SPE 144089

Sand Control Completions for Ormen Lange Big-Bore, High-Rate Gas

Development, Evaluation of Concept Selection, Qualification, Execution and
Well Performance
A. van Kranenburg, SPE, J. Twycross, SPE, C. Combe, SPE, K. Hals, SPE, A/S Norske Shell

Copyright 2011, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE European Formation Damage Conference held in Noordwijk, The Netherlands, 7–10 June 2011.

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

Ormen Lange is the second-largest gas field in Norway, currently providing 20% of the domestic gas consumption for the UK. As
one of the world's largest subsea big-bore gas wells - with production rates up to 350 MMscf/d per well, a 120-km tieback to shore,
subzero seabed temperatures and a design life of more than 30 years - these wells provide unique technical challenges. Due to the
low reservoir rock strength, high velocities and the life expectancy of these wells, sand control was identified as a prerequisite
from day one.

This paper clarifies the logic of the selected sand control method and describes the extensive testing required to qualify the sand
control hardware and associated completion fluids for the expected operating conditions. The execution of thirteen successful sand
control completion installations will be discussed, including both ends of the success spectrum, and well performance will be
covered.

Throughout the four-year installation campaign to date, a number of new technologies were qualified and successfully integrated
into the lower completions. These included cableless gauges to provide data closer from the reservoir inflow area, tagged proppant
for individual subsea well fingerprinting if proppant is found at the gas plant, and water-tracer technology to aid in identifying the
subsea source of potential formation water breakthrough.

This subject explains the extent of the effort to progress an ambitious field development based on big-bore gas wells only (to
reduce well count) from the drawing board to flawless execution and world-class production. Relevance also relates to other major
big-bore, high-rate gas subsea developments with sand control requirements that are currently being pursued in similar harsh
environments (deep water, Arctic conditions, sub-zero seabed) that might have interest in the application of Ormen Lange's
successfully implemented technology.

Introduction
Ormen Lange is the second-largest gas field in Norway, with recoverable gas volumes estimated in the order of 400 GSm3 or
14 Tcf. The field is located approximately 100 km off the west coast of Norway and covers an area of approximately 325 km2;
about 40 km long and 8 to 10 km wide.

The field is the first deepwater development in Europe, with water depths ranging from 850 to 1100 m (2,600 to 3,300 ft) in a
rugged seabed terrain that is the remnants of the Storegga slide (Fig 1.). Seawater temperatures at the seabed are approximately
-2° C, lower than the 4° C that is common in most other deepwater developments. These cold temperatures provided significant
challenges for the drilling and completion fluids design and the well-testing operations to mitigate flow-assurance risks from
hydrates.
2 SPE 144089

Figure 1- Ormen Lange field layout

Ormen Lange has been developed as a subsea-to-beach development. To date, three 8-slot templates have been installed at a water
depth of 880 m. The wet gas is produced back through either one of the two 30-in. flowlines to the new onshore gas plant built on
Aukra, a small island off the coast of central Norway. Two 6-in. monoethylene glycol (MEG) supply lines provide continuous
MEG injection to the three templates for hydrate prevention purposes. Total production system capacity is 70 MSm3/d or
2.5 Bcf/D. A fourth template is currently being constructed that will be installed in the summer of 2011 to develop the
mid-north section of the field.

The Ormen Lange main reservoir unit, at a depth of approximately 2700 to 2900 m msl, has a thickness of 50 to 60 m and has very
good reservoir quality, with porosities around 25% and an average permeability of 500 mD.

Big-bore wells were selected as a concept for the development of Ormen Lange to reduce well count and cost, with a nominal well
capacity of 10 MSm3/d (353 MMscf/D) per well. The key objective for these wells was to deliver sustainable, high production
rates without any intervention requirements for the 30-year lifetime of the wells. The first ten development wells from the two
mid-field templates were drilled and completed at a maximum 45-degree angle through the reservoir. The second-phase
development wells from the southern template were drilled and completed as horizontal wells due to the expected shallower water
contact requiring a larger standoff to mitigate the risk of water coning over time.

Key features of the selected well and completion concept are (Fig 2.):
• 7-in. subsea horizontal christmas tree
• 9⅝-in. 13Cr tubing
• Dual hot-stacked 7-in. TR-SSSV
• Two permanent downhole gauges (PDG) – one above the SSSV, one above the production packer
• 9⅝-in. x 7-in. production packer
• 9⅝-in. 13Cr flow wetted production liner
• Openhole gravelpack, with 300-micron direct wrap sand screens and 16/30 proppant
SPE 144089 3

Figure 2 - Ormen Lange Completion Design

During the course of the project, new technologies were qualified and implemented in the wells such as cableless gauges in the
openhole section (discussed in the technology section).

Drilling of the Ormen Lange wells was performed using a third-generation dynamic-positioned drillship. To date, 12 wells have
been completed, of which 10 have been unloaded to the rig and 2 have been directly unloaded to the host. Production startup of the
Ormen Lange field took place on 13 September 2007, after which the wells were slowly ramped up to their current stable
production capacity of 10 MS m3/d or 335 MMscf/D for each well. Total field production is currently around 70 M m3/d, all gas
being exported to the UK through the purposely built 1200-km Sleipner export line, providing up to 20% of UK domestic gas
consumption.

Formation Strength and Sand Production Risk

The Ormen Lange wells are completed on the Egga reservoir unit. Future wells may also be completed on the deeper Jorsalfare
formation. Based on appraisal well core data, extensive rock mechanical test were conducted, leading to the following key
observations.

