the TLP.
Phase 2 Operational SummaryMill Nipple Restriction from 3.895- to 4.393-in. ID to Allow 4.3
OD Straddle to Pass Through
1. Handover well from production operations.
2. Mobilize, spot, and rig up CT equipment.
3. Make up milling bottom-hole assembly (BHA) and run to depth.
4. Mill out nipple restriction using information gathered from SIT.
5. Continue to run in hole (RIH) through nipple restriction.
6. Make up straddle drift and drift through nipple restriction to 20 m below the proposed lower packer
setting depth.
7. Rig down CT.

12

SPE-170789-MS

Figure 15E-line rig-up with scaffold.

8. Handover well to production operations.

Straddle Installation
Straddle installation was the last phase of the operation. After the crews returned from time off, a
wellwork-on-paper exercise was performed to familiarize the crews with the upcoming work before
traveling offshore. After rig-up, the straddle installation was conducted during 24-hour operations using
e-line and tractors.
Phase 3 Operational SummaryInstalling GSO Straddle
1. Handover well from production operations.
2. Slickline drift.
3. Spot and rig up e-line equipment.
4. Run gamma/casing collar locator (CCL) for depth correlation.
5. Run and set bottom packer c/w mule shoe.
6. Run and set modular straddle Sections 1 through 4 (approximately 10-m long each) sequentially.
7. Run intermediate packer assembly.
8. Run and set modular straddle Sections 5 through 8 (approximately 10-m long each) sequentially.
9. Run intermediate packer assembly.
10. Run and set modular straddle Sections 9 through 12 (approximately 10-m long each) sequentially.
11. Run upper packer assembly.
12. Rig down equipment and hand back well to production operations.
Fig. 14 illustrates the lower packer assembly. Fig. 15 shows the e-line rig-up with scaffolding. Fig. 16
illustrates the stroker and tractor being laid down.

Workover 1 Results
Post-workover, well performance illustrated that the GSO successfully reduced unwanted gas production.
Well 1 demonstrated significantly improved well performance, with a much lower flowing wellhead
pressure (WHP) and a higher wellhead temperature, indicative of lower GOR production. The stabilized
test rate over several days averaged approximately 10,000 BOPD at 1,000 GOR on a 36% choke. This

SPE-170789-MS

13

represents an average production increase of 8,500

BOPD with a 5,000 GOR reduction. As shown by
the production plot in Fig. 17, production improved
dramatically because of the GSO workover. Performance to date confirms a durable change to the
completion inflow profile, with a GOR increase in
mid-2013 related to continued gas cap expansion.
Almost six million barrels of oil have been produced since the GSO.
An indication of GSO performance effectiveness
can be drawn from the post-workover GOR. Comparing the producing GOR of 1,000 to a current
solution GOR of 700, free gas production of three
million Mcf/D is observed. Presuming this is moving through the GSO straddle/OHGP assembly, the
calculated GSO isolation efficiency is approximately 90%.

Workover Candidate SelectionWell

2
After the success of Well 1, efforts were focused on
a similar GSO opportunity within Well 2, which is
located in a relative updip position, offsetting crestal GI. Downdip WI began shortly after reservoir
production startup in 2004. GI was initiated in August of 2007 and production from Well 2 began in
November of 2007. As shown in Fig. 18, early
production reached 20,000 BOPD before increasing
GOR led to well cycling in 2010. Production cur- Figure 16 Laying down tractor and stroker after setting straddle.
tailment and cycling were used from 2010 through
2013 to mitigate GOR growth, which had by then reached 6,000 GOR. Before the workover, Well 2
operated approximately 30% of the time with minimal water production.
Well 2 was completed across multiple sands, as shown in Fig. 19. Three primary sands are present
within the 300-m completion, represented as the Upper, Middle, and Lower sands, just as in Well 1. The
sands are also present in the updip GI well. All were considered to have good permeability and
connectivity to the GI well. Based on well performance and an offset producer PLT, it was expected that
free gas was being produced in the Upper sands, providing the opportunity to restore Well 2 to a lower
GOR, similar to that implemented in Well 1.
A PLT was run to confirm the gas entry source in the well. The wellbore trajectory for Well 2 has a
1000-m horizontal sail section, thus requiring an e-line tractor. As shown on the PLT GR track in Fig. 20,
Well 2 was confirmed to have a high gas influx in the Upper sand. No free gas was observed from the
Middle and Lower sands. PLT results also suggested that production from the Middle and Lower sands
was significantly reduced by the high gas production from the top of the completion. It was concluded
that, by shutting off the Upper sand and increasing drawdown on the Middle and Lower sands, the well
could be restored as a low GOR producer.
After the production log was analyzed, an enhanced well performance model was created using well
performance analysis software to determine optimal straddle length and effectiveness. This enhanced

