Flow Assurance Guidelines

SHELL NIGERIA EXPLORATION AND PRODUCTION COMPANY Ltd.

Bonga FPSO Plant Operating Procedures Manual Volume 2D FLOW ASSURANCE GUIDELINES

OPRMOPRM-20032003-0302D Version: 1.1

This document is confidential. The Copyright of this document is vested in Shell Nigeria Exploration and Production Company Limited. All rights reserved. Neither the whole nor any part of this document may be reproduced, stored in any retrieval system or transmitted in any form or by any means (electronic, mechanical, reprographic, recording or otherwise) without the prior written consent of the copyright owner.

S Shheelll N Niiggeerriiaa EE && PP CCoom mppaannyy LLttdd..

2.0

U Unnrreessttrriicctteedd

PURPOSE The purpose of this document is to provide guidance on the safe, efficient and environmentally aware operation of the Subsea Facilities, Flowlines and Risers. It is one Volume within an overall suite of Volumes, which comprise the Bonga FPSO Plant Operating Procedures Manual (POPM). The full listing of Volumes is as follows: Volume 1 Volume 2A Volume 2B Volume 2C Volume 2D Volume 3 Volume 4 Volume 5 Volume 6 Volume 7 Volume 8 Volume 9 Volume 10 Volume 11 Volume 12 Volume 13 Volume 14 Volume 15 Volume 16 Volume 17 Volume 18 Volume 19 Volume 20 Volume 21 Volume 22 Volume 23 Volume 24 Volume 25 Volume 26 Volume 27 Volume 28 Volume 29 Volume 30 Volume 31 Volume 32 Volume 33 Volume 34 Volume 35

OPRM-2003-0302D

Field and Facilities Overview Subsea Production System Subsea Waterflood System Subsea Control System Flow Assurance Guidelines Oil Separation and Treatment Oil Storage, Handling and Ballast Systems Oil Metering and Export System Vapour Recovery Compression System Field Gas Compression System Gas Dehydration/Glycol Regeneration Systems Gas Export/Import/Lift Systems Flare and Vent Systems Produced Water Treatment Systems Waterflood System Chemical Injection and Methanol Injection System Fuel Gas System Heating Medium System Drainage Systems Sewage Treatment Systems Bilge and Oily Water Separation Systems Inert Gas System Nitrogen Generation System Seawater System Fresh and Potable Water Systems Diesel Fuel System and Incinerator Aviation Fuel System Instrument and Utility Air System Deck Hydraulic Systems Fire Protection Systems and Equipment Safety and Lifesaving Equipment PSCS and ESS Power Generation and Distribution Systems Black Start Procedures HVAC Systems Deck Machinery and Mechanical Handling Systems (Cranes, etc) Telecommunications Ancillary Living Quarters (ALQ)

Page iii of xi

30-April-2006


3.0


SCOPE This document provides detailed reports and studies carried out to provide guidelines for the safe operation of the Bonga subsea facilities. The studies also include step-by-step guidance on the operation of the system under both normal and abnormal operation.

4.0

TARGET TARGET READERSHIP All SNEPCO staff who may be involved in the operation of the Subsea Systems onboard the Bonga FPSO.

5.0

SPECIAL NOTE Not applicable.

6.0

DEFINITIONS AND ABBREVIATIONS The definitions and abbreviations used within this document are listed at the end of these introductory pages.

7.0

REFERENCE INFORMATION/SUPPORTING DOCUMENTATION The primary reference/supporting documents, which have been either used or referred to in the development of this document, are listed at the end of these introductory pages. These are part of the available Operational Documentation, which SNEPCO Offshore Operations (OO) has in place to support its day-to-day operations. These and many other documents are available within the SNEPCO Livelink System. Where appropriate, these documents have been cross-referenced within this document.

OPRM-2003-0302D

Page iv of xi

30-April-2006



Definitions and Abbreviations Definitions Arrival Temperature

Flowing temperature of the fluids at the FPSO boarding valve.

Backpressure

Pressure on back of valve against which equalising pressure is applied to reduce differential

Blowdown

Action performed to depressurise the flowline, designed to reduce the maximum flowline pressure and thus reduce the risk of hydrates at ambient conditions (4°C) in the event of an extended shutdown.

Bubble Point

The bubble point is the pressure at which gas first comes out of hydrocarbon liquid phase for a given temperature.

Cloud Point

The cloud point is the temperature at which wax crystals begin to precipitate in the fluid. This is commonly taken to be the temperature for the onset of wax deposition, also called the Wax Appearance Temperature.

Cold Earth Start

Start-up in which the wellbore, wellbore fluids and all subsea equipment are initially at ambient temperature.

Equalising Pressure

Pressure applied to equalise pressure across the valve (ideally this should be greater than the downstream pressure).

Forward Pressure

Pressure on front of valve prior to equalising pressure being applied.

Gas Void Fraction

Technically defined as the ratio of the gas volume to the flowline volume, but it is more appropriately defined as the minimum gas volume required to achieve a successful flowline blowdown.

Hot Oiling

Precirculating heated dry hydrocarbons or diesel around a flowline loop to warm the flowlines and manifold prior to a cold well startup.

Hydrate Dissociation/ Formation Temperature

The temperature at a given pressure above which hydrates will not form or the temperature at a given pressure below which hydrates will form.

No-touch Time

The period of time following a shut-in during which the equipment is allowed to cool and production may be restarted without the need to inhibit the system.

Pour Point

The pour point of a petroleum fluid is the lowest temperature at which the fluid ceases to flow when brought to the temperature under specified conditions.

Safe Condition

The condition at which the subsea system has attained the desired temperature required to achieve minimum cooldown time.

OPRM-2003-0302D

Page v of xi

30-April-2006



Safe Condition Temperature

The temperature at which any section of the subsea system has the minimum specified cooldown time (8 hours for wellbore and 12 hours for the rest of the subsea system).

Safe Condition Time

The time taken to reach safe condition temperature.

Warm-up Time

The time that it takes the systems to reach a temperature sufficient to give the desired number of hours of cool down.

Abbreviations API ASTM

American Petroleum Institute American Society for Testing and Materials

Ba BaSO4 BIST BLPD BoD BOOR BS&W BSET

Barium Baryte Bonga Integrated Studies Team Barrels Liquid Per Day Basis of Design Bonga Oil Offloading Riser Base Sediment and Water Bonga Systems Engineering Team

CaCO3 CIV CPM CWDT

Calcite Chemical Injection Valve Cross-polar Microscopy Critical Wax Deposition Temperature

DTI

Department of Trade and Industry

EPIC ESDV

Engineer, Procure, Install and Construct Emergency Shutdown Valve

FAST FDP FEAST FPSO FPT FWHP FWHT

Flow Assurance Sub-team, Houston Field Development Plan Fluids Evaluation and Stability Testing Floating Production, Storage and Offloading Field Planning Tool Flowing Wellhead Pressure Flowing Wellhead Temperature

GLIV GLR GoM GOR

Gas Lift Injection Valve Gas Lift Riser Gulf of Mexico Gas/Oil Ratio

HDP HDT HRGC HS&E HSE HTGC

Hydrate Dissociation Pressure Hydrate Dissociation Temperature High Resolution Gas Chromatography Health, Safety and Environment Health and Safety Executive High Temperature Gas Chromatography

ID ITT

Inside Diameter Invitation to Tender

KHI

Kinetic Hydrate Inhibitor

LDHI LP

Low Dosage Hydrate Inhibitor Low Pressure

OPRM-2003-0302D

Page vi of xi

30-April-2006



MBLPD MBOPD MBWPD MEG MeOH MIV MMBO MoC MPT

Thousand Barrels Liquid Per Day Thousand Barrels Oil Per Day Thousand Barrels Water Per Day Monoethylene Glycol Methanol Methanol Injection Valve Million Barrels Oil Management of Change Model Pipeline Test

NORM NLNG

Naturally Occurring Radioactive Material Nigerian Liquefied Natural Gas

OD OGGS OPEX

Outside Diameter Offshore Gas Gathering Plant Operating Expenditure

PFL PID PIP PIV PM PMV POPM POV PP PPD PSDV psia PU PVT PWV

Production Flowline Proportional Integral Derivative Pipe-in-pipe Pigging Isolation Valve Production Manifold Production Master Valve Plant Operating Procedures Manual Ported Orifice Valve Pour Point Pour Point Depressant Pipeline Shutdown Valve Pounds Per Square Inch Absolute Polyurethane Pressure/Volume/Temperature Production Wing Valve

SBHP SC SCF SCSSV SIEP SITP SOI SPM SRTCA SSSV STB SWV

Shut-in Bottomhole Pressure Safe Condition Standard Cubic Feet Surface Controlled Subsea Safety Valve Shell International Petroleum Maatschappij Shut-in Tubing Pressure Shell Offshore Incorporated (SEPCo) Single Point Mooring Shell Research and Technology Center, Amsterdam Subsurface Safety Valve Stock Tank Barrels Sacrificial Wing Valve

TEG THF

Triethylene Glycol Tetrahydrofuran

UTH

Umbilical Termination Header

VIT

Vacuum Insulated Tubing

WHP

Wellhead Pressure

OPRM-2003-0302D

Page vii of xi

30-April-2006



WSV WTC

Well Switching Valve Westhollow Technology Center

XOV

Crossover Valve

OPRM-2003-0302D

Page viii of xi

30-April-2006



Reference Information/Supporting Documentation (1)

Bendiksen, KH, Malnes, D, Moe, R and Nuland, S (1991), ‘ The Dynamic Twofluid Model OLGA: Theory and Application’ , Soc of Petro Engr, May 1991, Page 171.

(2)

Ellison, BT and Kushner, DS (1998) Subsea Oil Production System Design and Operations Methodology. Shell TIR (BTC-3534).

(3)

Granherne (1998) Bonga (7471-BON-TN-C-00037).

(4)

Granherne (1999) Riser Gas-lift System: Option Review and Recommendation (7471-BON-TN-U-00062).

(5)

Mehta, A (1998) E-mail communication to BSET Team.

(6)

Wasden, FK (1995) Mars Phase I Subsea Flowline Thermal Design Study. Shell TPR (BTC 9-95).

(7)

Ratulowski, J et al 1999 Asphaltene Stability, Waxy Fluid Properties and Wax Deposition Potential of Crude Oils from the Bonga Prospect, Nigeria.

(8)

Schoppa, W, Wilkens, RJ and Zabaras, GJ (1998), Simulation of Subsea Flowline Transient Operations. Facilities 2000 Proceedings, New Orleans, October 2627.

(9)

Van Gisbergen, S (1999) Email communication to BSET Team.

Major:

Technical

Note

–

Flow

Assurance

(10) Zabaras, GJ (1987) A New Vertical Two-phase Gas-liquid Flow Model for Predicting Pressure Profiles in Gas-lift Wells. Shell TPR (WRC 223-87). (11) Westrich, JT, Predicting Wax-related Fluid Properties Away from Well Control at Bonga, Report number SIEP.99.6096, August 1999. (12) Ratulowski, J, G Broze, J Hudson, N Utech, P O’ Neal, J Couch and J Nimmons. Asphaltene Stability, Waxy Fluid Properties and Wax Deposition Potential of Crude Oils from the Bonga Prospect, Nigeria. SEPTCo, Houston, March 1999. (13) Broze, G, N Utech, P O’ Neal and J Nimmons, Summary Report: Waxy Fluid Properties of Crude Oil from the B1 well, 803 Sand of the Bonga Prospect, Nigeria. SEPTAR, Houston, July 1999. (14) Bonga Integrated Studies Team. SDS-SNEPCo Bonga Joint Venture, Integrated Development Plan, Field Development Plan, Rev 5, December 2001. (15) Schoppa, W, Flow Assurance Constraints for Bonga Production Forecasting: Wrap-up. SGSUS, May 2002. (16) Schoppa, W and A Kaczmarski, Bonga Dynamic Flow Assurance Analysis – Evaluation of Conceptual Design. SGSUS, Technical Progress Report, February 2001. (17) Stankiewicz, Artur, Matt Flannery, Pat O’ Neal, Nancy Utech and George Broze, Asphaltene Stability and Wax Properties of the Crude Oil from the OPL 212 Prospect, Well W6, Bonga, Nigeria, SGSUS, October 2001.

OPRM-2003-0302D

Page ix of xi

30-April-2006



(18) George Broze, Bonga Oil Offloading Riser Wax Deposition. Memo to Ram Gopalkrishnan of SDS, September 2000. (19) Steve C Tsai, George Broze and Sabi Balkanyi, Bonga Production Flowline Wax Assessment. Shell Global Solutions, Houston, Texas, March 2003. (20) Bonga Oil Offloading Risers Conceptual Designs Summary (SD 991080). Revision R1, September 1999. (21) Pigging of Pipelines, State-of-the-Art, EP 95-2580, SIEP, The Hague, 1995. (22) SOI Deepwater Flowline Pigging Guidelines (similar to the guidelines for pigging section in the DEP 31.40.00.10 report). (23) Bonga System-wide Functionality Review in Amsterdam (Nov 2001) and email communications from H Duhon and A Kaczmarski. (24) Tsai, A, Broze, G and S Balkanyi, Bonga Production Flowline Wax Assessment. Shell Global Solutions, April 2003. (25) Westrich, JT, Predicting Wax-related Fluid Properties Away from Well Control at Bonga, Report No SIEP.99.6096, August 1999.

OPRM-2003-0302D

Page x of xi

30-April-2006



Main Table of Contents

Document Status Information Definitions and Abbreviations Reference Information/Supporting Information/Supporting Documentation Section 1

Dynamic Flow Assurance Analysis

Section 2

Flow Assurance Production Constraints

Section 3

Hydrate Remediation Guidelines

Section 4

Production Flowline Wax Assessment

Section 5

Offloading Riser Wax Assessment

Section 6

Pour Point Depressant Risk Assessment

Section 7

Scale Review

Section 8

RiskRisk-based Evaluation of Scaling Tendencies for the Subsea System

OPRM-2003-0302D

Page xi of xi

30-April-2006

Shell Nigeria E & P Company Ltd.

Unrestricted

Section 1 Dynamic Flow Assurance Analysis

Table of Contents 1.0

2.0

3.0

4.0

5.0

6.0

EXECUTIVE SUMMARY...............................................................................................5 1.1

Hardware Design ...............................................................................................5

1.2

Operational Procedures .....................................................................................5

ITEM OVERVIEW AND SPECIFICATIONS ..................................................................6 2.1

Introduction........................................................................................................6

2.2

Reservoir Fluid...................................................................................................7

2.3

Wellbore Characteristics ....................................................................................7

2.4

Subsea Flowline Details.....................................................................................9

2.5

Operating Conditions and Constraints..............................................................10

2.6

Objectives........................................................................................................10

2.7

Computational Approach..................................................................................11

COLD WELL START-UP: HYDRATE PREVENTION STRATEGIES .........................18 3.1

Cold Earth Well Start-up ..................................................................................18

3.2

Well Safe Condition Analysis ...........................................................................20

3.3

Flowline Hot-oiling............................................................................................21

STEADY-STATE PRODUCTION ................................................................................26 4.1

Steady-state Thermal Performance: Wellbore and Flowline.............................26

4.2

Terrain-induced (Severe) Slugging ..................................................................27

4.3

Riser Gas Lift: Thermal Considerations............................................................30

4.4

Umbilical-based Design ...................................................................................31

4.5

Large-bore Riser Design..................................................................................31

SUBSEA SYSTEM SHUTDOWN: HYDRATE PREVENTION STRATEGIES .............41 5.1

Cooldown Performance of Subsea Facilities ....................................................41

5.2

Flowline Blowdown ..........................................................................................44

5.3

Gas Lift-assisted Blowdown .............................................................................45

CONCLUDING REMARKS AND PRELIMINARY OPERATING LOGIC .....................60


OPRM-2003-0302D

Page 1 of 89

30-April-2006


Unrestricted

Table of Contents (cont’d) TABLES Table 1.1 – Riser Gas Lift Requirements for Terrain Slug Suppression ................................29 Table 1.2 – Cooldown Time as a Function of PU Foam Thickness Within ‘Pipe-in-pipe’ Flowlines ...........................................................................43 FIGURES Figure 1.1 – Production Forecast for Bonga Phase I Development (refer to Bonga Basis of Design).........................................................................................................13 Figure 1.2 – Bonga Subsea Field Layout..............................................................................14 Figure 1.3 – Bonga Production Well Design, Used for All Thermal-hydraulic Analysis..........15 Figure 1.4 – Production Flowline Topography for (a) 10in West-side Flowlines, and (b) 12in East-side Flowlines....................................................................................16 Figure 1.5 – Insulation Systems for 10in and 12in Pipe-in-pipe Flowlines (Left Panel), and Steel Catenary Risers (Right Panel) ...........................................................17 Figure 1.6 – Definition of Well Start-up Terminology.............................................................22 Figure 1.7 – Wellhead Warm-up Time to HDT, for Cold Earth Start-up of the Field’s Coldest Well (702p7) at 0% Watercut................................................................22 Figure 1.8 – Treatable Liquid Rate for 18gpm MeOH Injection (Mehta, 1999) ......................23 Figure 1.9 – Well Warm-up Time of 702p7: Dependence on Water Cut ...............................23 Figure 1.10 – Safe Condition Time for 8-hour Wellbore Cooldown .......................................24 Figure 1.11 – Influence of Watercut on Well Safe Condition Time for 702p7 ........................24 Figure 1.12 – Safe Condition Time for 12-hour Cooldown of Tree/Jumper/Manifold, Based on Time for Wellhead Temperature to Reach 120°F............................25 Figure 1.13 – Hot-oiling Performance: Return Temperature for 50MBOPD Circulation of 150°F Source Oil ........................................................................................25 Figure 1.14 – Flowing Wellhead Temperatures Calculated for Initial-life Wells and the Field’s Coldest Well (702p7) with 0% Water Cut.......................................33 Figure 1.15 – Arrival Temperatures Calculated for All Initial-life Wells with 0% Water Cut....33 Figure 1.16 – Cumulative Arrival Temperature for Initial-life Well Production, Relative to the 98°F Arrival Temperature Constraint for Waste Heat Capacity .............34 Figure 1.17 – Influence of Riser Gas Lift on Riser Froude Number, as a Means to Eliminate Riser Instability and Terrain Slugging Shown for the 12in East-side Risers .............................................................................................34 Figure 1.18 – Riser Base Gas Lift Required for Complete Suppression of Terrain Slugging for 10in West-side Flowlines ............................................................35 Figure 1.19 – Riser Base Gas Lift Required for Complete Suppression of Terrain Slugging for 10in East-side Flowlines .............................................................35


OPRM-2003-0302D

Page 2 of 89

30-April-2006


Unrestricted

Table of Contents (cont’d) FIGURES Figure 1.20 – Riser Base Gas Lift Required to Limit Terrain Slugging to Within 50bbl Slugs for 12in East-side Flowlines ..................................................................36 Figure 1.21 – Slug Volumes Calculated for 12in East-side Flowlines and 50% Water Cut as a Function of Gas Lift Rate ........................................................................36 Figure 1.22 – Separator Level Fluctuation Calculated for 12in East-side Flowlines and 50% Water Cut as a Function of Gas Lift Rate.........................................37 Figure 1.23 – Effect of Cold (40°F) Gas Lift Injection on Arrival Temperature for 10MBOPD Production and 25MMscfd Gas Lift for Slug Suppression .............37 Figure 1.24 – Gas Injection Temperatures at Mudline for Prior Umbilical-based Gas Lift Design...............................................................................................38 Figure 1.25 – Dependence of Gas Injection Temperature on Gas Lift Riser Diameter for an Insulating Value of U = 4W/m2-C ...............................................................38 Figure 1.26 – Dependence of Gas Injection Temperature on Gas Lift Riser Insulating Value for a 3.5in Tube Diameter .....................................................................39 Figure 1.27 – System Temperature Summary for Base-case Flexible Riser-based Gas Lift Design...............................................................................................40 Figure 1.28 – Definition of Contributions to Cooldown Time .................................................46 Figure 1.29 – Downtime Duration Statistics for Unplanned Shutdowns in GoM ....................47 Figure 1.30 – Wellbore Cooldown at Wellhead for Hottest and Coldest 702 Wells ...............47 Figure 1.31 – East-side 12in Riser Cooldown Performance for (a) 2in Carazite and (b) 4in Carazite ....................................................................................................48 Figure 1.32 – West-side 10in Riser Cooldown Performance for (a) 2in Carazite and (b) 4in Carazite ....................................................................................................49 Figure 1.33 – Pipe-in-pipe Cooldown for East-side 12in Flowlines .......................................50 Figure 1.34 – Pipe-in-pipe Cooldown for East-side 10in Flowlines .......................................50 Figure 1.35 – Pipe-in-pipe Cooldown for 10in West-side Flowlines ......................................51 Figure 1.36 – Illustration of Non-unique Relationship Between U Value and Cooldown........51 Figure 1.37 – Blowdown Performance: 10in West-side and Full Line-pack...........................52 Figure 1.38 – Blowdown Performance: 10in West-side and Immediate Choke Closure........53 Figure 1.39 – Blowdown Performance: 12in East-side and Full Line-pack............................54 Figure 1.40 – Blowdown Performance: 12in East-side and Immediate Choke Closure.........55 Figure 1.41 – Blowdown Performance for 50% Watercut, Illustrating Unsuccessful Blowdown for All Scenarios ............................................................................56 Figure 1.42 – Blowdown Performance with Riser Gas Lift Assist, for 12in East-side Flowlines.........................................................................................57 Figure 1.43 – Blowdown Performance with Riser Gas Lift Assist, for 10in East-side Flowlines.........................................................................................58


OPRM-2003-0302D

Page 3 of 89

30-April-2006


Unrestricted

Table of Contents (cont’d) FIGURES Figure 1.44 – Pressure and Temperature Evolution During Cold Gas Lift-assisted Blowdown ...................................................................................59 Figure 1.45 – Benefit of Depressurisation for Unsuccessful Blowdown in Providing 24 Hours of Additional Cooldown Time...........................................................60 Figure 1.46 – Cold Start-up ..................................................................................................61 Figure 1.47 – Additional Well Start-up ..................................................................................62 Figure 1.48 – Interrupted Start-up ........................................................................................63 Figure 1.49 – Planned or Unplanned Shutdown from Steady-state ......................................64 Figure 1.50 – Blowdown .......................................................................................................65 APPENDICES Appendix 1A – Reservoir Fluid Properties ............................................................................66 Appendix 1B – Wellbore Modelling Summary and Production Forecast ...............................71 Appendix 1C – Production Flowlines: Topography and Ambient Temperature Data .............79


OPRM-2003-0302D

Page 4 of 89

30-April-2006


1.0

Unrestricted

EXECUTIVE SUMMARY Using validated analytical and computational techniques, the dynamic thermalhydraulic performance of the Bonga conceptual subsea system is evaluated with regard to Shell guidelines for flow assurance in deepwater applications, with particular focus on hydrate management. Through simulation of worst-case (albeit realistic) operational scenarios, the principal objective of this work is to ensure a robust design of the Bonga subsea system, to enable efficient, hydrate-free operations. Analysis presented herein validates the Bonga conceptual design with respect to hydrate management, upon implementation of the following modifications to hardware design and operational procedures.

1.1

1.2

Hardware Design •

Replacement of gas lift umbilical with flexible riser and addition of gas lift heating (MoC 16)

•

Increase of carazite riser insulation thickness from 2in to 4in

•

Increase of polyurethane foam thickness in pipe-in-pipe flowlines from 0.6in to 1.0in

•

Inclusion of cooldown in riser/flowline thermal performance specifications (MoC 59)

•

Replacement of 2in topsides blowdown valve with two-stage valve train with large orifice

•

Added capability to isolate individual flowlines for dry-oil circulation

•

Added riser base pressure/temperature sensors (MoC 64)

Operational Procedures •

Identified need for well tubing Methanol (MeOH) bullheading for cold-earth start-up

•

Developed separate well start-up procedures for low and high watercut

•

Revealed that slug control not required for west-side flowlines, above 10MBLPD

•

Identified that well MeOH bullheading to Subsurface Safety Valve (SSSV) required only for long shut-ins (> 2 days)

•

Revealed that blowdown unsuccessful for watercuts 50% and higher

•

Illustrated that success of gas lift assist blowdown is not guaranteed

•

Developed dual strategy for lengthy shutdowns: primary blowdown and secondary oil circulation


OPRM-2003-0302D

Page 5 of 89

30-April-2006


Unrestricted

2.0

ITEM OVERVIEW AND SPECIFICATIONS

2.1

Introduction Bonga is a deepwater Nigerian oil prospect in Block OPL 212 in 1000m water depth, operated by Shell Nigeria Exploration and Production Company Limited in a joint venture with Esso (20%), Elf (12.5%) and Agip (12.5%). Bonga will be developed as a subsea network, with 1.9 to 9.2km tiebacks to a permanently moored Floating Production, Storage and Offloading vessel (FPSO). Anticipated peak production rates are 225MBOPD oil, 170MMscfd gas (including recycled riser gas lift) and 100MBWPD produced water (refer to production function in Figure 1.1). Reservoir pressure will be maintained via 16 subsea waterflood wells with a 300MBWPD total water injection capacity. Produced oil will be stored on the FPSO (2MMBO storage capacity) for tanker offloading, while Bonga gas will be exported 90km via a 16in pipeline to Riser Platform A of the Offshore Gas Gathering System (OGGS), which feeds the Bonny Nigerian Liquefied Natural Gas Plant (NLNG) plant. The initial phase Bonga Field layout (refer to Figure 1.2) consists of four reservoirs (690, 702, 710/740, 803; roughly one half of reserves within 702) and 20 subsea production wells. Production wells contain a subsea tree (enabling surface controlled isolation valves, production choke and chemical injection valves) connected via short well jumpers to five subsea production manifolds. The subsea wells are produced through four pairs of piggable dual flowlines (three 10in pairs and one 12in pair), with pipe-in-pipe flowlines and externally insulated steel catenary risers. Each flowline is connected to a dedicated gas lift riser delivering up to 25MMscfd riser base gas lift. Riser base gas lift is critical for several Bonga operations, enabling: •

Riser unloading during start-up and blowdown

•

Severe slug suppression

•

Production enhancement

As a subsea production system of unprecedented complexity in a new deepwater operating environment, Bonga entails several key flow assurance and systems engineering challenges. Additionally, unlike typical Shell Deepwater Gulf of Mexico (GoM) projects, independent EPIC (Engineer, Procure, Install and Construct) Contractors are responsible for the detailed design, construction and installation of all Bonga facilities. However, Shell has chosen to retain ‘ownership’ of flow assurance via design specifications in each EPIC contract, based on flow assurance analysis performed in-house within the Bonga Systems Engineering Team (BSET). Thus, the completeness of in-house analysis and the communication of results with (and among) contractors (facilitated by BSET) are key success factors for Bonga. The principal objective of this report is to validate the Bonga conceptual design with respect to Shell Deepwater Flow Assurance Guidelines (Ellison and Kushner, 1998), and to outline the Management of Change (MoC) identified by this analysis.


OPRM-2003-0302D

Page 6 of 89

30-April-2006


2.2

Unrestricted

Reservoir Fluid The fluid composition and properties for each Bonga reservoir (690, 702, 710/740 and 803) are summarised in Appendix 1A Table 1A.1. The reservoir fluids exhibit the following variability in properties: •

Bubble point at reservoir temperature (145 to 190°F) = 3335 to 5015psia

•

Stock tank oil gravity = 29 to 33° API

•

Gas/oil ratio = 550 to 1200 SCF/STB (single-stage flash)

Unless otherwise noted, simulations here are based on compositional Pressure/Volume/Temperature (PVT) models tuned to match the properties of the dominant 702 reservoir. All transient simulations in OLGA are based on the phase diagram shown in Figure 1.46, calculated for the 702 reservoir fluid. For purposes of analysis, the oil gravity and gas: oil ratio (not to be confused with the gas:liquid ratio) are relatively constant over the field life at 600SCF/STB. Based on the production forecast (refer to Figure 1.1), watercuts of 0%, 50%, and 80% are assumed for early, mid and late-life scenarios, respectively. Hydrate dissociation curves (pressure (HDP) vs temperature (HDT)) for the 702 and 803 fluids are presented in Appendix A, calculated using MULTIFLASH (Mehta, 1998). The expected salinity is that of the seawater (due to significant waterflood), ie approximately 3wt % salt. As a result of this low salinity, compared to the typical 15% salinity of subsea GoM fields, hydrate management for Bonga is particularly challenging (ie HDT approximately 10°F higher). For conservatism, the hydrate dissociation conditions of the 803 fluid with 0% salinity (refer to Figure 1.48) are used as a worst-case for all flowline analysis in this report. At the minimum seabed temperature (40°F), this translates to a blowdown target pressure of HDP = 150psia. For subsea facilities (tree, well jumper and manifold) a target hydrate temperature of HDT = 74°F is used for the 702 wells considered here, corresponding to the maximum design shut-in pressure (4600psia).