The sandstone units are generally characterised by low-to-moderately-low rock strength (UCS 2 to 5 MPa), primarily
depending on the porosity. The mudstones, and occasionally the calcite-cemented sandstones, are characterised by moderate-to-
high-strengths (7 to 20+ MPa). Occasional well-cemented sandstone layers with calcite or quartz as cement were identified, having
rock strengths in excess of 15 MPa. The Poisson's ratio varied between 0.19 and 0.25, where the higher values generally correlated
with the mudstone horizons. The constrained modulus (formation compressibility) varied from 1.5 to 5 (0.3) GPa and was similar
to that measured for the Troll reservoir sand, which also required sand control.

Rock mechanics testing identified the risk of sand production and concluded that sand-free production from a vertical well
would be limited by the weakest reservoir intervals to less than 60-bar depletion at 2700 mTVD and less than 35-bar depletion at
2900 mTVD. For horizontal production wells, sand-free production would be limited to only 10-bar depletion (assuming
4 SPE 144089

low <5 bar drawdown). Turbulent flow due to high flow rates would theoretically reduce the sand-free operating window even
further, resulting in formation failure at higher reservoir pressures.

Due to the operating environment of ultra-high rate subsea gas wells and the design goal of an intervention-free 30-year design
life for the wells, the rock mechanical studies concluded that active sand control would be required from day one.

Lower Completion Concept Select

From rock mechanical studies performed on core material taken from appraisal wells in the Ormen Lange field, it was apparent
that sand control would be a must-have in the development of the gas field. More importantly, once the selected concept for the
field development plan included big-bore subsea wells producing up to 10 MSm3/d (353 MMscf/D) per well, a robust sand control
system became even more relevant as the consequences of sand production would compromise project delivery.
Sand control selection was based on the rock mechanical study and available particle size data from the appraisal wells
obtained at various locations from the field. The following sand control retention techniques were reviewed, and a number of these
were lab tested to produce a conclusive recommendation on the preferred sand retention technique and filter size.

Openhole Standalone Screens

Openhole standalone screens were found to be unsuitable, due to the poorly sorted and high-fines content nature of the
formation. Erosion potential for any screen, mesh or wire wrap, was considered to be the main risk for this type of retention
technique due to plugging and hot spotting.

Expandable Sand Screens

Expandable sand screens were rejected as an alternative. At the time of concept selection in 2003 they had a limited track
record on longevity performance and no track record in any high-rate gas environment requiring sand control from day one. In
addition, as 1 to 2-percent compaction was expected across the reservoir due to depletion, the collapse resistance of expandables
screens would not withstand the loads without losing their retention capability. Lastly, the fine woven structure of the mesh was
considered a risk for plugging and erosion in areas where compliant expansion could not be guaranteed due to openhole
irregularities such as washouts in combination with high annular flow velocities.

Alternate Path Gravel Packing

Alternate path gravel packing (shunt-tube technology or predrilled liner with screens) was examined, but it was considered to
be more applicable for areas where the risk of incomplete gravel packing was high due to hole stability problems, reactive shales,
long heterogeneous intervals, or where a margin to fracturing was limited. None of the above was considered to be a risk for the
initial development wells in the centre area of the field. For the southern horizontal wells completed in 2010, however, a
pre-drilled liner with screens inside was available as contingency because hole stability was considered a potential risk. For the
mid-north development, shunt-tube technology has been technically endorsed as the reduced margin to fracturing due to depletion
effects and constraints of complete packing of a full horizontal well. Use of either alternate path system does reduce the maximium
allowable basepipe size for the screens, adding a frictional component in the system given the initial high gas flow rates.

Frac Packs
Sufficient reservoir kH was the key reason not to pursue frac and packs. As most fracpacks are installed in a cased hole
environment, the concentrated high flow rates coming through a limited open area of perforations was also considered to be an
erosional risk that would be better handled in an openhole completion. As the water contact was not fully understood throughout
the field, fracing into any water-bearing zone would be unacceptable because this would pose a flow-assurance risk.

Openhole Gravel Packs

Openhole gravel packs were selected based on their broad retention capability, fines-tolerant design (if sized correctly),
stabilisation of the openhole, and proven track record in other high-rate gas wells. The risk of incomplete packing was mitigated
for the phase 1 wells by limiting the angle through the reservoir to a maximum of 45 degrees. Even though openhole gravel
packing is considered to be an established and mature technology, advances in gravel pack tools, novel completion and breaker
fluids, and robust high-strength sand screens have increased the reliability and installation success of this completion type over the
last few years.
After the gravel size was selected, based on extensive particle size distribution (PSD) analysis and full-scale retention testing,
the screen type and filter size had to be selected. As there are no field analogues within the same flux rate (Fig. 3.) as Ormen
Lange would produce, the screens had to be tested as close as possible to expected in-situ conditions.
SPE 144089 5

Figure 3 - Comparison of gas flux rates in other high-rate gas developments

A full-screen qualification program was setup with Det Norske Veritas (DNV) to test a shortlist of suitable screens from
various vendors. Screens were tested on erosion (high flux rates through screens), susceptibility to plugging (high percentage of
fines content), direct formation retention (keep out formation in case of voids) and mechanical strength (during installation and
future reservoir compaction).
Three screens qualified after the testing program out of which a high-strength, direct-wrap type of screen was selected. Other
key components were tested to ensure the integrity of the lower completion would not be compromised during and after
installation. These tests included:
• A scaled-down test in a gravel pack simulator to optimise position of top of screens with respect to the openhole shoe
interface (Fig 4.)
• Standoff tests in an API test cell with modified bow-type centralisers to ensure sufficient clearance of screens inside
the underreamed openhole
• Bridging tests to ensure the limited clearance between the formation isolation valve and the cableless gauge mandrel
inside the 9⅝-in. liner would not prematurely stall placement of gravel during gravel packing operations.