14

SPE-170789-MS

Figure 17Pre and post-GSO production for Well 1.

model took into account considerations from Well 1. The modeling workflow was similar to work
performed for the first GSO workover, but employed a simplified straddle geometry and used software
tools available to all engineers. Globally, enhanced well simulation models are used by reservoir and
subsurface engineers to optimize existing well production and to initiate new opportunities. The enhanced
well model takes into account the reservoir properties, fluid properties, upper and lower completion
hardware, well path, and flowline pressure to determine the modeled production. A well model was
developed and calibrated with well tests and the production log. Sensitivities were performed for the
straddle length and setting depth to quantify the input data uncertainty. Straddle setting depth was
determined based on reduction to gas, change in oil production, and consideration of operational risks
associated with installation of a long straddle.

Planning Efficiencies for the Second Installation

Several efficiencies were realized during planning for the second GSO workover. Although the first GSO
workover was completed several years after the production log that confirmed the opportunity, the second
GSO workover was completed just five months after its production log was run. This was accomplished
because of efficiencies in modeling, equipment testing, equipment availability, procedure development,
and return of the same crews.
Changes between the first and second workover were minimized. The crews performed testing of the
tool strings and practiced equipment integration before traveling offshore. The procedure from the first
workover was easily adapted for the second workover because the same straddle and conveyance
equipment were used. Even though the offshore crews were largely similar, a wellwork-on-paper exercise
was conducted to prepare the crew before traveling offshore.

Workover 2 Execution
Like the first workover, the second was completed without any safety incidents. Well 2 had a similar
completion to Well 1, but several differences affected wellwork planning:

SPE-170789-MS

15

Figure 18 Well 2 daily production before workover.

Figure 19 Well 2 openhole log and completion interval.

Figure 20 Well 2 production log.

Well 2 did not have the tubing restriction, so milling was unnecessary.
The well path was much more challenging for e-line with a 1000-m long horizontal sail section
above the straddle installation working depth.
The straddle length was shorter, requiring eight straddle sections and three packers to isolate the
gas zone.
Because of the sail angle, additional attention was paid to the e-line modeling. Like the first workover,
the installation of the packers and straddle sections proceeded per plan. Tool strings and equipment were
identical to the first workover, with the exception of the total straddle length. The service companies
provided almost all same crew for the second workover. Wellwork-on-paper exercises were conducted
during the planning stages in addition to immediately before traveling offshore. The straddle installation
was conducted in 20 days using 24-hour operations.

16

SPE-170789-MS

Figure 21Pre and post-GSO production for Well 2.

Operational SummaryInstalling the GSO Straddle

1. Handover the well from production operations.
2. Slickline drift.
3. Spot and rig up e-line equipment.
4. Run gamma/CCL for depth correlation.
5. Run and set bottom packer c/w mule shoe.
6. Run and set modular straddle Sections 1 through 4 (approximately 10-m long each) sequentially.
7. Run intermediate packer assembly.
8. Run and set modular straddle Sections 5 through 8 (approximately 10-m long each) sequentially.
9. Run upper packer assembly.
10. Rig down equipment and hand back well to production operations.