2.3

Wellbore Characteristics The November 1999 well design basis (Appendix 1B) indicates the following range of wellbore parameters: •

702 Wells –

Water depth = 990 to 1105m

–

Measured depth = 1770 to 2315m below mud line

–

True vertical depth = 1360 to 1730m below mud line

–

Tubing = 4.89in ID x 5.5in OD or 5.92in ID x 6.625in OD: bare tubing

–

Reservoir pressure (average) = 2520 to 4200psia

–

Reservoir temperature = 128 to 162°F

–

Productivity index (average) = 20 to 110BLPD/psia


OPRM-2003-0302D

Page 7 of 89

30-April-2006


•

•

•

Unrestricted

690 Wells –


–


–


–

Tubing = 4.89in ID x 5.5in OD or 5.92in ID x 6.625in OD: bare tubing

–


–


–

Productivity index (average) = 7 to 14 BLPD/psia

710 Wells –


–


–


–

Tubing = 5.92in ID x 6.625in OD: bare tubing

–


–


–


803 Wells –


–


–


–


–


–


–


For conceptual design evaluation, we focus here on wells 702p7 (coldest) and 702p4 (hottest), which represent the flowing wellhead temperature extremes for the dominant 702 reservoir. Note: Results here effectively bracket the thermal-hydraulic performance of all producing wells, which will be analysed individually as part of future detailed design and operability analysis.


OPRM-2003-0302D

Page 8 of 89

30-April-2006


Unrestricted

The well casing and annulus fluid design summarised in Figure 1.3 (from Van Gisbergen, 1999) is used for all transient and steady-state thermal wellbore analysis. A linear geothermal temperature gradient (from mid-perfs to mudline) is specified for the ambient formation temperature. The well specifications analysed herein are summarised as follows: •

•

2.4

702p7 (coldest) –

Measured depth = 1870m below mud line

–

True vertical depth = 1380m below mud line

–


–

Reservoir pressure = 3200psia (early life) to 2200psia (late life)

–

Reservoir temperature = 128°F

–

Productivity index (average) = 30BLPD/psia

–

Watercut = 0% (early life) to 80% (late life)

702p4 (hottest) –

Measured depth = 2280m below mudline

–

True vertical depth = 1760m below mud line

–


–

Reservoir pressure = 4800psia (early life) to 3600psia (late life)

–

Reservoir temperature = 162°F

–

Productivity index (average) = 80BLPD/psia

–

Watercut = 0% (early life) to 80% (late life)

Subsea Flowline Details The conceptual design evaluation presented here is based on the 10in west side and 12in east side flowline topographies (refer to Figure 1.4), which capture the essential terrain features on either side of the FPSO. Note: The significant difference in offset distance between the East (3.9 and 5.8 miles) and West (1.2 and 1.5 miles) flowlines (refer to Appendix 1C). The riser gas lift injection is located 1150m horizontal distance upstream from the FPSO, at the flowline/riser connection (refer to Figure 1.4). In Appendix 1C, further flowline details are summarised, including individual flowline topographies, the catenary riser profile and profiles of (ambient) sea temperature and current. With reference to the field layout in Figure 1.2, all production flowlines are of 10in nominal diameter, with the exception of the 12in east side flowlines PFL 3/4/5/6 (the ‘East-East’ flowline). As illustrated in Figure 1.5, pipe-in-pipe insulation is used for all production flowlines, with an insulating value of UOD=2.0 W/m2-C (0.352 Btu/hr-ft2-F) or better. Note: In Figure 1.5, U values as low as 1.4W/m2-C can be attained by filling the entire annulus space with foam (as recommended here based on cooldown considerations). Based on both steady-state and cooldown performance, a 4in carazite (or equivalent) insulation has been specified for all production risers (refer to Figure 1.5).


OPRM-2003-0302D

Page 9 of 89

30-April-2006


2.5

Unrestricted

Operating Conditions and Constraints As a tieback comprised of numerous subsea wells and flowlines, Bonga entails several key flow assurance constraints on system design and operation, including: •

12-hour minimum cooldown time for flowline and riser

•

8-hour minimum cooldown time for wellbore, subsea tree, well jumper and manifold

•

Target minimum turndown rate of 10MBLPD per well and per flowline

•

Target blowdown pressure of 145psia

•

Minimum boarding temperature of 98°F (@ maximum production)

•

Maximum boarding temperature of 153°F

•

Separator pressure = (300, 150, 150) psia for (early, mid, late) field life

In addition to general Shell subsea operating guidelines:

2.6

•

Operation outside of stable hydrate region at all times, with chemical inhibition otherwise

•

No wax deposition in the wellbore

Objectives The principal objective of this report is to evaluate the conceptual design of the Bonga subsea system with respect to flow assurance, topsides and subsea system constraints, and operability. The main focus here is on hydrate prevention during all expected operating scenarios; detailed wax and asphaltene analysis appears separately in Ratulowski et al, 1999. In particular, detailed thermal hydraulic multiphase flow simulations (described in Paragraph 2.7) are used to analyse the following critical flow assurance issues: •

Well cold start-up

•

Well safe condition time

•

Steady-state flowing wellhead temperature

•

Well cooldown

•

Steady-state arrival temperature

•

Flowline cooldown

•

Flowline blowdown

Riser gas lift requirements: •

Slug suppression

•

Riser unloading

•

Injection temperature

For limitations identified in the conceptual design, possible design improvements are suggested and evaluated. Preliminary operating logic charts, consistent with this conceptual design analysis, are also developed.


OPRM-2003-0302D

Page 10 of 89

30-April-2006


Unrestricted

2.7

Computational Approach

2.7.1

Steady-state and Transient Wellbore For all wellbore analysis, the WELLTEMP software developed by ENERTECH is used. WELLTEMP fully models wellbore flow using Shell two-phase flow models, and both conductive and convective heat transfer in casing annuli are explicitly modelled. Heat transfer in the surrounding formation (eg 50ft radius) is simulated directly using finite-difference methods, coupled to finite-volume (ie conservation form) representations of multiphase flow in the well tubing and heat transfer in the casing strings. Refined wellbore pressure modelling is performed using the Shell NEWPRS software, which is also based on the Shell GZM two-phase flow model (described below) and allows bubble point specification.

2.7.2

Steady-state Flowline The process simulation software HYSYS, marketed by HYPROTECH, is used for steady-state predictions of thermal-hydraulic multiphase flow in the Angus flowlines. Extensive testing has shown that HYSYS PVT thermodynamic modelling is superior to other marketed packages, and the Shell GZM two-phase flow model (Zabaras, 1987) is incorporated into HYSYS for proprietary use by Shell. The GZM model uses Taitel and Dukler phase transition criteria, combined with empirical correlations for interphase friction, entrainment, holdup and wall-wetted fraction.

2.7.3

Flowline/Riser Cooldown Flowline cooldown results are obtained with the Shell COOLDOWN software (Wasden, 1995), which solves the full transient heat conduction equation for axisymmetric, radial heat transfer, including multiple insulation layers. Axial heat conduction within the fluid and pipe are neglected, since axial temperature gradients (ie heat fluxes) are generally orders of magnitude smaller than radial gradients. Average thermophysical properties of the fluid are obtained with HYSYS, and selected cases are validated using full transient thermal-fluid simulations (OLGA).

2.7.4

Transient Flowline To model time-dependent two-phase flow in the subsea flowlines, the OLGA software marketed by SCANDPOWER is used. OLGA solves a set of six coupled first-order, non-linear, one-dimensional partial differential equations: three continuity equations (gas, liquid film and liquid droplets), two momentum equations (liquid film, and a combined gas and liquid droplet field) and a mixture energy equation. For numerical solution, a staggered mesh finite difference method is used for spatial discretisation, with semi-implicit time stepping. The momentum equations are mechanistic in nature, requiring correlations of friction factor, wetted perimeter, entrainment, and deposition, along with flow regime specification based on a minimum-slip concept (ie regime with minimum slip velocity chosen). Although the total fluid composition is constant within a given pipeline branch, the liquid and gas compositions (thus, liquid and gas physical properties) can change continuously, eg during a flash.


OPRM-2003-0302D

Page 11 of 89

30-April-2006


Unrestricted

Transient mass transfer between phases is modelled using a Taylor-series expansion of the equilibrium gas mass fraction in terms of pressure and temperature. Non-equilibrium gas fractions (eg gas pockets above the bubble point in shut-in wellbores) may be specified as initial conditions and will subsequently vary according to the mass transfer rate. Simulations fully account for important elements such as flowline topography, multi-layered pipe insulations (including wellbore casings), heat storage in pipe walls and buried earth, and time-dependent valve openings, boundary conditions, and source flowrates, among others. Additionally, the proximity of instantaneous pressure and temperature values to hydrate dissociation conditions can be tracked both in space and time. For further details of the OLGA modelling approach and transient flow assurance applications, refer to Bendiksen et al (1991) and Schoppa et al (1998).


OPRM-2003-0302D

Page 12 of 89

30-April-2006


Unrestricted

Figure 1.1 – Production Forecast for Bonga Phase I Development (refer to Bonga Basis of Design)


OPRM-2003-0302D

Page 13 of 89

30-April-2006


Unrestricted

Figure 1.2 – Bonga Subsea Field Layout


OPRM-2003-0302D

Page 14 of 89

30-April-2006


Unrestricted

0.50 psi/ft water-based

0.54 psi/ft oil-based

0.52 brine

psi/ft

Figure 1.3 – Bonga Production Well Design, Used for All Thermal-hydraulic Analysis


OPRM-2003-0302D

Page 15 of 89

30-April-2006


Unrestricted

0

-200

Elevation (m)

-400

-600

-800 Gas Lift

-1000 -1100 0

500

1500

1000

2000

2500

Length (m)

0 -100 -200 -300

Elevation (m)

-400 -500 -600 -700 -800 Gas Lift -900 -1000 -1100 0

1000

2000

3000

4000

5000

6000

7000

8000

Length (m) OPRM20030302D_001.ai

Figure 1.4 – Production Flowline Topography for (a) 10in West-side Flowlines and (b) 12in East-side Flowlines


OPRM-2003-0302D

Page 16 of 89

30-April-2006


Unrestricted

10in Production Flowline

10in Production Riser 10.75in OD x 1.0in Steel

Flowline 10.75in OD x 0.937in Steel PU Foam

4in Carazite (or equivalent)

Air Gap 14in OD x 0.563in Steel

12in Production Flowline

12in Production Riser 12.75in OD x 1.063in Steel

12.75in OD x 1.126in Steel

PU Foam

Air Gap

4in Carazite (or equivalent)

16in OD x 0.625in Steel OPRM20030302A_011.ai

Figure 1.5 – Insulation Systems for 10in and 12in Pipe-in-pipe Flowlines (Left Panel), and Steel Catenary Risers (Right Panel)


OPRM-2003-0302D

Page 17 of 89

30-April-2006


3.0

Unrestricted

COLD WELL START-UP: HYDRATE PREVENTION STRATEGIES For flow assurance in the subsea wells, the hottest (702p4 – horizontal) and coldest (702p7 – conventional) 702 wells (described in Paragraph 2.3 and Appendix 1B) are evaluated with regard to: (i) cold-earth start-up, (ii) safe condition requirements and cooldown performance, and (iii) steady-state flowing wellhead temperature. All wellbore thermal analysis is performed using WELLTEMP, for the casing designs in Figure 1.3 and a linear geothermal temperature profile, from the reservoir temperature to 40°F at the wellhead. Production rates over the range 2.5 to 40MBLPD are considered for early, mid, and late-life conditions (0%, 50%, 80% watercut). A sample WELLTEMP input file, summaries of simulation cases and results appear in Appendix 1B Tables 1B.1 to 1B.5. For wellbore transients, the relevant terminology illustrated in Figure 1.6 is defined as follows:

3.1

•

Cold Earth Start-up – Well start-up in which the wellbore, tree and well jumper are initially at ambient temperature

•

Well Warm-up Time – Elapsed time upon start-up required for the Flowing Wellhead Temperature (FWHT) to exceed HDT (HDT = 74°F at well shut-in pressure)

•

Safe Condition (SC) Temperature – FWHT which must be reached after start-up such that 8 hours of cooldown time is available

•

Safe Condition Time – Elapsed time upon start-up for safe condition temperature to be reached

Cold Earth Well Start-up A critical aspect of well flow assurance for Bonga is cold earth well start-up, in which the wellbore and surrounding formation are at ambient (geothermal) temperature, either at initial start-up or after an extended shut-in (ie longer than 1 week). In contrast to the common use of Vacuum Insulated Tubing (VIT) to provide fast warm-up of deeper subsea wells in the GoM, bare tubing is used for all Bonga wells. Although the relatively shallow depth of the Bonga wells makes bare tubing viable, careful evaluation is required of the relative hydrate risk at start-up. As a worst case, the start-up of the coldest well (702p7) is considered first for early life conditions. As shown in Figure 1.7, the well warm-up time to HDT = 74°F is moderately lengthy, particularly at low start-up rates. Note: Although rapid well ramp-ups are anticipated for Bonga (eg 10MBLPD within 1/2 hour), a more moderate start-up rate (eg 5MBPLD average) is analysed as a design case. At a start-up rate of 5MBLPD, the wellhead region is temporarily in the hydrate region for 80 minutes (refer to Figure 1.7). Note: As a general guideline, based on operating experience and preliminary hydrate kinetics research (which must be used carefully), a hydrate exposure longer than 60 minutes with greater than 10°F, subcooling is considered an unacceptable risk for subsea wells (with significant cost of intervention/ remediation).


OPRM-2003-0302D

Page 18 of 89

30-April-2006


Unrestricted

As a possible operational solution, bullheading of MeOH into the entire wellbore prior to start-up significantly reduces the hydrate risk, as reflected by the MeOH residence time (time required for one well pass) in Figure 1.7 (eg hydrate exposure time reduced from 80 minutes to 40 minutes at 5MBLPD). Notes: (1)

Although the current well and subsea system design permit bullheading of MeOH past the SSV, it is undesirable to expose the bottomhole hardware to MeOH. Thus, precise operating and MeOH monitoring procedures will be required for whole-well bullheading.

(2)

The MeOH volumes required: 150bbl for 4.9in ID well tubing and 250bbl for 5.9in ID.

In summary, the well warm-up times for cold earth start-up do pose a hydrate concern, but the risk is relatively small at expected start-up rates and can be reduced significantly by whole-well MeOH bullheading, if necessary (yielding hydrate exposure times comparable to currently operating GoM subsea wells). The decision whether to bullhead MeOH into the entire wellbore or only to the SSSV will be made on a well-by-well basis, as a part of ongoing operability and hydrate kinetics analysis (conducted in-house). In summary, the wellbore hydrate exposure times for each bullheading option are: •

0% watercut: Bullheading Option

•

Hydrate Exposure (5MBLPD)

No MeOH in well

80 minutes

MeOH to SSSV (50 to 75bbl)

65 minutes

MeOH to perfs (150 to 250bbl)

40 minutes

50% watercut: Bullheading Option

Hydrate Exposure (5MBLPD)

No MeOH in well

50 minutes

MeOH to SSSV (50 to 75bbl)

35 minutes

MeOH to perfs (150 to 250bbl)

10 minutes

At higher watercuts, an additional issue that arises is the maximum start-up rate for which the resulting water production is treatable by MeOH delivery capacity (ie 18gpm per subsea tree). That is, whereas faster well start-up is desirable from a wellbore hydrate viewpoint (refer to Figure 1.7), at significant watercuts (50 to 80%), the MeOH rate becomes insufficient to protect the tree and well jumper. The treatable liquid rate at 18gpm MeOH injection is illustrated in Figure 1.8 as a function of watercut (based on MULTIFLASH calculations, Mehta, 1999). For the anticipated average start-up rate of 5MBLPD, Figure 1.8 indicates a watercut limit of ~20% for sufficient MeOH protection of the tree and jumper.


OPRM-2003-0302D

Page 19 of 89

30-April-2006


Unrestricted

At 50 to 80% watercut, an MeOH injection rate of 54 to 90gpm would be required, which is infeasible from an umbilical delivery viewpoint (leading also to significant MeOH production contamination). Thus, an additional factor to be considered during future operability analysis is whether or not to constrain start-up rates at high watercut, to protect the tree and jumper at the expense of the wellbore. Preliminary operability analysis suggests a possible dual start-up strategy: (1) Low watercut (below 20%): constrained start-up rate to yield MeOH-treatable (at 18gpm) water rates, with full hydrate protection of the tree and jumper. (2) High watercut (above 20%): unconstrained (fast) start-up rate (ie limited only to prevent well/reservoir impairment) to ‘outrun’ the finite time hydrate kinetics in the wellbore, tree and jumper. Note: For the fortunate result in Figure 1.9, the well warm-up time to HDT is much faster at higher watercut (due to higher heat capacity of water), which reduces the relative hydrate risk of the high watercut strategy. Further development and testing of low dosage hydrate inhibitors will also be undertaken to further reduce the hydrate risk in the tree and jumper, for watercuts up to 80% and subcoolings up to 30°F (Mehta, 2001).

3.2

Well Safe Condition Analysis The concept of a well safe condition is motivated by the risk of hydrate formation in the wellbore in the event of an aborted start-up. In this way, operations staff can determine whether immediate MeOH treatment is required in the event of an aborted start-up. Before well safe condition has been reached during a well start-up, immediate operator action (eg well bullheading) is required before safe condition (without any no-touch time), in contrast to the full 8-hour cooldown period available after safe condition. Note: The SC definitions are based on 8 hours of required cooldown time for the wellbore, tree and well jumper (eg 3 hours no-touch time + 5 hours MeOH treatment time). During well start-up, hydrate inhibition via MeOH injection at the tree is generally recommended until the SC time is reached (Ellison and Kushner, 1998). Note: If MeOH usage/storage is a concern, special operating guidelines may be developed to treat until 5 hours of cooldown are available (the MeOH treatment time), or even only to the (shorter) warm-up time to HDT. These less conservative procedures are based on the idea that in an aborted start-up of a single well, no-touch time is unnecessary and only the well being started must be treated immediately. For Bonga, the condition for termination of MeOH injection at the tree will be determined as part of future operability analysis. For the coldest (702p7) and hottest (702p4) 702 wells, and early-life conditions, the wellbore SC times are shown in Figure 1.10 as a function of the average rate during start-up. For a moderate start-up rate of 5MBOPD, these results bracket the SC times for all 702 wells to between 5 to 10 hours, translating to 130 to 260bbls MeOH volume per well, for an 18gpm injection rate. At higher watercuts (eg greater than 50%), the SC time is significantly reduced (ie faster warm-up due to higher heat capacity of water), as illustrated in Figure 1.11 for the 702p7 well.


OPRM-2003-0302D

Page 20 of 89

30-April-2006


Unrestricted

The corresponding SC analysis for the tree and well jumper is based on the specification (for the subsea contractor) that these components must provide at least 12 hours of cooldown from 120°F to 73°F. Note: This tree/jumper cooldown period is longer than the 8-hour cooldown allotted to the well tubing, to allow an additional operational margin for the field complexity of Bonga. The SC temperature for the tree and well jumper is 120°F, for which the corresponding SC time is shown in Figure 1.12. Note: The steady-state FWHT for well 702p7 does not reach 120°F, so its SC temperature in Figure 1.12 is modified to 110°F for purposes of comparison (an exception for 702p7 to be accounted for in operability analysis). Owing to the rather lengthy tree/jumper SC times (eg greater than 10 hours at 5MBOPD), operating procedures for less than 12 hours of cooldown (ie more immediate action upon aborted start-up) may be necessary in lieu of MeOH injection until the tree/jumper SC time is reached. Note: For treatment until a 12-hour SC, production at higher watercuts would have to be constrained for several hours to maintain a MeOH-treatable water rate, with the additional cost of deferred production.

3.3

Flowline Hot-oiling Flowline preheating via hot-oiling is an effective means to prevent hydrate risk in the flowlines during cold start-up. Topsides hot-oiling facilities provide two oil circulation pumps capable of delivering 50MBOPD each, with heating of the (dry) supply oil to 150°F. The maximum oil supply pressures, based on 5mph circulation of an initially ambient flowline, are calculated as 520psia for the west-side flowlines and 770psia for the east-side flowlines (for a 250psia flowline outlet pressure). In Figure 1.13, the hot-oiling performance is shown for 50MBOPD circulation of 150°F source oil. For the west-side flowlines, a return temperature of 140°F is attained in 3 hours, with 130°F reached in 7 hours for the east-side flowlines. Preliminary start-up analysis indicates that hot-oiling provides at least 6 hours of cooldown (reaction) time in the event of an aborted start-up, provided that a steady state is established within 8 hours after hot-oiling.


OPRM-2003-0302D

Page 21 of 89

30-April-2006


Unrestricted

Wellhead Temperature (ºF)

120 SC Temperature 100

80 HDT Warm-up Time

60

SC Time

40 0

2

4

8

6

10

12

14

Time (hours) OPRM20030302D_002.ai

Figure 1.6 – Definition of Well Start-up Terminology

300 Time for one well pass Time to HDT Time After Start-up (minutes)

250

200

Hydrate Exposure Time

150

Wellbore Outside Hydrate Region

100

50

0 0

2000

4000

6000

8000

10,000

Average Start-up Rate (BLPD) OPRM20030302D_003.ai

Figure 1.7 – Wellhead Warm-up Time to HDT, for Cold Earth Start-up of the Field’s Coldest Well (702p7) at 0% Watercut


OPRM-2003-0302D

Page 22 of 89

30-April-2006


Unrestricted

Treatable Liquid Rate (oil + water) (BLPD)

10,000

8000

6000

4000

2000

0 0

20

40

60

80

100

Watercut (%) OPRM20030302D_004.ai

Figure 1.8 – Treatable Liquid Rate for 18gpm MeOH Injection (Mehta, 1999) 240

Time to Reach HDT (minutes)

Time for one well pass 50% wc 0%wc 180

120 Wellbore Outside Hydrate Region 60

0 0

2000

4000

6000

8000

10,000


Figure 1.9 – Well Warm-up Time of 702p7: Dependence on Water Cut


OPRM-2003-0302D

Page 23 of 89

30-April-2006


Unrestricted

12 702p4 horiz 702p7

SC Time (hours): Guarantee 8-hour Cooldown

10

8

6

4

2

0 4

6

8

10

12

14

16

Average Start-up Rate (MBOPD) OPRM20030302D_006.ai

Figure 1.10 – Safe Condition Time for 8-hour Wellbore Cooldown (refer to Figure 1.6 for definition)

25 50% wc 0% wc Well SC Time (hours): Guarantee 8-hour Cooldown

20

15

10

5

0 0

5000

10,000

15,000

20,000


Figure 1.11 – Influence of Water Cut on Well Safe Condition Time for 702p7


OPRM-2003-0302D

Page 24 of 89

30-April-2006


Unrestricted

20

SC Time (hours): Guarantee 8-hour Cooldown

702p7 702p4 horiz 15

10

5

0 0

5

10

15

20

25

30

Average Start-up Rate (MBOPD) OPRM20030302D_008.ai

Figure 1.12 – Safe Condition Time for 12-hour Cooldown of Tree/Jumper/Manifold, Based on Time for Wellhead Temperature to Reach 120°F

160 West Arrival Temperature (ºF)

140 East 120

100

80

60

40 0

2

4

6

8

10


Figure 1.13 – Hot-oiling Performance: Return Temperature for 50MBOPD Circulation of 150°F Source Oil


OPRM-2003-0302D

Page 25 of 89

30-April-2006


4.0

Unrestricted

STEADY-STATE PRODUCTION Steady-state system modelling typically focuses on the hydraulic capacity of the well/flowline system in delivery of the production forecast, which for Bonga has been addressed extensively using PIPESIM (refer to Granherne, 1998; Hartwik and Lindsey, 2000). Additionally, several key aspects of flow assurance are linked to steady-state system performance, including:

4.1

•

Arrival temperatures in relation to topsides oil heating capacity

•

Riser base temperatures governing available flowline/riser cooldown time

•

Slugging

•

Production fluid cooling by riser base gas lift

Steady-state Thermal Performance: Wellbore and Flowline Since prior wellbore analysis (Granherne, 1998) has been based on the approximation of constant U value for the wellbore (ie U=2.0Btu/hr-ft2-F), the more rigorous thermal modelling in WELLTEMP is used here to obtain refined FWHT predictions. The range of FHWT predicted for the six initial-life production wells is shown in Figure 1.14, along with 702p7, the field’s coldest well (which fortunately produces through the short-offset West flowline PFL11). At the minimum well production rate of 10MBLPD specified in the Basis of Design (BoD), the FWHT lies in the range 115 to 165°F. The lower end of this FWHT range is noticeably colder than that typical of (deeper) GoM subsea oil wells, which should be accounted for in building upon GoM subsea operating experience. Note: Production rates lower than 10MBLPD (eg as low as 5MBLPD) are also operable from a thermal point of view, although well stability must also be accounted for in specifying the minimum turndown rate. Later in field life, the FWHT increases slightly for all flowrates (eg by 5°F for 80% watercut), due to the enhanced thermal heat capacity of water (which may be offset to some degree by reservoir cooling due to waterflood). With regard to the thermal performance of the coupled well/flowline system, there are three key constraints which govern the minimum operable arrival temperature for steady-state production: •

Flowline operation outside of hydrate regime: arrival T > 60°F

•

Minimum 12-hour cooldown of riser/flowline: arrival T > 90°F

•

Sufficient topsides oil temperature for available waste heat capacity at high production rates (~200MBOPD): arrival T > 98°F

In Figure 1.15, the arrival temperatures for the six initial-life wells/flowlines are shown as a function of production rate. Note: Each initial-life well produces through a dedicated flowline, with an initially inactive West flowline pair PFL8/9. For all pipe-in-pipe flowlines, an overall heat transfer co-efficient of Uod = 2W/m2-C is used, corresponding to a polyurethane foam thickness of ~0.6in (leaving ~0.4in of air gap, refer to Figure 1.4).


OPRM-2003-0302D

Page 26 of 89

30-April-2006


Unrestricted

Note: Cooldown requirements will likely require an entirely foam-filled gap (ie MoC 59, discussed in Paragraph 5.1), for which the arrival temperatures will be slightly higher than those reported here (particularly at low production rates). The 12-hour cooldown constraint (detailed analysis presented in Paragraph 5.1) translates to a minimum turndown production rate of approximately 10MBOPD for the four east-side flowlines. Although slightly lower production rates may be possible for special operations which are manageable with less than 12 hours of cooldown, production rates less than 5MBOPD are inoperable due to onset of flowing conditions in the hydrate regime. With regard to the topsides heat requirement at high production rates, the cumulative oil temperature for all six initial-life wells/flowlines (with equal production from each; refer to Figure 1.16) indicates that the 98°F constraint is met even at turndown conditions (ie >50MBOPD), with a 20°F margin in arrival temperature at flowrates greater than 150MBOPD. Thus, the available topsides waste heat for oil heating is not of concern at initial field life, which serves as the worst case since oil production will subsequently decrease (accompanied by increasing water production).