Figure 4 - Top of screen positioning test setup

6 SPE 144089

The final lower completion design (Fig. 5) consisted of a washdown shoe to allow for fluids displacement and washdown
capability, 7-in. direct-wrap screens, with two non-perforated screen joints across and inside the 9⅝-in. casing, an expansion joint
to absorb some of the strain during future compaction, a fluid isolation valve to isolate the formation after the lower completion is
installed and prevent losses into the formation, and a gravel pack packer with all the standard extensions required to enable
circulating below and above the gravel pack packer. A straddle was installed on a separate run across the closed sliding sleeve as
insurance if the sliding sleeve were to open due to vibration over the lifetime of the well.

Figure 5 - Lower completion base components

System Qualification

Fluids Lower Completion Concept Selection

Once it was established that the likely candidate for the lower completion would be an openhole gravel pack, work began to
find a non-damaging, water-based reservoir drill-in fluid (RDIF) and other fluids required in the gravel pack operations. (Much of
this work is described in SPE 112292.)
The lower completion fluids had to meet a number of complex criteria:
• Minimum formation damage
• Flowback of fluids and filter cake through sand screen and gravel pack
• Compatible with all the gravel pack fluids (solids-free mud, brine and breaker)
• Identical densities of all fluids
• Minimum corrosion rates (< 0.006 mm/yr) for fluids in the wellbore and also back production to the facilities
• Hydrate inhibition at the Ormen Lange water depths (-2° C at a hydrostatic pressure of 108 bar)
• Stability during transportation and storage in winter conditions
• Contain only “green or yellow” chemicals (best environmental status, Norsok regulations)

Achieving all the above was challenging as meeting one criterion was often detrimental to another.

RDIF Design
The mud had to provide the capabilities of good hole cleaning and borehole stability. It also had to be acid soluble. The mud
was optimised to provide minimal formation damage and inhibit hydrates.
It was found that the lowest achievable mud weight that provided hydrate inhibition and contained sufficient bridging solids to
minimise formation damage was 1.24 s.g.
However, this resulted in a large, 53-bar (770 psi) overbalance for the first wells.
SPE 144089 7

Solids-Free Drill-In-fluid (SFDIF)

The openhole was displaced to SFDIF to provide good displacement of mud from the openhole and to minimise losses when
running sand screens.

Clear Brine
An hydrate-inhibiting clear brine was used as the displacement fluid in the cased hole above the SFDIF in the completion phase
and used to circulate the well during the gravel pack operations. Again, the density had to be the same as the RDIF, be compatible
with all other fluids, and result in no damage to the reservoir.

Carrier Fluid
The carrier fluid was used to transport the gravel into the wellbore-sand screen annulus and contained a slow-acting, breaker-
treatment filter cake removal to enhance cleanup and productivity. The treatment comprised enzyme breakers to break down
polymers and starches and an in-situ acidizing system to dissolve the calcium carbonates in the filter cake.
Initially, the carrier fluid was designed to be the same density as the RDIF and other fluids, but this was relaxed due to high
corrosion rates in a low pH environment. Further optimisation to minimise corrosion resulted in the addition of sodium acetate as
a pH buffer to slow the speed of acid generation and to prevent pH from falling too low. Removal of the MEG (the hydrates risk in
the carrier fluid was felt to be low), improved the post-treatment efficiency. The density of the carrier fluid is 1.19 s.g., but with
the addition of the gravel the density is 1.24 s.g., thus meeting density requirements.

Packer Fluid
The packer fluid had to be non-corrosive and remain in the A-annulus for the lifetime of the well. The fluids had to be hydrate
inhibitive at -2° C and 290 bar. A 1.17-s.g. MEG NaCl formulation was found to provide the required hydrate inhibition.
Corrosion inhibition was provided by pH adjustment to pH 9.5 to 10 and the addition of an oxygen scavenger. See Table 1 for all
fluid formulations.

Compatibility Testing
Compatibility testing was performed on all fluids that would come into contact with each other at downhole and BOP
conditions to ensure no unexpected reactions, such as salt dropout, would be experienced. No major incompatibilities were seen.
The true crystallization temperature (TCT) was determined for base brines and blended to ensure no dropout during
transportation in winter conditions.

Formation Damage Testing

Full-sequence coreflood testing was performed, in which all fluids were applied to the Ormen Lange core to replicate the same
manner in which they will be used in the field during the completion phase.
Tests were run with and without carrier fluids to demonstrate that the breaker treatment increased the return permeability.
Long-term suspension tests were also performed, with extended soak periods of 13 and 26 weeks to establish that no
troublesome reactions would take place if there were delays in cleaning the lower completion fluids (see Table 2).

