SPE-Number175440-MS

Diver-Less Removal Of The Teal Riser

Trevor Jones , Shell UK Ltd; Craig Walker, Proserv UK Ltd

Copyright 2015, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Offshore Europe Conference & Exhibition held in Aberdeen, United Kingdom, 811 September 2015.

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
Objective/Scope In Q4 2012, the 8 Teal dynamic riser required replacement. A project team was
assembled to remove the existing riser and install a replacement. The riser was originally installed in the
1990s by divers based on the Anasuria Floating Production Storage Offloading (FPSO). The FPSO
facilities that were used to provide diver access into the chain table area to secure the bend restrictor
were no longer operational. Previous diver intervention at the chain table had experienced high levels of
non-productive time, so a remote removal method to cut and lower the riser and bend restrictor to the
seabed was pursued.

Methods, Procedures, Process The project team initiated a concept study to set the challenge to
release the riser from the FPSO remotely to a) reduce HSE risk; b) de-risk project by separating release
campaign from recovery campaign and c) perform the work offshore using a very small team (FPSO
bedding constraints was one of the reasons the team didnt use an air dive team to do it from the FPSO).
The concept selected was largely based on a method used for removal of other similar risers in 1999
albeit that all that remained from that campaign was a PowerPoint presentation. The majority of
engineering calculations and drawings could not be located.

Results, Observations, Conclusions The tooling was designed, fabricated and tested in less than 6
months from award of contract to offshore operations. During the offshore campaign, various issues
were encountered (e.g. incorrect as built documentation) but the remote removal campaign was a
success with both the riser and its bend restrictor being removed from the FPSO without requiring any
diver intervention, thus reducing the project costs and HSE exposure.

Novel/Additive Information Lessons learned from the offshore campaign are being captured with the
concept being further refined to allow the removal of the riser and bend restrictor without first having to
cut and remove the topside riser end fitting. This should reduce the offshore operational time for any
future riser removals and remove the risks associated with unknown internal condition of the riser.
2 SPE-Number175440-MS

Background Trevor Jones

The Anasuria FPSO was installed in 1995 to allow production from the Teal, Teal South and Guillemot
developments. The risers and bend restrictors were pulled in through the chain table. Divers that were
based on the FPSO secured the bend restrictors to the chain table with a combination of nuts, studs and
C plates. The risers were then pull though the upper I-tubes and secured at the main deck level. See
Figure 1.

Figure 1 - Turret GA
SPE-Number175440-MS 3

Concept Development Trevor Jones

In Q4 2012 an 8 production riser had to be replaced. The project faced various challenges. Diving
from the FPSO was ruled out due to the complexity of putting a dive system onto a producing asset with
limited bed space. Shallow air diving from a daughter craft was investigated but again discounted as a
HSE risk and high cost exposure as a previous daughter craft campaign for anode installation had very
high levels of non-production time.

In 1999 two similar 6 risers had been successfully removed using remote tooling but the tooling no
longer existed and the engineering calculations and drawings for the tooling could not be located. A
concept study was initiated to look at the feasibility of removing the 8 riser and bend restrictor without
the use of divers to; a) reduce HSE risk; b) de-risk project by separating release campaign from recovery
campaign and c) perform the work offshore using a very small team (FPSO bedding constraints was one
of the reasons an air dive team on the FPSO was not used in 1999). An additional constraint was that
the I-tubes could not be damaged as part of the removal process. The study considered various methods
to remove the 8 riser and its bend restrictor and concluded with the selection of a method similar to that
previously used in 1999.

In parallel the project team contacted the engineers from the 1999 project, brought them out of
retirement and they joined the team. They provided useful insight on the lessons learned from the
previous project.

An Invitation To Tender (ITT) with initial calculations and drawings was issued in early December
2013. After the normalisation of the bids, close out of the technical and commercial qualifications, a
contract was awarded mid February 2014.

The tooling development/fabrication had to be ready for System Integration Testing (SIT) in June and
the team ready to mobilise offshore in mid-July so the winning vendor was on the fast track from the
day of award.
4 SPE-Number175440-MS

Tooling Development Craig Walker

The design pack received from the client consisted of the following tool concepts, as highlighted in
Figure 2:

Riser Gripper (Tool ID - RG01) required to lift the full 15Te of the Teal riser prior to the end
fitting being severed;
Installation Spigot Gripper (ISG01) required to lift the bend restrictor and spigot once all M33
studs had been severed;
Cutting Manipulator (CM01) Rotating cutting tool consisting of a shoulder, elbow and wrist
joint for cutting the M33 studs holding the bend restrictor spigot in place.