4.2

Terrain-induced (Severe) Slugging The phenomenon of severe slugging induced by undulations in flowline terrain is predicted to be significant at Bonga in the absence of mitigating control, due to: •

Significant downhill flow near the riser base for (~30m elevation drop, refer to Figure 1.49)

•

Production of high watercuts (80 to 90%)

•

Large diameter flowlines (10in to 12in)

•

Significant water depth (~1000m)

east-side flowlines

Note: The distinction between shorter hydrodynamic slugs (up to ~50 diameters in length) in locally horizontal or uphill flow and longer terrain slugs (proportional to the length of downhill flow), which are more problematic for topsides facilities and process control. That terrain slugging is outside the scope of steady-state simulations, which cannot capture at all the adverse effects of higher well backpressure and order-of-magnitude fluctuations in liquid production rate. In the following, Olga2000 is used to define the terrain slugging operability envelope, including detailed assessment of slug suppression via riser gas lift. For terrain slugging to occur in a flowline/riser system, three necessary conditions must be satisfied simultaneously (Vreenegoor, 1999): (1) The Pots slugging number less than order unity in the flowline:

π ss =

zRT m& g < O(1) αLg m& l

(2) The densimetric Froude number less than order unity in the riser:

Fr = U sg

ρg (ρ l − ρ g ) gD

< O(1)

(3) Stratified flow pattern in the riser base region of the flowline. Section 1 Dynamic Flow Assurance Analysis

OPRM-2003-0302D

Page 27 of 89

30-April-2006


Unrestricted

Physically, the slugging number condition: •

Reflects the fact that sufficient gas compressibility (‘capacitance’) is required for slugs not to be expelled from the flowline. The Froude number condition

•

Indicates that unstable riser flow (ie liquid surging and fallback in the riser) is necessary to initiate a flow blockage at the riser base

•

Enables growth of the liquid slug

For representative Bonga conditions at 10MBLPD and 50% watercut, the slugging numbers for each flowline are: • East 12in: πss = 0.3 • East 10in: πss = 0.2 • West 10in: πss = 0.7 Furthermore, the Froude number (without gas lift) is on the order of 0.05 and stratified flow is predominant in the downhill flow regions near the riser base. Thus, based on this simple conceptual analysis, severe terrain slugging is predicted at Bonga without riser gas lift, particularly for the east-side flowlines. Although it has not yet been field-proven for large-diameter deepwater risers, a potentially effective slug control technique involves gas injection at or near the riser base. With reference to the necessary conditions for terrain slugging, gas lift can eliminate the riser instability required for slug initiation (ie Froude number greater than order unity). For the 12in east Bonga flowline, riser gas lift increases the Froude number from order 0.05 to order 1 (refer to Figure 1.17), and hence is expected to be effective in slug suppression. In the following, Olga2000 computations are used to investigate in detail the effectiveness of riser gas lift in suppressing terrain slugging. In Figures 1.18 to 1.20, the gas lift required to suppress terrain slugging is shown as a function of liquid production rate. In Olga2000, terrain slugging can be isolated from smaller, less problematic hydrodynamic slugs (ie by switching Slugtracking off), to yield a sharp transition from terrain to hydrodynamic slugging. For all west-side 10in flowlines (refer to Figure 1.18), no riser gas lift is required at the minimum turndown rate of 10MBLPD, even at 80% watercut. This result is in contrast to prior studies (Granherne, 1998), which indicated that 5MMscfd gas lift was required, apparently due to inaccurate modelling of the riser-base topography. Note: Slugging may be suppressed at turndown rates as low as 5MBLPD, by gas lift rates up to 10MMscfd (refer to Figure 1.18). For the 10in east-side flowlines, 5 to 10MMscfd gas lift is required to eliminate slugging for the minimum rate of 10MBLPD at 0 to 80% watercut (refer to Figure 1.19). Due to the more adverse east-side topography, the gas lift requirement for flowrates lower than 10MBLPD is much more significant for the 10in east flowlines, compared to the 10in west results. Thus, even at the gas lift capacity of 25MMscfd per flowline, signficant slugging will occur for the east-side flowlines for turndown rates lower than approximately 8MBLPD (refer to Figure 1.19).


OPRM-2003-0302D

Page 28 of 89

30-April-2006


Unrestricted

For the 12in east flowlines, terrain slugging cannot be totally eliminated by feasible riser gas lift rates. Hence, for these 12in flowlines, the gas lift required to reduce the terrain slug size to <50bbl is shown in Figure 1.20. Due to the larger diameter, the gas lift requirements are more stringent compared to the 10in east-side results. In particular, gas lift approaching 20MMscfd is required at the minimum rate of 10MBLPD and 80% watercut (refer to Figure 1.20). In addition, marginal gas lift volumes are needed at higher production rates as well (eg ~5MMscfd at 20MBLPD), suggesting that gas lift (albeit at varying rates) may be frequently required for the 12in east flowlines, even very early in field life. The required gas lift volumes at the minimum anticipated turndown rate of 10MBLPD are summarised for all flowlines in Table 1.1.

Watercut

Minimum Stable Production Without Gas Lift

Gas Lift Required for 10MBLPD Production

10in West

0%

10MBLPD

0

PFL 8/9

50%

10MBLPD

0

PFL 11/12

80%

10MBLPD

0

10in East

0%

30MBLPD

5MMscfd

PFL 1/2

50%

35MBLPD

8MMscfd

80%

35MBLPD

10MMscfd

12in East

0%

30MBLPD

10MMscfd

PFL 3/4/5/6

50%

35MBLPD

17MMscfd

80%

40MBLPD

17MMscfd

Flowline

Note: The requirements for the 12in east flowline are based on a maximum slug volume of 50bbl, while results for other flowlines reflect complete terrain slug suppression. Table 1.1 – Riser Gas Lift Requirements for Terrain Slug Suppression To address severe slugging and the mitigating effect of riser gas lift in greater detail, an Olga Slugtracking Analysis was performed for the 12in east flowline, which exhibits the worst-case slugging (refer to Table 1.1). The Olga Slugtracking model captures the accumulation at the riser base of smaller hydrodynamic slugs generated in the flowline, which may enhance terrain slugging. Additionally, the effect of slugging on topsides vessel level control is modelled as an inlet separator attached to the flowline outlet, with the following specifications (in accord with the Bonga topsides conceptual design): •

132in diameter x 50ft seam-seam inlet separator (reflecting one of two available separators)

•

75MBLPD oil dump capacity (qualitative surge capacity for oil train)


OPRM-2003-0302D

Page 29 of 89

30-April-2006


•

Unrestricted

Oil dump valve Proportional Integral Derivative (PID) controller: –

Gain = 10

–

Integral const = 60 s (fast-acting level control)

For the 12in east flowline at 50% watercut, significant slug volumes on the order of 800bbl are predicted at turndown rates of 10 to 20MBLPD in the absence of riser gas lift (refer to Figure 1.21). As a consequence of these large slug volumes (without gas lift), separator level fluctuations of 10 to 20% occur at 10 to 20MBLPD (refer to Figure 1.22), magnitudes considered by Bonga topsides engineers to be unacceptably large for efficient separation and overall process control. Riser gas lift is seen to be particularly effective in reducing the slug volume, as manageable slug volumes of 50 to 100bbl and separator level fluctuations of 2 to 3% are attained with only 10MMscfd gas lift (Figures 1.21 and 1.22). Note: There is no benefit of gas lift rates higher than 10MMscfd, due to smaller (~50bbl) hydrodynamic slugs generated in the flowline and accelerated through the riser. In summary, modest gas lift rates on the order of 10MMscfd per flowline are predicted to manage severe slugging at Bonga to an acceptable degree, for a minimum turndown rate of 10MBLPD. Nevertheless, it is important to apply a significant design margin to these results, noting the modelling complexity and lack of field data for riser gas lift in deepwater systems. In particular, further experimental studies are clearly needed for gas lift in large-diameter risers, to confirm the effectiveness of gas lift in lifting riser liquid during flowing conditions (ie extending recent experimental analysis of gas lifting of a static liquid column; Zabaras and Schoppa, 2001). Additionally, the ‘resonance’ of multiphase flow in the flowline with topsides process flows (shown to intensify severe slugging in recent industry publications) is a detailed design issue beyond the scope of this report. Such coupling of subsea/topsides flows is the subject of extensive ongoing dynamic simulation work for Bonga (Duhon and Schoppa, 2001).

4.3

Riser Gas lift: Thermal Considerations Continuous riser base gas lift will be used during steady-state production for two expected operational scenarios: •

Slug suppression at turndown rates (particularly for the east-side flowlines)

•

Production enhancement at high watercut (eg as high as 80 to 90% at late life)

Thus, the thermal impact of (potentially cold) gas lift injection at the riser base is considered here in detail, with respect to available riser cooldown time and arrival temperature. The thermal limitations of a prior umbilical-based gas lift design are outlined, and an improved large-bore design is presented (ie MoC 16). In prior conceptual analysis (Granherne, 1998), it was incorrectly assumed that the injected gas would have negligible influence on the production fluid temperature. For example, for a gas lift injection rate of 25MMscfd (for terrain slug suppression) and 10MBLPD production, cold gas injected at 40°F reduces the production fluid temperature by 20°F throughout the riser (ie see temperature drop at gas lift location in Figure 1.23).


OPRM-2003-0302D

Page 30 of 89

30-April-2006


Unrestricted

With reference to Figure 1.15, such a 20°F temperature drop would significantly reduce available design margins in arrival temperature, with respect to waste-heat capacity and especially riser cooldown. For instance, to maintain 12 hours of available cooldown time, the minimum flowline turndown rate would have to be increased to 15 to 20MBLPD (refer to Figure 1.15). In turn, such a reduction in the subsea operability envelope would obviously have a detrimental impact on the overall production system availability. As shown in Figure 1.23, the reduction in fluid temperature by gas lift can be limited to 5°F by increasing the gas-injection temperature at the mudline to 90°F. In the following, we analyse the effectiveness of such gas lift heating for two system designs:

4.4

•

Umbilicals with multiple small-diameter tubes

•

Integrated risers with a single large bore, dedicated to each flowline

Umbilical-based Design In a prior conceptual design (Granherne, 1999), the base-case design for gas lift consisted of four umbilicals connected to each flowline pair, containing 6 x 1.5in tubes for gas lift (ie three tubes dedicated to each flowline). The umbilical would contain no external insulation, other than its outer sheath, with an overall heat transfer coefficient estimated on the order of 10W/m2-C. Due to umbilical material constraints, the inlet umbilical temperature could not exceed 140°F. Under typical operating conditions, the gas lift source will be at the gas-export pressure of 2320psia (upstream of the topsides gas lift choke), with a production riser base pressure of approximately 400psia. In Figure 1.24, the gas injection temperature (at mudline) is shown for a gas flowrate of 6.7MMscfd per tube, corresponding to a total gas lift rate of 20MMscfd for 3 x 1.5in tubes per flowline. Interestingly, topsides heating of the gas lift stream to 140°F (maximum at the umbilical inlet) is of little benefit, as the gas-injection temperature is approximately 50°F for expected umbilical U values (U~10W/m2-C). Note: Even with improved umbilical insulation, the gas injection temperature target of 90°F cannot be met. As indicated by results for a perfectly insulated tube (ie 60°F injection at U = 0), the cold injection temperatures observed result from significant Joule-Thomson cooling. Due to their relatively small size, each umbilical tube exhibits a significant pressure drop at higher rates, resulting in choke-like cooling. Due to manufacturing limitations, the hydraulic capacity of this six-tube umbilical design could not be significantly increased to reduce this choking effect (Granherne, 1999). In light of the severe thermal implications (refer to Figure 1.23) of the 40 to 50°F gas lift temperatures observed with the umbilical design, an alternative design was deemed necessary in order to provide a 90°F injection temperature.

4.5

Large-bore Riser Design In response to the unacceptable degree of Joule-Thomson gas cooling revealed above for the umbilical design, an alternative design was developed, consisting of eight large-bore risers (ie one gas lift line per production riser) with integrated electric-hydraulic service lines and methanol injection tubes. The principal advantage of this design from a flow assurance perspective is the capability for a larger hydraulic capacity (ie larger flow area) to reduce the Joule-Thomson gas cooling.


OPRM-2003-0302D

Page 31 of 89

30-April-2006


Unrestricted

The thermal performance of the large-bore gas lift design is evaluated below with respect to the following constraints: •

Minimum injection temperature for 5 to 25MMscfd gas lift > 90°F

•

Inlet temperature to gas lift riser (downstream of topsides choke) < 160°F

•

Gas heater temperature (upstream of topsides choke) < 200°F

As illustrated in Figure 1.12, for a 25MMscfd gas lift rate, the riser diameter strongly influences the gas injection temperature, as a 3.5in riser produces a 15°F higher injection temperature compared to a 3in riser. This is due to the fact that for smaller diameters, less topsides choking is required (more pressure drop in riser) and the gas heater temperature must be reduced to satisfy the 160°F riser inlet temperature constraint. At the minimum gas lift rate of 5MMscfd, the riser insulation dominates the thermal performance, for which an insulating value of approximately U = 4W/m2-C is needed to attain the 90°F injection target (refer to Figure 1.26). This U value corresponds to a 2.5in carazite insulation thickness (or equivalent) applied externally to the gas lift riser. In summary, the recommended design parameters, serving as a base case to be optimised during detailed design, are a 3.5in ID central gas lift pipe with an effective U value of 4W/m2-C. As illustrated in Figure 1.27, this large-bore riser design satisfies all requirements for gas lift, providing a gas injection temperature of at least 90°F over the entire range of gas rates. In this design, topsides heating of the gas lift stream is an effective approach to prevent a significant gas lift cooling penalty on arrival temperature and riser cooldown. This analysis culminated in the preparation and acceptance of MoC 16, which specified the gas lift heating requirements and large-bore riser design described above.


OPRM-2003-0302D

Page 32 of 89

30-April-2006


Unrestricted

702p15/PF11

180 Flowing Wellhead Temperature (ºF)

803p1/PF12 170

702p4/PF1 690p1/PF2

160

702p9/PF3 702p10/PF6

150

702p7/PF11 140 130 120 110 0

5

10

15

20

25

30

35

40

Rate (MBOPD) OPRM20030302D_010.ai

Figure 1.14 – Flowing Wellhead Temperatures Calculated for Initial-life Wells and the Field’s Coldest Well (702p7) with 0% Water Cut

Arrival Temperature per Flowline (ºF)

160 702p15/PF11 803p1/PF12

140

702p4/PF1 690p1/PF2 120

702p9/PF3 702p10/PF6

100

80

60 0

10

20

30

40

50

60


Figure 1.15 – Arrival Temperatures Calculated for All Initial-life Wells with 0% Water Cut


OPRM-2003-0302D

Page 33 of 89

30-April-2006


Unrestricted

130

Bulk Arrival Temperature (ºF)

120

110

100

90

80

70 0

50

100

150

200

250

Cumulative Rate (MBOPD) OPRM20030302D_012.ai

Figure 1.16 – Cumulative Arrival Temperature for Initial-life Well Production, Relative to the 98°F Arrival Temperature Constraint for Waste Heat Capacity

1

Froude#

FR < 0(1): Riser instability and possible slugging

0.1

0.01 0

5

10

15

20

25

30

35

Riser Gas Lift (MMSCFD) OPRM20030302D_013.ai

Figure 1.17 – Influence of Riser Gas lift on Riser Froude Number, as a Means to Eliminate Riser Instability and Terrain Slugging Shown for the 12in East-side Risers


OPRM-2003-0302D

Page 34 of 89

30-April-2006


Unrestricted

20

Required Gas Lift (MMSCFD)

0%wc 50%wc 15

80%wc

10

5

0 0

10

20

30

40

Liquid Production Rate (MBLPD) OPRM20030302D_014.ai

Figure 1.18 – Riser Base Gas Lift Required for Complete Suppression of Terrain Slugging for 10in West-side Flowlines

40


0%wc 50%wc 30

80%wc

20

10

0 0

10

20

30

40


Figure 1.19 – Riser Base Gas Lift Required for Complete Suppression of Terrain Slugging for 10in East-side Flowlines


OPRM-2003-0302D

Page 35 of 89

30-April-2006


Unrestricted

30 0%wc Required Gas Lift (MMSCFD)

50%wc 80%wc 20

10

0 0

10

20

30

40


Figure 1.20 – Riser Base Gas Lift Required to Limit Terrain Slugging to Within 50bbl Slugs for 12in East-side Flowlines

800

Maximum Slug Volume (bbl)

10MBLPD 20MBLPD

600

40MBLPD

400

200

0 0

5

10

15

20

25

Gas Lift Rate (MMSCFD) OPRM20030302D_017.ai

Figure 1.21 – Slug Volumes Calculated for 12in East-side Flowlines and 50% Water Cut as a Function of Gas Lift Rate


OPRM-2003-0302D

Page 36 of 89

30-April-2006


Unrestricted

Maximum Separator Level Fluctuation (%)

20

10MBLPD 20MBLPD

15

40MBLPD

10

5

0 0

5

15

10

20

25

Gas Lift Rate (MMSCFD) OPRM20030302D_018.ai

Figure 1.22 – Separator Level Fluctuation Calculated for 12in East-side Flowlines and 50% Water Cut as a Function of Gas Lift Rate

130

120

110

ºF

100

90

80

70

60 0

1000

2000

3000

4000

5000

6000

7000

Horizontal Length (m) OPRM20030302D_019.ai

Figure 1.23 – Effect of Cold (40°F) Gas Lift Injection on Arrival Temperature for 10MBOPD Production and 25MMSCFD Gas Lift for Slug Suppression


OPRM-2003-0302D

Page 37 of 89

30-April-2006


Unrestricted

80 Tin = 120ºF

Gas Injection Temperature (ºF)

70

Tin = 140ºF

60

50

40

30

20 0

5

10

15

20

25

Effective U of Each Umbilical Tube (W/m^2-C) OPRM20030302D_020.ai

Figure 1.24 – Gas Injection Temperatures at Mudline for Prior Umbilical-based Gas Lift Design


120

115

110

105

100 2

2.5

3

3.5

4

4.5

5

Gas Lift Tube ID (in) OPRM20030302D_021.ai

Figure 1.25 – Dependence of Gas Injection Temperature on Gas Lift Riser Diameter for an Insulating Value of U = 4W/m2-C


OPRM-2003-0302D

Page 38 of 89

30-April-2006


Unrestricted


100

95

90

85

80

75

70 2

3

4

5

6

7

8

Effective U (W/m^2-C) OPRM20030302D_022.ai

Figure 1.26 – Dependence of Gas Injection Temperature on Gas Lift Riser Insulating Value for a 3.5in Tube Diameter


OPRM-2003-0302D

Page 39 of 89

30-April-2006


Unrestricted

From Gas Heater Heater T

Production Riser Riser Inlet T

Topsides Subsea

Gas Lift Riser 3.5in ID UID = 4W/m2-C Injection T

Heater T (ºF) Riser Inlet T (ºF) Injection T (ºF) 220


200 180 160 140 120 100

Spec = 90ºF

80 0

5

10

15

20

25

30

Gas Rate (MMSCFD) OPRM20030302D_023.ai

Figure 1.27 – System Temperature Summary for Base-case Flexible Riser-based Gas Lift Design Section 1 Dynamic Flow Assurance Analysis

OPRM-2003-0302D

Page 40 of 89

30-April-2006


5.0

Unrestricted

SUBSEA SYSTEM SHUTDOWN: HYDRATE PREVENTION STRATEGIES A critical aspect of hydrate management for deepwater subsea systems is prevention of hydrate formation by system cooling during shut-ins of widely varying duration. Operationally, subsea shut-ins are inherently complex with multiple decision gates (particularly for a subsea network of the scope of Bonga), with operating procedures which depend on the shutdown duration.

5.1

Cooldown Performance of Subsea Facilities To aid Operations staff, who must simultaneously work to troubleshoot the shutdown and to protect the subsea system from hydrates, subsea facilities must be designed with sufficient cooldown time. In general terms, cooldown is defined as the time required for the inner wall of the flowpath to reach the hydrate formation temperature, somewhere in the system. The contributions to the cooldown time anticipated for Bonga (refer to Figure 1.28) consist of: •

‘No-touch’ time

•

Time to treat the well tubing and wellhead equipment

•

Time allotted for flowline blowdown

The no-touch time is defined as the time during which Operations staff can act to rectify the shutdown cause, without having to undertake operations to protect the subsea system from hydrates. The 3-hour no-touch time specified for Bonga is based on GoM platform statistics for unplanned shutdowns (refer to Figure 1.29), which indicate that 80% of typical process and instrumentation interrupts were analysed and corrected within 3 hours. Figure 1.29 indicates a rapidly diminishing benefit of no-touch times longer than 3 hours. 5.1.1

Well Tubing Based on the timing illustrated in Figure 1.28, the well tubing must provide at least 8 hours of cooldown time, accounting for a well MeOH treatment time of 5 hours (ie well tubing cooldown time > 3-hour no-touch + 5-hour MeOH well treatment). An important benefit of bare well tubing is the lengthy wellbore cooldown provided by thermal energy generated in the surrounding formation during (steady-state) production. As shown in Figure 1.30, for early-life production at minimum rate (10MBOPD), at least 48 hours of cooldown are available in the wellbore (eg 100ft depth and below). Thus, MeOH bullheading of the well to the SSSV will be required only for very lengthy shut-ins, ie greater than 2 days (expected to be rare). For shorter duration shut-ins, only the top portion of the wellbore (a few hundred feet) have to be topped with MeOH during the allotted 8-hour well cooldown time. For these shut-ins, less than 2 days will be required and they are expected to be much more frequent (refer to Figure 1.29). The required MeOH treatment time will generally be less than the 5 hours allotted. As an added benefit, this surplus time due to quicker MeOH treatment may be used to increase the no-touch time and/or the flowline blowdown time.


OPRM-2003-0302D

Page 41 of 89

30-April-2006


5.1.2

Unrestricted

Subsea Tree/Jumper/Manifold As for the well tubing, the subsea tree, well jumper and manifold must provide at least 8 hours of cooldown, accounting for 5 hours allotted for MeOH displacement of these components. Although the chemical injection system is sized to treat all wells within 5 hours, 12 hours of cooldown time are specified for the wellhead facilities in the Subsea ITT as an added margin to assist Operations. In particular, the following gas cooldown specification appears in the Subsea Invitation to Tender (ITT). •

Upstream of choke (subsea tree) –

•

120°F (49°C) to 73°F (23°C) in no less than 12 hours

Downstream of choke (subsea tree + well jumper + manifold) –


The starting wellhead temperature of 120°F is satisfied for all initial-life wells at rates greater than 5MBOPD (refer to Figure 1.14). However, the field’s coldest well (702p7) does not reach 120°F at any rate and hence will require well-specific operating procedures. The final temperatures reflect the HDT at the well shut-in pressure (4600psia) upstream of the choke and the anticipated flowline shut-in pressure downstream of the choke. 5.1.3

Flowline and Riser For both the pipe-in-pipe flowlines and steel catenary risers, a 12-hour cooldown is specified in the flowline/riser ITT, for gas-filled (methane) components at 28bara: •

West-side 10in flowlines –

•


East-side 10in and 12in flowlines –


The work presented herein culminated in approval of MoC 59, which specifies that both this cooldown requirement and a U value requirement of Uod ≤ 2.0W/m2-C must be met for the cylindrical cross-sections of the flowline and riser. Note: The more conservative specification of gas cooldown is based on restart considerations, ie the hydrate risk of wet fluid passing through cold, originally gas-filled sections upon restart. The starting temperatures for cooldown are based on the minimum anticipated riser base temperatures for 10MBOPD production, including margins for cooling by riser gas lift and possible reservoir cooling by waterflood. With these conservative margins, the starting riser base temperatures are comparable to the arrival temperatures at 10MBOPD (refer to Figure 1.15). The west-side starting temperature is 11°F than the east-side flowlines due to the significantly shorter offsets (hence lesser heat losses) of the west-side flowlines. The final temperatures are based on the HDT at the flowline shut-in pressure, using the hydrate dissociation conditions of the 803 fluid with 0% salinity for conservatism. Furthermore, the effect of a 10-minute choke closure time on the flowline shut-in pressure is explicitly accounted for. Due to their longer offsets, the east-side flowlines experience less partial packing and hence a lower shut-in pressure, which is why the final temperature for east-side cooldown is lower (61°F for east-side versus 66°F for west-side).


OPRM-2003-0302D

Page 42 of 89

30-April-2006


Unrestricted

For the steel catenary risers, prior conceptual analysis (Granherne, 1998) had specified a 2in carazite insulation for (liquid-filled) cooldown. However, Figure 1.31(a) and 1.32(a) indicate that 2in of carazite does not satisfy the gas-filled cooldown requirement (typical for deepwater GoM), even at higher production rates. Figures 1.31(b) and 1.32(b) demonstrate that a 4in carazite (or equivalent) riser insulation is required to attain 12 hours of cooldown at the minimum turndown rate of 10MBOPD per flowline. The added benefit of a Ported Orifice Valve (POV) upstream of the choke is not accounted for, which will yield lower flowline shut-in pressures and hence longer cooldown times (ie results closer to the immediate choke closure curves in Figures 1.31 and 1.32). At anticipated production rates of 30 to 40MBOPD (according to the production function), 18 to 20 hours of gas cooldown are available, providing Operations staff additional time to react and/or secure the system against hydrates. For the base-case pipe-in-pipe flowline design (refer to Figure 1.5), the U = 2.0W/m2-C requirement can be met by filling only 0.6in of the ~1in annular gap with polyurethane foam. However, the cooldown analysis presented here indicates that the annular gap must be filled with foam (at marginal additional cost) to meet the 12-hour gas cooldown requirement. In Figures 1.33 to 1.35, the cooldown performance of each pipe-in-pipe flowline is shown for 0.6in (U = 2.0W/m2-C) and 1in (foam-filled annulus) foam thicknesses. As summarised in Table 1.2, 10 to 11.5 hours of cooldown are attained with a 0.6in foam thickness. In each case, a foam-filled annular gap (with a 5mm tolerance for manufacturing) is required to meet the 12-hour gas cooldown specification. In summary, this analysis reveals that the base case flowline with U = 2.0W/m2-C (without foam filling of the annular gap) does not satisfy the 12-hour cooldown requirement. The U value requirement is based only on steady-state thermal performance, which does not uniquely determine the cooldown performance. That is, significantly different cooldown performance can occur for the same U value, depending on the ‘thermal mass’ of the pipe and insulation system. As illustrated in Figure 1.36, a carrier pipe with a 0.94in wall thickness meets the 12-hour cooldown target, while a 0.75in wall provides only 10 hours of cooldown, although the corresponding U values are identical. The situation is complicated further for alternative pipe diameters and wall thicknesses, which may be explored in the detailed design process. Thus, to ensure adequate flowline/riser cooldown performance, MoC 59 specifies that both the U value and cooldown specifications shall be satisfied simultaneously. East 12in

East 10in

West 10in

0.6in PU foam (U = 2W/m2-C)

11.5 hours

10.5 hours

10 hours

1in PU foam (foam-filled gap)

13.5 hours

13 hours

12.5 hours

Table 1.2 – Cooldown Time as a Function of PU Foam Thickness Within ‘Pipe-in-pipe’ Flowlines


OPRM-2003-0302D

Page 43 of 89

30-April-2006


5.2

Unrestricted

Flowline Blowdown With reference to Figure 1.28, for subsea shutdowns lasting longer than 8 hours, depressurisation of producing flowlines (blowdown) must commence to secure the continuously cooling flowline against hydrate formation. To remain within the 12-hour cooldown window (the minimum cooldown at turndown rates), all flowlines must be blown down within approximately 4 hours (ie 12-hour cooldown ≥ 3-hour no-touch + 5-hour well MeOH treatment + 4-hour blowdown). Due to the lengthy well tubing cooldown, the well treatment may take only 3 to 4 hours, which will allow 5 to 6 hours of blowdown time. The precise breakdown of the available cooldown time will be the subject of future operability analysis. The principal objective of blowdown is to prevent hydrate formation in the flowlines, for lengthy shut-ins. By reducing the flowline pressure to below the HDP at the ambient seafloor temperature of 40°F, the flowline will be secured against hydrate formation for an indefinite shut-in. For conservatism, a blowdown target of HDP = 145psia (10bara) is used throughout this analysis, based on the worst case of 803 fluid with 0% salinity (refer to Appendix 1A Table 1A.3). This target is ~70psia lower than the dominant 702 fluid production (with HDP~220psia), a depressurisation margin which is necessary for successful hydrate remediation. The flow assurance and topsides constraints on blowdown are summarised as follows: •

Maximum flowline pressure after blowdown < 145psia

•

Blowdown time < 4 hours (all eight flowlines)

•

Gas flare rate (instantaneous radiant heat capacity) < 200MMscfd

•

Oil carryover rate (instantaneous flare scrubber capacity) < 75MBOPD

In Figures 1.37 to 1.40, the blowdown performance for the 10in west-side and 12in east-side flowlines is summarised in terms of pressure, gas outlet rate and liquid carryover, for initial-life conditions at 0% watercut. Results are shown for the following scenario, with both immediate choke closure and full line-packing considered to bracket the full design range: 40MBOPD steady-state → Shut-in (immediate or full line-pack) → 3-hour cooldown → Blowdown to 20psia @ topsides (0.5in to 2in blowdown valve) Note: The line-packing cases capture the maximum design gas and liquid rates during blowdown, while the immediate choke closure cases reflect the typical operating scenario. With respect to the topsides facility constraints, none of the blowdown cases in Figures 1.37 to 1.40 exceed the 200MMscfd gas flare capacity or the 75MBOPD oil scrubber capacity. For the west-side flowlines, the maximum gas and liquid rates for the worst-case line-packing scenario are 27MMscfd and 45MBOPD (refer to Figure 1.37). For the 12in east-side flowlines, the peak rates are 40MMscfd and 70MBOPD (refer to Figure 1.39). Due to the short duration of these peak rates, simultaneous blowdowns of multiple flowlines may be pursued, provided that each blowdown is staggered by at least 30 minutes.


