OTC-26471-MS

A Successful Attempt to Insert Coiled Tubing Through Parted Production

Tubing with a Malfunctioned Downhole Safety Valve to Perform
Extralightweight Cement Treatment to Regain Pressure Integrity for Plug
and Abandonment
Jae Seok Yi, Jong Yong Lee, Hyun Jung Oh, Tan Khoa Nguyen, and Tran Thang Nguyen, Korea National Oil
Company; Tuanangkoon Daohmareeyor, Reawat Wattanasuwankorn, and Kritsana Kritsanaphak, Halliburton
Prapas Phayakrangsee Gagie Corporation

Copyright 2016, Offshore Technology Conference

This paper was prepared for presentation at the Offshore Technology Conference Asia held in Kuala Lumpur, Malaysia, 22–25 March 2016.

This paper was selected for presentation by an OTC 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 Offshore Technology Conference and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Offshore Technology Conference, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the
written consent of the Offshore Technology Conference 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 OTC copyright.

Abstract
This paper discusses an offshore coiled tubing (CT) application to address a complex well situation and
well control issue. Multi-engineering designs and 100% yard testing were used successfully during this
campaign.
Parted production tubing was identified some 50 m from the surface. The control line of the downhole
safety valve failed and one barrier was lost in the well. This paper describes efforts to regain pressure
integrity for a plug and abandonment (P&A) procedure in this live gas well. CT was inserted through the
parted production tubing in an attempt to regain pressure integrity.
A carefully planned cement placement strategy was critical to operational success. Key challenges and
solutions included the following:
● Wellhead pressure (WHP), friction, and a highly deviated trajectory affected CT pulling capabil-
ities. Therefore, extra-lightweight cement was used to reduce weight inside the CT.
● The downhole safety valve failed to close, so the downhole tool was redesigned to help ensure that
the tool would not stick.
100% Yard testing for the application was performed to simulate onsite, downhole operations.
Well pressure integrity is a primary priority for global operators. An option to regain pressure integrity
was introduced and involved running CT and placing cement into the live well. A combination of
multi-engineering designs aided this successful CT campaign. The CT was run through the parted
production tubing and the failed downhole safety valve. The well was killed and extra-lightweight cement
was successfully placed. The operator was able to regain pressure integrity and perform a P&A procedure.
The success of this case study indicated that this procedure could be used for future downhole equipment
and tubing failures in live gas wells.
2 OTC-26471-MS

Objective
The objective of this operation was to use CT to kill a live gas well with a parted tubing inside and shut
it off using extra-lightweight cement in a perforated interval for P&A.

Well History
● This was a conventional gas well completion with 13 3/8-in. surface casing, 9 5/8-in. intermediate
casing, and a liner hanger set with a 7-in. openhole slotted liner to total depth (TD). With 5 1/2-in.
production tubing from the surface to 13 m below liner hanger in an openhole section. A 5 1/2-in.
tubing parted 51.2 m from the surface was observed (Fig. 1).

Figure 1—Well completion schematic.

● A downhole camera revealed that both the tubing and a 5 1/2-in. downhole safety valve control
line parted. Then, a downhole safety valve was lost from the surface (Fig. 2).
OTC-26471-MS 3

Figure 2—Downhole camera showed parted tubing requiring the design of a CT downhole assembly.

● A downhole pressure survey recording showed the failure of the downhole safety valve, which still
maintained pressure from below without leaking out from the flapper valve, and pressure was built
up to the wellhead (Fig. 3).

Figure 3—Memory downhole pressure survey showed safety valve still maintains pressure from below,
4 OTC-26471-MS

CT Program Design
Enter Parted Tubing Stump
● Challenge: production tubing was parted and a gap created of ~50 m; coil entering could become
caught if it takes a trip down the wrong path of the tubing annulus. Figs. 4 and 5 illustrate the
downhole conditions.

Figure 4 —Simulation of potential coil stuck in annulus, top view.

Figure 5—Side view.

● Design solution: the bottomhole assembly (BHA) had been designed with an outer diameter (OD)
larger than the annulus gap to help eliminate the risk of coil tripping in the wrong path. Coil
entering the correct path also could be confirmed by tagging the coil to the downhole safety valve
at the known depth. An intensive yard test was performed by setting up different BHA configu-
OTC-26471-MS 5

rations to simulate the optimized design of the BHA (Figs. 6 through 9).

Figure 6 —Design of coil BHA with large OD figure.

Figure 7—Identification of coil entering correct path by tagging onto SSSV figure.
6 OTC-26471-MS

Figure 8 —Yard testing setup.

Figure 9 —Simulated CT run in downhole conditions.

