SPE/IADC 67746
Case History of the Removal of a Hydrate Plug Formed During Deep Water Well
Testing
E. M. Reyna, SPE, Conoco U.K. LTD, and S. R. Stewart, SPE, Halliburton U. K. LTD
Copyright 2001, SPE/IADC Drilling Conference
This paper was prepared for presentation at the SPE/IADC Drilling Conference held in
Amsterdam, The Netherlands, 27 February–1 March 2001.
This paper was selected for presentation by an SPE/IADC Program 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 Society of Petroleum Engineers or the
International Association of Drilling Contractors and are subject to correction by the author(s).
The material, as presented, does not necessarily reflect any position of the SPE or IADC, their
officers, or members. Papers presented at the SPE/IADC meetings are subject to publication
review by Editorial Committees of the SPE and IADC. Electronic reproduction, distribution, or
storage of any part of this paper for commercial purposes 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 where and by whom the paper was presented. Write
Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
Abstract
The potential for gas hydrate production is always present in
deep water operations. The negative impact of gas hydrate
formation on both drilling and completions activity in deep
water has been documented previously1,2. Another speciality
operation that holds equal hazardous potential is drill stem
testing in deep water. This paper describes the cause and
solution to a hydrate plug that formed during a deepwater well
test operation off the UK Continental Shelf, West of
Shetlands. During well test operations in a water depth of
2750’, an 800’ hydrate plug formed in the string inside the
riser above the mud line, prohibiting communication with the
well below the SSTT. Diagnostic tests were carried out to
determine the source of the blockage, and two potential
culprits were initially identified - paraffin or hydrates. Coiled
Tubing was mobilised to attempt removal using a heated
glycol / brine solution. After several diagnostic operations, a
coiled tubing jetting assembly and finally a mill and mud
motor was successful in safely removing the hydrate blockage.
This paper will discuss:
¾ How hydrate formation was masked by other aspects of
the well test
¾ The technical and operational issues considered while
developing the procedure for removal of the hydrate plug
¾ The outline of the final procedure
¾ The result of the execution
¾ The lessons learned
Introduction
Conoco UK Ltd. became involved in a deepwater well testing
operation in Quad 204 in 2,750’ of water in September 1998.
The well was located in deep water on the UK Atlantic
Margin, West of the Shetland Islands. Strong currents and
environmental sensitivity in the area added to the complexity
of rig operations. While bringing the well in, a hydrate plug
developed in the landing tubing string between the sea floor
and the surface. This paper will provide details of why we
believe the hydrate plug formed and what steps were
considered to safely remove it. The execution phase of the
operation is shared fully.
Background
Quad 204 lies west of Shetlands on the edge of the Faroes
Trough in an environmentally sensitive area. A well test was
designed for the first well drilled on the block in the event that
hydrocarbons were found. In order to minimise the potential
for an unplanned hydrocarbon release; the test was based on a
specific cumulative volume to be extracted from the reservoir
at a minimal flow rate. All hydrocarbons were to be contained
at surface and shipped to shore for disposal. No oil was to be
burned to the atmosphere. After logging the well, the decision
was made to proceed with the well test. During the initial
clean up period of the well test, a hydrate plug began to form
in the landing string above the Sub-sea Test Tree (SSTT), and
solidified during the initial pressure build up period of the test.
The proprietary nature of the well does not allow full
discussion of the details of the test, but this paper is at liberty
to discuss the trends that took place to enable hydrates to be
formed.
Hydrate Formation during the Well Test
Though hydrates is a risk in any deepwater operation, paraffin
production was identified to be more of a risk during this test
and remedial plans were put in place, including injection of
chemical inhibitors at the SSTT. The well was perforated
using Tubing Conveyed Perforating (TCP) guns and an
attempt was made to clean up the well.
During the flow back period, the well cleaned up poorly. The
gross production rate fell along with an associated dropping
trend in flowing tubing pressure. Well test personnel assumed
2 E. M. REYNA, S. R. STEWART
that the rate reduction was a result of start of the paraffin
formation that had been identified as the primary risk
identified in the planning stages. Typical technical parameters
were monitored during this phase of the operation. Since
hydrates and paraffin form under similar pressure,
temperature, and chemical properties, field assumptions were
weighted by the primary risk identified in the planning stage -
paraffin. This focus masked the actual hydrate blockage.
Hydrates are ice-like solids that form under the specific
conditions of temperature and pressure. Water is the “host”
and forms a hydrogen-bonded lattice as a three dimensional
cage like structure. Natural gas is the “guest” which enters the
lattice and stabilises the structure, which precipitates as a
solid. Low temperature and high pressure can favour hydrate
formation. Those conditions were observed during the test.