OPRM-2003-0302D

Page 44 of 89

30-April-2006


Unrestricted

For the flow assurance targets, blowdown is successful for the 10in west-side flowlines provided that the blowdown valve size is at least 2in (Figures 1.37(a) and 1.38(a), to enable sufficient liquid removal from the flowline. For the west flowlines, the blowdown is completed within 1 hour. In contrast, blowdown for the 12in east-side flowline is unsuccessful for immediate choke closure (refer to Figure 1.40(a), with a final pressure of 600psia which is well above the 145psia target. The counterintuitive result that blowdown is successful for a line-packed east flowline is due to the additional liquid carryover driven by the higher shut-in pressure. Significantly, for a 50% watercut (which will be attained early in field life), blowdown is unsuccessful for all scenarios, as indicated in Figure 1.41. Thus, to secure flowlines for indefinite shut-ins, alternatives to a traditional, totally passive blowdown must be considered (eg riser gas lift assist or dry-oil circulation).

5.3

Gas Lift-assisted Blowdown In light of the unsuccessful blowdowns predicted for the 12in east-side flowlines and the 10in west-side flowlines at higher watercuts, the possibility of riser gas lifting to remove riser liquid during blowdown is now considered. The specific worst-case scenario analysed below consists of: 30MBLPD production (50% watercut) → Immediate shut-in at time of maximum riser liquid during severe slugging → 3-hour cooldown → Open 2in to 10in blowdown valve (@ t = 4 hours) → Inject riser gas lift pulse of 10MMscfd for 1 hour → Stop gas lift → 7-hour flowline/riser settle-out Gas lift blowdown results for the 12in east-side flowlines are shown in Figure 1.42, indicating the counter-intuitive result that riser gas lift does not guarantee blowdown success (refer to Figure 1.42a). If the blowdown valve is not sufficiently large, back-pressure at the flowline outlet prevents slug-like removal of riser liquid, which instead falls back to the riser base resulting in churn-like flow. To attain pressures below 145psia, a very rapid blowdown with a 10in valve is required, with an associated peak liquid outlet rate of 200MBLPD (refer to Figure 1.42b). Although this exceeds the flare scrubber capacity, any overflow will empty (by gravity feed) into a 24,000bbl slop tank. The peak outlet gas rate of 70MMscfd is well within the instantaneous flare capacity (200MMscfd). Note: After gas lift ceases (@ t = 5 hours in Figure 1.42), the flowline pressure slowly increases to approximately 170psia as liquid in the flowline and riser settles out. Similar results are obtained for the 10in east-side flowlines (refer to Figure 1.43), with a more effective blowdown (final pressure near 155psia) and lesser liquid volumes resulting from a smaller riser diameter. A potential concern for gas lift assisted blowdown is the hydrate risk of injecting 40°F lift gas into the flowline, which contains wet fluids which have cooled several hours (near the end of the 12-hour cooldown period). To address this concern, the hydrate condition tracking feature of OLGA is applied to the following scenario: 15MBLPD production (0 to 50% watercut) → Immediate shut-in → 10-hour cooldown → Open blowdown valve → Inject 10MMscfd gas lift @ 40°F for 2 hours → Stop gas lift


OPRM-2003-0302D

Page 45 of 89

30-April-2006


Unrestricted

As shown in Figure 1.44(a), the cold gas lift injection causes a local pressure/temperature within 1°F of hydrate conditions. The fact that hydrate condition subcooling does not occur is due to the rapid riser-base depressurisation by gas lifting (refer to Figure 1.44(c). Due to the residual heat in the flowline liquid and pipe wall, this depressurisation outruns the gas lift cooling (refer to Figure 1.44(b), preventing a local hydrate condition. In light of this depressurisation effect, it is critical that the topsides blowdown valve is fully open before the gas lifting operation commences, as a significant (~20°F) subcooling of wet fluids at the riser base will occur otherwise. In summary, although riser gas lift can significantly reduce the flowline pressure, several additional design and operability modifications were required to enable hydrate-free indefinite-length shut-ins. In particular, the requirement of a large blowdown valve orifice for effective gas lift assisted depressurisation resulted in replacement of the prior fixed 2in blowdown valve with a two-stage blowdown valve train containing a smaller variable choke and a large fixed orifice. Furthermore, it was revealed that gas lift-assisted blowdown does not guarantee successful blowdown below 145bara, due to pressure recovery resulting from liquid settle-out in the flowline and riser. Hence, a backup strategy was formulated for more lengthy shut-ins, consisting of flowline displacement by dry oil circulation at 3 to 5mph. Associated topsides modifications were also made to improve the timing and control of the dry-oil circulation operation. Additionally, a pressure/temperature sensor was added to each riser base (at the gas-injection tee) to enable Operations to accurately determine the effectiveness of gas lift assisted blowdown operations (captured by MoC 64). Since blowdown is marginally effective for the east-side 12in flowline, it is logical to question whether a primary dry-oil circulation strategy should be used in place of gas lift-assisted blowdown. There are two key advantages of blowdown as a primary shut-in strategy. First, it is an essentially passive operation which can be performed under unexpected or emergency topsides shutdowns. Secondly, even an unsuccessful blowdown provides significant extra reaction time for trouble-shooting and a secondary dry-oiling operation if necessary (eg blowdown to 250psia provides 24 hours of additional cooldown time; refer to Figure 1.45). 3 hours

3 to 5 hours

4 to 6 hours

“No-touch”

Well MeOH Treating

Blowdown

Figure 1.28 – Definition of Contributions to Cooldown Time


OPRM-2003-0302D

Page 46 of 89

30-April-2006


Unrestricted

120

Percentages (%)

100 80 60 40 20 0 1

2

3

4

5

6

7

8

9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Hours OPRM20030302D_024.ai

Figure 1.29 – Downtime Duration Statistics for Unplanned Shutdowns in GoM

Minimum Wellbore Temperature (ºF)

150 702p4 702p7

140 130 120 110 100 90 80

HDT

70 0

10

20

30

40

50

Time After Shut-in (hours) OPRM20030302D_025.ai

Figure 1.30 – Wellbore Cooldown at Wellhead for Hottest and Coldest 702 Wells


OPRM-2003-0302D

Page 47 of 89

30-April-2006


Unrestricted

Production Rate (MBOPD) 10

20

40

14

Gas Cooldown Time (hours)

12 Target 10 8 6 4 Immediate Choke Closure 10-minute Closure

2

Full Line-pack 0 60

70

80

90

100

110

120

130

140

Initial Riser Base Temperature (ºF)


20

40

30


25

20

15 Target 10 Immediate Choke Closure 5

10-minute Closure Full Line-pack

0 60

70

80

90

100

110

120

130

140

Initial Riser Base Temperature (ºF) OPRM20030302D_026.ai

Figure 1.31 – East-side 12in Riser Cooldown Performance for (a) 2in Carazite and (b) 4in Carazite


OPRM-2003-0302D

Page 48 of 89

30-April-2006


Unrestricted


10

20

40

16


14 Target 12 10 8 6 4

Immediate Choke Closure 10-minute Closure

2

Full Line-pack 0 80

90

100

110

120

130



10

20

40


25

20

15 Target 10

Immediate Choke Closure 5

10-minute Closure Full Line-pack

0 80

90

100

110

120

130


OPRM20030302D_027.ai

Figure 1.32 – West-side 10in Riser Cooldown Performance for (a) 2in Carazite and (b) 4in Carazite


OPRM-2003-0302D

Page 49 of 89

30-April-2006


Unrestricted

35 U=2

30 Gas Temperature (ºC)

Foam-filled 25 20 15 10 5 0 0

2

4

6

8

10

12


Figure 1.33 – Pipe-in-pipe Cooldown for East-side 12in Flowlines

35 U=2

30 Gas Temperature (ºC)

Foam-filled 25 20 15 10 5 0 0

2

4

6

8

10

12


Figure 1.34 – Pipe-in-pipe Cooldown for East-side 10in Flowlines


OPRM-2003-0302D

Page 50 of 89

30-April-2006


Unrestricted

40 U=2

35

Gas Temperature (ºC)

Foam-filled 30 25 20 15 10 5 0 0

2

4

6

8

10

12


Figure 1.35 – Pipe-in-pipe Cooldown for 10in West-side Flowlines

40 Initial T = 36ºC

0.94in wt and U = 1.4W/m^2-K 0.75in wt and U = 1.4W/m^2-K

Gas Temperature (ºC)

35

30

Minimum CDT = 12 hours

25

20

15 0

2

4

6

8

10

12


Figure 1.36 – Illustration of Non-unique Relationship Between U Value and Cooldown


OPRM-2003-0302D

Page 51 of 89

30-April-2006


Maximum Flowline Pressure (psia)

Line-pack

Unrestricted

Blowdown

5150

4150

3150 1in valve 2150 0.5in valve 1150 2in valve

Target: HDP = 145psia

0

2

4

6

8

10

Time (hours) 50

Outlet Oil Rate (MBOPD)

2in valve: 480bbl 40

30

20 1in valve: 360bbl 10 0.5in valve: 160bbl 0 0

2

4

6

8

10

Time (hours) 30

Outlet Gas Rate (MMSCF)

2in valve 25 20 15 1in valve 10 0.5in valve

5 0 0

2

4

6

8

10

12


Figure 1.37 – Blowdown Performance: 10in West-side and Full Line-pack Section 1 Dynamic Flow Assurance Analysis

OPRM-2003-0302D

Page 52 of 89

30-April-2006


Maximum Flow Pressure (psia)

Shut-in

Unrestricted

Blowdown

950

750

550 0.5in valve 350

1in valve


2in valve 0 0

2

4

6

8

10

8

10

Time (hours)


50

40

30

20 2in valve: 45bbl 10

0 0

2

4

6

Time (hours)

Outlet Gas Rate (MMSCF/D)

20

15

10 2in valve

5 1in valve 0.5in valve 0 0

2

4

6

8

10

12


Figure 1.38 – Blowdown Performance: 10in West-side and Immediate Choke Closure


OPRM-2003-0302D

Page 53 of 89

30-April-2006


Unrestricted

Line-pack Blowdown 5150


4150

3150 1in valve 2150 0.5in valve 1150 2in valve


0

2

4

6

8

10

12

14

16

Time (hours) 70 2in valve 1360bbl Outlet Oil Rate (MBOPD)

60 50 40 30 1in valve: 910bbl 20 0.5in valve 620bbl

10 0 0

2

4

6

8

10

Time (hours) 40 Outlet Gas Rate (MMSCF/D)

35 2in valve 30 25 20 15

1in valve

10 0.5in valve 5 0 0

5

10 Time (hours)

15

20

OPRM20030302D_035.ai

Figure 1.39 – Blowdown Performance: 12in East-side and Full Line-pack Section 1 Dynamic Flow Assurance Analysis

OPRM-2003-0302D

Page 54 of 89

30-April-2006


Unrestricted

Shut-in

Blowdown


950

750

550

350

Target: HDP = 145psia 0

2

4

6

8

10

8

10

Time (hours)


50

40

30

20

10

0 0

2

4

6

Time (hours)


25

20

15

10 2in valve 5

1in valve 0.5in valve

0 0

5

10

15

20


Figure 1.40 – Blowdown Performance: 12in East-side and Immediate Choke Closure Section 1 Dynamic Flow Assurance Analysis

OPRM-2003-0302D

Page 55 of 89

30-April-2006


Unrestricted

1150

950

Maximum Flowline Pressure (pisa)

West – Immediate closure West – Full line-pack

750

East – Full line-pack East – Immediate closure

550

350

Target: 145pisa

0

5

10

15

20


Figure 1.41 – Blowdown Performance for 50% Water Cut, Illustrating Unsuccessful Blowdown for All Scenarios


OPRM-2003-0302D

Page 56 of 89

30-April-2006


Unrestricted


500 450 400

2in valve 4in valve

350

10in valve

300 250 200 150

Target HDP = 145psia 100 4

5

6

7

8

9

10

11

12

Time (hours) 250

Outlet Oil Rate (MBLPD)

2in valve 4in valve

200

10in valve 150

100

50

0 4

4.5

5

5.5

6

Time (hours)


70 2in valve

60

4in valve 50

10in valve

40 30 20 10 0 -10 4

4.5

5

5.5

6


Figure 1.42 – Blowdown Performance with Riser Gas Lift Assist, for 12in East-side Flowlines


OPRM-2003-0302D

Page 57 of 89

30-April-2006


Unrestricted


500

2in valve 4in valve

450

8in valve

400 350 300 250 200 150

Target HDP = 145psia 100 4

5

6

7

8

9

10

11

12

Time (hours) 100

Outlet Oil Rate (MBLPD)

4in valve 2in valve

80

8in valve 60

40

20

0 4

4.5

5

5.5

6

Time (hours) 50


4in valve 40

2in valve 8in valve

30 20 10 0 -10 4

4.5

5

5.5

6


Figure 1.43 – Blowdown Performance with Riser Gas Lift Assist, for 10in East-side Flowlines


OPRM-2003-0302D

Page 58 of 89

30-April-2006


Unrestricted

Hydrate

Proximity to Hydrate Condition: Flowline Maximum (HDT-T) (ºF)

0 No Hydrate

50% Water Cut

-10

0% Water Cut

-20 -30 -40 -50 Shut-in

Gas Lift On

Gas Lift Off

-60 0

5

10

15

20

Time After Shut-in (hours)

Temperature at Gas Lift Location (ºF)

120 50% Water Cut

110

0% Water Cut

100 90 80 70 60 50 40 0

5

10

15

20

Temperature at Gas Lift Location (psia)

Time After Shut-in (hours) 1400 50% Water Cut 1200

0% Water Cut

1000 800 600 400 200 0 0

5

10

15

20

Time After Shut-in (hours) OPRM20030302D_038.ai

Figure 1.44 – Pressure and Temperature Evolution During Cold Gas Lift-assisted Blowdown


OPRM-2003-0302D

Page 59 of 89

30-April-2006


Unrestricted

90

Riser Gas Temperature (ºF)

85 80 75 70 65

Shut-in HDT

60 55 HDT after Blowdown to 250pisa

50 45 40 0

5

10

15

20

25

30

35

40

45

50

55


Figure 1.45 – Benefit of Depressurisation for Unsuccessful Blowdown in Providing 24 Hours of Additional Cooldown Time

6.0

CONCLUDING REMARKS AND PRELIMINARY OPERATING LOGIC In summary, detailed thermal-hydraulic analysis validates the Bonga conceptual design with respect to hydrate management, for the most extreme anticipated operating conditions. The modifications to hardware design and operating procedures identified have been addressed and fully implemented within the appropriate Bonga teams. As the first step toward development of detailed subsea operating procedures, preliminary operating logic charts, consistent with the flow assurance analysis documented here, are shown in Figures 1.46 to 1.50 for the following: •

Cold start-up

•

Additional well start-up

•

Interrupted start-up

•

Planned or unplanned shutdown

•

Blowdown

As flow assurance efforts progress into detailed design and the development of subsea operating procedures, further dynamic thermal-hydraulic studies are recommended for the following areas: •

Gas buy back at OGGS-RPA and Bonga for initial start-up and shutdown

•

Minimum operating temperature analysis for the subsea system and topsides interfaces

•

Development and check-out of operating procedures via coupled well/flowline/ topsides dynamic modelling

•

Definition of subsea transient operability envelope for on-demand operational decisions


OPRM-2003-0302D

Page 60 of 89

30-April-2006


Unrestricted

Hot oil circulate 150ºF hot oil at 3 to 5mph

Cold Start-up Start-up of cold well into blown down flowine

System Conditions

Is arrival temperature at FPSO > 95ºF?

Wells bullheaded with MeOH Trees, jumpers and manifolds flushed w/MeOH

No

Yes

Flowlines stabilised (blown down or dry-oiled) Stop hot oil circulation. Close pigging iso valve at manifold

Sufficient MeOH available on FPSO

Start riser base gas lift as appropriate: 10MMSCFD – 10in flowlines 20MMSCFD – 20in flowlines

Start-up lowest wc well, as per Start-up Guidelines: Start MeOH injection upstream choke Open subsea choke and start specified ramp-up

Is the FWHT >95ºF?

No

Continue MeOH injection

Yes

Stop MeOH injection and continue well ramp-up

Full system cooldown not available. No Shutdown requires immediate action: Go to 'Interrupted Start-up'

FWHT >120ºF and Arrival Temperature > 85ºF?

5-hour wellbore cooldown available

Steady-state operation. Yes For additional wells: Go to 'Additional Well Start-up'

OPRM20030302D_041.ai

Figure 1.46 – Cold Start-up


OPRM-2003-0302D

Page 61 of 89

30-April-2006


Additional Start-up

Unrestricted

Line up subsea equipment for new well start-up

Start-up of new (cold) well into producing flowline

System Conditions

Adjust riser base gas lift as appropriate for new well

Cold well bullheaded with MeOH Cold tree and jumper flushed w/MeOH Start-up new well, as per Start-up Guidelines:

Flowline producing at steady-state: Arrival T > 85ºF Producing wells FWHT > 120ºF

Start MeOH injection upstream of choke

Sufficient MeOH available on FPSO

Open subsea choke and start specified ramp-up

Is the FWHT > 95ºF?

No

Continue MeOH injection

Yes

Stop MeOH injection and continue well ramp-up

5-hour wellbore cooldown available

Full system cooldown not available.

Steady-state operation. No

Shutdown requires immediate action: Go to 'Interrupted Start-up'

FWHT (all) > 120ºF and Arrival Temp > 85ºF?

Yes For additional wells: Go to Top

OPRM20030302D_042.ai

Figure 1.47 – Additional Well Start-up


OPRM-2003-0302D

Page 62 of 89

30-April-2006


Unrestricted

System Conditions Flowlines hot-oiled prior to well start-up: Untreated water present: Wellbore, jumper, manifold and/or flowline Interrupted Start-up Full cooldown not available System shutdown prior to steady-state

Immediate action required

5-hour wellbore cooldown available Tree temperature > 95ºF?

No

Continuous MeOH injection at tree Flowline inhibited: treated water and dry oil

Yes

Optional: Blow down flowlines (contain dry oil and inhibited fluid only)

Tree temperature > 120ºF?

Displace tree and bullhead well with MeOH ASAP (refer to MeOH table)

No

Wellbore and flowline uninhibited Wellbore, tree, jumper, manifold, flowline cooldown not secured

Optional: Displace jumpers and manifold with MeOH (already treated)

Displace tree, jumpers and manifold with MeOH ASAP

Yes

Blow down Flowlines ASAP Go to 'Blowdown'

Bullhead well with MeOH Complete within 5 hours

MeOH Table Arrival temperature > 85ºF?

No

Blow down Flowlines ASAP Go to 'Blowdown'

Jumper, Tree and Manifold MeOH

GPM

9

18

Duration

hours

2

1

Vol jumper

bbls

2

2

Vol manifold

bbls

20

20

24

24

Yes

Steady-state condition: Go to 'Shutdown from steady-state'

Bullhead wells and displace tree, jumper, manifold with MeOH. Complete within 8 hours

Total bbls used Well Treatment MeOH

GPM

9

18

Duration/well

hours

3.88

1.94

bbls

50

50

Two Wells – Total bbls Used

100

100

Total System bbls Used

124

124

Vol/well

OPRM20030302D_043.ai

Figure 1.48 – Interrupted Start-up


OPRM-2003-0302D

Page 63 of 89

30-April-2006

Minimum available cooldown times: • Wellbore: 48 hours • Tree, jumper, manifold: 8 hours • Flowline: 16 hours • Riser: 12 hours

Planned Shutdown

Unplanned Shutdown

Close tree chokes and PSDVs

Auto-close boarding valves

Allow flowlines to evacuate to LP separator

Stop riser gas lift Close POVs and subsea chokes (each tree)

Stop riser gas lift MeOH Table Jumper, Tree and Manifold MeOH

GPM

9

18

Duration

hours

2

1

Vol jumper

bbls

2

2

Vol manifold

bbls

20

20

24

24

Total bbls used

Can production be resumed within 3 hours? (5 hours of treatment time alotted for wells, jumpers and manifold)

Yes


Planned or Unplanned Shutdown for Steady-state

FWHT (all wells) > 120ºF Arrival T > 85ºF Topsides facilities and export available

Go to 'Restart' (start up without utilising methanol)

No Well Treatment MeOH

GPM

9

18

Duration/well

hours

3.88

1.94

bbls

50

50

Two Wells – Total bbls Used

100

100

Total System bbls Used

124

124

Vol/well

Displace tree, jumpers and manifold with MeOH. Complete within 8 hours of shutdown (refer to MeOH table)

Bullhead wells with MeOH (refer to MeOH table)

Can production be resumed within 8 hours?

Yes

Go to 'Warm Start' (start up utilising methanol as necessary)

No No

Go to 'Cold Start'

Yes

Can production be resumed within 48 hours?

Blow down all flowlines within 12 hours of shutdown Go to 'Blowdown' OPRM20030302D_044.ai

Unrestricted

30-April-2006

Figure 1.49 – Planned or Unplanned Shutdown from Steady-state

Page 64 of 89


OPRM-2003-0302D

System Conditions (Steady-state)


Unrestricted

Blowdown

System Conditions

Secure flowlines for indefinite shut-in

Flowlines isolated at platform and tree Manifold pigging iso valve closed Untreated water present in flowline Flowline at/near 8-hour cooldown

Line up topsides for blowdown to flare system

Open appropriate blowdown valves Depressure until gas/ liquid rates diminish

Manifold pressure > 10bara? (or gas-assist known to be necessary?

No

Yes

Initiate gas lift at each riser base: 15MMSCFD for 1 hour

Initiate dry-oil circulation at 3 to 5mph Option: Launch pig

Yes

Manifold pressure > 10bara? (or gas-assist known to be insufficient? No

Close boarding valve Flowline secure for indefinite shut-in OPRM20030302D_045.ai

Figure 1.50 – Blowdown


OPRM-2003-0302D

Page 65 of 89

30-April-2006


Unrestricted

Appendix 1A Reservoir Fluid Properties Table of Contents TABLES Table 1A.1 – Measured Fluid Properties for Each Reservoir (from Bonga BoD) ...................67 Table 1A.2 – Hydrate Dissociation Data for 702 Reservoir Fluid (from A Mehta, 1998) ........69 Table 1A.3 – Hydrate Dissociation Data for 803 Reservoir Fluid (from A Mehta, 1998) ........71

FIGURES Figure 1A.1 – Phase Envelope for 702 Reservoir Fluid, Calculated in OLGA .......................67 Figure 1A.2 – Hydrate Dissociation Curves for 702 Reservoir Fluid (Data in Table 1A.2) .....68 Figure 1A.3 – Hydrate Dissociation Curves for 803 Reservoir Fluid (Data in Table 1A.3) .....70

Section 1 Appendix 1A Reservoir Fluid Properties

OPRM-2003-0302D

Page 66 of 89

30-April-2006


Unrestricted

Table 1A.1 – Measured Fluid Properties for Each Reservoir (from Bonga BoD)

Figure 1A.1 – Phase Envelope for 702 Reservoir Fluid, Calculated in OLGA


OPRM-2003-0302D

Page 67 of 89

30-April-2006


Unrestricted

Wate r Fres h

8000

3 wt %S

10 w

9000

alt

t% S alt

10000

Pressure, psia

7000 6000

Hydrate Stability Region

5000 4000 3000 2000 Non-Hydrate Region

1000 0 40

45

50

55

60

65

70

75

80

85

Temperature, F

Figure 1A.2 – Hydrate Dissociation Curves for 702 Reservoir Fluid (Data in Table 1A.2) Section 1 Appendix 1A Reservoir Fluid Properties

OPRM-2003-0302D

Page 68 of 89

30-April-2006


Unrestricted

Fresh Water P (p s ia)

T (F)

218.1

40.0

500

50.8

750

55.9

1000

59.4

2000

67.2

3000

71.4

4000

73.2

5000

74.9

6000

76.6

7000

78.2

8000

79.8

9000

81.3

10000

82.8

3 wt% Salt 252.0

40.0

500

48.9

750

54.0

1000

57.5

2000

65.2

3000

69.4

4000

71.2

5000

72.9

6000

74.6

7000

76.1

8000

77.7

9000

79.3

10000

80.8

10 wt% Salt 418.8

40.0

500

42.3

750

47.3

1000

50.7

2000

58.2

3000

62.2

4000

64.0

5000

65.8

6000

67.5

7000

69.1

8000

70.7

9000

72.3

10000

73.8

Table 1A.2 – Hydrate Dissociation Data for 702 Reservoir Fluid (from A Mehta, 1998)


OPRM-2003-0302D

Page 69 of 89

30-April-2006


Unrestricted

a lt

10000

ter h Wa

8000

Fres

10 w

3 wt %S

alt

t% S

9000

Pressure, psia

7000 6000 Hydrate Stability Region

5000 4000 3000 2000

Non-Hydrate Region

1000 0 40

45

50

55

60

65

70

75

80

85

90

Temperature, F

Figure 1A.3 – Hydrate Dissociation Curves for 803 Reservoir Fluid (Data in Table 1A.3) Section 1 Appendix 1A Reservoir Fluid Properties

OPRM-2003-0302D

Page 70 of 89

30-April-2006


Unrestricted

Fresh Water 159.5

40.0

500

55.2

750

60.3

1000

63.7

2000

71.0

3000

75.0

4000

77.9

5000

80.5

6000

82.7

7000

84.5

8000

86.2

9000

87.8

10000

89.4

3 wt% Salt 184.1

40.0

500

53.3

750

58.4

1000

61.8

2000

69.0

3000

72.9

4000

75.8

5000

78.4

6000

80.6

7000

82.4

8000

84.1

9000

85.7

10000

87.3

10 wt% Salt 301.7

40.0

500

46.6

750

51.6

1000

54.9

2000

62.0

3000

65.8

4000

68.7

5000

71.2

6000

73.4

7000

75.2

8000

76.9

9000

78.6

10000

80.2

Table 1A.3 – Hydrate Dissociation Data for 803 Reservoir Fluid (from A Mehta, 1998)


OPRM-2003-0302D

Page 71 of 89

30-April-2006


Unrestricted

Appendix 1B Wellbore Modelling Summary and Production Forecast Table of Contents 1.0

RESERVOIR PRESSURE AND TEMPERATURE SUMMARY ..................................76

2.0

WELL PRODUCTION SUMMARY ............................................................................. 77

3.0

DESIGN BASIS AND PRODUCTION FORECAST: 702 RESERVOIR ......................78

4.0


5.0


6.0


TABLES Table 1B.1 – Sample WELLTEMP Input File, for Well 702p4 ...............................................73 Table 1B.2 – WELLTEMP Input Data for 702p4, Representing the Hottest 702 Well............74 Table 1B.3 – Wellhead Temperatures Calculated in WELLTEMP for 702p4, for Cold-earth Start-up (t = 0 to 1440 hours) and Cooldown (t = 1440 to 1488 hours)..................................................................................74 Table 1B.4 – WELLTEMP Input Data for 702p7, Representing the Coldest 702 Well...........75 Table 1B.5 – Wellhead Temperatures Calculated in WELLTEMP for 702p7, for Cold-earth Start-up (t = 0 to 1440 hours) and Cooldown (t = 1440 to 1488 hours)..................................................................................75

Section 1 Appendix 1B Wellbore Modelling Summary and Production Forecast

OPRM-2003-0302D

Page 72 of 89

31-December-2004


TITLE: Bonga Well 1000 BOPD CSE Bare Tubing 6.625in VERSION 3.4 TUBING 1 2 5.9 6.625 2000 2000 5.9 6.625 7477 7477 CASING 4 1 8.670 9.625 7477 6477 1 12.330 13.375 3637 2637 1 18.710 20.000 2000 0 1 27.000 30.000 200 0 WELLBORE 3 6 0 0 2000 2000 2200 2212 2400 2438 2700 2828 5780 7477 INITIAL TEMP 2 40 0 162 5780 PVYP FLUIDS 3 1 1 10.0 1 2 2 10.4 10 3 1 9.63 14 ASOLID 3 7 488 0.113 8 180 0.5 0.5 9 0.001 0.25 0.005 NATURAL GAS 1 10 0.7885 0.0663 0.0671 PRINT OPTIONS 1 0 1 1 1000. PF 0 25 0 OPTIONS 3 3 0.0006 1 0 1 0 0 END CHANGE 0.5 'HR' SINGLE FLOW 2 2 10 162