Risk of Becoming Stuck Because of Potentially Damaged Tubing

● Challenge: the buckled or bent/twisted tubing in such an event can cause the friction between the
coil tripping in/out and the tubing to be significantly high, thus causing the coil to stick easily.
● Design solution: the coil running procedure was programmed to trip in the tubing in the hesitation
manner while checking coil weight every short distance to minimize the risk of the CT becoming
stuck (Figs. 10 and 11).
OTC-26471-MS 7

Figure 10 —Risk of coil stuck because of unknown tubing conditions.

Figure 11—Design to trip coil in with regular check weight every short distance.
8 OTC-26471-MS

Risk of Damaging SSSV Prematurely: Yard Test Simulation

● Challenge: the SSSV is a ⬙failed close⬙ design and its flapper tends to close against the coil body
when tripping in/out. This could potentially cause the coil to stick and/or damage the coil or the
SSSV itself in the event that operation becomes too complicated, leading to a high possibility of
operation failure.
● Solution: the risk was unavoidable by all means. The precaution was evaluated by intensive yard
testing with a duplicate SSSV and the coil metal lost/damage was justified as acceptable with
minimal movement of the coil across the SSSV (Figs. 12 and 13).

Figure 12—Flapper position during running of BHA.

Figure 13—Yard testing by pulling coil through SSSV.

Retrieval of Coil with Interference to SSSV: BHA Design

● Challenge: BHA contained components that have upset profiles (centralizer, cement big bore
nozzle, knuckle joint etc.); these components would be impossible to retrieve from the well against
the SSSVs flapper.
● Solution: BHA was designed to ⬙drop off⬙ components in the well (Fig. 14) and only the upper
section with slick OD design was to be retrieved after cement placement. Actual disconnecting tool
(Fig.15)
OTC-26471-MS 9

Figure 14 —Downhole assembly diagram.

Figure 15—downhole disconnect tool.

Cement Design: Challenge and Solution

Cement Placement Method with Restriction of Coil Tensile Strength

● A good cement treatment would require a proper placement method. Coil is necessary to remain
at the well’s TD to lay cement in the wellbore. Coil can then be picked up and the cement spotted
across the perforated interval to help prevent contamination and/or gas migration through the
cement (Fig. 16).
10 OTC-26471-MS

Figure 16 —CT cement placement positions.

● Challenge: normal cement is too heavy for the coil to pump and pick up from within deep depths
with full cement tensile strength of the coil (normal cement has minimum density of 12.8 lbm/gal).
Various coils were simulated for the tensile, but the weight simulation still showed similar results
of limitations; this is typical for pumping large volumes of cement through coil (Fig. 17).

Figure 17—CT force simulation shown exceeds yield from 12.8 lbm/gal cement density.

● Solution design: design a special lightweight cement with low density to allow the coil to be picked
up from the bottom with a full volume of cement. Advanced engineering had to be performed
because cement density had to be very low (9 lbm/gal) and still meet all requirements (density,
OTC-26471-MS 11

compressive strength, setting time, etc.) (Fig. 18).

Figure 18 —Force simulation for lightwight cement (9 lbm/gal).

Cement with Low Density Design

● Lightweight cement can be composed in one of three ways—water extended, injection of gas
(foamed cement), or by adding low-specific-gravity microspheres or other enhancing additives.
X Water extended: adding additional water to the cement slurry is a common means of reducing
slurry density by way of reducing necessary cement density to 9 lbm/gal, which is just slightly
higher than water (8.3 lbm/gal); and this method cannot achieve this low level of density.
X Foamed cement: injecting gas (nitrogen) into the slurry provides the benefit of increased
slurry compressibility, increased set-cement elasticity, and the flexibility to vary density
during operations; analysis also revealed that a complex pumping operation would be
necessary because foamed fluid is compressible and volume tracking is variant and pressure-
dependent; this is a major challenge to control/measure.
X Microspheres: several types of material are available; the lightest additive would be hollow
glass bubbles, which have specific gravity (SG) of approximately 0.32 to 0.61; the slurry
would provide the density necessary while being incompressible.
● With a detailed evaluation of these cement types, an engineering team selected the program of
cement with microsphere additives (glass beads) because it met the low density requirement while
being similar to conventional cement in terms of pumping operations (Table 1).

Table 1—Cement type design.

Option Description Density 9 lbm/gal Achievable? Complexity of Pumping Operation

Option 1 Water extended cement No Normal

Option 2 Foamed cement Yes Complex
Option 3 Cement with microsphere additives Yes Normal
12 OTC-26471-MS

Cement with High Pumping Pressure Necessary and Solution

● Challenge: at least 45 bbl of cement was necessary to seal off the perforated interval (~300 m); this
is a large amount to pump through coil. Cement has high friction and would require high pressure
to pump through coil; this is significant when the coil becomes full of cement (a typical cement
through coil operation pumps approximately 5 to 10 bbl of cement) (Fig. 19).