Testing personnel also noted that the well was producing more
water than expected and was not cleaning up effectively, but
they did not initially identify the affect on hydrate formation
from such unexpected volume waters.
The following is a summary of the trends identified during the
clean up.
Water Depth: 2750’
Water Production: Water production was present during the
entire flow back period, partly due to mud filtrate, but it was
also caused by water encroachment from outside of the
production interval due by poor primary cementation of the
production liner. During the flow period water salinity
decreased with time, reducing its hydrate inhibiting properties.
As indicated previously, water became the host for hydrate
formation as gas production increased during the clean up with
constant water production.
Temperature: The seafloor temperature during the test was
31.6oF. Produced fluids during the well clean up had to pass
through a very frigid environment at relatively low rates of
speed. The heat transfer was rapid, causing flowing
temperatures at the SSTT to be lower than the threshold for
hydrate production. The maximum flowing temperature
recorded at the SSTT was 46.0o F, much lower than the
calculated range of 56.0oF - 65.0oF (at the depth of the SSTT)
required to inhibit hydrate production under the pressure (600
- 1000 psi – static conditions) and measured gas gravity
experienced at the range of depth that the plug was observed.
The very low flow rates contributed to this temperature
environment. Had the production rate been doubled, the
flowing temperature at the SSTT would have been at least
twice the value observed which would have inhibited hydrate
plug formation during clean up.
Flowing Tubing Pressure: Stable Flowing Tubing Pressures
(FTP) were not achieved throughout the testing operations. A
slight decline trend was observed throughout. This pressure
decline was associated with a relatively consistent water to oil
SPE/IADC 67746
ratio. As indicated, paraffin production was identified as a
higher risk than hydrate plug production during the design of
the well test. The pressure trend supported both, but the focus
was on paraffin and the decline was initially attributed to
paraffin build up, not to hydrate production.
Chemical Inhibition: Chemical inhibition was designed for
the specific conditions expected during the well test, but
water production was greater than anticipated. Incorrectly,
this excess water production was viewed as part of the clean
up process. Chemical injection was designed for the expected
flowing conditions, and did not take into account the
possibility of excess water production. With the additional
water contribution to the clean up, the chemical injection
designed was insufficient to properly inhibit the production
stream for hydrates for the water volumes observed.
Subsequent estimates indicate that the rate of methanol
injection should have been 5 times higher that the 1.5
litres/minute used during the test.
Subsequent Operations after Hydrate Formation
The well was shut in for the initial pressures build up period as
per the programme. Ten hours later, a 2” drift was run on
slickline in to the landing string, encountering an obstruction
at 1,650’. A multiple operating circulation valve, between the
mudline and the packer, was cycled in to the open position.
Pressure was applied into the drill stem test string, which
confirmed that there was no communication between the
annulus and the tubing. A wax dissolving solution (Diesel + a
high flashpoint aromatic solvent) was injected at the Subsea
Lubricator Valve (SSLV) with no effect. Believing that
paraffin was still the culprit, a 1-½” hydrostatic bailer was run
in the well to bail through what was assumed to be a paraffin
plug. The bailer was run twice to a depth of 1700’ with no
recovery.
One approach to moving paraffin out of tubing is to apply
cyclic pressure to the plug. This force would allow the plug to
plastically deform and initiate movement. Though, this would
be a reasonable approach towards removing a paraffin plug, it
would only exacerbate a hydrate plug and assist in moving it
into a more highly compressed state. Pressuring up on a
hydrate plug will allow the “wet hydrates” above the hard
sections to dehydrate and harden into a more solid form.
With this assumption, pressures in excess of 5,000 psi were
placed above and 3,000 psi below the plug. We know now
that the hydrate plug was strengthened by this application of
pressure. Once paraffin was eliminated as the cause, hydrates
became the primary culprit and a team was gathered to prepare
a plan to remove it from the landing string.
Operational Planning
The design basis for hydrate plug removal has been previously
documented4. Any one or combinations of the following
techniques have proven to be successful for removal:
SPE/IADC 67746 CASE HISTORY OF THE REMOVAL OF A HYDRATE PLUG FORMED DURING DEEP WATER WELL TESTING
Mechanical Force – Direct mechanical force by drilling out
the hydrate plug with a bit or mill.
Depressurisation – The pressure over the hydrate plug is
reduced below the hydrate equilibrium pressure at the
prevailing temperature. The hydrate blockage starts to
dissociate at the boundary subjected to the pressure reduction.
Chemical Inhibitors – Chemicals such as methanol, ionic
salts or glycol inhibit the formation of hydrates by reducing
the temperature at which the hydrate will form for a given
pressure.