Workover 2 Results
As with Well 1, post-workover well performance confirmed a successful GSO. Well tests indicated Well
2 produced 10,000 BOPD at approximately 500 GOR on a 50% choke, representing a production increase
of 8,000 BOPD with a 5,000 GOR reduction. The production plot in Fig. 21 shows the results of the
mid-2013 GSO workover. Again, performance indicates that the workover provided a durable change to
completion inflow, with the modestly increasing GOR trend related to continued gas cap expansion. Over
three million barrels of oil have been produced since the GSO. Post-workover production is at solution
GOR, indicating the GSO effectiveness was 100%.

Conclusions
To date, two GSO workovers have been safely and successfully completed in wells with OHGP
completions. The straddle installations were executed per plan and have proved effective in terms of GSO,
as demonstrated by more than one year of production data for the first well and approximately one year
for the second well. Post-workover, both wells have returned to continuous production and well tests

SPE-170789-MS

17

reveal a significant reduction in GOR with higher oil rates. Post-workover, the first well has produced
almost 6 million barrels of oil and the second well more than 3 million barrels to date.
Considerable planning was involved during the preparation for the first workover using mathematical
analysis for the GSO concept, well candidate selection, equipment SITs, vendor engagement, and
procedure development. Considerable planning efficiencies were realized during the second workover.
Throughout this work, keys to success were identified as follows:

Concessionaire support to increase production and add reserves at low cost.

Well performance modeling to prove the GSO concept in an OHGP.
SITs to understand equipment operation, potential failure modes, and manage interfaces.
Use of a global engineering organization (Houston, Aberdeen) to plan special projects where
individual business units have smaller technical staff groups.
Excellent service company support both locally and globally to help ensure the success of this
first-of-its-kind wellwork.
Excellent wellwork supervisor planning and coordination. The same two wellwork supervisors
managed all offshore operations for both jobs.
Teamwork and communication between the operator and service companies to identify and resolve
problems during the planning stages.
Additional GSO workovers are currently being planned and executed by the operating company.
We based the information on data believed to be reliable on the date compiled, but we do not represent,
warrant, or otherwise guarantee, expressly or impliedly, the merchantability, fitness for a particular
purpose, suitability, accuracy, reliability, or completeness of this information or the products, materials,
or processes described. The recipient is solely responsible for all determinations regarding any use of
material or product and any process in its territories of interest. This presentation may include forwardlooking statements. Actual future conditions (including economic conditions, energy demand, and energy
supply) could differ materially due to changes in technology, the development of new supply sources,
political events, demographic changes, and other factors discussed herein. We expressly disclaim liability
for any loss, damage, or injury directly or indirectly suffered or incurred as a result of or related to anyone
using or relying on any of the information in this document. Exxon Mobil Corporation has numerous
affiliates, many with names that include ExxonMobil, Exxon, Esso and Mobil. For convenience and
simplicity in this paper, those terms and terms like corporation, company, our, we and its may
sometimes be used as abbreviated references to specific affiliates or affiliate groups. ExxonMobil, the
ExxonMobil Logo, the Interlocking X Device, and all other product names used herein are trademarks
of ExxonMobil unless indicated otherwise.

Acknowledgements
There were many significant contributors to this project from the initial idea through the design and
implementation of this program. The authors specifically thank Mauro Carvalho, Gordon Kornfeld, David
Howell, Mark Gardner, Mirick Cox, Ali Farah, Kean Seng Lee, Jairo Lima, Ginga Mateus, Alioskin
Sambo, Steve Brown, Keith Chalmers, Jose Paulo, James Hicks, Edson Dos Santos, Cheryl Trudell,
Marcos Vela, John Reedy, and the ExxonMobil Upstream Research Company.
Thanks are also extended to the state petroleum company and co-venturers for their project support and
approval of this paper.