Hydrate Inhibition
In the event of a gas influx, it was stipulated that the water-based mud and the lower completion fluids would be hydrate
inhibitive. Hydrate inhibition was provided by a combination of saturated salt and MEG. Concentrations were selected by
modelling and were confirmed by testing at an independent hydrate testing laboratory. This effort produced the hydrate phase
boundary for all the hydrate-inhibiting fluids using Ormen Lange gas composition (Fig. 7).

Corrosion Testing
Corrosion testing was performed on fluids that would contact the wellbore metallurgy for an extended period. Testing
identified issues with 13Cr L80 material when exposed to a NaCOOH based carrier fluid, even for a short period of time (Fig. 8).
Further testing at downhole conditions produced a better understanding of pH behaviour and corrosion rates, enabling fluids to be
optimised. The final carrier fluid formulation, with a sodium acetate buffer to prevent severe pH drop, gave an average for
13Cr L80 material of 1.4 mm per year when tested at downhole conditions in the absence of filter cake. The corrosion rates on the
higher alloy (Inconel 825 and 316L) material used on the other completion material were negligible.
8 SPE 144089

QA/QC of Fluids in the Field During the Campaign

An extensive QA/QC programme was initiated to ensure the laboratory formulations were reproduced in the field throughout
the campaign. Emphasis was placed on minimising contamination during mixing at the onshore plant, the supply vessel transport
and at the rigsite. A dedicated supply vessel used throughout the drilling and completion campaign supplied the dedicated tanks
and lines for the various fluids. All brines were filtered as they come on board the drillship and the rig’s surface lines and pits were
thoroughly cleaned in advance once we moved to batch completion phase.
Laboratory testing offshore included pore plugging apparatus (PPA) to ensure correct bridging with the calcium carbonate
bridging agent sizes in the received RDIF from the onshore plant, production screen tester (PST) to ensure all used fluids could
pass the sand screen if they were left downhole, and carrier fluid internal breaker-specific tests (enzyme activity, pH over time
profile, filter cake breakthrough tests) to ensure the carrier fluid could break down the filter cake as designed.

2008-2009
By 2009, six wells (A2, A3, A6 and A7; B2 and B3) had been drilled and completed on Ormen Lange as per the above
formulations.
However, it was recognised that reservoir depletion effects prompted a requirement to:
a) Lower the mud weight and gravel pack completion fluids to 1.17 s.g. due to high overbalance,
b) Lower the packer fluid density (this was left in the production tubing, and when the FIV was opened it effectively killed
the well when the workover riser was in place).

Work was initiated in 2008 to look at these changes. The specification for the fluids remained the same as for the original
concept selection.

Lower Mud Weight RDIF and Carrier Fluids

Initial design work revealed that a lower mud weight could not be achieved without compromising hydrate inhibition.
Laboratory work was initiated, but it soon became apparent that hydrate inhibition would be severely compromised. The lighter-
weight carrier fluids also showed significantly increased corrosion rates due to the increased free water in the formulation.
Research to find lighter-weight RDIF and gravel pack fluids for the lower completion was aborted.

Lower-Weight Packer fluid

The new packer weight fluid design had to be hydrate inhibitive and again have low corrosion rates. The fluid selected was a
MEG freshwater blend for which the hydrate inhibition properties are well documented and understood. Initial concern over MEG
stability for the lifetime of the well was relaxed and a 1.06-MEG- freshwater blend was proposed. Again, corrosion inhibition was
provided by pH adjustment from 9.5 to 10 and an oxygen scavenger.
This fluid was used in 2009 on Wells B6, B7, and A4, A5 and the wells were successfully flowed back to the West Navigator
drillship and into full production at Nyhamna.

2010 – Lightweight Packer Fluid and Flow Back to Host

In 2010, the desire to flow back the wells directly to host required an even lighter fluid weight. A base oil with a 0.82 s.g. was
selected (based on density and HS&E). Due to compatibility concerns with the BOP elastomers and HS&E concerns (including
spills to the environment), it was decided not to circulate the compete well to base oil before running the production tubing, but to
displace base oil inside the production tubing before setting the production packer.

Qualification of the Base Oil for Flow Back to Host

The change to base oil initiated further testing, as there were concerns of formation damage and emulsion problems at the plant
at Nyhamna. Formation damage could occur as base oil might enter the formation when the FIV was opened as the well was being
unloaded. Core flood testing including a full sequence test indicated base oil entering the formation would be no issue. (Table 3).
The lower completion fluids flowing back to the plant at Nyhamna could potentially cause emulsion problems and initiate a
process trip, posing a risk to the environment. Emulsions can be formed when subjected to high shear conditions such as
experienced due to pressure drop across a choke.
The lower completion fluids contain calcium ions and formation fines, all of which can stabilise emulsions. Fluid testing was
performed to confirm compatibility with all fluids that could mix downhole at both pipeline (-1° C) and plant conditions (40° C).
Screening was performed at the same time to identify a suitable demulsifier if emulsions were formed.
Laboratory test results showed that stable emulsions could only be formed under very high shear conditions. A suitable
demulsifier was identified, and this was utilised on the first D-Template well D7, which flowed back to host. For the second well,
D2, the risk was felt to be low, so fluids flowed back without demulsifier injection without any problems.
SPE 144089 9