Thermal Energy – An external source of heat is used to raise
the hydrate plug temperature above the equilibrium
temperature.
The engineering plan to remove the plug incorporated or
addressed all of the above, except for Chemical Inhibition.
The approach taken relied on Mechanical force as the primary
mode of removal. Depressurisation and Heat transfer were
also important areas that were taken into account. The
following is a summary of the major equipment and
techniques selected that were important for the ultimate
success of the operation.
Coiled Tubing - The landing string of the DST assembly was
composed of 3-½” 15.8 lb/ft PH-6. The coiled tubing OD was
selected based on an effective working clearance inside the ID
of 2.548” of the 3-½” landing string. A 1-½” OD QT 1000
reel gave over 1” clearance between OD of the coil and the ID
of the landing string. This size of coil provided an acceptable
circulation rate of the designed fluid while providing a
relatively large cross sectional area outside of the coil to
accept large pieces of hydrate that may break from the main
body and travel up the annulus before changing into gas. Two
reels were provided to location. Both were fully inspected.
Prior to the operation, a complete coiled tubing analysis was
carried out covering the following points and the design was
found to be acceptable:
Axial drag simulations – To ensure that the coiled tubing
to be used does not exceed its buckling limits for the well
conditions and proposed well operations.
Collapse pressure – To determine the minimum internal
pressure to prevent collapse for a given well condition and
tubing condition. A large range of well conditions was
considered.
Combined loading calculations – To confirm that the
combined effects of pressure and axial load acting
simultaneously would not exceed the tubing collapse pressure
and stress load limits.
3
Stress Calculation – To ensure that the calculated stress
does not exceed 72% of the pipe minimum yield strength for
normal and worst case conditions.
BHA Design – Two approaches were readily available with
coiled tubing: jetting and milling. Neither approach was
considered to be totally reliable in removing the plug and both
held different risks. The decision was taken to progress both
and to have each available on location.
BHA #1 - was a jetting assembly. The principle factor in
selecting jetting as a first the desire to remove the plug with as
little impact force as possible and creating conditions that
would be less likely to break off hydrate pieces from the main
body. The assembly consisted of a Jetting Sub, 2-¼” OD
wash nozzle, 2-¼” Motor-head assembly, 1-½” tubing
connector (Figure 3).
BHA #2 - was designed to apply the highest level of
mechanical and impact force to the plug. The assembly
consisted of a 2-1/8” Tapered Pilot Mill with tungsten carbide
inserts, a 1-¾” Motor head assembly, 1-11/16” positive
displacement motor (Figure 3). The motor was capable of 6oo
rpm at a circulation rate of 1.1 bpm.
Drilling Fluid Selection: The circulating fluid needed to
remove a hydrate plug had to be both compatible with
hydrates and further production of and inhibit re-formation of
hydrates once released into the Coiled Tubing x Landing
String Annulus.
Fluid Type: The workover fluid was selected on the ability
to inhibit hydrate production. Glycol was readily available
and could be provided in a timely manner to the rig. A typical
mixture of 25% BV Glycol and 75% BV Brine was selected
based on logistics considerations, the need to minimise glycol
storage to manage deck space, and ease of use. Based on the
Hammerschmidt equations, this mixture would successfully
reduce hydrate formation temperature by 12.5°F.
Temperature: Any fluid used to remove the plug required
to be heated to as high a temperature as possible. Heating was
accomplished by re-routing the return stream of fluid from
discharged from the well through the well test heater, prior to
discharging to the pit. The minimum temperature needed to
fully inhibit hydrate production was 65oF under the conditions
of the well. The circulation temperature of the workover fluid
returned to surface varied between 126 oF when the CT BHA
was at surface to 100 oF when the BHA was at the SSTT. The
discharge temperature from the heater varied from 184 oF at
the beginning of the operation to 140 oF at the end. It appeared
that the system temperature was high enough, throughout its
circulation, to prevent hydrates from re-forming once the BHA
removed them.
4 E. M. REYNA, S. R. STEWART
Circulating Procedures: Though most of the discharge from
the well was designed to be the circulating fluid and the
gaseous phase of the hydrate plug, the circulating procedures
focused on mitigating the risk of solid hydrates moving up the
annulus and either damaging the surface equipment or sticking
the coiled tubing inside the landing string.
Circulation Rate – The design circulation rate was limited
by the coil tubing size and the bottom hole assembly used.
The maximum circulation rate for the milling work was 38
gallons per minute (associated with a circulation pressure of
4000+ psi). An operational limit of 32 gpm was maintained
through the positive displacement motor when the BHA was
in the well.