Figure 2 Concept of Tooling at Project Award

A full design review was required, concept modifications, procurement, assembly, testing as well as the
design of a topside hydraulic control system to operate each of the tools. Additionally, all topside
equipment was to be provided by the vendor including a water abrasive cutting system from the existing
fleet.

All design reviews, concept modifications and topside control systems were performed by in-house
vendor services. Whilst a major design and procurement exercise was ongoing, with an overall
programme of 16 weeks until testing completion, several key yard tests were required to establish
suitability of topside equipment.
SPE-Number175440-MS 5

M33 Stud Cutting

Due to the spatial and zoning restrictions on the FPSO topside, immediate testing was devised to
establish the most suitable topside cutting system to provide the most effective cutting performance but
still remaining within the operational parameters. The bespoke tooling would then be adapted and
planned with the identified topside cutting spread in mind.

The system selected was an AMU-1200 which uses high pressure abrasive water cutting as cutting
medium. Cutting at pressures of up to 1200 bar, the system is certifiable for Zone 2 environments and
has a footprint of approximately 1 square metre. A Zone 2 electric pump capable of delivering 27 litres
per minute at approximately 700 bar was utilized with the AMU-1200 to minimise footprint.

Water abrasive test cuts with M33 studs proved successful and confirmed the topside equipment. By
trialling different cutting speeds and pressures, a cutting time of approximately 6 minutes per stud was
found to be repeatable. The offshore cutting targets were a mix of perpendicular cuts for studs at the lip
of the spigot flange and 45 degree cuts for studs on the inner ring of the spigot flange. Testing for both
was successful. See Figure 3 and Figure 4.

Figure 3 M33 Cutting Test Rig Figure 4 Cut M33 Stud

Riser Cutting Trials

A length of 8 riser was sourced by Shell UK and free-issued to Proserv for yard trials. Although not
identical, the specification was considered similar enough so that onshore cutting trial results were of
worth to the project.
6 SPE-Number175440-MS

Yard trials were set up at Proserv using the same AMU-1200 abrasive cutting system that successfully
cut the M33 studs. Additionally a guideband and a circumferential manipulator were used to keep the
cutline fixed at the chosen elevation.

One of the key offshore procedural challenges was cutting the Teal riser whilst rigging and control lines
were running inside it. These items could not incur any damage during the abrasive cutting therefore
yard testing initially focussed on tangential cutting so that the centre of the riser was never exposed to
the full cutting pressure of the AMU-1200. Simple 4air hose was placed within the riser test pieces
during the cutting trials to ensure that any cutting damage was recorded.

Again, cutting pressures and speeds were experimented with until repeatable tangential cutting was
achieved with no damage to the sacrificial 4 air hose within the riser. This was proven and a cutting
time of approximately 55 minutes taken forward in planning. See Figure 5 and Figure 6 and note the
pink Garnet abrasive once dry post-cutting in Figure 5. The sacrificial air hose (yellow) can also be seen
in Figure 5.

Figure 5 Severed 8 Riser Showing Abrasive and Air Hose Figure 6 Tangential Cutting Detail

Site Survey

Approximately 3 months before the main offshore scope was executed, a site survey was conducted on
the Anasuria FPSO. This primarily established working conditions, opened dialogue with the key FPSO
staff, noted service points and confirmed dimensions of the allotted laydown area. The contacts made
with the FPSO staff led to onshore visits to the testing set up in order to further de-risk the workscope
and troubleshoot any remaining queries.
SPE-Number175440-MS 7

Footprint Investigation

A laydown area next to the turret was identified and planned in for the main scope. The main section
was an L-shaped area with access and egress challenges. In order to best prepare for the offshore scope
as well as ensuring that the identified topside spread was compatible with the FPSO, the laydown area
footprint was marked out as part of the onshore SIT and the topside kit laid out as per the deck plan.
This proved the viability of the spread as well as increasing operator familiarity and confidence in the
method. Equally, the client was able to witness onshore the exact offshore topside proposal which
increased confidence further. See Figure 7.

Figure 7 Yard Equipment Trialling Within Marked Out Offshore Laydown Area

System Integration Testing (SIT) June 2014

A full SIT was carried out at the vendors Birchmoss facility, UK, in June 2014. All critical aspects of
the offshore workscope were trialled, separately at first, and then in sequential order as they would be
once offshore. A full scale test piece to represent the bend restrictor spigot, C-plates and M33 studs was
procured for the SIT to establish cutting performance, rotational timings, camera positioning and
visibility on the cutting manipulator.