RDIF SFDIF Brine Carrier fluid Packer fluid Packer fluid

2007-2008 2009-2010
Component Units Concentration Concentration Concentration Concentration Concentration Concentration
Fresh water g/l 774 710.5 703 816.6 667 450
NaCl g/l 232 170 105.5 268.6 210 ---
MEG g/l 89 134 134.5 --- 289 550
NaCOOH g/l 200 200 297.3 --- --- ---
Xanthan Gum g/l 2,9 2.9 --- --- --- ---
Modified Starch g/l 20,0 20 --- --- --- ---
Calcium g/l 59,9 --- --- --- --- ---
Carbonate A
Calcium g/l 59,9 --- --- --- --- ---
Carbonate B
Surfactant g/l 2,9 2.9 --- --- --- ---
Acid Generator g/l --- --- 59 --- ---
Enzyme A g/l --- --- --- 5.9 --- ---
Enzyme B g/l --- --- --- 11.7 --- ---
Enzyme C g/l --- --- --- 5.9 --- ---
Sodium Acetate g/l --- --- --- 23.8 --- ---
NaHCO3 g/l --- --- --- --- 0.3 0.3
Oxygen g/l --- --- --- --- 0.1 0.1
Scavenger
NaOH g/l --- --- --- --- Approx.1.5 Approx. 1.5
Density@20° C 1.24 1.24 1.24 1.19 1.17 1.06
Table 1 - Fluids Formulation Summary

Figure 6 - Filter Cake Destruction by Carrier Fluid with Enzyme and Acid Treatment

Core Sample Fluids Applied Base Permeability (mD) Final Permeability (mD)
(% Return Perm )
A RDIF 562 503 (90%)
B RDIF, SFDIF, Carrier fluid 432 187 (43%)
gravel pack and screen
applied, Carrier fluid soak
–No enzyme breaker
C RDIF, SFDIF, Carrier fluid 469 355 (76%)
gravel pack and screen
applied, Carrier fluid soak
–with enzyme breaker
D As C but soak for 13 475 284 (65%)
weeks
E As C 482 264 (55%)
Table 2 - Return Permeability Results
10 SPE 144089

Figure 7 - Experimental derived hydrate phase boundary lines for main drilling and completion fluids and water for
Ormen Lange multicomponent gas

Figure 8 - Corrosion test results on 13 Cr coupons with different brine-based breaker systems

Core Sample Fluids Applied Base Permeability (mD) Final Permeability (mD)
(% Return Perm )
G 0,82 Base Oil 705 759
(+7,66%)
H RDIF, SFDIF, Carrier fluid gravel pack and screen 778 778
applied, Carrier fluid soak –Base oil application (+5,66%)
I RDIF, SFDIF, Carrier fluid gravel pack and screen 707 778
applied, Carrier fluid soak –Remove Filtercake (+5,94%)
–Base oil application
Table 3 - Coreflood Testing with Base Oil
SPE 144089 11

Job Execution Overview: 2007-2010 Campaign

Over the course of the four-year campaign to date, 13 lower completion openhole gravel pack completions have been
successfully installed on Ormen Lange with full packs on each. During the first phase of the project, 10 wells were completed in
the central area of the field, with a maximum 45-degree angle through the reservoir. Wells completed during the second phase of
the development in the southern area of the field were completed as horizontal openhole gravel packs. The main driver for
changing to horizontals was to increase the standoff to the water contact, which is shallower in the south, and thus mitigate the risk
of water coning, which would pose a serious flow assurance risk due to insufficient hydrate inhibition.
All wells throughout the campaign were typically drilled and completed in batch mode, where the top holes and intermediate
section into the top reservoir were completed in the autumn and winter time frame as these were less weather-sensitive activities.
In early spring until mid-summer, after drilling out the reservoir section and installing the horizontal trees, the lower and upper
completion would be installed in batch mode. This activity would be followed by deployment of the workover riser to flow back
and unload the wells to the rig, and the handover of the wells to production. In 2010 two wells were successfully unloaded directly
to the host. These will be discussed later in this paper.
After re-entering the suspended wells andthe 9⅝-in. liner was set into the top reservoir, the following steps were conducted:
• Retrieve suspension plugs
• RIH to TOC, displace well from inhibited seawater to reservoir drill in fluid (RDIF)
• Drill out shoe track and drill 8½-in. hole through reservoir to TD, POOH
• Underream to 12¼-in. (only for phase 1 wells), POOH
• Perform wellbore cleanout trips with dedicated gravel pack string
• Displace openhole to solids-free drill-in fluid, pull back inside casing shoe
• Pump soap pill around to clean casing and workstring
• Displace casing and riser to filtered brine within cleanliness specification, POOH
• Make-up screens and lower completion assemblies
• RIH to TD
• Displace openhole from solids-free drill-in fluid to filtered brine
• Set gravel pack packer and release service tool
• Locate and test gravel pack tool positions (circulate, reverse and test position)
• Perform circulation test to confirm sufficient returns
• Forward-circulate the well above the gravel pack packer to filtered brine
• Test gravel pack packer
• Mix carrier fluid containing delayed breaker while conducting rehearsal screenout drill
• Establish circulating parameters in circulate position
• Perform gravel packing operation
• Recover workstring through formation isolation valve and close the same
• Pressure test formation isolation valve from above
• POOH with service tools to surface
• Run straddle across gravel pack closure sleeve
• Release running tool and displace well to packer fluid packer fluid
• POOH
• Run upper completion
The following section provides some details on the individual well executions performed during the four-year timeframe.