0 BWPD

Unrestricted

15000 BOPD

0 BWPD

1 1 2 3 8 8

0 7 7

60 60 60

24.8

0.0391 0.0

0.0158

0

1

1

-8.83+08

15000 'BPD'

0

4612.5 9

0

29

Table 1B.1 – Sample WELLTEMP Input File, for Well 702p4 (refer to the schematic in Figure 1.3)


OPRM-2003-0302D

Page 73 of 89

31-December-2004


Unrestricted

INPUT---------------------------------------------------------------------------------------------------------------------common filename

MBPD

2.5 5 7.5 10 15 25 40 2.5 5 7.5 10 15 25 40

-----> 702p4

WC (%)

0 0 0 0 0 0 0 70 70 70 70 70 70 70

P-res (PSI)

4800 4800 4800 4800 4800 4800 4800 3600 3600 3600 3600 3600 3600 3600

GOR PI (SCF/STB) (BLPD/PSI)

°API

29 29 29 29 29 29 29 29 29 29 29 29 29 29

600 600 600 600 600 600 600 600 600 600 600 600 600 600

80 80 80 80 80 80 80 80 80 80 80 80 80 80

BHT (°F)

162 162 162 162 162 162 162 162 162 162 162 162 162 162

bare tubing insulated file number file number

1 2 3 4 5 6 7 8 9 10 11 12 13 14

15 16 17 18 19 20 21 22 23 24 25 26 27 28

BOPD

2500 5000 7500 10000 15000 25000 40000 750 1500 2250 3000 4500 7500 12000

BWPD

0 0 0 0 0 0 0 1750 3500 5250 7000 10500 17500 28000

MMSCFD

1.500 3.000 4.500 6.000 9.000 15.000 24.000 0.450 0.900 1.350 1.800 2.700 4.500 7.200

FBHP (PSI)

4769 4738 4706 4675 4613 4488 4300 3569 3538 3506 3475 3413 3288 3100

Table 1B.2 – WELLTEMP Input Data for 702p4, Representing the Hottest 702 Well BARE TUBING RESULTS: time T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) (hr) 702p41 702p42 702p43 702p44 702p45 702p46 702p47 702p48 702p49 702p410 702p411 702p412 702p413 702p414 2.5 MBPD 5 MBPD 7.5 MBPD 10 MBPD 15 MBPD 25 MBPD 40 MBPD 2.5 MBPD 5 MBPD 7.5 MBPD 10 MBPD 15 MBPD 25 MBPD 40 MBPD 0.5 56.53 64.42 71.52 78.06 90.04 110.15 130.59 55.69 66.75 76.70 85.73 101.93 126.55 143.12 1 61.33 73.25 84.02 93.81 110.58 132.58 146.36 63.45 80.70 95.49 108.66 128.47 144.83 151.31 2 68.54 86.39 101.68 114.27 131.00 144.83 151.86 75.30 100.33 119.99 132.45 143.64 150.39 153.55 3 74.08 95.87 112.99 125.00 137.92 147.64 153.21 84.04 113.64 131.30 139.65 146.67 151.61 154.20 6 85.27 111.99 127.14 135.15 143.38 150.24 154.65 100.95 129.74 139.74 144.54 149.26 152.96 155.00 12 97.36 123.02 134.10 140.03 146.35 151.95 155.71 116.03 136.38 143.51 147.20 150.94 153.92 155.59 24 107.66 129.25 138.22 143.13 148.39 153.17 156.49 124.34 140.06 145.92 149.00 152.12 154.63 156.04 48 114.10 132.99 140.78 145.07 149.75 154.01 157.02 128.74 142.34 147.50 150.15 152.90 155.11 156.34 96 118.62 135.87 142.83 146.76 150.91 154.71 157.49 131.91 144.19 148.82 151.19 153.60 155.54 156.61 120 119.95 136.68 143.54 147.22 151.23 154.93 157.62 132.91 144.77 149.21 151.49 153.81 155.67 156.69 1440 129.83 142.97 148.07 150.78 153.69 156.44 158.59 139.98 148.82 152.01 153.63 155.28 156.56 157.25 1441 127.92 141.11 146.26 148.99 151.91 154.68 156.83 138.40 147.17 150.33 151.93 153.56 154.81 155.46 1443 122.90 135.67 140.66 143.31 146.14 148.82 150.88 133.32 141.67 144.68 146.19 147.72 148.88 149.43 1446 115.88 127.79 132.45 134.94 137.56 140.06 141.97 125.78 133.47 136.24 137.63 139.04 140.08 140.53 1448 112.07 123.47 127.93 130.31 132.82 135.23 137.04 121.59 128.94 131.58 132.91 134.26 135.25 135.68 1451 107.49 118.26 122.47 124.72 127.10 129.38 131.08 116.52 123.45 125.95 127.21 128.49 129.44 129.85 1452 106.20 116.78 120.92 123.14 125.48 127.72 129.40 115.09 121.90 124.36 125.60 126.85 127.79 128.20 1464 95.58 104.59 108.13 110.03 112.03 113.94 115.39 103.18 109.00 111.11 112.18 113.26 114.08 114.47 1488 84.49 91.58 94.45 95.99 97.63 99.17 100.35 90.46 95.18 96.89 97.77 98.65 99.33 99.67

Table 1B.3 – Wellhead Temperatures Calculated in WELLTEMP for 702p4, for Cold-earth Start-up (t = 0 to 1440 hours) and Cooldown (t = 1440 to 1488 hours)


OPRM-2003-0302D

Page 74 of 89

31-December-2004


Unrestricted

INPUT---------------------------------------------------------------------------------------------------------------------common filename

MBPD

2.5 5 10 15 20 2.5 5 10 15 20 2.5 5 10 15 20

-----> 702p7

WC (%)

0 0 0 0 0 50 50 50 50 50 80 80 80 80 80

P-res (PSI)

3200 3200 3200 3200 3200 3200 3200 3200 3200 3200 2200 2200 2200 2200 2200

GOR PI (SCF/STB) (BLPD/PSI)

°API

29 29 29 29 29 29 29 29 29 29 29 29 29 29 29

600 600 600 600 600 600 600 600 600 600 600 600 600 600 600

30 30 30 30 30 30 30 30 30 30 30 30 30 30 30

BHT (°F)

128 128 128 128 128 128 128 128 128 128 128 128 128 128 128

bare tubing insulated file number file number

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

BOPD

2500 5000 10000 15000 20000 1250 2500 5000 7500 10000 500 1000 2000 3000 4000

BWPD

0 0 0 0 0 1250 2500 5000 7500 10000 2000 4000 8000 12000 16000

MMSCFD

1.500 3.000 6.000 9.000 12.000 0.750 1.500 3.000 4.500 6.000 0.300 0.600 1.200 1.800 2.400

FBHP (PSI)

3117 3033 2867 2700 2533 3117 3033 2867 2700 2533 2117 2033 1867 1700 1533

Table 1B.4 – WELLTEMP Input Data for 702p7, Representing the Coldest 702 Well

BARE TUBING RESULTS: time T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) T(°F) (hr) 702p71 702p72 702p73 702p74 702p75 702p76 702p77 702p78 702p79 702p710 702p711 702p712 702p713 2.5 MBPD 5 MBPD 10 MBPD 15 MBPD 20 MBPD 2.5 MBPD 5 MBPD 10 MBPD 15 MBPD 20 MBPD 2.5 MBPD 5 MBPD 10 MBPD 0.5 54.19 60.88 71.75 80.23 86.59 54.76 64.21 79.62 90.98 98.94 55.91 66.79 83.99 1 57.90 67.76 83.06 93.28 99.50 60.68 74.82 94.75 105.48 110.97 62.80 78.73 99.47 2 63.71 77.95 96.19 104.68 108.49 69.89 89.01 107.82 114.18 116.73 73.37 93.81 111.00 3 68.23 84.90 102.31 108.63 111.00 76.65 97.13 112.11 116.31 118.01 80.90 101.60 114.22 6 77.33 95.37 108.12 111.90 113.13 88.71 106.32 115.41 118.08 119.22 93.32 109.13 116.65 12 86.01 101.88 111.00 113.61 114.33 97.78 110.62 117.13 119.14 120.00 101.64 112.57 118.05 24 92.84 105.56 112.70 114.71 115.19 103.09 112.96 118.25 119.91 120.54 105.98 114.50 118.96 48 96.90 107.74 113.83 115.45 115.74 105.99 114.40 119.00 120.40 120.93 108.33 115.69 119.57 96 99.49 109.33 114.71 116.07 116.21 107.96 115.60 119.63 120.82 121.25 110.13 116.67 120.08 120 100.36 109.77 114.99 116.25 116.35 108.66 115.98 119.80 120.94 121.34 110.62 116.99 120.23 1440 105.98 113.23 116.86 117.55 117.31 112.98 118.40 121.08 121.81 122.00 114.31 119.01 121.29 1441 104.18 111.53 115.24 115.96 115.73 111.45 116.86 119.53 120.24 120.40 112.64 117.27 119.46 1443 100.17 107.31 110.92 111.62 111.39 107.39 112.54 115.08 115.72 115.83 108.30 112.63 114.61 1446 95.14 101.79 105.17 105.83 105.62 101.93 106.69 109.04 109.63 109.71 102.72 106.69 108.50 1448 92.54 98.91 102.15 102.79 102.60 99.05 103.60 105.86 106.42 106.51 99.84 103.64 105.39 1451 89.46 95.49 98.56 99.17 99.01 95.63 99.92 102.06 102.61 102.70 96.43 100.03 101.71 1452 88.58 94.51 97.53 98.13 97.98 94.64 98.87 100.98 101.52 101.62 95.45 98.99 100.66 1464 81.21 86.29 88.89 89.42 89.31 86.39 90.01 91.84 92.31 92.42 87.17 90.23 91.69 1488 73.02 77.13 79.25 79.70 79.61 77.20 80.16 81.63 82.04 82.13 77.88 80.38 81.58

Table 1B.5 – Wellhead Temperatures Calculated in WELLTEMP for 702p7, for Cold-earth Start-up (t = 0 to 1440 hours) and Cooldown (t = 1440 to 1488 hours)


OPRM-2003-0302D

Page 75 of 89

31-December-2004


1.0

Unrestricted

RESERVOIR PRESSURE AND TEMPERATURE SUMMARY S Van Gisbergen, A Hartwijk and S Lindsey (1999). Medium Skin P50 702 T@midperfs Initial Pavg 702p2 142 3421 702p3 132 2518 702p4 162 4503 702p5 153 3366 702p6 136 2830 702p7 128 2648 702p9 148 4317 702p10 148 4312 702p15 139 4183

690 T@midperfs Initial Pavg b690p1 164 4586 b690p2 147 3826 b690p3 156 3722 b690p4 138 4201 b690p5 139 3138

803 T@midperfs Initial Pavg 803p1 179 5211 803p2 186 5299

710 T@midperfs Initial Pavg 710p1 146 4455 710p2 134 4238 710p3 144 4464 710p4 158 4649

High Skin P50 702 T@midperfs Initial Pavg 702p2 142 3679 702p3 132 2690 702p4 162 4503 702p5 153 3252 702p6 136 2987 702p7 128 2862 702p9 148 4317 702p10 148 4312 702p15 139 4183

690 T@midperfs Initial Pavg 690p1 164 4586 690p2 147 4042 690p3 156 3739 690p4 138 4201 690p5 139 3118

803 T@midperfs Initial Pavg 803p1 179 5211 803p2 186 5315

710 T@midperfs Initial Pavg 710p1 146 4455 710p2 134 3964 710p3 144 4197 710p4 158 4468

Low Skin P50 702 T@midperfs Initial Pavg 702p2 142 702p3 132 702p4 162 4503 702p5 153 702p6 136 702p7 128 702p9 148 4317 702p10 148 4312 702p15 139 4183

690 T@midperfs Initial Pavg 690p1 164 4586 690p2 147 690p3 156 690p4 138 690p5 139

803 T@midperfs Initial Pavg 803p1 179 5211 803p2 186

710 T@midperfs Initial Pavg 710p1 146 710p2 134 710p3 144 710p4 158


OPRM-2003-0302D

Page 76 of 89

31-December-2004


2.0

Unrestricted

WELL PRODUCTION SUMMARY S Van Gisbergen, A Hartwijk and S Lindsey (1999).

Phase 1 Wells well 702p4 702p9 702p10 702p15 690p1 803p1

year Q1 2003 Q1 2003 Q1 2003 Q1 2003 Q1 2003 Q1 2003

flowline PF1 PF3 PF6 PF11 PF2 PF12

months 0 0 0 0 0 0

Phase 2 Wells well 710p1 702p2 690p2 803p2 710p4 710p3 702p5 710p2 690p3 702p6 702p7 690p4 702p3 690p5

year Q1 2004 Q1 2004 Q2 2004 Q2 2005 Q4 2005 Q1 2006 Q4 2006 Q1 2007 Q2 2007 Q1 2008 Q1 2008 Q1 2008 Q3 2008 Q1 2009

flowline PF8 PF12 PF4/PF3 PF6/PF5 PF12 PF8/PF9 PF3 PF9 PF2 PF11 PF11 PF11 PF8 PF5

months 8 11 14 26 29 32 44 47 48 59 60 62 65 71

Well 702p2 702p3 702p4 702p5 702p6 702p7 702p9 702p10 702p15

Year Q1 2004 Q3 2008 Q1 2003 Q4 2006 Q1 2008 Q1 2008 Q1 2003 Q1 2003 Q1 2003

Max. rate 22000 20000 54000 24000 20000 20000 50000 50000 50000

b690p1 b690p2 b690p3 b690p4 b690p5

Q1 2003 Q2 2004 Q2 2007 Q1 2008 Q1 2009

20000 20000 17000 16000 18000

803p1 803p2

Q1 2003 Q2 2005

24000 27000

710p1 710p2 710p3 710p4

Q1 2004 Q1 2007 Q1 2006 Q4 2005

30000 30000 28000 30000


OPRM-2003-0302D

Page 77 of 89

31-December-2004


3.0

Unrestricted

DESIGN BASIS AND PRODUCTION FORECAST: 702 RESERVOIR Refer to the Field Development Plan Rev 5 for production profiles.

4.0


5.0


6.0



OPRM-2003-0302D

Page 78 of 89

31-December-2004


Unrestricted

Appendix 1C Production Flowlines: Topography and Ambient Temperature Data Table of Contents 1.0

TEMPERATURE AND SALINITY PROFILES.............................................................86

TABLES Table 1C.1 – West-side Flowline Topography Data Extracted from Rev D Field Layout (Corresponding to Figure 1.49) ........................................82 Table 1C.2 – Steel Catenary Riser Profile Data (Corresponding to Figure 1.50: Phifer 1998) ...................................................85 Table 1C.3 – Representative Ambient Sea Temperature Profile...........................................87 Table 1C.4 – Salinity and Density Profiles (Parts per Thousand)..........................................88 Table 1C.5 – Anticipated Bonga-area Water Current Velocities............................................89

FIGURES Figure 1C.1 – Flowline Topography for West-side Flowlines (Rev D Layout)........................80 Figure 1C.2 – Steel Catenary Riser Profile (Phifer 1998)......................................................84

Section 1 Appendix 1C Production Flowlines: Topography and Ambient Temperature Data

OPRM-2003-0302D

Page 79 of 89

31-December-2004


Unrestricted

Flowline Topography PFL - 11/12 (West South)-10" - 960 - 980

Water Depth, m

-1 000 -1 020 -1 040

Rev. D

-1 060 -1 080 -1 100 -1 120 0

500

1 000

Riser Base

1 500

2 000

2 500

Distance, m

Flowline Topography PFL - 08/09 (West North)-10" - 960 - 980

Water Depth, m

-1 000 -1 020 -1 040

Rev. D

-1 060 -1 080 -1 100 -1 120 0 Riser Base

500

1 000

1 500

2 000

2 500

3 000

Distance, m

Figure 1C.1 – Flowline Topography for West-side Flowlines (Rev D Layout) Section 1 Appendix 1C Production Flowlines: Topography and Ambient Temperature Data

OPRM-2003-0302D

Page 80 of 89

31-December-2004


Unrestricted

Flowline Topography PFL - 01/02 (East West)-10" - 960 - 980

Water Depth, m

-1 000 -1 020 Rev. D

-1 040 -1 060 -1 080 -1 100 -1 120 0

1 000

2 000

3 000

Riser Base

4 000

5 000

6 000

7 000

8 000

9 000

10 000

Distance, m

Flowline Topography PFL - 03/04/05/06 (East East)-12" - 960 - 980

Water Depth, m

-1 000 -1 020 Rev. D

-1 040 -1 060 -1 080 -1 100 -1 120 0

1 000

Riser Base

2 000

3 000

4 000

5 000

6 000

7 000

Distance, m

Figure 1C.1 – Flowline Topography for East-side Flowlines (Rev D Layout) (cont’d)


OPRM-2003-0302D

Page 81 of 89

31-December-2004


0 105.2632 184.2105 263.1579 315.7895 394.7368 447.3684 500 578.9474 631.5789 657.8947 710.5263 736.8421 763.1579 789.4737 815.7895 868.4211 1105.263 1263.158 1421.053 1447.368 1473.684 1500 1552.632 1578.947 1894.737 2000

West South, PFL - 11/12 Rev. D -1028 manifold -1026 -1024 -1022 -1020 -1018 -1016 -1014 -1012 -1011 -1012 -1012 -1010 -1008 -1006 -1004 -1002 -1000 -998 -998 -1000 -1002 -1000 -998 -996 -994 -994 riser base

Unrestricted

0 131.5789 473.6842 1052.632 1184.211 1236.842 1315.789 1342.105 1368.421 1421.053 1447.368 1500 1526.316 1578.947 1657.895 1710.526 1763.158 1789.474 1815.789 1842.105 1868.421 1921.053 1947.368 1973.684 1973.684 2000 2026.316 2078.947 2263.158 2289.474 2342.105 2368.421 2394.737

West North, PFL - 08/09 Rev. D -1000 manifold -998 -996 -998 -997 -998 -1000 -1002 -1004 -1004 -1002 -1002 -1000 -998 -996 -996 -998 -1000 -1002 -1004 -1006 -1006 -1004 -1002 -1000 -998 -996 -994 -995 -994 -992 -990 -988 riser base

Table 1C.1 – West-side Flowline Topography Data Extracted from Rev D Field Layout (Corresponding to Figure 1.49)


OPRM-2003-0302D

Page 82 of 89

31-December-2004


X, m 0 131.5789 263.1579 368.4211 447.3684 578.9474 657.8947 815.7895 894.7368 1000 1052.632 1131.579 1236.842 1342.105 1447.368 1526.316 1605.263 1684.211 1763.158 1868.421 1921.053 2000 2105.263 2210.526 2342.105 2447.368 2473.684 2500 2526.316 2578.947 2631.579 2684.211 2710.526 2736.842 2815.789 2894.737 2921.053 2947.368 2973.684 3000 3052.632 3105.263 3157.895 3210.526 3315.789 3394.737 3473.684 3552.632 3684.211 3789.474 3973.684 4052.632 4131.579 4210.526 4289.474 4368.421 4447.368 4552.632 4684.211 4815.789 4868.421 4947.368 5131.579 5184.211 5289.474 5394.737 5526.316 5710.526 5815.789 5973.684 6052.632 6157.895 6236.842

East West, PFL - 01/02 Rev. D -1106 manifold -1104 -1102 -1100 -1098 -1096 -1094 -1092 -1090 -1088 -1086 -1084 -1082 -1080 -1078 -1076 -1074 -1072 -1070 -1068 -1066 -1064 -1062 -1060 -1058 -1060 -1062 -1062 -1060 -1058 -1056 -1054 -1052 -1050 -1052 -1048 -1046 -1044 -1042 -1040 -1038 -1036 -1034 -1032 -1030 -1028 -1026 -1024 -1022 -1020 -1018 -1016 -1014 -1012 -1010 -1008 -1006 -1004 -1002 -1000 -998 -996 -994 -992 -990 -988 -986 -985 -986 -984 -982 -981 -982

Unrestricted

0 78.94737 236.8421 342.1053 447.3684 552.6316 631.5789 684.2105 710.5263 763.1579 789.4737 842.1053 947.3684 1078.947 1184.211 1289.474 1421.053 1526.316 1657.895 1789.474 1842.105 1868.421 1921.053 2000 2026.316 2052.632 2131.579 2184.211 2210.526 2236.842 2263.158 2289.474 2315.789 2342.105 2368.421 2421.053 2552.632 2605.263 2657.895 2684.211 2710.526 2736.842 2789.474 2868.421 3000 3289.474 3578.947 3684.211 3736.842 3789.474 3815.789 3842.105 3868.421 3894.737 3921.053 3973.684 4026.316 4131.579 4236.842 4315.789 4342.105 4394.737 4447.368 4631.579 4763.158 4868.421 4947.368 5000 5026.316 5078.947 5131.579 5210.526 5368.421

East East, PFL - 05/06 Rev. D -1038 manifold -1036 -1034 -1032 -1030 -1028 -1027 -1028 -1029 -1028 -1026 -1024 -1022 -1020 -1018 -1016 -1014 -1012 -1010 -1012 -1010 -1008 -1006 -1004 -1006 -1008 -1009 -1008 -1006 -1004 -1002 -1000 -998 -996 -994 -992 -990 -992 -990 -988 -986 -984 -982 -980 -978 -976 -978 -980 -982 -984 -986 -988 -990 -992 -994 -996 -998 -996 -994 -996 -998 -999 -998 -1000 -1002 -1004 -1006 -1008 -1010 -1008 -1009 -1008 -1010

Table 1C.1 – East-side Flowline Topography Data (cont’d)


OPRM-2003-0302D

Page 83 of 89

31-December-2004


6289.474 6342.105 6421.053 6473.684 6526.316 6578.947 6631.579 6684.211 6763.158 6947.368 7078.947 7131.579 7210.526 7342.105 7578.947 7736.842 7815.789 7868.421 7947.368 8131.579 8394.737 8657.895 8842.105 8921.053 9052.632 9105.263 9131.579 9157.895 9184.211 9210.526 9236.842 9263.158 9289.474

Unrestricted

-984 -986 -988 -990 -992 -994 -996 -998 -1000 -1002 -1002 -1003 -1002 -1001 -1002 -1004 -1006 -1008 -1009 -1010 -1012 -1010 -1008 -1006 -1004 -1002 -1000 -998 -996 -994 -992 -990 -988 riser base

5421.053 5447.368 5473.684 5552.632 5578.947 5605.263 5631.579 5657.895 5684.211 5736.842 5789.474 5894.737 5947.368 6105.263 6157.895 6184.211 6210.526 6236.842 6263.158 6289.474 6315.789 6342.105

-1012 -1014 -1014 -1016 -1018 -1016 -1014 -1012 -1010 -1008 -1006 -1004 -1002 -1000 -998 -996 -994 -992 -990 -988 -986 -984 riser base 0

Table 1C.1 – East-side Flowline Topography Data (cont’d)

3,500.00

3,000.00

2,500.00

Elevation, Feet

2,000.00

1,500.00

1,000.00

500.00

0.00 0

1000

2000

3000

4000

-500.00 Horizontal Distance, Feet

Figure 1C.2 – Steel Catenary Riser Profile (Phifer 1998) Section 1 Appendix 1C Production Flowlines: Topography and Ambient Temperature Data

OPRM-2003-0302D

Page 84 of 89

31-December-2004


Unrestricted

Riser top angle = 9.5 degrees from vertical Riser top elevation = 3230 feet

X (ft) 0 3.16 6.32 9.5 12.69 15.9 19.13 22.37 25.62 28.9 32.19 38.82 45.52 52.29 59.13 66.04 73.03 80.1 87.25 94.47 101.78 109.17 116.64 128.02 139.6 151.4 163.41 175.65 188.12 200.84 213.81 227.04 240.55 254.34 268.43 282.83 297.56 312.62 328.03 343.82 359.99 370.99 382.18 393.57 405.15 416.93 428.93 441.15 453.59 466.27 479.19 492.36 505.8 519.51 533.5 547.78 562.36 577.27 592.5 608.07 624

Y (ft) 3,230.00 3,211.14 3,192.28 3,173.43 3,154.58 3,135.73 3,116.89 3,098.05 3,079.21 3,060.37 3,041.54 3,003.88 2,966.23 2,928.60 2,890.98 2,853.37 2,815.77 2,778.19 2,740.63 2,703.08 2,665.55 2,628.03 2,590.53 2,534.31 2,478.14 2,422.01 2,365.92 2,309.88 2,253.90 2,197.97 2,142.10 2,086.29 2,030.55 1,974.87 1,919.27 1,863.75 1,808.32 1,752.97 1,697.73 1,642.58 1,587.55 1,550.93 1,514.37 1,477.87 1,441.42 1,405.05 1,368.74 1,332.51 1,296.35 1,260.27 1,224.28 1,188.39 1,152.59 1,116.89 1,081.30 1,045.83 1,010.49 975.27 940.2 905.28 870.52

X (ft) 640.3 656.99 674.08 691.6 709.55 727.96 746.85 766.24 786.15 806.61 827.64 849.26 871.51 894.4 917.97 942.24 967.24 993.01 1,019.56 1,046.92 1,075.12 1,104.17 1,134.10 1,164.91 1,196.60 1,229.16 1,262.58 1,296.82 1,331.79 1,367.40 1,385.39 1,403.46 1,421.57 1,439.67 1,457.68 1,475.50 1,492.99 1,509.98 1,526.20 1,541.31 1,544.16 1,546.94 1,549.65 1,552.28 1,554.82 1,557.28 1,559.64 1,561.89 1,564.04 1,566.06 1566.06 1569.94 1573.81 1577.68 1581.56 1585.43 1589.3 1593.18 1597.05 1600.93 1604.8 1953

Y (ft) 835.93 801.52 767.32 733.33 699.56 666.05 632.81 599.85 567.21 534.9 502.96 471.43 440.33 409.7 379.59 350.04 321.11 292.86 265.34 238.64 212.81 187.96 164.16 141.52 120.14 100.14 81.62 64.72 49.55 36.24 30.33 24.92 20.04 15.7 11.91 8.67 5.98 3.83 2.21 1.08 0.91 0.75 0.61 0.48 0.37 0.27 0.19 0.11 0.05 0 0 -0.09 -0.16 -0.23 -0.28 -0.33 -0.37 -0.4 -0.43 -0.45 -0.46 -0.46

Table 1C.2 – Steel Catenary Riser Profile Data (Corresponding to Figure 1.50: Phifer 1998)


OPRM-2003-0302D

Page 85 of 89

31-December-2004


1.0

Unrestricted

TEMPERATURE AND SALINITY PROFILES G FORRISTALL (1998) Temperature and salinity profiles were constructed by averaging all of the profiles on the US National Oceanographic Data Center CD-ROM for the area between 4° to 6°N and 4° to 6°E. Our experience is that deepwater temperatures do not vary much over such an area. All of the profiles were averaged over depth bins, and the standard deviation of the temperature in each bin was also found. The columns in Table 1C.3 give the mean depth in the bin, the mean temperature, the standard deviation of the temperature, the mean +/- the standard deviation and n, the number of observations in the depth bin. There are many more observations at shallow depths than deep in the water, but the standard deviations of the observations are also much higher at shallow depths. This variability is natural, largely due to seasonal effects in the temperature and river runoff in the salinity. The average temperatures and salinities are, for engineering purposes, nearly constant at great depth, and the average values in the tables can be used with confidence despite the small numbers of observations. Average values of seawater density were computed from the average temperature, salinity and depth, and are given in the last column of Table 1C.4. The density is given in units of kg/m3. Temperatures, salinities and densities at other depths can be found by interpolation in the table.