Additionally, pull tests were carried out using the free-issued section of riser provided by the client as an
anchor in order to certify the two internal gripping tools to 5Te and 22.5Te respectively. See Figure 8 to
Figure 13 inclusive.

The key testing carried out was as follows:

In air M33 stud cutting on the test piece;

Wet M33 stud cutting on the test piece;
8 SPE-Number175440-MS

In air 8 riser cutting without damaging internal control lines;

Full run through of the offshore sequence for severing a M33 stud;
C-plate removal;

Finally, a contingency gripping tool for RG01, a ballgrab type tool based on ball and taper design, was
procured fully load tested from a third party specialist. This tool travelled offshore as the primary option
for this part of the workscope however was ultimately not used due to unpredicted bore restrictions
within the Teal riser.

Figure 8 RG01 and ISG01 Ready for Testing Figure 9 CM01 Ready for Testing
SPE-Number175440-MS 9

Figure 10 ISG01/CM01 Engaged in Test Piece Figure 11 Detail of CM01 Arm Targeting M33 Stud

Figure 12 Detail of CM01 Cutting Head and M33 Stud Figure 13 Detail of Cut Studs after Testing
10 SPE-Number175440-MS

The SIT was successfully signed off in late June 2014 as per the original tendered schedule. The
equipment travelled offshore in early July 2014.

Offshore Execution Craig Walker

Prior to the onsite arrival of vendor equipment, the insulating cladding, ESV and all associated pipework
was removed from the end fitting of the Teal Riser. A blank cap was installed and all C-plate retaining
bolts were changed out so that during the actual operation the removal of the blank cap would not be
problematic. The site area was prepared and the laydown area cleared to accommodate the vendor
equipment. The Proserv equipment was set up as per the onshore SIT orientations. See Figure 14.

Figure 14 All Containers Shown are Vendors. Laid Out as per SIT Deck Plan

Phase 1
Separating the end fitting from the riser

The blank cap was removed from the Teal Riser end fitting in order to facilitate the insertion of the
bespoke Ballgrab tool within the bore of the Teal Riser. It was immediately noted that the internal bore
stepped in by around 5mm radially approximately 1 metre into the Teal Riser. The internal condition of
the riser was also considered more degraded than expected with the internal carcass layer seemingly
parted and slipped around 100mm down the length of the inside of the riser. See Figure 15 and Figure
16.
SPE-Number175440-MS 11

Figure 15 Teal Riser Internal View Slipped Liner Figure 16 Teal Riser Bore Showing Stepped Restriction

The tolerances of the ballgrab tool did not allow the tool to be lowered passed the blockage so the
planned operation could not proceed. Modifications to the ballgrab were not considered possible
however the contingency tool, RG01, presented opportunity for simple modification to reduce the outer
diameter by 6mm to allow this tool to pass the bore restriction within the Teal Riser and engage with the
riser internal wall at the preferred elevation of 4000mm below the end fitting.

A bespoke lifting clamp utilising a standard Grayloc end fitting clamp was installed and then RG01 was
passed into the riser. The hydraulic control lines for RG01 were encased in a steel sheath in order to
fully protect them from the cutting force of the abrasive water cutting system that would sever the riser
below the end fitting. This protective system had been fully tested onshore with the same cutting angle,
pressure and duration as planned with acceptable results and no damage to the control lines.

Once this was in place, the tool was engaged, test pulls were conducted as per the procedure and whilst
the Teal riser load of 15Te was taken up by the lifting clamp and 25Te winch, the C-plates holding the
Teal Riser in place in the turret were removed. The Teal Riser was then raised 1.5m out of its seat in the
deck and the winch locked off to allow the external cutting equipment to be fitted, guards put in place
and hydraulic function tests to be carried out. See Figure 17, Figure 18 and Figure 19.
12 SPE-Number175440-MS

Figure 17 End Fitting Lifted by RG01 Pre-Cutting Figure 18 Post-Cut Showing Degraded Riser and Cutline

Figure 19 Fully Separated End Fitting Awaiting De-Rigging and Shipment to Shore. Vendor End Clamp still fitted.
SPE-Number175440-MS 13

The cutting operation proceeded as planned using a topside water abrasive cutting system. Cutting at a
maximum system pressure of 700 bar, approximately 10000 psi, and monitored remotely, the cumulative
cutting time required was 2 hours within a period of 4 hours allowing for abrasive refills of the systems
and basic maintenance. The degradation of the outer layers of the Teal Riser required modifications to
the external cutting equipment prior to installation as the outer diameter of the riser was less than
expected due to missing layers.