2007
In April 2007, the first well A7 was successfully gravel packed. After the gravel pack packer was set, the service tool
prematurely released during the pull-push test. Post-job analysis on the service tool revealed two shear screws of the release sleeve
were missing. After the screenout, after picking up to reverse position, the service tool was stuck for several hours before the
string was worked free. The stuck-tool incident was attributed to a combination of poor communication and clear guidance for the
driller on maximum pull limits at surface. This led to decisions being made without considering the consequences of proppant
falling on top of the packer. Clear maximum surface pull figures were prepared for all future jobs and clearly communicated to the
driller and more emphasis on screenout rehearsal drills were conducted. No further incidents occurred in the remaining campaign,
even though some very high overpulls were noted on several wells after screenout to pull the service tool back into the reverse
position.
12 SPE 144089

On the second well A2, a surface leak developed on the low-pressure side of the mixing equipment, and some proppant spilled
onto the deck. As sufficient proppant was in the workstring once the incident occurred, the job was continued. After the screenout,
the spilled proppant was collected for mass-balance check to confirm sufficient proppant was placed.
On the third well, A3, the gravel pack service tool would not move back down to the circulate position after the packer was
tested. The service tool was pulled back to surface. Observations of seal damage indicated there might be some small junk in the
well, restricting free movement of the service tool seals. The tool was redressed and rerun, and the job was successfully pumped.

2008
On well B2, the service tool prematurely released during the gravel pack setting sequence. The service tool was retrieved to
surface followed by a run to retrieve the packer and screen assembly. The weather deteriorated, so activity at the well was
temporarily suspended. After reentering the well six months later, a check trip to TD indicated the hole was still in good condition,
and the well was successfully gravel packed. The long suspension of the openhole to solids-free drill-in-fluid did have a negative
impact on initial productivity, but over time the well cleaned up within the same range of productivity indexes (PI’s) as the other
wells. A thorough investigation into the gravel pack tool failure, resulted in a rigorous revised QA/QC assembly plan for the gravel
pack tools after. No further gravel pack tool incidents have occurred in the remainder of the campaign. Two other wells, A6 and
B3, were gravel packed without any noticeable incidents.

2009
During the 2009 campaign, four wells were completed without any incidents. Tagged proppant was used from this year
onwards to enable identification of any proppant being produced back to the onshore gas plant to individual subsea wells in case of
a sand control failure. In addition, two wells were completed with cableless gauges above the screens that enabled the collection of
pressure and temperature data close to the reservoir and provided more accurate data on actual flowing bottomhole pressure, which
was used to fine tune the used-lift curve correlations. More details can be found in the technology section of this paper.

2010
During the 2010 campaign, three horizontal openhole gravel packs were successfully performed with full alpha beta packs on
wells D7, D2 and D3. The last well D3, was initially suspended for three months due to issues with the riser slip joint, which had
to be repaired. After re-entering the well, the openhole was reamed to 9¾ in. before installation of the lower completion, to ensure
a fresh filtercake would be in place, enabling better cleanup action of the breaker system. Two additional cableless gauges were
installed, and a water-tracer screen pup was installed in the straddle assembly. This assisted in detection of any formation water
breakthrough as this carried a higher flow-assurance risk in the southern section of the field due to the smaller standoff to the water
contact. Further details on this technology can be found in the new technology section.
Throughout the campaign, the carrier fluid formulation contained a delayed breaker system that assisted in breakdown of the
filtercake during the suspension time of the wells before they were produced back. A large number of failures after the closure of
the gravel pack sliding sleeve, mitigated by the installation of the straddle, led to a redesign of the sleeve. Changing out the seal
design from an O-ring-based to a molded seal type of design proved successful, as no additional failures occurred after this
modification. For the horizontal wells a contingency system was available consisting of a 7-in. pre-drilled liner with 5-in. screens
that would have been deployed if wellbore stability issues occurred, requiring an increase in mud weight. All jobs were pumped
directly through the rig’s cement line and topdrive, mitigating any manriding and time-consuming rigup of chiksan lines, which is
the typical setup on most other rigs. Before the start of each year’s completion campaign, the rig’s fluid system was thoroughly
cleaned by a dedicated company to mitigate the risk of contamination of completion fluids. Gravel pack quality was measured
through post-job analysis of washpipe-deployed memory gauges and through accurate mass-balance check on the volume of
proppant placed downhole. A purposely built sand-catch tank was placed at the shakers to catch and measure all returned proppant
after reverse out. Washpipe gauge data was also used to analyse if any additional cycles would have been applied on the formation
isolation valve. Pump schedules were based on overgauge hole sizes and 30% excess in slurry design to account for washouts and
other uncertainties and to ensure a positive screenout would be achieved and no top-up job would have to be performed.

Cleanups and F2H

Ormen Lange Well Clean-up Operations
From 2007 to 2009, the Ormen Lange wells were cleaned up through a temporary well test package on board the West
Navigator drillship. The objectives of the clean up were to efficiently remove all residual drilling mud, completion fluid, to
displace the well to hydrocarbons ready for production, to gather representative surface samples, to verify the integrity of the
gravel pack and completion, and to assess well performance.
Initially, the formation isolation valve (FIV) was cycled open by pressurizing a fluid-full workover riser, and then nitrogen was
injected through the annulus line to unload the 1.17-sg packer fluid. This resulted in high liquid losses in the reservoir. To
remediate the losses and also due to a depleted reservoir pressure, the unload methodology was slightly modified. As described in
SPE 144089 13