Back Pressure – Hydrate plugs are not believed to be
homogeneous. They can be can have variable compressive
strength throughout and pockets of gas can be interspersed
with solid hydrates. Milling or jetting into a pocket of trapped
gas will cause immediate expansion of the gas as it is exposed
to a potentially lower hydrostatic pressure of the fluid column.
Hydrates in solid form above these pockets (that have not yet
changed form) can become projectiles and rocket upward
towards surface equipment. The resulting damage can vary
from minor or no damage to high impact damage up to and
including collapse of the coil. For these operations, back
pressure was to be maintained on the surface system in order
to regulate expansion of these gas pockets and lessen the affect
of any solid hydrates being propelled to surface. No
reasonable pressure design criteria had been established at the
time of these events and no reasonable algorithms could be
created within the short time allotted, so a recommendation of
1000 psi was based on field experience was used.
Circulation Loop: Maximum use was made of the surface
well testing facilities available. Two main objectives were to
heat the fluid to as high of temperature as possible, and
maintain the back pressure needed to discharge the fluid from
the well, and control any unexpected discharges of hydrate
plug material. Circulation was initiated through the cement
unit, which took its suction from a dedicated mud pit. The
fluid in the pit was heated to as high a temperature as possible
prior to the start of operations. The fluid from the cement unit
was pumped down the well through the coiled tubing unit.
The fluid was passed through the end of the coil, back up the
annulus to surface. The discharge from the well was passed
through the choke, into a surge tank on surface. Gas was
vented, and the fluid phase was directed to the flowstream
heater on surface. Discharge from the heater was sent back to
the dedicated pit for continued use during circulation (see
figure 2 for description of flow loop).
Execution of Hydrate Removal Using Coiled Tubing
The coiled tubing unit was rigged up and pressure tested. The
jetting assembly (see figure 3) was rigged up to wash through
the obstruction. The head was designed to direct fluid force
SPE/IADC 67746
down (70%) and sideways (30%). Circulation of the brine
was established using the rig’s cement unit via the coiled
tubing unit as indicated in the previous section. During
circulation, 1000psi backpressure was maintained at surface
with choke at all times. Circulation continued until the
temperature was seen to reach the following values: Pit (178o
F), Choke (126o F), returns after heat treatment (184o F).
Glycol was added to the brine while circulating so that the
final ratio remained at 25% BV Glycol, 75% BV Brine.
The assembly was run into the well and tagged the top of the
plug at 1700’+. Normal precautionary measures for coiled
tubing were taken during the washing operation. The coil was
used to wash down at 20’+/minute, at a circulation rate of 0.9
barrel per minute (bpm). The coil was pulled up the hole 50’
for every 250’ progressed to ensure the tubing was free while
running in. Temperature of brine / glycol fluid was
maintained as high as the field rig up could make it. (Pit
165oF, Choke 100oF, returns after heat treatment 140oF).
Initial progress was good, making 490’ in 1 hour 40 minutes.
The plug then became more highly compressed and the last
12’ took 4 hours 30 minutes to wash through. At that depth
(2211’), the jetting operation was suspended in favour of an
attempt to mill.
The coil assembly was pulled from of the well and the BHA
was changed to a mud motor and a Taper Pilot Mill assembly
(see figure 3). After securing to the motor head assembly the
motor was functioned tested on the surface and run into the
hole. The new BHA was run into the hole to 200’ and
circulation was initiated using the heated brine / glycol fluid at
0.75 bbl/minute at 2300 psi in the coiled tubing while again
holding a back pressure of 1000 psi on the choke at surface.
The string was run into the hole and the obstruction was
tagged at 2210’. A glycol pill was placed into the hole above
the plug in order to ensure that no re-formation of the hydrate
plug would occur above the CT bottom hole assembly during
initial circulation. The mill was rotated while pumping and
cut through the hydrate plug easily. The assembly
encountered no reported obstructions and was able to reach
2647’ in approximately 4 hours at a controlled penetration rate
of 1.75 ft. per minute. While milling the hydrate plug, a solid
piece of hydrate was flowed to surface and struck the surface
system. In this instance, because backpressure was being
maintained at adequate levels, damage was minimal to the
system and well control integrity was maintained throughout.
Care was taken at this point by stopping circulation and
pulling back 100’ to ensure that the coiled tubing was not
trapped by falling debris. The downstream choke was
temporarily blocked and the operation was suspended while it
was cleaned out.
The milling operation continued until the bottom of the plug
was reached at approximately 2705’ when the coiled tubing
could move freely without having to apply weight to the BHA.