Lowering Teal Riser to the seabed and laydown for recovery

After the end fitting was confirmed to be fully severed, the 25Te winch was instructed to pay out and the
Teal Riser, minus the end fitting, was lowered to approximately -78m in 40 minutes whilst held by
RG01. A pan and tilt camera secured to the end of a deployment frame was lowered down an empty
neighbouring upper I-Tube into the chain table in order to monitor the lowering of the Teal Riser
through the bend restrictor. At approximately -78m the hydraulic pressure was released allowing the
RG01 gripping dogs to loosen and release the Teal Riser to the seabed from a height of approximately
10 metres. RG01 was recovered to deck and demobilised. The severed end fitting was bagged according
to the FPSO procedure for transport to shore, decontamination and recycling.

Phase 2
Tooling Engagement with Bend Restrictor

Following the complete removal of the Teal Riser, the second and final phase of the offshore workscope
was to engage with and remove the bend restrictor within the lower I-Tube which was secured by a
spigot and 44 M33 studs at the upper end of the lower I-tube in the chain table. Unlike Phase 1, this
operation was fully subsea and remotely monitored using onboard cameras and the pan and tilt camera
deployed via a neighbouring I-tube. All tooling was operated remotely from a topside control cabin and
operating system.

The Installation Spigot Gripper (ISG01) was lowered to engage and grip the spigot to allow the Cutting
Manipulator (CM01) to sever the M33 studs securing the bend restrictor to the chain table. As planned,
these two tools were lowered as a single assembly. See Figure 20.
14 SPE-Number175440-MS

Figure 20 ISG01/CM01 Assembly docked together awaiting deployment

It was found that ISG01 would not dock into the spigot as expected and following camera surveys of the
exposed spigot, something that was not possible prior to the Teal Riser being removed, it was found that
the two halves of the fabricated spigot were misaligned resulting in a reduction in the internal diameter
to the extent that ISG01 would not engage.

Following discussions between the client, FPSO crew and main contractor, the onshore vendor team
designed, procured, assembled, load tested and mobilised a series of drift gauges to remotely assess the
actual bore of the spigot. See Figure 21 and Figure 22.
SPE-Number175440-MS 15

Figure 21 Gauge disks awaiting shipment to FPSO Figure 22 Gauging tool body with no disks

Simultaneously, ISG01 was demobilised to shore so that modifications could take place to reduce the
diameter where possible.

Following the above and once back onsite, the ISG01/CM01 assembly was deployed and engaged with
the spigot. Pull tests were carried out with ISG01 engaged and the go ahead was given to proceed with
the removal of the M33 studs.

Onboard cameras confirmed at this point that an unrecorded anode was positioned hard against one side
of the spigot that was likely to be an obstacle to the cutting operation.

Removal of Bend Restrictor

CM01 removed 19 of the M33 studs according to the procedure in 4 hours once cutting operations
began. Averaging 6 minutes of cutting per stud with an additional 6 minutes of repositioning the cutting
head per stud, the cutting times and tooling operation performed as expected following the successful
onshore SIT. The onboard cameras during this period confirmed that an anode, unrecorded on any as-
built or survey information received, was positioned hard against the outer ring of M33 studs blocking
the access required for CM01 to sever nine of the stud targets.

Schedule delay was experienced as a number of restrictive factors on the FPSO ruled out mobilising a
more powerful cutting system in the time available so CM01 was utilized to begin cutting away the
portions of the anode that were blocking access. This process had never been trialled and CM01 was not
closely suited to this task so real time trial and error was required. See Figure 23 and Figure 24.
16 SPE-Number175440-MS

Figure 23 Cutting of M33 Bolt with CM01. Anode behind. Figure 24 Detail of Severed Bolts, Cut Anode and CM01

Access was fully achieved after several days however the additional tool operation was far in excess of
the 24 hour operation that CM01 and associate spares pack were tested for. As a result additional spares
were required to be mobilised. The weak point in CM01 as operational hours increased was found to be
the micro electric motor gearing assembly controlling the rotation of the tool and all spares were used.
The extended cutting time coupled with the volume of anode particles entering CM01 during the anode
cutting meant that CM01 required recovery to deck on several occasions to flush out anode particles and
abrasive from the gearing assembly.

Additional components were procured and mobilised to the worksite to assist in completing the
workscope. Testing to destruction was not included in the onshore testing as the information on the
workscope as well as information from an identical workscope in 1999 suggested that all cutting would
be complete within 24 hours. This was corroborated by the 19 of 44 studs removed in 4 hours prior to
removing the anode.