a previous section, a lighter packer fluid, 1.06 sg, was used. The workover riser was also displaced to nitrogen before the FIV was
cycled open. The latter was then cycled open via the annulus rather than via the workover riser. After the FIV was open, the well
was unloaded by injecting nitrogen via the annulus.
The wells were unloaded to approximately 2.8 MSm3/d, which corresponds to the handling capacity of the surface well test
package.
In 2010, in an effort to eliminate flaring on the drillship and reduce HS&E exposure, it was decided to clean up the wells
directly to the Nyhamna plant in Aukra. This opportunity applied to three wells by reducing rig time by 20 days for each well clean
up. However, several risks were associated with this opportunity such as corrosion, emulsion, flow assurance and plant operability.
Intensive fluid testing, described in previous section, and multiphase flow simulations performed to mitigate some risks. To kick
off the well, the tubing had to be displaced to base oil and the subsea pressure had to be drastically reduced. However, no change
on the plant operating pressure was required. The well was unloaded to 3 MSm3/d in less than 24 hours with the drillship still
connected to the well to cycle-open the FIV and perform the annulus depressurization during the unloading. No major issue
occurred at the plant. Provisions were made to inject demulsifier at the inlet of the plant, but this was not required.
Future unloads to the plant will involve new challenges due to depleted reservoir pressure such as decreasing the plant pressure
beyond what has been performed to date and displacing the tubing to nitrogen.
Ormen Lange Well performance
All Ormen Lange wells are completed with two downhole gauges, one above the tubing retrievable subsurface safety valve
(TRSSSV) and one above the production packer. The most recent wells also have a cableless gauge, as described latter in this
paper, located at the top reservoir. Having three to four pressures data points in the well provide a very high level of confidence in
the outflow modeling. The outflow is very well-matched with the lift correlation used, without any adjustment required. The
productivity is higher than originally predicted, with the improvement mainly due to higher-than-expected permeability and
thickness values combined with lower skins than expected. These results confirm that the design and execution of the lower
completion contributed to the well performance. To date, no decrease in productivity on the Ormen Lange wells has been observed
(Fig. 8). Some Ormen Lange wells have even shown an increase in productivity (Fig. 9).

Ormen Lange Productivity vs. Time

Productivity Index

0 200 400 600 800 1000 1200

Elapsed Time since start up, days

Actual PI Expected PI

Figure 8 Example of an Ormen Lange well with constant productivity

14 SPE 144089

Ormen Lange Well Productivity vs. Time

Productivity Index

0 100 200 300 400 500 600 700 800

Elapsed Time since start up, days

Actual PI Expected PI

Figure 9 Example of an Ormen Lange well with an increase in productivity

New Technology Phased in Throughout Campaign

Throughout the four-year completion campaign, several new technologies were qualified and successfully integrated into the
lower completion.

Tagged Proppant
All individual subsea wells were fitted with acoustic detectors and intrusive erosion probes to detect any sand production if
sand control failure occurred. If any of these sand detection systems would fail to pickup signs of a failed completion and proppant
was found back at the gasplant, the source well could not be identified. For this reason, a tagged version of the proppant was
introduced that enabled tracing the unique tag back to an individual well. The properties of the proppant remained unchanged as
the tracer was completely inert. The tracer consisted of boosting one trace element in the commonly occurring earth oxides during
manufacturing of the ceramic proppant up to a detectable level. Spectral lab analysis of individual proppant particles could then
identify the unique tag.

Cableless Gauges
The big-bore, high-flowrate completion designs being used on Ormen Lange make it impractical to install traditional, cabled
permanent downhole gauge (PDG) systems in close proximity to the producing sandface. With an along-hole separation distance
in excess of 1,000 meters between the PDG and the sandface, and with frictional pressure drop gravity head differences and
temperature effects to contend with, there is uncertainty in how the pressure measurements being recorded by the PDG relate to the
actual sandface flowing pressure (figure xx, showing inaccuracy of extrapolation).
Access to high-accuracy sandface flowing pressure data enables determination of skin damage and the possibility of
monitoring the sand face completion efficiency and integrity. Actual sand face pressure data also reduces the uncertainty of the
drawdown at the sand face in constrained wells compared to using multiphase flow correlations. An improvement of 1 to 2 percent
in the calculated drawdown can improve production by 1 to 2 percent in drawdown constrained wells. The cableless gauge data
enabled fine-tuning of the multiphase flow correlations (used to calculate the frictional pressure drop throughout the completion),
resulting in operating standard wells at their full potential (maximum drawdown limitation) as well.
Resulting from an engineering development program extending over more than two years, the gauge large-bore mandrel
system was deployed as part of the lower completion assembly just above the sand screens. The incorporation of duplex
functionality into the mandrel system enabled the frequency of data transmissions to be varied on demand from the central control
room located at Nyhamna, onshore Norway. In addition, there was also an option to trigger high-density data capture for pressure
build-up analysis. The system comprises of the following three main elements

Gauge Mandrel
This mandrel system is a concentric full-bore system that is deployed as an integral part of the lower completion assembly. The
mandrel electronics package consists of a two-way wireless communications module, quartz transducer module and battery
SPE 144089 15

package. Downhole data can be acquired via a pre-programmed schedule or by the user-controlled data-acquisition modes that can
be initiated via commands from the onshore control room.

Downhole Pickup (with ¼-in. electrical control line)

The cableless gauge downhole pick-up (DHPU) is an assembly that is used as a means to pick up the cableless gauge data
signal downhole. It consists of a clamp-type arrangement to which a ¼-in. OD tubing encased conductor (TEC) line is connected.
This line runs to the subsea tree and is terminated at the cableless gauge subsea transceiver package.