Figure 19 —Simulated CT pressure before added friction reducer.

● Solution: the engineering design tailored the blend recipe with sufficient friction reducer so it
could be pumped through 1.75-in. coil [~1.45-in. inner diameter (ID)] and still not exceed the
maximum surface pressure (Fig. 20).

Figure 20 —Pumping pressure after added friction reducer, tremendous pressure reducer.
OTC-26471-MS 13

Cement with Long Pumping Time Necessary and Solution

● Challenge: as pumping pressure was limited to the surface equipment rating, the pump rate was
achievable and relatively small (0.5 to 0.75 bbl/min), leading to a lengthy pumping time for a large
volume of cement (Table 2).

Table 2—Total pumping time design table.

Volume of cement (A) 45 bbl
Coil volume (B) 30 bbl
Pump rate (⬙C) 0.5 bbl/min
Time necessary to pump cement through coil ⫽(A ⫽ B) / C 2.5 Hours
Mixing time necessary 3.0 Hours
Extra time for safety factor 3.0 Hours
Total pumpable time necessary 8.5 Hours

Figure 21—Cementing thickening time.

Cement Stability
● Challenge: because the cement slurry was a combination of standard cement Class G (or
equivalent), which has density of 15.9 lbm/gal and lightening additive (hollow glass bead), which
has a density as low as 3 lbm/gal, a solution of those mixtures had the potential to separate and
the heavy material could drop out of the solution and ruin the cement property.
14 OTC-26471-MS

●Solution: an engineering design was used to optimize the viscous component to keep the cement
blend as a proper homogeneous solution until set; every test was performed with a proper visual
check and measurement.
Multiple Laboratory Testing with Associated Quality Assurance/Quality Control (QA/QC) to
Back Up the Design
● All of the designs had been backed up by multiple laboratory tests conducted in downhole
conditions (temperature and pressure); the cement properties were improved until all of the criteria
met the strict requirement. A yard test involving actual pumping of cement through coil to check
the cement was also performed (Table 3 and Fig. 22).

Table 3—Cement design had been improved after the test by test.
Tuned Light Cement Slurry

Parameter Test No. 1 Test No. 2 Test No. 3 Test No. 4

Slurry density 9.0 9.0 9.0 9.0

Cementing thickening time (hours) 5.30 5.30 6.00 9.00
CT weight at TD over yield limit (Y/N) Y Y N N
Pumping pressure (psi) 5,500 4,900 4,500 2,700
Selection result (pass/fail) Fail Fail Fail Pass

Figure 22—Cement 9 lbm/gal mixture testing with associated laboratory test and QA/QC.

Onsite Operation
Begin Tripping Coil into 5 1/2-in. Tubing Stump
● Run in hole (RIH) coil at low speed and regular pickup check weight; enter the coil into the lower
section of 5 1/2-in. tubing without any significant restriction.
● Coil the softly tagged SSSV (slack-off weight ~1k lbm) and confirm entering the correct path (5
1/2-in. tubing stump).
OTC-26471-MS 15

● Fill the void space in the annulus of the 5 1/2-in. tubing with kill fluid [no lost circulation material
(LCM)] with returns open to production; this is to minimize LCM particles from getting into the
annulus during a later stage of the operation (Fig. 23).

Figure 23—Coil operational log for entering parted tubing.

Kill Well by Pumping Through Coil at SSSV

● Parking the coil across the SSSV, perform an injection test at high pump rates (~3.7 bbl/min);
observe the low pumping pressure through the coil and annulus, indicating that no restriction to
flow and the lower section of tubing is feasibly accessible.
● Pump 85 bbl of LCM and displace with killing fluid to push the LCM pill to the perforation
interval.
● Kill the well by ⬙pumping and bleeding off⬙ repeatedly until WHP bleeds off to 0 psi and no
pressure buildup is observed in six hours; this indicates the well kill is under control and fluid loss
rate is minimal (⬍0.1 bbl/min) (Figs. 24 and 25).
16 OTC-26471-MS

Figure 24 —Coil operational log performing well kill.

OTC-26471-MS 17

Figure 25—Coil operational log performing well kill (continued).

Run Coil Through 5 1/2-in. Tubing to Well’s TD

● RIH with check weight every 30 m to minimize the risk of the coil sticking and coil weight exhibits
normal friction between the coil and tubing (even less than conservative prediction); this confirms
that the tubing does not have significant damage (Table 4).

Table 4 —Friction coefficient matching.