See Figure 25 from the post operation site survey showing the anode, crescent cut portion performed by
CM01 and proximity to the now bare chain table flange. Note the new riser is also in place in this image.
SPE-Number175440-MS 17

Figure 25 As found, ROV Photo Showing Exposed Flange, new riser (yellow) and Cut Anode following Vendor Operations to
remove its obstruction

Laydown of Bend Restrictor for Recovery

Following the removal of the anode portions blocking CM01 the final M33 studs were severed. ISG01
was then raised to free the spigot from the chain table, allowing the locking C-plates to fall away.

ISG01, gripping the bend restrictor, was then lowered through the lower I-Tube and out of the body of
the FPSO. This operation was monitored at the chain table and hull exit by the pan and tilt camera in the
empty neighbouring I-tube slot. The bend restrictor was then lowered to the seabed and released
approximately 10m above the seabed. Note that ISG01 and CM01 were separately rigging to lifting
devices to allow ISG01 to go to the seabed and CM01 to remain in the chain table. This was to remove
the need to provide additional length to the CM01 umbilical and worked as tested.

ISG01/CM01 were recovered to deck and demobilised.

The Teal Riser bend restrictor was separately recovered to the deck of a DSV after set down on the
seabed. See Figure 26 and Figure 27.
18 SPE-Number175440-MS

Figure 26 Post-operation Bend Restrictor Recovered to DSV

Figure 27 Post-operation Detail of Severed M33 Bolts on Bend Restrictor Spigot

SPE-Number175440-MS 19

Conclusions - Trevor Jones

Despite the short time span that the project vendor team had between contract award and offshore
mobilisation, they carried out a detailed drawing pack/design review, produced fabrication drawings and
material take-offs, offshore site survey, procurement of all items, assembled the tools, carried out a
Factory Acceptance Test (FAT), the SIT was completed with the equipment placed in the exact foot
print available offshore and a full set of procedures and documentation were prepared.

Significant cutting trials were carried out on the studs and sections of a recently recovered 8 dynamic
riser.

In the offshore campaign, despite the problems encountered, the remote removal was a success with
both the riser and its bend restrictor being removed from the FPSO without requiring any diver
intervention, thus reducing the project costs and HSE exposure. The fallback option would have been to
use a DSV with daughter craft to allow divers to access the chain table. Previous daughter craft activities
had significant non-productive time. There were no accidents or incidents during the entire removal
campaign.

When unforeseen obstacles arose offshore, equipment was shipped onshore for modification and
recertification (via the project holdback vessel) and turned around in the minimum amount of time.

The biggest problem encountered was the as-built information on an anode (that that had been installed
the year before) didnt match as-found. If it had been in the proper location, the cutting of the 28 outer
and 16 inner M33 studs on the bend restrictor would have been carried out in a 12 hour shift. This was
another example of how the vendor was able to manage the unexpected challenges they encountered
offshore and their team offshore adapting and finishing the job they started.

Both the riser and bend restrictor were available for recovery when the DSV arrived in the field ensuring
the replacement schedule was met.

To optimise the design further we are currently investigating modifications to the tooling so the bend
restrictor can be removed without cutting off the end fitting and lowered to the seabed with the riser in
one operation. This should reduce the offshore operational time for any future riser removals and
remove the risks associated with unknown internal condition of the riser.
20 SPE-Number175440-MS

Lessons Learned - Trevor Jones and Craig Walker

Due to the fast track nature of the project, issues did arise that had not been foreseen and these had to be
overcome using the management of change process during the project offshore execution phase. If we
were going to carry out a similar project in future, we would invest more time in surveying the chain
table and riser and surrounding area to mitigate against errors in as-built documentation and highlight
unknowns. For a future similar project we would add the following to the program.

Carry out a pre survey with a suitcase ROV. The team was given an as-built drawing of the
anode locations but that was not accurate;
Ensure the external riser cutting system can manage changes in the outside diameter of the riser
Carry out a drift run of the internal diameter of the riser;
Ensure the internal riser gripper tool has inter-changeable components to allow for changes of
internal diameter of the riser;
Replace the camera on a stick with a suitcase ROV. The camera on a stick moved with the
swell which made viewing difficult;
Replace the bespoke miniature electric motor/gearing assemble with off the shelf hydraulic
motor system;
Once the riser has been cut and lowered, carry out a drift run of the internal diameter of the bend
restrictor;
Ensure the bend restrictor gripper component has inter-changeable components to allow for
greater changes of internal diameter.