Subsea Transceiver
The subsea transceiver collects, decodes and stores data from the downhole system. The data are transmitted onwards to the
client’s own data interface. The transceiver also transmits commands received from the control room via an IWIS Option 3
Interface to the downhole mandrel.
The battery life is limited to the number of data transmissions performed, which is based on the schedule setup for the gauges.
With the current schedule consisting of one daily high-resolution pressure and a low-resolution temperature data point
transmission and four to five pressure buildups, battery life is estimated to be two years. Four gauges have been successfully
installed, two in the 2009 campaign and two in the 2010 horizontal completion campaign. Interestingly, although one of the 2010
wells has no upper completion installed yet, through the downhole pickup in one of the adjacent wells the readings from the
cableless gauges from the lower completion have been recorded, providing valuable information on reservoir pressure response.
One of the first installed gauges has stopped transmitting data after approximately 20 months of service.

Water Tracer
Formation water breakthrough poses a serious risk to flow assurance as hydrate inhibition at each individual wellhead is
limited to a maximum MEG injection rate. Current water-detection level of the wet-gas meter installed at each tree could only
pickup a massive water influx event. However, a gradual increase of water production (due to formation water breakthrough)
below the detection limit of the wet-gas meter (0.05% equating to +/- 20 m3/d additional water influx at 10 Msm3/d well, becomes
progressively less as the rate drops towards 5-8 Msm3/d range), is unlikely to be detected.
The sampling program at the gas plant, which includes water ion-analysis, could possibly detect the onset of formation water,
but as this is performed on a bi-weekly basis, a delay between actual formation water breakthrough and analysed lab data would
occur. Water ion-analysis can not detect the source well of formation water production.
A qualification study was started to investigate the application of tracer technology to detect formation water from individual
Ormen Lange wells. Tracer technology had been successfully deployed in oil wells to identify water breakthrough or to detect
individual zones contributing to oil production in long horizontals, but the technology had never been applied in gas wells.
The technology is based on selective diffusion, a function of the surrounding liquids, governed by solubility and affinity. The
polymers are doped with non-radioactive tracer molecules and act as the carrier and remain in the well permanently. Released
tracers are transported to the surface by produced fluid flow, and through the routine analysis onshore they can be detected. The
analysis technique has a detection level of parts per billion (ppb).
During the technology qualification, an additional complication was identified as any released tracer would not be removed
from the rich MEG phase after flowing through the regeneration process onshore. Any released tracer during the suspension phase
of the well in brine prior to startup of the well would, therefore, remain in the closed MEG loop, a system volume of 17,000 m3.
To test the behaviour of the tracers in the MEG loop at Ormen Lange a full-scale field test was performed. Some 10 g of the
water tracer RWT-002 was added as a shot into the full 17-million liter MEG system volume, to simulate if the expected initial
release of tracer, whilst suspended in completion fluid before the well is flowed back to host, could be detected.
In the second phase of the test, 1 g of the same tracer RWT-002 was injected as a shot along with 10 g of a new water tracer
RWT-017. The objective of this test was to determine if an additional release of tracer could be detected above the established
baseline measurement which was positively confirmed (Fig. 10).
16 SPE 144089

Figure 10 Tracer concentration vs. time for full scale MEG flowloop test

The carrier used for the tracer system is a standard direct wrap screen pup joint (Fig 11.) positioned above the lower
completion, acting as a type of vent screen. The wetting of the systems by droplets in the well stream is crucial to enable tracer
release, and this was verified by CFD simulations. Tracer monitoring systems were developed, and two unique tracer systems were
installed in one tracer carrier per individual well. During the 2010 campaign, tracer systems were successfully installed on all three
completed wells.

Figure 11 Preparation of Tracer Pup Joint

SPE 144089 17

Conclusion
The Ormen Lange development is an example for the industry that subsea deepwater developments in the challenging and
harsh Norwegian Sea conditions can be performed safely and efficiently. Through a thorough qualification program, focusing on
assurance that the selected sand control method and all associated lower completion components would be able to perform within
the 30-year lifetime expectancy and within the very challenging operating envelope, thirteen highly productive lower completions
have been delivered, producing close to 70 million Msm3/d. A key component contributing to the success of these wells
performance has been the focus on the completion fluids selection, qualification and integrating these with the overall well
requirements and objectives. Through a phased introduction of new technology throughout the four-year campaign, valuable data
has been added allowing wells to be operated more efficiently (cableless gauges) and introducing additional detection capability to
ensure flow assurance is not compromised (tagged proppant and water tracer technology). A major contribution to successful
project delivery has been the clear prioritization of key performance indicators in the following order of importance: safety,
quality, schedule and cost. Any challenges faced during the design and execution phase offshore where always solved with quality
in mind ranking higher than cost as failure of any of the prolific high-rate wells was not an option. Lastly, having a ring fenced
team with a number of dedicated highly gifted technical professionals working on the concept selection, qualification and
execution of the wells and keeping a number of these key individuals together in the team for more than six years has been a key
enabler of the success story of Ormen Lange.

Acknowledgement
The authors would like to express thanks to the management of A/S Norske Shell and the other partners in the Ormen Lange
venture for permission to publish this paper and to the many colleagues who have contributed to the success of this project.