The assembly continued to wash its way down through the
SPE/IADC 67746 CASE HISTORY OF THE REMOVAL OF A HYDRATE PLUG FORMED DURING DEEP WATER WELL TESTING
wellhead, pressure was applied below the SST to equalise
pressure, the SSTT was opened, and the coiled tubing was run
to 3300’ (just above the SSTT). The coiled tubing was then
pulled to surface while maintaining circulation. The multiple
operation circulating valve in the DST string below the BOPs
valve was opened to the circulating position and the 65 bbls of
9.5 ppg brine was reversed through the kill line, into the
annulus, through the circulating valve and returned to surface
indicating that the string was clear to surface.
At this stage, the obstruction was removed, mechanical
integrity established to the wellbore and test string, and well
testing operations were continued. The following is a
summary of the steps taken during the operation.
Execution Steps
1. RIH with wire line drift using 2” drift. Hung up at 1650’
2. Cycle multi-operation circulating valve, confirm no
communication between annulus and tubing.
3. Inject pour point suppressant at SSTT and in between the
two Subsea Lubricator Valves in the landing string
4. Inject a solution of Diesel + high flashpoint aromatic
solvent at SSTT and in between the two Subsea
Lubricator Valves in the landing string.
5. RIH with 1.5” Bailer, hung up at 1764’, POOH empty.
6. RIH with Hydrostatic Bailer, POOH empty.
7. Pressure up tubing to 7500 psi.
8. Open multi operation circulation valve in DST string and
pressure up below SSTT to 3,000 psi.
9. Rig up Coiled Tubing
10. Commence circulation of heated glycol / brine solution.
11. RIH with CTU jetting assembly.
12. Wash 502’ from 760’ to 2212’ in 6hr. 10 min.
13. POOH and change out BHA to 2.129” Tapered Mill and
Motor
14. RIH and Mill the remaining 587’ of hydrate plug in 4-hr.
33 min.
15. Stop at the SSTT at 2797’
16. Equalise across and open the SSTT. TIH to 3300’
17. Open multi operating circulation valve and pump glycol /
brine solution with gas to surface.
18. POOH with Coiled Tubing.
Conclusions
The operation to remove the hydrate plug with coiled tubing
was successful. Several learning from both the planning and
execution stages were key to this success:
1. Hydrates and Paraffin form under similar pressure,
temperature, and chemical trends. Field observations
must take both into consideration when data is collected.
Water production, in any form, filtrate clean up or
produced water, has the potential of forming hydrates and
must be accounted for in well testing operations and
identified properly. Once identified, contingency
planning should be in place to respond effectively to a
5
situation where the potential for hydrate production is
occurring.
2. While the washing operation was successful in removing
relatively unconsolidated hydrate material, milling was
required to remove the more consolidated section of the
plug. In future operations, milling should be considered
as the primary course of action.
3. Insulated tubing from the SSTT to the surface could have
reduced the heat loss of the produced fluids as they
flowed through the heat sink created from the seafloor tot
he surface. In this particular case, even with insulated
tubing, hydrate formation could have been possible due to
the low flow rates designed for the well test.
4. The 25%BV Glycol / 75%BV Brine solution was
effective in the washing and milling operation to remove
the hydrate plug.
5. Back pressure should be maintained on the system while
washing and milling to minimise the affect on small
pieces of hydrate plug breaking away from the main body
and being launched up the annulus towards the surface
control system. The 1000 psi used was an arbitrary figure
used based on field experience of similar situations.
Future operations should consider developing stronger
criteria for back pressure selection, though the authors
would strongly suggest using the 1000 psi value as a
minimum for such work.
6. Keep the fluid as warm as possible. A closed loop system
with a steam heater was effective in elevating the
temperature of the workover to above the temperature
required to prevent hydrate formation. Future operations
should consider a more effective flow loop system where
the heater is placed in the flow stream between the suction
pit and the coiled tubing unit.
Acknowledgements
The authors would like to thank both Conoco UK LTD and
Halliburton for their support and permission to prepare and
present this paper.
References
1. Barker, J. W. and Gomez, R. K.: “Formation of Hydrates
During Deepwater Drilling Operations”, JPT, March
1989, 297-301.
2. Castro, G. T., Terry, A. P., Ferreira, L. V, Ribeiro, G.
S.:Albacora Manifold Hydrate Plug formation: Cause and
Solution”, OTC 8777
3. Yousif, M. H., Dunayevsky, V. A., Hale, A. H.., ‘Hydrate
Plug Remediation: Options and Applications for Deep
Water Drilling Operations’, SPE/IADC 37624
6 E. M. REYNA, S. R. STEWART SPE/IADC 67746

Figure 1 Figure 2

Figure 3