OPRM-2003-0302D

Page 86 of 89

31-December-2004


Unrestricted

Depth

Avg

Avg+std

Avg-std

Std

n

1.98

27.90

29.26

26.54

1.36

129.00

12.70

27.72

29.13

26.32

1.40

112.00

22.40

26.94

28.70

25.18

1.76

124.00

32.58

24.32

26.90

21.74

2.58

123.00

43.14

21.59

24.25

18.92

2.67

108.00

52.48

19.53

21.78

17.29

2.24

89.00

62.84

17.98

19.70

16.27

1.72

69.00

73.09

17.25

18.91

15.58

1.66

77.00

82.76

16.54

18.15

14.92

1.61

63.00

93.25

16.14

17.64

14.65

1.50

67.00

118.00

15.31

16.78

13.85

1.46

235.00

170.49

14.52

15.81

13.23

1.29

98.00

222.63

12.88

14.16

11.61

1.28

88.00

269.80

11.31

12.50

10.12

1.19

71.00

323.66

10.07

11.03

9.11

0.96

41.00

371.86

9.42

10.99

7.85

1.57

37.00

421.94

8.38

9.27

7.50

0.88

32.00

475.56

7.32

7.66

6.97

0.34

34.00

523.77

6.78

7.11

6.45

0.33

22.00

574.00

6.35

6.74

5.96

0.39

22.00

626.60

5.87

6.20

5.55

0.33

20.00

676.75

5.58

5.85

5.31

0.27

20.00

722.00

5.33

5.66

5.01

0.33

19.00

766.44

5.02

5.24

4.79

0.22

16.00

830.33

4.82

4.86

4.79

0.03

3.00

978.33

4.43

4.48

4.37

0.05

3.00

1000.00

4.43

4.43

4.42

0.00

2.00

Table 1C.3 – Representative Ambient Sea Temperature Profile


OPRM-2003-0302D

Page 87 of 89

31-December-2004


Unrestricted

Depth

Avg

Avg +std

Avg -std

Std

n

Avg Density

0.19

32.27

37.37

27.17

5.10

27.00

1020.4

11.78

33.70

35.47

31.94

1.77

18.00

1021.6

21.69

34.12

35.32

32.93

1.19

29.00

1022.1

31.81

35.30

35.71

34.88

0.42

26.00

1023.9

42.18

35.59

35.72

35.46

0.13

11.00

1024.9

50.75

35.69

35.73

35.64

0.05

12.00

1025.7

61.67

35.83

35.83

35.83

0.00

3.00

1026.2

74.36

35.73

35.74

35.72

0.01

11.00

1026.4

80.00

35.80

35.80

35.80

0.00

1.00

1026.6

95.40

35.76

35.76

35.76

0.00

5.00

1026.7

117.11

35.63

35.64

35.62

0.01

19.00

1026.9

168.38

35.49

35.50

35.49

0.00

13.00

1027.2

217.33

35.33

35.34

35.31

0.01

12.00

1027.5

270.77

35.13

35.14

35.12

0.01

13.00

1027.6

300.83

35.00

35.00

35.00

0.00

6.00

1028.4

381.00

34.83

34.84

34.83

0.00

8.00

1028.6

400.00

34.82

34.82

34.81

0.00

4.00

1028.6

483.80

34.71

34.71

34.71

0.00

5.00

1029.3

515.33

34.68

34.68

34.68

0.00

3.00

1029.6

585.00

34.69

34.69

34.69

0.00

1.00

1029.9

682.33

34.55

34.55

34.55

0.00

3.00

1030.3

700.00

34.57

34.57

34.57

0.00

2.00

1030.5

978.33

34.69

34.69

34.69

0.00

3.00

1032.0

1000.00

34.69

34.70

34.69

0.00

2.00

1032.0

Table 1C.4 – Salinity and Density Profiles (Parts per Thousand) Section 1 Appendix 1C Production Flowlines: Topography and Ambient Temperature Data

OPRM-2003-0302D

Page 88 of 89

31-December-2004


Unrestricted

Depth (m)

Current (m/s)

1100m

0.18

800

0.17

500

0.19

200

0.35

100

0.37

0

0.70

Table 1C.5 – Anticipated Bonga-area Water Current Velocities


OPRM-2003-0302D

Page 89 of 89

31-December-2004


Unrestricted

Section 2 Flow Assurance Production Constraints


OBJECTIVES ...............................................................................................................3

2.0

WELL STABILITY.........................................................................................................3

3.0

WELL KICK-OFF..........................................................................................................6

4.0

WAX DEPOSITION.......................................................................................................9 4.1

Flowline Wax Management..............................................................................11

4.2

East Flowlines..................................................................................................11

4.3

West Flowlines.................................................................................................11

5.0

WELLHEAD COOLDOWN .........................................................................................12

6.0

FLOWLINE/RISER COOLDOWN ...............................................................................12

7.0

FLOWLINE SLUGGING..............................................................................................13

8.0

CONCLUDING REMARKS .........................................................................................14

TABLES Table 2.1 – Minimum Well Production Rates for Stable, Controllable Flow.............................4 Table 2.2 – Manifold Pressures for Various Hot-oiling Scenarios, With and Without Gas Lift .6 Table 2.3 – Flowing Wellhead Temperatures .......................................................................10 Table 2.4 – Wax Pigging Frequencies for Turndown 1 Well/1 Flowline Production (Tsai et al, 2002) ................................................................................................11 FIGURES Figure 2.1 – Illustration of Multiple Solution Behaviour Associated with Well Instability ..........3 Figure 2.2 – Reservoir Pressure Required for Well Start-up to Stable Flowrates: Wells in Manifolds PM3 and PM4 (East-East Flowlines 3, 4, 5 and 6) ................7 Figure 2.3 – Reservoir Pressure Required for Well Start-up to Stable Flowrates: Wells in Manifold PM5 (East-West Flowlines 1 and 2) ........................................7 Figure 2.4 – Reservoir Pressure Required for Well Start-up to Stable Flowrates: Wells in Manifold PM 1 (West-North Flowlines 8 and 9) .....................................8 Figure 2.5 – Reservoir Pressure Required for Well Start-up to Stable Flowrates: Wells in Manifold PM2 (West-South Flowlines 11 and 12)..................................8


OPRM-2003-0302D

Page 1 of 25

30-April-2006


Unrestricted

Table of Contents (cont’d) FIGURES Figure 2.6 – Arrival Temperature as a Function of Rate, for 1 Well/1 Flowline Production Scenarios .......................................................................................12 Figure 2.7 – Riser Gas Lift Required for Slug Control: West Flowlines .................................13 Figure 2.8 – Riser Gas Lift Required for Slug Control: East 10in flowlines............................14 APPENDICES Appendix 2A – Well Design Basis – FDP Rev 5 ...................................................................15


OPRM-2003-0302D

Page 2 of 25

30-April-2006


1.0

Unrestricted

OBJECTIVES The principal objective’ of this study is to quantify flow assurance constraints for production forecasting, particularly minimum acceptable flowrates per well and flowline. It is important to note that results herein represent the absolute edge of the flow assurance envelope, with essentially all conservatism in analysis removed. As such, this analysis is intended for Shell Bonga project staff only and should not be shared with Engineer, Procure, Install and Construct (EPIC) contractors, who could misinterpret/misuse these results as a basis for systems design. The key end-users of these results are: •

Bonga reservoir engineering staff (Bonga Integrated Studies Team (BIST)), to enable assessment and risking of production forecasts with respect to flow assurance

•

Bonga operations staff, to outline the operating envelope for relevant flow assurance risks

Noting that well stability is found to be the governing constraint for minimum well flowrate, the following analysis approach is used:

2.0

(1)

Identification of minimum well rates for stable flow on a well-by-well basis.

(2)

Verification of flow assurance requirements for wax, hydrate and slugging at the minimum stable rates.

WELL STABILITY With respect to minimum well production rates, a key consideration is well stability, particularly so for the larger tubing of the Bonga wells (5 1/2in and 6 5/8in). As illustrated in Figure 2.1, multiphase wells exhibit multivalued behaviour at lower production rates (ie two possible flowrates at the same applied pressure drop). The low flowrate solution represents a liquid loaded well (usually with slugging at the wellhead), while the high flowrate solution has less liquid hold-up and a larger frictional pressure drop. Hence, rates below the instability threshold (the minimum in Figure 2.1) are generally not controllable, as the past history of the well’s liquid loading will determine whether the low or high-flowrate solutions are attained. In general, if the well flowrate is reduced (from a higher rate) to below the instability threshold (by choking), the well will load-up with liquid and shut-in if the wellhead pressure is not reduced.

Well ∆p

Instability Production rate

Figure 2.1 – Illustration of Multiple Solution Behaviour Associated with Well Instability Section 2 Flow Assurance Production Constraints

OPRM-2003-0302D

Page 3 of 25

30-April-2006


Unrestricted

Well

Tubing Size

Minimum Rate for Stability

690p1 (horizontal)

5 1/2in

2.5MBLPD

690p2

5 1/2in

3.0

690p3

5 1/2in

2.0

690p4

5 1/2in

2.0

702p2

5 1/2in

5.0

702p4 (horizontal)

6 5/8in

7.0

702p3

5 1/2in

4.5

702p5

5 1/2in

4.5

702p6

5 1/2in

4.5

702p7

5 1/2in

5.0

702p9 (horizontal)

6 5/8in

6.5

702p10 (horizontal)

6 5/8in

7.0

702p14

5 1/2in

2.0

702p15 (horizontal)

6 5/8in

7.0

710p1

5 1/2in

5.0

710p2

5 1/2in

2.0

710p3

5 1/2in

3.5

710p4

5 1/2in

4.5

803p1

5 1/2in

4.5

803p2

5 1/2in

3.0

803p3

5 1/2in

5.0

Table 2.1 – Minimum Well Production Rates for Stable, Controllable Flow


OPRM-2003-0302D

Page 4 of 25

30-April-2006


Unrestricted

In this report, well stability for Bonga was assessed on a well-by-well basis using Olga2000, with thermal well modelling tuned to match WellTemp predictions. Well design parameters (ie productivities, deviation profiles, reservoir pressures etc) are based on Bonga Field Development Plan (FDP) Rev 5 (summarised in Appendix 2A). The procedure for assessing well stability is as follows: (1)

Initial conditions consist of a gas-filled well at ambient geothermal conditions.

(2)

Reduce flowing wellhead pressure in 25psi increments until sustained production occurs. If the Flowing Wellhead Pressure (FWHP) is too high, the well will shut-in after liquid travels up the wellbore.

(3)

The minimum acceptable flowrate for a well is the smallest sustained production that can occur as calculated in Step (2). Note:

Production rates below the minimum rate for stability may simply be unattainable (even if sufficient reservoir pressure exists), as additional choking can cause the well to load-up and abruptly shut-in. That is, intermediate rates below the threshold are unstable and may not be observable in practice (much like the inherent instability of a pin balanced on its tip).

As shown in Table 2.1, the minimum well rates for stability vary between 2 to 7MBLPD. Notes: (1)

The key discriminator between the lower and higher thresholds is the well tubing, since lower gas velocities obtained for the larger 6 5/8in tubing are more conducive to well load-up and instability.

(2)

The only wells with 6 5/8in production tubing are 702p9, 702p15, 702p10, 702p4, which are also horizontal completions (690p1 is the only other horizontal well, but with 5 1/2in tubing).

Noting the complexities in modelling multiphase flow and the discrete 25psi WWellhead Pressure (WHP) steps used in analysis, the limiting rate for both tubing sizes is interpreted as the stability threshold for controllable steady-state production: •

5 1/2in: 5MBLPD minimum rate for stability

•

6 5/8in: 7MBLPD minimum rate for stability

These thresholds are consistent with previous steady-state analysis (analogous to Figure 2.1), summarised in van Gisbergen, 1999.


OPRM-2003-0302D

Page 5 of 25

30-April-2006


3.0

Unrestricted

WELL KICK-OFF Noting the relatively low Bonga reservoir pressures and the importance of waterflood for pressure maintenance, well kick-off requirements are evaluated with respect to depletion predictions for each well. Figures 2.2 to 2.5 show the reservoir pressure required to start each well against a minimum attainable wellhead pressure of 600psi, relative to the minimum reservoir pressure (over the field life) predicted by GAP (GAP is a subsurface software used to model wells and flowline networks). The minimum reservoir pressures tend to occur in mid-life; assuming effective waterflood, the reservoir pressure rises later in field life. Note that a wellhead backpressure of 600psi requires availability of gas lift if starting up into a hot oiled (or high water cut) flowline. The manifold pressures obtained during hot-oiling of the (worst-case) east flowloops are summarised in Table 2.2. To obtain manifold pressures in the range of 600psi, the hot-oiling rate will have to be turned down (eg to 10MBOPD) if a well is started up while hot-oiling. Further, gas lift (of the return riser) is also required to reduce the riser hydrostatic head.

Flowloop

Hot-oiling Rate

Gas Lift

Manifold P

E-E (12in)

50MBOPD

0MMscfd

1640psia

E-E

50

10

1025

E-E

10

0

1545

E-E

10

20

500

E-W (10in) 50

0

1870

E-W

50

10

1280

E-W

10

0

1550

E-W

10

10

500

Table 2.2 – Manifold Pressures for Various Hot-oiling Scenarios, With and Without Gas Lift


OPRM-2003-0302D

Page 6 of 25

30-April-2006


Unrestricted

5000 4500

Required reservoir P for stable flow [psia]

4000 3500 3000

0% wc 50% wc 80% wc Min Reservoir P

2500 2000 1500 1000 500 0 702p14

702p5

702p9

702p10

803p2

Figure 2.2 – Reservoir Pressure Required for Well Start-up to Stable Flowrates: Wells in Manifolds PM3 and PM4 (East-East Flowlines 3, 4, 5 and 6)

Figure 2.3 – Reservoir Pressure Required for Well Start-up to Stable Flowrates: Wells in Manifold PM5 (East-West Flowlines 1 and 2)


OPRM-2003-0302D

Page 7 of 25

30-April-2006


Unrestricted

Figure 2.4 – Reservoir Pressure Required for Well Start-up to Stable Flowrates: Wells in Manifold PM 1 (West-North Flowlines 8 and 9) 4000 3500

Required reservoir P for stable flow [psia]

3000 2500

0% wc 50% wc 80% wc Min Reservoir P

2000 1500 1000 500 0 690p4

702p2 702p6 702p7 702p15 710p4 803p1 803p3

Figure 2.5 – Reservoir Pressure Required for Well Start-up to Stable Flowrates: Wells in Manifold PM2 (West-South Flowlines 11 and 12)


OPRM-2003-0302D

Page 8 of 25

30-April-2006


Unrestricted

As shown in Figures 2.2 to 2.5, all wells except for 803p2 are able to start up against a 600psi wellhead backpressure, for the worst-case scenario of an initially liquid-loaded well and the minimum reservoir pressures of each body over the field life. The liquid-loaded initial condition is based on the restart scenario for a well which falls below the stability threshold and loads up with liquid. For most wells, 500 to 750psi of ‘spare’ reservoir pressure capacity is available, with lesser margin for 702p2, 702p5, and the 803 wells. For 803p2, extra surveillance attention will be needed to avoid loading it with liquid, as it may not be restarted at the minimum 803 reservoir pressure. Also, the phasing of 803p2 with respect to stronger wells should be assessed to assure that its production will not be backed out. These results underscore the importance of effective waterflood for reservoir pressure maintenance, as assumed in the GAP predictions. In early field life, all wells are strong enough to start-up against a dead-oil filled riser (with the possible exception of 803p2, which has a minimal pressure margin). In fact, this additional riser hydrostatic head is needed for chilly choke management in early life. Thus, an important surveillance activity will be to track the backpressure requirements of individual wells, which will be necessary whenever wells in different phases of life are to be started up and produced into the same flowline.

4.0

WAX DEPOSITION The basic wax management strategy for Bonga is to flow above the Critical Wax Deposition Temperature (CWDT) in the wellbore and to pig flowlines during planned shutdown operations. Recent wax analysis (Tsai et al 2002) indicates a maximum CWDT of 43°C (109°F) for B2ST3-702, at a (minimum) wellhead pressure of 400psi. As shown in Table 2.3, at the minimum rates for well stability (5MBLPD for 5 1/2in tubing; 7MBLPD for 6 5/8in tubing), several wells are at or near the onset point for wellhead wax deposition: 690p4, 702p3, 702p6, 702p7, 710p2. Hence, long-term turndown production (ie below 10MBLPD) should be avoided for these wells. Noting the relatively low deposition rate characteristic of the Bonga fluids, production of these lower-T wells may be tolerable for shorter-term durations to accommodate transient operations such as well testing or well flowline switching. Note: In Table 2.3, all other wells are outside the wax deposition envelope at the minimum rates for stable flow.


OPRM-2003-0302D

Page 9 of 25

30-April-2006


Unrestricted

Well

Minimum Production Rate (MBLPD)

Turndown Wellhead T (°F)

690p1

5

132

690p2

5

114

690p3

5

119

690p4

5

108

702p2

5

115

702p3

5

106

702p4 (6 5/8in)

7

136

702p5

5

117

702p6

5

109

702p7

5

99

702p9 (6 5/8in)

7

121

702p10 (6 5/8in)

7

123

702p14

5

117

702p15 (6 5/8in)

7

116

710p1

5

121

710p2

5

105

710p3

5

112

710p4

5

126

803p1

5

140

803p2

5

141

803p3

5

139

Table 2.3 – Flowing Wellhead Temperatures Table 2.3 gives the flowing wellhead temperatures (24 hours after warm-up) at minimum stable production rates of 5MBLPD (5 1/2in tubing wells), and 7MBLPD (6 5/8in tubing wells). Temperatures below CWDT = 109°F are highlighted.


OPRM-2003-0302D

Page 10 of 25

30-April-2006


4.1

Unrestricted

Flowline Wax Management Regarding flowline/riser wax management, the basic operating strategy is to pig flowlines for wax during scheduled or planned shut-ins, during hot or dry-oiling operations. Based on the updated wax analysis in Tsai et al, 2002, pigging frequency requirements at turndown conditions are shown in Table 2.4. The Flowing Wellhead Temperature (FWHT) values of 100°F and 120°F are based on the minimum FWHT observed at rates of 5MBLPD and 7MBLPD, respectively (refer to Table 2.3, with slight exception of 116°F for 702p15 at 7MBLPD). Recall that the Bonga Basis of Design (BoD) specifies a minimum turndown rate of 10MBLPD per flowline, so that these results apply to operations outside the design envelope.

4.2

East Flowlines For wells with FWHTs in the order of 100°F, extended turndown production at 5MBLPD (one well into one flowline) is not feasible for both East flowline loops, as 8 to 10 piggings per year would be required. This would likely involve system shut-ins (or temporary well curtailment) solely for wax management, if planned shutdowns are less frequent than once per month (as is expected in availability analysis). At 7MBLPD, the pigging frequency decreases to six per year (East 10in) and 4 per year (East 12in), due to both the shorter residence time in the flowline and the higher wellhead temperature (120°F, refer to Table 2.4). The feasibility of such pigging frequencies will have to be determined based on operating experience and shutdown statistics (ie number of pigging opportunities). During surveillance, wax analysis of the 690 wells (690p1, 690p2, 690p3) producing into the (worst-case) East 10in flowlines PFL 1 and 2 can be used (along with thermal model benchmarking) to further refine the 690 specific pigging requirements.

4.3

West Flowlines Due to their much shorter offset, the West flowlines’ wax requirements are less severe, with four piggings per year required for 5MBLPD production (one well into 1 flowline). Note: In Table 2.3, the wellhead temperature for most wells exceeds 100°F at 5MBLPD, so that this pigging frequency represents the upper limit. As discussed above, post-start-up wax analysis should be included in the surveillance programme, especially if such turndown production is anticipated for some wells. Rate (MBLPD)

FWHT (°F)

Pigging Frequency (No per Year)

East 10in (PFL 1, 2)

5

100

10

7

120

6

East 12in (PFL 3, 4, 5, 6)

5

100

8

7

120

4

West 10in (PFL 8, 9, 11, 12)

5

100

4

7

120

1

Flowline

Table 2.4 – Wax Pigging Frequencies for Turndown 1 Well/1 Flowline Production (Tsai et al, 2002) Section 2 Flow Assurance Production Constraints

OPRM-2003-0302D

Page 11 of 25

30-April-2006


5.0

Unrestricted

WELLHEAD COOLDOWN Wellhead area equipment (tree, jumper, manifold) is to be insulated to meet a cooldown specification: 120°F to 73°F in 12 hours (upstream of choke) and 120°F to 63°F in 12 hours (downstream of choke). As shown in Table 2.3, roughly half of the wells will meet the 120°F start temperature, even at turndown conditions (5 to 7MBLPD), and hence provide 12+ hours of wellhead area cooldown. Based on analogy with cooldown performance of cylindrical components (eg well jumpers), a start temperature of 100°F will provide roughly 8 hours of cooldown. Since the colder wells produce at 100°F or higher at turndown, these wells will provide at least 8 hours of cooldown. Noting that the cooldown criteria is designed to provide time for wellhead Methanol (MeOH) flushing of up to 16 wells, it is expected that 8 hours of cooldown is sufficient for a limited number of producing wells, with the exact number based on actual MeOH treatment times determined via surveillance.

6.0

FLOWLINE/RISER COOLDOWN The production flowlines and riser are governed by the following cooldown specifications: •

West-side 10in flowlines: –

•


East-side 10in and 12in flowlines: –


Arrival Temperature per Flowline (ºF)

160 702p15/PF11 803p1/PF12

140

702p4/PF1 690p1/PF2 120

702p9/PF3 702p10/PF6

100

80

60 0

10

20

30

40

50

60


Figure 2.6 – Arrival Temperature as a Function of Rate for 1 Well/1 Flowline Production Scenarios


OPRM-2003-0302D

Page 12 of 25

30-April-2006


Unrestricted

Note: The West and East start temperatures differ due to offset differences, while the end temperatures differ due to the effects of line-packing (with an assumed 10 minimum choke closing time). Arrival temperatures as a function of production rate for representative early-life one well/one flowline combinations are shown in Figure 2.6. Note: The required riser base start temperatures translate to arrival temperatures of approximately 80°F (East) and 90°F (West). In Figure 2.6, the 12-hour cooldown requirement corresponds to minimum rates of approximately 5MBLPD (West) and 7MBLPD (East). These results are also consistent with the generalised thermal modelling in Tsai et al, 2002, for a wellhead temperature of 120°F.

7.0

FLOWLINE SLUGGING A key consideration for turndown production at Bonga is control of terrain slugging, noting recent slug-induced operational difficulties in the Gulf of Mexico (GoM). For Bonga, riser gas lift with up to 25MMscfd for a given riser is available for slug control at turndown, but it is important to note that the total gas lift compression capacity is 65MMscfd (Bonga BoD). Hence, only a limited number of flowlines may be operated simultaneously in an extended turndown condition. As illustrated in Figures 2.7 and 2.8, terrain slug control requires the 25MMscfd gas lift capacity at production rates of 5MBLPD (West) and 7MBLPD (East). Note: The minimum flowrate for the East 12in flowlines is also approximately 7MBLPD, with residual 50bbl slugs observed even at high gas lift rates (compared to complete slug suppression for the other flowlines).

20


0% wc 50% wc 15

80% wc

10

5

0 0

10

20

30

40


Figure 2.7 – Riser Gas Lift Required for Slug Control: West Flowlines


OPRM-2003-0302D

Page 13 of 25

30-April-2006


Unrestricted

40


0% wc 50% wc 30

80% wc

20

10

0 0

10

20

30

40


Figure 2.8 – Riser Gas Lift Required for Slug Control: East 10in flowlines

8.0

CONCLUDING REMARKS To better quantify the Bonga operating envelope, key flow assurance issues have been analysed at turndown production conditions, including: well stability, well and flowline wax deposition, wellhead and flowline cooldown, and flowline slugging. Collectively, the edge of the operating envelope is defined by the following production constraints, which must be satisfied simultaneously: •

5 1/2in wells: Rates per well ≥ 5MBLPD

•

6 5/8in wells: Rates per well ≥ 7MBLPD

•

West flowlines (PFL 8, 9, 11, 12): Rates per flowline ≥ 5MBLPD

•

East flowlines (PFL 1, 2, 3, 4, 5, 6): Rates per flowline ≥ 7MBLPD

Interestingly, each of these flow assurance requirements tends to involve a similar minimum rate constraint, indicating that a variety of operational difficulties may occur if these constraints are violated. In production forecasting, lower rates may be feasible but should be risked for flow assurance. Noting the high cost of deferment/intervention, it is recommended to maintain (through operational solutions and careful well sequencing) the original minimum design rates of 10MBLPD per well and 10MBLPD per flowline. Finally, it is important to note that these results are based on complex predictive modelling – surveillance, sampling and model benchmarking will be required to precisely define the actual operating envelope.