Design Friction Coefficient(conservative) Actual Friction Coefficient

Well Path RIH POOH RIH POOH

RD-7P Tubing/casing 0.46 0.38 0.4 0.25

● Ensure coil tag TD at 4332 m (coil depth) and coil weight is within the yield strength, allowing
cement treatment to be performed per the ⬙balance plug⬙ method (Figs. 26 and 27).
18 OTC-26471-MS

Figure 26 —Coil operational log for tripping coil trip in.

Figure 27—Coil weight record showing normal friction coefficient between coil and tubing.
OTC-26471-MS 19

Release the Lower Section of BHA at Well’s TD

● As per the design, the lower section of BHA, which contains the upset profiles, has to be ⬙dropped
off⬙ so the coil and remaining upper section of BHA can be retrieved from the well against the
⬙failed close⬙ SSSV after cement P&A is completed.
● With coil parked at 2 m off TD, the BHAs hydraulic shear was activated by means of a ⬙ball drop⬙
method; the lower section of BHA was left at TD; the disconnection was indicated by the response
in pumping pressure diagnostic (event No. 39 on the log in Fig. 28).

Figure 28 —Coil operational log— disconnect lower BHA section at Well’s TD.

● The upper section of BHA remained with the coil, the only dimple connector, and double flapper
check valves (Fig. 28).

Clean the Wellbore Before Cement Treatment

● Continue circulating kill fluid to clean the wellbore to help ensure the well is clean before cement
mixing.
● Increase the fluid loss rate (~0.3 bbl/min) and then proceed further to cement placement.
● Mix 48 bbl of cement (45 bbl pumpable) and 85 bbl of spacer (Fig. 29).
20 OTC-26471-MS

Figure 29 —Coil operational log— clean the wellbore before the cement treatment.

Place 45 bbl Cement Across the Perforated Interval: Balance Plug Method
● With coil at 4315 m (some m off TD to help prevent the coil from becoming stuck); pump the
cement through the coil; as the cement is laid in the wellbore, pick up the coil to spot cement across
the perforated interval; pump 45 bbl of 9 lbm/gal cement.
● Pick up the coil to 3870 m (safe depth) and wait on cement (WOC) for 48 hours (Fig. 30).
OTC-26471-MS 21

Figure 30 —Coil operational log—pump 45 bbl cement (9 lbm/gal cement).

Pressure Test Cement Plug and Top Up 7 bbl of Conventional Cement

● Pressure test the cement plug—failed as pressure decreased significantly.
● Run CT down to tag the top of cement (TOC)—TOC observed at 4052 m (top of perforation at
4038 m).
● Top up cement by pumping 7 bbl of additional cement and lay cement from 4052 to 3871 m; then,
squeeze 1 bbl of cement into the formation to create the cement knots at the perforated holes (Fig.
31).
22 OTC-26471-MS

Figure 31—Coil operational log—test cement plug and top up 7 bbl conventional cement.

● WOC for 24 hours.

Tag TOC and Pressure Test Cement Plug: Passed (Fig. 32)
● Run down CT to tag TOC—TOC observed at 3996 m.
● Perform pressure test of the cement integrity at 500 psi (Events 16 and 17)—passed without leak
observed during 30 minute test.
● Proceed to pull coil out of well while placing the packer fluid in the wellbore.
OTC-26471-MS 23

Figure 32—Tag TOC and pressure test cement plug, passed.

Successful Retrieval Coil Out of Well (Fig. 33)

● On the way out of well, coil weight was observed to be normal, except for some over-pulls that
had to be made when pulling the tip of coil through the SSSV. When the whole coil returned to
the surface, a cut was made to the dimple connector by the ⬙close⬙ SSSV (Fig. 34).
24 OTC-26471-MS

Figure 33—Retrieved coil from well.

Figure 34 —Whole coil retrieved on surface (left) with only a cut to the dimple connector (right).

● The coil on surface had no damage and no section of it was left in hole.

Conclusions
● Tubing stump condition has been confirmed with no severe damage (bends, twists, etc.).
● The well has been Plugged & Abandoned successfully following government regulations.
X Cement plug: 4360 m measured depth (MD) (well TD); TOC plug: 3996 m MD covered the
entire perforated interval.
OTC-26471-MS 25

X The cement plug passed the integrity test at 500 psi for 30 minutes (Fig. 35).

Figure 35—Pressure test cement plug—WHP kept stable for 30 minutes.

X WHP is being monitored closely on a daily basis and no pressure buildup has been observed.
26 OTC-26471-MS

Appendix

Figure A1—CT stack up.

OTC-26471-MS 27

Figure A2—Site layout.

28 OTC-26471-MS

Figure A3—Surface line schematic.

OTC-26471-MS 29

Figure A4 —BHA diagram.