OPRM-2003-0302D

Page 14 of 25

30-April-2006


Unrestricted

Appendix 2A Well Design Basis – FDP Rev 5 Compiled by Kelda McFee. 690 Wells 690p1 Well Trajectory MD(ft) SS WD 3581 2 3600

TVD (ft) SS 3581 3600

Inc (deg) 0 0

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61

3700 3800 3900 4000 4100 4200 4300 4400 4500 4600 4700 4800 4900 5000 5100 5200 5300 5400 5500 5600 5700 5800 5900 6000 6100 6200 6299.7 6399.1 6497.6 6594.2 6688.4 6780.5 6872.9 6965.2 7057.6 7150 7242.4 7334.8 7427.1 7519.5 7611.9 7704.3 7796.7 7889.1 7981.4 8073.8 8166.2 8258.6 8351 8435.6 8443.3 8533.6 8619.8 8701.3 8777.4 8847.5 8911.2 8967.9 9017.3

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2.5 5 7.5 12.5 17.5 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.51 22.93 27.93 32.93 37.93 42.93 47.93 52.93 57.93 62.93

3700 3800 3900 4000 4100 4200 4300 4400 4500 4600 4700 4800 4900 5000 5100 5200 5300 5400 5500 5600 5700 5800 5900 6000 6100 6200 6300 6400 6500 6600 6700.3 6800 6900 7000 7100 7200 7300 7400 7500 7600 7700 7800 7900 8000 8100 8200 8300 8400 8500 8591.6 8600 8700 8800 8900 9000 9100 9200 9300 9400

hz Summary Profile PI@PSSS Initial Pavg

20 4511

bbl/day psia psia

Initial GOR T@midperfs

605 160

scf/bbl °F

SSSV Depth ML (ft) ID (in) Length (ft) TVD (ft) SS

2300 4.56 9.72 9447.2

Tubing size (I) AHD ft (SS) 12931

Roughness Geothermal profile Heat transfer coefficient

ID (in) 4.892

OD (in) 5 1/2

0.0018 Linear between reservoir and seabed 2

Section 2 Appendix 2A Well Design Basis – FDP Rev 5

OPRM-2003-0302D

Page 15 of 25

31-December-2004


Unrestricted

690 Wells (cont’d) 690p1 Well Trajectory MD(ft) SS

TVD (ft) SS

Inc (deg)

62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103

9058.8 9092.3 9117.7 9138.2 9159 9180.1 9200.6 9221.4 9240.1 9242.2 9258.8 9268.3 9278.8 9289.4 9299.9 9310.4 9320.9 9323.7 9325.1 9325.4 9330.9 9340.6 9350.4 9360.1 9369.9 9379.7 9389.4 9399.2 9406.5 9408.7 9411.2 9413.2 9416.8 9420.5 9424.1 9427.8 9431.4 9435.1 9438.7 9442.4 9446 9447.2

67.93 72.93 78 78 78 78 78 78 78 78.5 83.97 83.97 83.97 83.97 83.97 83.97 83.97 83.97 84.54 84.4 84.4 84.4 84.4 84.4 84.4 84.4 84.4 84.4 84.4 85.65 87.91 87.91 87.91 87.91 87.91 87.91 87.91 87.91 87.91 87.91 87.91 87.91

Well Trajectory MD(ft) SS WD 3581.0 2 5500.0 3 6400.0 4 7789.4

TVD (ft) SS 3581.0 5500.0 6310.3 7292.7

Inc (deg) 0.00 0.00 45.00 45.00

5 1 2 3 4

7521.0 8286.2 8404.3 8678.0 8775.0

65.00 65.00 83.46 83.46 83.46

9500 9600 9701.3 9800 9900 10001.3 10100 10200 10289.9 10300 10409.6 10500 10600 10700 10800 10900 11000 11026.7 11041.4 11044.1 11100 11200 11300 11400 11500 11600 11700 11800 11874.8 11900 11945.5 12000 12100 12200 12300 12400 12500 12600 12700 12800 12900 12931.2

hz

690p2

8189.4 10000.0 10435.0 12837.6 13689.2

Summary Profile Tubing size (I) PI@PSSS Initial Pavg Initial GOR T@midperfs

25 4279 605 146


2300 4.56 9.72 5881.0

bbl/day psia psia scf/bbl °F

AHD ft (SS) 13689


ID (in) 4.892

OD (in) 5 1/2

0.0018 linear 2


OPRM-2003-0302D

Page 16 of 25

31-December-2004


Unrestricted

690 Wells (cont’d) S690p3 Well Trajectory MD(ft) SS

TVD (ft) SS

Inc (deg)

Sidetrack of 702p4 WD 3568.0 1 7300.0 2 7686.6 3 10589.6

3568.0 7236.6 7489.2 9023.0

0.00 40.00 58.11 58.11

Summary Profile

S690p4 Well Trajectory MD(ft) SS

TVD (ft) SS

Inc (deg)

Sidetrack of 702p2 WD 3369.0 1 6150.0 2 7415.4 3 10841.9

3369.0 6147.6 7090.1 8268.0

0.00 7.50 69.89 69.89

Tubing size (I) PI@PSSS Initial Pavg Initial GOR T@midperfs

18 4365 605 155


2300 4.56 9.72 5868.0


AHD ft (SS) 10590


ID (in) 4.892

OD (in) 5 1/2

0.0018 linear 2

Summary Profile Tubing size (I) PI@PSSS Initial Pavg Initial GOR T@midperfs

19 4415 605 137

SSSV Depth ML (ft) ID (in) Length (ft) Depth (ft) SS

2300 4.56 9.72 5669.0


AHD ft (SS) 10842


ID (in) 4.892

OD (in) 5 1/2

0.0018 linear 2

702 Wells 702p2 Well Trajectory MD(ft) SS WD 3369.0 2 5850.0

TVD (ft) SS 3369.0 5850.0

Inc (deg) 0.00 0.00

3 4 5 6 7 8 9

6149.1 6838.1 8121.9 8387.4 8467.0 8544.0 8838.0

7.50 47.03 47.03 30.00 30.00 30.00 30.00

TVD (ft) SS

Inc (deg)

6150.0 6940.6 8824.1 9164.7 9256.6 9345.5 9685.0

S702p3 Well Trajectory MD(ft) SS Sidetrack of 710p1 WD 3276.0 1 5950.0 2 7095.7 3 9150.2

Summary Profile PI@PSSS Initial Pavg

100 4201


589.57 scf/bbl 143 °F

SSSV Depth ML (ft) ID (in) Length (ft) TVD SS (ft)

2300 4.56 9.72 5669.0

bbl/day psia psia



ID (in) 4.892

OD (in) 5 1/2


Summary Profile

3276.0 5948.7 6915.0 8050.0

0.00 7.50 56.46 56.46

PI@PSSS Initial Pavg Initial GOR T@midperfs

20 4052 589.57 130


2300 4.56 9.72 5576.0




ID (in) 4.892

OD (in) 5 1/2



OPRM-2003-0302D

Page 17 of 25

31-December-2004


Unrestricted

702 Wells (cont’d) 702p5 Well Trajectory MD(ft) SS 1 3343.0 2 5500.0 3 6333.2 4 9831.1

Summary Profile TVD (ft) SS 3343.0 5500.0 6261.7 8875.0

Inc (deg) 0.00 0.00 41.66 41.66

S702p6 Well Trajectory MD(ft) SS Sidetrack of 702p15 WD 3359.0 1 5600.0 2 6790.5

3359.0 5600.0 6586.5

0.00 0.26 59.64

3

8216.0

59.64

10014.0


35 4351 589.57 151


2300 4.56 9.72 5643.0




ID (in) 4.892

OD (in) 5 1/2


Summary Profile TVD (ft) SS

Inc (deg)

S702p7 Well Trajectory MD(ft) SS Sidetrack of 803p3 WD 3359.0 1 6500.0 2 7840.1 3 10528.0 4 10778.5 5 11349.1

3359.0 6494.6 7357.1 7850.0 7869.1 7851.0

0.00 17.50 79.43 79.43 91.82 91.82

702p14 Well Trajectory MD(ft) SS WD 3327 1 5500 2 6500 3 6909.6 4 7409.6 5 13957.9 6 14267

TVD (ft) SS 3327 5500 6377.8 6641.1 6870.2 8565 8645

Inc (deg) 0 0 50 50 75 75 75


15 4129 589.57 136


2300 4.56 9.72 5659.0




ID (in) 4.892

OD (in) 5 1/2


Summary Profile TVD (ft) SS

Inc (deg) PI@PSSS Initial Pavg Initial GOR T@midperfs

14 4032 589.57 120


2300 4.56 9.72 5659.0




ID (in) 4.892

OD (in) 5 1/2

0.0018 linear 2

Summary Profile Tubing size (I) AHD ft (SS) 14267


13 4247 589.57 144

SSSV Depth ML (ft)

2300

Roughness

0.0018

ID (in) Length (ft) TVD SS (ft)

4.56 9.72 5627.0

Geothermal profile Heat transfer coefficient

linear 2


ID (in) 4.892

OD (in) 5 1/2


OPRM-2003-0302D

Page 18 of 25

31-December-2004


Unrestricted

702 Wells (cont’d) 702p9 Well Trajectory MD(ft) SS WD 3331.0 2 3721.4 3 4003.9 4 4286.0 5 4568.8 6 4851.4 7 5133.7 8 5416.1 9 5599.8 10 5678.0 11 5764.0 12 5866.0 13 5960.0 14 6051.0 15 6143.0

TVD (ft) SS 3331.0 3721.4 4003.9 4286.0 4568.8 4851.4 5133.7 5416.1 5599.8 5678.0 5764.0 5866.0 5959.9 6050.4 6141.3

Inc (deg) 0.00 0.09 0.35 0.31 0.18 0.18 0.18 0.22 0.13 0.45 0.30 1.33 3.94 7.13 10.65


16 17 18

6240.0 6333.0 6426.0

6235.9 6324.2 6408.6

15.04 21.29 28.27

Well Design/ Tubing Size

19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70

6520.0 6615.0 6709.0 6804.0 6896.0 6989.0 7151.0 7244.0 7341.0 7434.0 7526.0 7620.0 7713.0 7807.0 7901.0 7995.0 8089.0 8184.0 8278.0 8374.0 8468.0 8561.0 8656.0 8666.2 8750.0 8820.0 8846.0 8920.0 9020.0 9120.0 9220.0 9320.0 9420.0 9520.0 9620.0 9720.0 9820.0 9853.7 9920.0 10020.0 10120.0 10195.5 10216.0 10220.0 10266.0 10320.0 10392.2 10420.0 10520.0 10620.0 10720.0 10820.0

6487.6 6560.2 6629.5 6700.0 6767.0 6831.4 6935.9 6991.1 7044.7 7095.2 7146.9 7201.9 7259.1 7320.1 7383.7 7448.9 7516.0 7585.1 7655.0 7729.5 7804.2 7878.5 7955.3 7963.4 8029.3 8085.1 8106.3 8165.8 8241.8 8312.2 8376.7 8434.8 8486.1 8530.2 8567.0 8596.1 8617.3 8622.7 8632.4 8647.0 8661.7 8672.7 8675.7 8676.3 8683.0 8689.7 8694.7 8695.8 8699.7 8703.6 8707.5 8711.4

37.12 43.15 41.96 42.11 44.43 47.94 51.95 55.29 57.73 56.47 55.16 53.17 51.00 48.10 46.70 45.47 43.32 43.37 40.65 37.46 37.36 36.49 35.73 37.92 38.51 35.78 34.77 38.21 42.86 47.51 52.16 56.81 61.46 66.11 70.76 75.41 80.05 81.59 81.59 81.59 81.59 81.59 81.59 81.59 81.59 84.23 87.77 87.77 87.77 87.77 87.77 87.77

Summary Profile PI@PSSS Initial Pavg Initial GOR T@midperfs

100 4292 589.57 147

SSSV Depth ML (ft) ID (in) Length (ft)* TVD SS (ft)

2300 4.56 9.72 5631.0

AHD ft (SS) 3545 5545 5576.72 10576.72 12084.7

ID (in) 4.892 5.921 4.562 5.921 4.892



OD (in) 6 7.191 7.99 7.191 6.05

Length (ft) 45 2000 31.72 5000 1508.0

Description Tubing Hanger Tubing SSSV* Tubing Excluder Screens


OPRM-2003-0302D

Page 19 of 25

31-December-2004


Unrestricted


TVD (ft) SS

Inc (deg)

71 72 73 74 75 76 77 78 79 80 81 82 83

8715.3 8719.1 8723.0 8726.9 8730.8 8734.7 8738.6 8742.5 8746.4 8750.3 8754.2 8758.1 8760.6

87.77 87.77 87.77 87.77 87.77 87.77 87.77 87.77 87.77 87.77 87.77 87.77 87.77

702p15 Well Trajectory MD(ft) SS WD 3359.0 2 3369.0 3 3420.0 4 3520.0 5 3620.0

TVD (ft) SS 3359.0 3369.0 3420.0 3520.0 3620.0

Inc (deg) 0.00 0.00 0.00 0.00 0.00


135 4168 589.57 140

6 7 8 9 10 11

3720.0 3820.0 3920.0 4020.0 4120.0 4220.0

3720.0 3820.0 3920.0 4020.0 4120.0 4220.0

0.00 0.00 0.00 0.00 0.00 0.00


2300 4.56 9.72 5659.0

12 13 14 15 16

4320.0 4420.0 4520.0 4620.0 4720.0

4320.0 4420.0 4520.0 4620.0 4720.0

0.00 0.00 0.00 0.00 0.00


17 18 19

4820.0 4920.0 5020.0

4820.0 4920.0 5020.0

0.00 0.00 0.00

Well Design/ Tubing Size

20 21

5120.0 5220.0

5120.0 5220.0

0.00 0.00

AHD ft (SS)

ID (in)

OD (in)

Length (ft)

Description

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53

5320.0 5420.0 5520.0 5620.0 5720.0 5820.0 5920.0 6020.0 6120.0 6220.0 6320.0 6420.0 6520.0 6620.0 6720.0 6820.0 6920.0 7020.0 7120.0 7170.8 7220.0 7320.0 7420.0 7520.0 7620.0 7720.0 7820.0 7920.0 8020.0 8120.0 8220.0 8320.0

5320.0 5420.0 5520.0 5620.0 5720.0 5820.0 5920.0 6020.0 6120.0 6220.0 6320.0 6420.0 6520.0 6620.0 6720.0 6820.0 6920.0 7020.0 7120.0 7170.8 7220.0 7319.6 7418.0 7514.6 7608.6 7699.2 7785.8 7867.8 7944.4 8015.1 8079.4 8136.8

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.46 7.46 12.46 17.46 22.46 27.46 32.46 37.46 42.46 47.46 52.46 57.46

3545 5545 5576.72 10576.72 11264.1

4.892 5.921 4.562 5.921 4.892

6 7.191 7.99 7.191 6.05

45 2000 31.72 5000 687.4

Tubing Hanger Tubing SSSV* Tubing Excluder Screens

10920.0 11020.0 11120.0 11220.0 11320.0 11420.0 11520.0 11620.0 11720.0 11820.0 11920.0 12020.0 12084.7

Summary Profile

Summary Profile


0.0018 Linear between reservoir and seabed 2 Btu/hr ft2/°F


OPRM-2003-0302D

Page 20 of 25

31-December-2004


Unrestricted


TVD (ft) SS

Inc (deg)

54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89

8186.9 8229.2 8263.5 8289.4 8299.4 8307.9 8325.3 8342.6 8347.0 8352.0 8360.7 8373.7 8378.2 8379.7 8384.1 8388.4 8392.7 8397.1 8401.4 8405.8 8410.1 8414.4 8418.8 8421.0 8424.1 8429.5 8436.0 8448.9 8461.8 8474.7 8487.6 8500.5 8513.5 8526.4 8539.3 8545.0

62.46 67.46 72.46 77.46 80.00 80.00 80.00 80.00 80.00 80.00 80.00 84.46 87.51 87.51 87.51 87.51 87.51 87.51 87.51 87.51 87.51 87.51 87.51 87.51 85.09 82.58 82.58 82.58 82.58 82.58 82.58 82.58 82.58 82.58 82.58 82.58

702p10 Well Trajectory MD(ft) SS WD 3178.0 2 3558.0 3 3808.0 4 4094.0 5 4379.0 6 4641.0 7 4946.0 8 5217.0 9 5450.0

TVD (ft) SS 3178.0 3558.0 3808.0 4094.0 4379.0 4641.0 4946.0 5217.0 5450.0

Inc (deg) 0.00 0.62 0.40 0.18 0.22 0.13 0.40 0.75 0.40

10 11 12 13 14 15 16 17

5517.0 5611.0 5706.0 5800.0 5895.0 5990.0 6083.0 6177.0

5517.0 5610.8 5704.5 5795.5 5884.4 5968.6 6046.2 6121.4

0.27 5.99 12.10 17.17 23.73 31.36 35.52 38.23


18 19 20

6271.0 6366.0 6458.0

6193.6 6263.9 6331.3

41.38 43.11 42.77

Well Design/Tubing Size

21 22 23 24 25 26 27 28 29 30

6555.0 6650.0 6744.0 6838.0 6932.0 7026.0 7121.0 7215.0 7307.0 7324.0

6402.7 6472.1 6539.7 6607.4 6676.6 6747.3 6818.9 6888.4 6954.5 6966.5

42.43 43.71 44.23 43.68 41.46 40.98 41.33 43.20 44.93 45.52

8420.0 8520.0 8620.0 8720.0 8770.8 8820.0 8920.0 9020.0 9045.2 9074.0 9124.4 9220.0 9285.3 9320.0 9420.0 9520.0 9620.0 9720.0 9820.0 9920.0 10020.0 10120.0 10220.0 10271.4 10320.0 10370.4 10420.0 10520.0 10620.0 10720.0 10820.0 10920.0 11020.0 11120.0 11220.0 11264.1

Summary Profile


60 4262 589.57 146

SSSV Depth ML (ft)

2300

ID (in) Length (ft)* TVD SS (ft)

4.56 9.72 5478.0

AHD ft (SS) 3545 5545 5576.72 10576.72 14019.3

ID (in) 4.892 5.921 4.562 5.921 4.892


0.0018 Linear between reservoir and seabed 2 Btu/hr ft2/°F

OD (in) 6 7.191 7.99 7.191 6.05

Length (ft) 45 2000 31.72 5000 3442.6

Description Tubing Hanger Tubing SSSV* Tubing Excluder Screens


OPRM-2003-0302D

Page 21 of 25

31-December-2004


Unrestricted


TVD (ft) SS

Inc (deg)

31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100

6982.0 7047.0 7112.1 7176.1 7237.5 7294.5 7348.2 7396.9 7440.7 7479.2 7514.6 7549.1 7582.9 7615.5 7647.0 7678.2 7709.3 7741.5 7773.8 7805.3 7826.3 7834.8 7863.7 7894.7 7925.7 7956.6 7987.6 8018.6 8049.5 8080.5 8111.5 8142.4 8173.4 8204.4 8235.3 8266.3 8297.3 8328.2 8359.2 8390.2 8421.1 8452.1 8483.1 8514.0 8545.0 8576.0 8607.0 8624.9 8634.3 8637.9 8649.8 8667.3 8688.9 8701.9 8706.3 8706.3 8706.6 8707.0 8707.3 8707.6 8707.9 8708.2 8708.5 8708.8 8709.2 8709.5 8709.8 8710.1 8710.4 8710.7

45.05 46.23 46.23 48.14 51.23 54.13 57.04 60.57 64.54 67.03 68.73 68.29 69.52 70.31 70.60 71.01 70.31 69.68 70.56 70.34 70.68 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 71.96 75.03 80.01 85.00 89.82 89.82 89.82 89.82 89.82 89.82 89.82 89.82 89.82 89.82 89.82 89.82 89.82 89.82 89.82 89.82

7346.0 7439.0 7533.0 7627.0 7722.0 7816.0 7911.0 8005.0 8100.0 8194.0 8288.0 8382.0 8476.0 8571.0 8665.0 8760.0 8854.0 8948.0 9043.0 9137.0 9200.0 9226.5 9320.0 9420.0 9520.0 9620.0 9720.0 9820.0 9920.0 10020.0 10120.0 10220.0 10320.0 10420.0 10520.0 10620.0 10720.0 10820.0 10920.0 11020.0 11120.0 11220.0 11320.0 11420.0 11520.0 11620.0 11720.0 11778.1 11808.3 11820.0 11858.3 11920.0 12020.0 12120.0 12216.9 12220.0 12320.0 12420.0 12520.0 12620.0 12720.0 12820.0 12920.0 13020.0 13120.0 13220.0 13320.0 13420.0 13520.0 13620.0


OPRM-2003-0302D

Page 22 of 25

31-December-2004


Unrestricted


TVD (ft) SS

Inc (deg)

101 102 103 104

8711.0 8711.4 8711.7 8712.0

89.82 89.82 89.82 89.82

702p4 Well Trajectory MD(ft) SS WD 3568.0 2 3915.0 3 4189.0

TVD (ft) SS 3568.0 3915.0 4189.0

Inc (deg) 0.00 0.22 0.18

4 5 6 7 8 9 10

4470.0 4748.0 5027.0 5317.0 5599.0 5791.0 5905.0

4470.0 4748.0 5027.0 5317.0 5599.0 5791.0 5905.0

0.18 0.97 0.13 0.22 0.22 0.26 0.99

11 12 13 14 15

6002.0 6097.0 6191.0 6289.0 6383.0

6001.9 6096.9 6190.8 6288.8 6382.8

1.41 1.92 1.62 1.73 1.85

16 17 18 19 20

6477.0 6571.0 6659.0 6758.0 6848.0

6476.7 6570.5 6658.0 6756.1 6844.6

2.41 5.03 6.54 8.85 12.39

21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

6945.0 7039.0 7132.0 7225.0 7299.0 7320.0 7349.0 7420.0 7520.0 7620.0 7647.8 7720.0 7820.0 7920.0 8020.0 8120.0 8220.0 8320.0 8420.0 8520.0 8620.0 8647.8 8720.0 8820.0 8858.8 8920.0 9020.0 9120.0 9138.7 9220.0 9320.0 9420.0 9520.0 9620.0 9720.0 9820.0 9920.0 9932.8 10020.0 10120.0 10132.8 10220.0

6938.1 7025.9 7109.2 7186.2 7242.8 7258.4 7279.9 7333.7 7413.4 7497.2 7521.1 7583.6 7670.2 7756.8 7843.4 7930.0 8016.6 8103.2 8189.9 8276.5 8363.1 8387.1 8448.5 8529.4 8559.4 8606.1 8682.4 8758.7 8772.9 8833.0 8901.3 8963.1 9017.7 9064.9 9104.2 9135.3 9158.0 9160.3 9175.4 9192.8 9195.0 9207.9

18.48 23.07 29.79 38.13 42.16 42.16 42.16 39.20 35.10 31.09 30.00 30.00 30.00 30.00 30.00 30.00 30.00 30.00 30.00 30.00 30.00 30.00 33.51 38.40 40.30 40.30 40.30 40.30 40.30 44.37 49.37 54.36 59.36 64.36 69.36 74.36 79.36 80.00 80.00 80.00 80.00 83.05

13720.0 13820.0 13920.0 14019.3

Summary Profile PI@PSSS Initial Pavg Initial GOR

70 bbl/day psia 4465 psia 589.57 scf/bbl

T@midperfs

161


2300 4.56 9.72 5868.0


°F

0.0018 Linear between reservoir and seabed 2 Btu//hr ft2/°F

Well Design/ Tubing Size AHD ft (SS)

ID (in)

OD (in)

Length (ft)

Description

3545 5545 5576.72 10576.72 11958.0

4.892 5.921 4.562 5.921 4.892

6 7.191 7.99 7.191 6.05

45 2000 31.72 5000 1381.3

Tubing Hanger Tubing SSSV* Tubing Excluder Screens


OPRM-2003-0302D

Page 23 of 25

31-December-2004


Unrestricted


TVD (ft) SS

Inc (deg)

63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83

9216.9 9219.9 9220.0 9220.7 9221.5 9222.2 9223.0 9223.7 9224.5 9225.2 9226.0 9226.3 9227.6 9234.7 9247.8 9267.0 9268.2 9289.5 9312.0 9334.5 9343.0

86.55 89.57 89.57 89.57 89.57 89.57 89.57 89.57 89.57 89.57 89.57 89.57 87.69 84.19 80.69 77.19 77.00 77.00 77.00 77.00 77.00

Well Trajectory MD(ft) SS WD 3276.0 2 5650.0 3 5950.0 4 6695.1 5 9544.7 6 9706.6

TVD (ft) SS 3276.0 5650.0 5949.1 6606.4 8630.0 8745.0

Inc (deg) 0.00 0.00 7.50 44.75 44.75 44.75

7

9038.6

44.75

Well Trajectory MD(ft) SS WD 3278.0 2 5200.0 3 6291.6 4 9332.2

TVD (ft) SS 3278.0 5200.0 6133.9 7896.0

Inc (deg) 0.00 0.00 54.58 54.58

5

8165.0

54.58

10320.0 10406.2 10420.0 10520.0 10620.0 10720.0 10820.0 10920.0 11020.0 11120.0 11220.0 11266.4 11320.0 11420.0 11520.0 11620.0 11625.5 11720.0 11820.0 11920.0 11958.0

710 Wells

10120.0


150 4306 1139.2 147



ID (in) 4.892

OD (in) 5 1/2



2300 4.56 9.72 5576.0


710p2

9796.3

Summary Profile

710p3 Well Trajectory MD(ft) SS WD 3278.0 2 5200.0

TVD (ft) SS 3278.0 5200.0

Inc (deg) 0.00 0.00

3 4 5

5874.9 8364.0 8605.0

36.08 36.08 36.08

5921.7 9001.6 9299.8


14 4152 1139.2 128


2300 4.56 9.72 5578.0

bbl/day psia psia scf/bbl F



ID (in) 4.892

OD (in) 5 1/2


Summary Profile Tubing size (I) AHD ft (SS) 9300

PI@PSSS Initial Pavg

20 4308


1139.2 scf/bbl 139 °F

SSSV Depth ML (ft) ID (in)

2300 4.56

Roughness Geothermal profile

0.0018 Linear between reservoir and seabed

Length (ft) TVD SS (ft)

9.72 5578.0

Heat transfer coefficient

2

bbl/day psia psia

ID (in) 4.892

OD (in) 5 1/2


OPRM-2003-0302D

Page 24 of 25

31-December-2004


Unrestricted

702 Wells (cont’d) 710p4 Well Trajectory MD(ft) SS TVD (ft) SS Inc (deg) Combined with 803p1 in a single wellbore WD 3362.0 3362.0 0.00 2 6150.0 6150.0 0.00 3 6730.0 6705.5 29.00 4 9387.6 9030.0 29.00


18 4527 1139.2 158


2300 4.56 9.72 5662.0




ID (in) 4.892

OD (in) 5 1/2


803 Wells 803p1 Well Trajectory MD(ft) SS WD 3362.0 2 6150.0 3 6730.0 4 9387.6 5 10770.0 6 10873.3 7 10988.5 8 12379.0

TVD (ft) SS 3362.0 6150.0 6705.5 9030.0 9910.9 9930.0 9949.6 10165.0

Inc (deg) 0.00 0.00 29.00 29.00 79.31 79.31 81.09 81.09

803p2 Well Trajectory MD(ft) SS WD 3195.0 2 5500.0 3 6400.0 4 15607.7 5 16007.7 6 16465.8

TVD (ft) SS 3195.0 5500.0 6310.3 12821.1 13049.4 13243.0

Inc (deg) 0.00 0.00 45.00 45.00 65.00 65.00


18 5142 1447 176


2300 4.56 9.72 5662.0



ID (in) 4.892

OD (in) 5 1/2

0.0018 Linear 2


803p3 Well Trajectory MD(ft) SS WD 3359.0 2 5700.0 3 6684.6 4 7584.6

TVD (ft) SS 3359.0 5700.0 6684.6 7494.9

Inc (deg) 0.00 0.00 0.00 45.00

5 6 7 8 9

10437.8 10675.1 10805.0 10955.0 11401.0

45.00 30.00 30.00 30.00 30.00

4 5176 1447 179



ID (in) 4.892

OD (in) 5 1/2


11746.5 12046.5 12196.5 12369.7 12884.7



2300 4.56 9.72 5495.0

0.0018 Linear 2


70 5381 956 198


2300 4.56 9.72 5659.0




ID (in) 4.892

OD (in) 5 1/2

0.0018 Linear 2


OPRM-2003-0302D

Page 25 of 25

31-December-2004


Unrestricted

Section 3 Hydrate Remediation Guidelines


2.0

3.0

4.0

INTRODUCTION...........................................................................................................3 1.1

Start-up..............................................................................................................3

1.2

Shutdown...........................................................................................................3

1.3

Steady-state.......................................................................................................4

HYDRATE CHARACTERISTICS OF THE BONGA FLUIDS ........................................6 2.1

Hydrate Curves..................................................................................................6

2.2

Methanol Treatment Curves...............................................................................9

2.3

Hydrate Plug Dissociation Times .....................................................................12

HYDRATE FORMATION RISK FOR SUBSEA SYSTEMS .........................................13 3.1

Start-up............................................................................................................16

3.2

Steady-state.....................................................................................................18

3.3

Shutdown.........................................................................................................18

3.4

Aborted Start-up ..............................................................................................19

HYDRATE PLUG DETECTION AND REMEDIATION ................................................20 4.1

Flowlines/Risers...............................................................................................21

4.2

Wellbore Jumper and Manifold.........................................................................29

4.3

Wellbore/Tree (Upstream of Inhibitor Injection Point) .......................................33

4.4

Umbilicals ........................................................................................................36

4.5

Gas Lift Riser ...................................................................................................38

4.6

Water Injection Wells .......................................................................................43

TABLES Table 3.1 – Hydrate Temperatures for the Bonga Fluids ........................................................8 Table 3.2 – Hydrate Dissociation Pressure at 4.4°C (40°F) ....................................................9 FIGURES Figure 3.1 – Hydrate Curves for the Bonga Fluids ..................................................................7 Figure 3.2 – Maximum Treatable Flowrate for the 702 Oil with a Methanol Rate of 14gpm ..................................................................................10 Figure 3.3 – Methanol Volume Requirement for 702 Fluid....................................................10


OPRM-2003-0302D

Page 1 of 64

30-April-2006


Unrestricted

Table of Contents (cont’d) FIGURES Figure 3.4 – Maximum Treatable Flowrate for the 710 Oil with a Methanol Rate of 14gpm ..................................................................................11 Figure 3.5 – Methanol Volume Requirements for the 710/740 Fluid .....................................11 Figure 3.6 – Hydrate Remediation Times for the 702 Reservoir Fluid, Dashed Curves 12in PIP Flowline, Solid Curves 10in PIP Flowline ..................12 Figure 3.7 – Hydrate Remediation Times for the 710 Reservoir Fluid, Dashed Curves 12in PIP Flowline, Solid Curves 10in PIP Flowline ..................13 Figure 3.8 – Risk Identification for Hydrate Plugging in Different Parts of the Subsea System for Bonga................................................................................15 Figure 3.9 – Schematic of Hydrate Plug in Flowline (Except PFL 03/04) ..............................22 Figure 3.10 – Schematic of Hydrate Plug in Flowline (PFL 03/04) ........................................23 Figure 3.11 – Schematic of Hydrate Plug in Riser ................................................................23 Figure 3.12 – Remediation Procedure for Hydrate Plug in Flowline ......................................27 Figure 3.13 – Remediation Procedure for Hydrate Plug in Riser...........................................28 Figure 3.14 – Schematic of Hydrate Plate in Jumper............................................................29 Figure 3.15 – Remediation Procedure for Hydrate Plug in Jumper/Manifold.........................32 Figure 3.16 – Schematic of Hydrate Plug in Wellbore...........................................................33 Figure 3.17 – Remediation Procedure for a Hydrate Plug in the Wellbore ............................35 Figure 3.18 – Schematic of Hydrate Plug in Umbilical Line...................................................36 Figure 3.19 – Remediation Procedure for a Hydrate Plug in an Umbilical.............................37 Figure 3.20 – Schematic of Hydrate Plug in Riser Gas Lift System (Between Methanol Line and Flowline) ............................................................38 Figure 3.21 – Schematic of Hydrate Plug in Riser Gas Lift System (Between Methanol Line and GLR Topsides) ..................................................39 Figure 3.22 – Remediation Procedure for a Hydrate Plug in the Gas Lift Riser (Between Methanol Line and the Flowline) ......................................................41 Figure 3.23 – Remediation Procedure for a Hydrate Plug in the Gas Lift Riser (Between Methanol Line and Topsides)...........................................................42 Figure 3.24 – Schematic of Hydrate Plug in Water Injection Line .........................................44 APPENDICES Appendix 3A – Pressure Tags ..............................................................................................45 Appendix 3B – Case Studies ................................................................................................49 Appendix 3C – Nomenclature...............................................................................................63


OPRM-2003-0302D

Page 2 of 64

30-April-2006


1.0

Unrestricted

INTRODUCTION Bonga is a deepwater development offshore Nigeria in Block OML 118, in approximately 1000m water depth. Shell Nigeria E&P (SNEPCO) will operate Bonga in a joint venture with Esso (20%), Elf (12.5%) and Agip (12.5%). Bonga is being developed as a subsea network with 1.9 to 9.2km tiebacks to a permanently moored Floating Production, Storage, and Offloading vessel (FPSO). Peak production rates are anticipated at 225,000 barrels of oil per day, 170MMSCF of gas per day (including recycled riser lift gas) and 100,000 barrels of produced water per day. Reservoir pressures will be maintained via subsea waterflood wells with up to 300,000 barrels water per day injection capacity. Bonga consists of four reservoirs (690, 702, 710/740 and 803) with roughly half of the total reserves in the 702 reservoir. The production system contains subsea trees – enabling Surface Controlled Subsurface Safety Valves (SCSSVs), production chokes, and chemical injection valves – connected via short well jumpers to five subsea production manifolds. The subsea wells are produced through four pairs of piggable dual pipe-in-pipe flowlines, with externally insulated steel catenary risers. Each flowline is connected to a dedicated gas lift riser delivering up to 25MMSCF per day. One of the biggest flow assurance challenges at Bonga is hydrate control. Bonga is expected to operate under the philosophy of hydrate avoidance during all phases of operation – start-up, shutdown and steady state. This is achieved by the following operational strategies:

1.1

Start-up The strategy is to hot oil the flowlines to protect them from hydrates. The strategy for the trees, well jumpers and manifolds is to inject methanol/Low Dosage Hydrate Inhibitor (LDHI). In the absence of any methanol injection downhole, the well is ramped up as quickly as practicable (notionally 5000 to 7000bpd, depending on water-cut and pressure) such that the flowing wellhead temperature is greater than the hydrate dissociation temperature (approximately 24°C (75°F), but exact temperature depends on fluid properties and pressure) within 30 minutes to 1 hour.

1.2

Shutdown The strategy is to blow down the flowlines to a pressure below the hydrate dissociation pressure at 4.4°C (40°F) before the cooldown period has expired (notionally 12 hours following production at minimum flowline flowrates of 10,000bpd). The well jumpers and manifolds are displaced with methanol, before the cooldown time has expired, to remove hydrateable fluids and replace them with methanol. The wellbore is also bullheaded with methanol to the SCSSV in order to protect it from hydrates during a shut-in that lasts longer than 2 days1.

1

Due to the bare tubing in the wellbore, cooldown times are much larger when the well has been operating at steady state. Cooldown times are typically of the order of 2 days and hence bullheading must be done only if shutdown is expected to last more than 2 days. However, the first 100ft of the wellbore must be treated immediately upon shut-in, since the cooldown time in this section is limited.


OPRM-2003-0302D

Page 3 of 64

30-April-2006


1.3

Unrestricted

Steady-state The strategy is to rely on the heat content of the system to avoid hydrates. The system flows well above hydrate forming temperatures and in fact flows at a temperature that guarantees at least 12 hours to cool down to the hydrate temperature during a shut-in. However, in spite of the above philosophies, there are four major reasons which makes hydrate control at Bonga particularly challenging. These are: Low Water Salinity

•

The expected produced water salinity at Bonga is ~3wt% while typical Gulf of Mexico (GoM) produced water is between 6 to 22wt%. Assuming an average of 10wt% salinity, the typical GoM system has a subcooling that is 3 to 4°C (6 to 7°F) less than Bonga, which means the system needs to be warmed 3 to 4°C (6 to 7°F) less than Bonga to move the system outside the hydrate region. Alternatively, the pressure requirement during blow down is increased by 7 to 10bar (100 to 150psi) for produced brines with a salinity of 10wt%. This has important implications for Bonga since current blowdown calculations with and without riser base gas lift indicate that the low blowdown pressure requirement challenges the limits of the blowdown system (transient report on blowdown has shown that the minimum blowdown pressure is 10bar±2bar (150psi± 30psi)2. Kinetics of Hydrate Formation in Bonga

•

The kinetics of hydrate formation is difficult to quantify since experimental data for black oil systems is limited. There are a number of different factors that determine the rate at which hydrates will form, including fluid properties, water cut and flow regime. In the case of the flow loop tests with the Bonga crude, plugging times were all very rapid. In all tests, the fluids were cooled from higher temperatures down into the hydrate region, and within a few degrees of cooling into the hydrate region the system was plugged. The formation of hydrates was so rapid that the waves in the oil phase (the system was operating in the wavy-stratified flow regime, with a water/oil emulsion) actually froze in place. This rapid hydrate formation has not been observed previously. While there are no other experiments done at the same conditions as these for Bonga, tests with other crude oil systems in the flow loop were more difficult to plug. •

System Limitations Bonga is expected to start producing water within 18 to 24 months of first oil and up to a water-cut of 80 to 90%. In order to completely prevent hydrate formation at such high water rates, methanol injection rates of nearly 60 to 90gpm are required. However, only 14gpm per well can be injected at Bonga (and it is certainly not practical to inject at rates of 60 to 90gpm). When operating at such high water rates, Bonga depends on a unique never-before-used methanol/LDHI cocktail to prevent hydrate formation in the trees, jumpers and manifolds during start-up. Although, laboratory tests have indicated that this combination will work at Bonga, it must be understood that this strategy has not been field tested.

2

Wade Schoppa ‘Bonga Dynamic Flow Assurance Analysis – Evaluation of Conceptual Design’, February 2001.


OPRM-2003-0302D

Page 4 of 64

30-April-2006


•

Unrestricted

Lack of Operating Experience at Bonga Experience has indicated that new facilities are prone to a lot of aborted start-ups and extended shutdowns during their first few years of operation (eg Auger had nearly 253 unplanned shutdowns and 79 planned shutdowns during its first 22 months of operation, while Mars had nearly 112 unplanned shutdowns and 29 planned shutdowns during its first 8 months of operations)3. The subsea system is most vulnerable to hydrates during start-up and shutdown (especially before steady-state is attained), and the probability of operational errors is greatest during these transient events.

In view of the above reasons, hydrate formation/plugging is a credible risk at Bonga and hence, maximum precautions must be taken to ensure hydrate free operations at Bonga. The main purposes of this document are to: •

Provide guidance to operations and surveillance staff on how to identify hydrate formation in the subsea system

•

Provide the first steps of blockage remediation to operators/surveillance staff in case a hydrate blockage forms so that operators/surveillance staff can safely secure the system and/or prevent the problem from getting worse

•

Define safe procedures to start remediation of the subsea hydrate blockage before expert help can be summoned

•

Provide examples from other fields (within and outside Shell) on how hydrates blockages were formed and were remediated along with important lessons learned

•

Provide an evergreen document that can be updated when operating conditions on the field significantly change (eg when LDHI comes on, LDHI charts should be added) and to include any Bonga-specific hydrate incidents

This document is not meant to: •

Provide detailed procedures on how to remediate hydrates from various parts of the subsea system. (We view hydrate plugging as an abnormal event requiring expert help. Each event is somewhat different and hence routine procedures cannot be written. Although some initial procedures can be initiated from the FPSO, we strongly recommend summoning expert help to complete remediation of a hydrate plug)

•

Provide operating strategies to avoid hydrates in the Bonga system. These will be covered in the POPMs, and control system warnings and interlocks will cover some of the critical operations. In fact, this document assumes that the reader is intimately familiar with Bonga’s operating strategies especially with respect to hydrate management

•

Bypass normal operating procedures (POPMs during normal field operations). Some of the recommendations given in this document may be in conflict with the POPMs and should only be followed if flow has stopped abruptly and hydrate formation is a strong suspect

3

Sada Iyer ‘Analysis for Full Thermal Cycles for Bonga Over a 20-year Period’, April 2003.


OPRM-2003-0302D

Page 5 of 64

30-April-2006


Unrestricted

This document is structured as follows:

2.0

•

Paragraph 2.0 describes the hydrate characteristics of the Bonga fluids. This includes hydrate curves, methanol treatment curves, blowdown pressures, and hydrate dissociation times

•

Paragraph 3.0 describes the hydrate formation risks for various parts of the subsea system

•

Paragraph 4.0 describes remediation methodologies for each section of the subsea system and contains guidelines for remediating hydrates once a plug is formed

•

Appendix 3A gives a table of all relevant pressure tags that are of interest in terms of hydrate detection and remediation

•

Appendix 3B describes several different case studies involving hydrate plug formation and remediation. The case studies used were chosen because of their general similarity to Bonga

•

Appendix 3C gives a listing of all abbreviations used in this report

HYDRATE CHARACTERISTICS OF THE BONGA FLUIDS This paragraph is intended as a summary of the information used in determining the hydrate curves and methanol requirements4. The hydrate curves are presented for all fluids to give some indication of the relative hydrate risk of each of the different oil systems. Methanol rates are only included for the 702 oil (best-case) and the 710 oil (worst-case) to bracket the potential methanol requirements at Bonga.

2.1

Hydrate Curves The hydrate curves define the stability of hydrates in the Bonga system. The hydrate curves are dependent on the salinity of the produced water. Hydrate curves are included for a produced brine with a salinity of 3wt%. Due to the waterflood, the expected salinity of the produced water is about 3wt%, hence Figure 3.1 and Table 3.1 should be used in determining if the system is operating in the hydrate region. For example, if the operating conditions are 200bar (2900psi) and 20°C (68°F), all four fluid systems are in the hydrate region (refer to Figure 3.1). If the pressure is decreased to 150bar (2175psi), the 690 and 702 fluids are no longer in the hydrate stability region, but the 803 and 710 fluids are still in the hydrate region. If the pressure is further reduced to 100bar (1450psi), the conditions are such that all fluids are now out of the hydrate region. Alternatively, if the temperature is increased from 20°C (68°F) to 25°C (77°F) and the pressure remains constant at 200bar (2900psi), all four fluid systems are in the non-hydrate region.

4

For a more detailed description, refer to Peters, D, et al ‘Bonga Hydrate Risk Assessment and Management Strategy’, report OG.03.80057, 2003.


OPRM-2003-0302D

Page 6 of 64

30-April-2006


Unrestricted

Temperature (ºF) 42

32

52

62

72

82 5000

690 300

4500

720 4000 803 3500

710/740 Hydrate Stability Region

200

3000 2500

150

2000

Pressure (psi)

Pressure (bar)

250

1500

100 Non-hydrate Region

50

1000 500 0

0 0

5

10

15

20

25

30

Temperature (ºC)

OPRM20030302D_063.ai

Figure 3.1 – Hydrate Curves for the Bonga Fluids


OPRM-2003-0302D

Page 7 of 64

30-April-2006


Unrestricted

Pressure (bar)

690 HDT (°C)

702 HDT (°C)

710/740 HDT (°C)

803 HDT (°C)

8.6

0.1

-0.4

3.1

1.6

11.6

2.2

1.7

5.2

3.8

15.7

4.3

3.8

7.3

6.0

21.1

6.4

6.0

9.5

8.2

28.5

8.5

8.1

11.6

10.3

38.4

10.6

10.3

13.6

12.5

51.7

12.7

12.4

15.6

14.5

69.7

14.6

14.4

17.5

16.5

94.0

16.5

16.3

19.3

18.3

126.7

18.3

18.1

20.9

20.0

170.7

20.0

19.9

22.5

21.6

230.2

21.2

21.5

24.1

23.2

310.3

22.5

22.6

25.5

25.0

Pressure (bar)

690 HDT (°F)

702 HDT (°F)

710/740 HDT (°F)

803 HDT (°F)

125

32.1

31.2

37.7

34.8

169

35.9

35.0

41.4

38.8

227

39.7

38.9

45.2

42.7

306

43.5

42.7

49.0

46.7

413

47.4

46.6

52.8

50.6

556

51.1

50.5

56.5

54.4

750

54.8

54.2

60.1

58.2

1011

58.4

57.9

63.6

61.7

1363

61.7

61.3

66.7

64.9

1837

64.9

64.6

69.7

68.0

2476

68.0

67.7

72.4

70.8

3338

70.2

70.7

75.3

73.8

4500

72.5

72.7

77.9

76.9

Table 3.1 – Hydrate Temperatures for the Bonga Fluids


OPRM-2003-0302D

Page 8 of 64

30-April-2006


2.1.1

Unrestricted

Blowdown and Plug Remediation Pressures An important aspect of the hydrate equilibriums curve is the Hydrate Dissociation Pressure (HDP) at the ambient seafloor temperature of 4.4°C (40°F). This is the pressure that will determine the blowdown requirement and the hydrate plug remediation pressure. One of the key components to hydrate control at Bonga is the ability to blow down the flowlines and move the flowline conditions outside of the hydrate region. Table 3.2 shows the hydrate equilibrium pressure at 4.4°C (40°F). Well

HDP (psi)

HDP (bar)

690

202

13.9

702

216

14.9

710/740

130

9.0

803

162

11.1

Table 3.2 – Hydrate Dissociation Pressure at 4.4°C (40°F)5 The hydrate dissociation pressure at 4.4°C (40°F) is also important in the process of hydrate plug remediation. The flowline pressure must be reduced below the HDP of the particular fluid in order for the hydrate plug to melt. During plug remediation, the flowline pressure should be reduced as low as possible to increase the rate at which the plug melts. If a flowline has fluids from more than one reservoir, then use the fluid with the lowest HDP.

2.2

Methanol Treatment Curves During early life, the hydrate strategy is to treat all produced water with methanol. Figures 3.2 and 3.4 show the highest treatable water cut that can be protected with the 702 and the 710 fluids. Figures 3.3 and 3.5 give the methanol requirements in a more general format that can be applied to any flowrate. The minimum flowrate during start-up is either 5000blpd or 7000blpd, depending on the well. At a given pressure and flowrate, these figures can be used to determine how much water can be treated. This is important in determining when to switch from the methanol-only strategy to the methanol/Kinetic Hydrate Inhibitor (KHI) strategy. For example, if the 702 fluid is being produced at a rate of 5000blpd, the flowline pressure is 150bar (2175psi) and the water cut is greater than 20%, then methanol alone (at 14gpm) is not enough to protect against hydrate formation and it is time to switch to the methanol/KHI strategy. Similarly, if the flowline is producing the 710 fluid at a rate of 5000blpd and a pressure of 150bar (2175psi), then the highest water cut that can be treated is 17%.

5

The hydrate equilibrium values for fresh water have been used. These pressures are required for hydrate plug remediation but give slightly conservative estimates for the blowdown pressure required to prevent hydrate formation.


OPRM-2003-0302D

Page 9 of 64

30-April-2006


Unrestricted

Pressure (psi)

Maximum Treatable % Water Cut

70

0

500

1000

1500

2000

2500

3000

3500

4000

4500

60 50

Fluids cannot be treated with methanol alone

40 30 5000blpd

20 Fluids can be treated with methanol

10

7000blpd

0 0

50

100

150

200

250

300

Pressure (bar) OPRM20030302D_049.ai

Figure 3.2 – Maximum Treatable Flowrate for the 702 Oil with a Methanol Rate of 14gpm 0.80

m3 Methanol/m3 Water

0.70 0.60 0.50 0.40 0.30 0.20 0.10 0

5

10

15

20

25

30

35

40

% Water Cut 34.5bar (500psia)

68.9bar (1000psia)

241.3bar (3500psia)

317.2bar (4600psia)

103.4bar (1500psia)

OPRM20030302D_050.ai

Figure 3.3 – Methanol Volume Requirement for 702 Fluid


OPRM-2003-0302D

Page 10 of 64

30-April-2006


0

500

Unrestricted

1000

1500

2000

2500

3000

3500

4000

4500

45

Maximum Treatable % Water Cut

40 35 Fluids Cannot be Treated with Methanol Alone

30 25 20

5000blpd 15 10 7000blpd

Fluids Can be Treated with Methanol

5 0

5

0

100

150

200

250

300

Pressure (bar) OPRM20030302D_071.ai

Figure 3.4 – Maximum Treatable Flowrate for the 710 Oil with a Methanol Rate of 14gpm 0.80

m3 Methanol/m3 Water

0.70

0.60

0.50

0.40

0.30

0.20

0.10 0

5

10

15

20

25

30

35

40

% Water Cut 34.5bar (500psia)

68.9bar (1000psia)

241.3bar (3500psia)

317.2bar (4600psia)

103.4bar (1500psia)

OPRM20030302D_064.ai

Figure 3.5 – Methanol Volume Requirements for the 710/740 Fluid


OPRM-2003-0302D

Page 11 of 64

30-April-2006


2.3

Unrestricted

Hydrate Plug Dissociation Times Figures 3.6 and 3.7 are intended to give approximate hydrate plug remediation times as a function of the pressure. In this case, approximate means an order of magnitude estimate. The intent of these predictions is to indicate if the plug melting time is days, weeks or months. This model has been compared with available information on hydrate plug removal and was found to give reasonable estimates of the plug melting time6. If the predicted dissociation time is a week, then the plug can be expected to take 1 week, plus or minus a few days. If the plug is predicted to take a month to melt, then the plug can be expected to take 1 month, plus or minus a week. These predictions are the amount of time required to establish pressure communication across the plug, not the time required to completely melt the plug. The two-sided depressurisation case (applicable generally to looped flowlines) assumes that there is a small (~3.5bar (50psi)) pressure drop across the plug. The one-sided case (applicable generally to wellbore jumpers and wellbores) assumes a pressure drop of > 70bar (1000psi) across the plug. These figures were generated for the Bonga flowlines (10in and 12in), but similar results were obtained for the melting times of hydrate plugs in a line with a 5in diameter. As was the case with the methanol predictions, the 702 and 710 reservoir fluids are used in the prediction of the hydrate remediation times. These two fluids bracket the possible remediation times that are expected for Bonga.

Downstream Pressure (psia) 30

87

100

6

7

113

126

139

152

165

178

191

204

Hydrate Dissociation Pressure

Remediation Time (Days)

25

217

20

15

10

5

0 8

9

10

11

12

13

14

15

Downstream Pressure (bar) 1-sided dissociation 2-sided dissociation OPRM20030302D_051.ai

Figure 3.6 – Hydrate Remediation Times for the 702 Reservoir Fluid, Dashed Curves 12in PIP Flowline, Solid Curves 10in PIP Flowline

6

Walsh et al ‘Hydrate plug dissociation model’, EP 2001-3018, June 2001.


OPRM-2003-0302D

Page 12 of 64

30-April-2006


Unrestricted

Downstream Pressure (psia) 44

54

64

74

84

94

104

114

124

134

30


Remedlation Time (Days)

25

20

15

10

5

0 3

4

5

6

7

8

9

Downstream Pressure (bar) 1-sided dissociation

2-sided dissociation OPRM20030302D_065.ai

Figure 3.7 – Hydrate Remediation Times for the 710 Reservoir Fluid, Dashed Curves 12in PIP Flowline, Solid Curves 10in PIP Flowline

3.0

HYDRATE FORMATION RISK FOR SUBSEA SYSTEMS When assessing the hydrate risk in the subsea system, there is an important distinction to be made between hydrate formation and the formation of a hydrate plug. In an uninhibited system, if the subsea temperature and pressure are in the hydrate formation region, hydrates will form. The formation of a solid hydrate plug is not predictable but, from laboratory and field experience, it is most likely to occur during a restart. If a plug is not formed immediately upon restart, continued operation inside the hydrate region greatly increases the risk that sufficient hydrates will accumulate and lead to the formation of a plug. How long the system can operate in the hydrate region before a plug is formed depends on a number of factors, including the kinetics of hydrate formation and the ‘stickiness’ of the hydrate particles that are formed. Unfortunately there is no accurate means of predicting when the hydrate particles will accumulate into a plug, but laboratory testing with the Bonga fluids showed the rapid formation of hydrate plugs once the system was inside the hydrate region. There are several examples of GoM systems that operate in the hydrate region without forming hydrate plugs even though hydrates are formed. However, these examples are typically gas condensate systems. For example, at South-east Tahoe, the formation of hydrates is detected as an increase in the pressure drop across the flowline. Once the pressure drop increases sufficiently, the production rate is curtailed while the methanol pumps continue to run at full flowrates to flush the hydrates out of the flowline. This strategy can be used at South-east Tahoe since production is at the end of the field life and the consequences of forming a plug are not severe (limited loss of production).


OPRM-2003-0302D

Page 13 of 64

30-April-2006


Unrestricted

The same is not true of oil systems. Currently, the steady-state operating strategy for oil systems is to operate outside of the hydrate region, thus there are no similar cases to compare. Most plugs in oil systems occur during restart after the system remained in the hydrate region for some amount of time. Also, based on flowloop tests, the plugs that are formed in an oil system usually occur much more catastrophically and the typical warning signs observed in gas condensate systems (such as gradual increase in pressure drop) are not observed. The Bonga fluids have shown rapid failure times in testing and hence it is assumed that the formation of hydrates will quite quickly lead to the formation of a plug. Therefore, the philosophy at Bonga is for hydrate avoidance during all phases of operation by keeping the operating conditions outside of the hydrate formation region (or by delaying the formation of hydrates for at least 12 hours by using LDHI in mid-life to late-life). For the special case of the wellbore during start-up (without downhole inhibitor injection capability at Bonga), the risk of hydrate plug formation is minimised by ‘bullheading’ the upper wellbore with methanol (or methanol/LDHI) after shut-in. There is some evidence from a North Sea oil field that signals are provided during plug formation and prior to complete blockage in an oil system. As was the case with South-east Tahoe, in the North Sea oil field after significant build-up in the line pressure drop, the methanol rate was increased and the hydrate restrictions cleared from the flowline. The methanol rate was reduced to initial levels and the cycle repeated. However, several key differences exist between this example and Bonga, in particular, the water cut (~10%) was much lower than is expected at Bonga. So, while it may be possible to observe early signs of hydrate plug formation, this will not be the operating strategy used at Bonga. With Bonga production, it is assumed that any hydrate formation will rapidly lead to the formation of a plug. It will be crucial to monitor the temperatures and pressures along the subsea system in order to determine where hydrates are forming. This will allow the location of a hydrate plug to be isolated to the riser, flowline, jumper or wellbore. These temperature and pressure measurements can be monitored using the various sensors on the subsea system. Appendix 3A shows the tag numbers for the various pressure sensors in different parts of the subsea system for different flowline loops, trees, manifolds and gas lift risers. The tags for the temperature sensors are not included since they have only a limited usefulness in detecting hydrate plugs. Before writing any paragraph on remediation, it is important to assess the risk of hydrate formation on various parts of the Bonga subsea system during all stages of operation. Figure 3.8 shows the risk associated with each part of the subsea system. However, the temperature readings are vitally important in determining if the system is in the hydrate formation region.


OPRM-2003-0302D

Page 14 of 64

30-April-2006


Unrestricted

Event Subsea System

Start-up

Steady State

Shutdown

Aborted Start-up

Wellbore Trees Well Jumper Production Manifold Flowlines Riser Gas lift Sled/Umbilicals Umbilicals Water Injection Flowlines Water Injection Tees Water Injection Trees High Risk

Medium Risk

Low Risk

Figure 3.8 – Risk Identification for Hydrate Plugging in Different Parts of the Subsea System for Bonga For the purposes of this document, we define the risk levels as follows: High Risk The system design itself does not guarantee protection against hydrates but hydrate control is achieved by a combination of design and active strategies. Examples include the tree and jumper where the addition of methanol/LDHI is used to prevent hydrates during start-up. If a methanol pump fails, it could lead to the formation of hydrates. Another example is that for the wellbore, there is not the ability to treat with methanol/LDHI to prevent hydrates during start-up and instead a strategy of minimising the time of operation within the hydrate region by rapid production ramp up and ‘outrunning’ the formation of a hydrate plug is used. Medium Risk The system design protects itself from hydrates, but that protection could be eroded by operational decisions. An example of this is deciding to start up into a warm flowline (without hot-oiling) that has partially cooled down. An aborted start-up at this stage would leave the flowline in a state with an unknown cooldown time and possibly allow the liquids in the system to cool inside the hydrate region prior to following the procedure for a ‘normal’ aborted start-up.


OPRM-2003-0302D

Page 15 of 64

30-April-2006


Unrestricted

Low Risk The system design protects itself from hydrates for normal operating conditions. For example, the Bonga subsea system is designed such that during steady state operation, all sections are outside of the hydrate region and that upon a shut-in, there is sufficient cooldown time provided by the insulation to give at least 3 hours during which no hydrate inhibition actions are required.

3.1

Start-up High Risk As shown in Figure 3.8, the highest risk during start-up is in the wellbore, tree, wellbore jumper, manifold and umbilicals. During start-up, these sections are typically at the coolest temperature and highest pressure in the subsea system. The highest risk among these sections is the wellbore since the system design does not include wellbore hydrate inhibitor injection, whereas the tree, manifold and jumper will be susceptible to hydrate formation if the injection of methanol or methanol/LDHI fails. The hydrate risk in the wellbore is difficult to quantify. Prior to the start-up, the wellbore should have been bullheaded with methanol to prevent any hydrate from forming during shut-in. Once the system is restarted, there is a fairly lengthy warm-up time and the wellbore may be operating in the hydrate region for up to 90 minutes1. There is a good chance hydrates will form and be pushed out of the wellbore to areas of the subsea system that have already been warmed and/or treated with methanol/LDHI without forming a hydrate plug in the wellbore. The risk of hydrate formation is high. The risk of forming a hydrate plug is less and is difficult to quantify. Based on GoM experience, the formation of a hydrate plug in the wellbore is low if the warm-up times are quick (ie less than 1 hour). However, there are several key differences between the GoM wells and the Bonga wells that make the Bonga wells a higher risk. GoM wells typically have Vacuum Insulated Tubing (VIT), which decreases the warm-up time considerably. An unknown in this process is the effect of watercut. All testing indicates that the higher the watercut, the higher the hydrate risk, but there are very few subsea flowlines producing oil at high watercut, which makes this risk difficult to quantify, hence, the overall risk is marked as high. There are cases when an oil system (eg Auger and Tahoe) has plugged with hydrates during start-up. However, both examples are a bit unusual in that very little produced water was present. At Auger, there may have been some additional unloading fluids present that increased the total watercut. In both cases, the systems were shut-in without any hydrate inhibition (since it was assumed the watercut was too low to form a hydrate plug). Likely hydrates were formed from the small amounts of water present in the oil stretched throughout the flowline but also in larger amounts from water settled and accumulated in low spots in the flowline. Upon restart, these hydrate particles grew to form a plug. This illustrates the need to protect the system against hydrates even when the watercut is low. As the watercut is increased, the likelihood of forming hydrates is only increased.7 1

Due to the bare tubing in the wellbore, cooldown times are much larger when the well has been operating at steady state. Cooldown times are typically of the order of 2 days and hence bullheading must be done only if shutdown is expected to last more than 2 days. However, the first 100ft of the wellbore must be treated immediately upon shut-in, since the cooldown time in this section is limited. 7

At Terra Nova, this problem with gas migration only appeared during the first 6 weeks of water injection. However, Terra Nova has a different water injection strategy (involving alternating periods of injection followed by extended shut-in) and the properties of the reservoir are different than at Bonga.


OPRM-2003-0302D

Page 16 of 64

30-April-2006


Unrestricted

The hydrate risk in the tree, well jumper and manifold is largely mitigated by the use of chemicals. During early life, methanol can be injected in sufficient quantities to prevent hydrates, so the only risk here is due to the failure of the methanol pumps. In this case, there are cold untreated fluids coming from the wellbore (possibly with hydrates already present) flowing into a cold untreated section with many areas where water can accumulate and this greatly increases the risk of hydrate plug formation. If adequate measures are taken to ensure pump reliability, then this risk is decreased. However, based on the definition of risk, this is considered a high risk since the insulation alone does not provide protection from hydrates during restart and transition to steady-state, but instead hydrate mitigation relies upon chemical treatment. Similarly, the chemical umbilical lines (especially flying leads) are susceptible to hydrates due to pressure fluctuations that occur during start-up (eg due to slugging). This might lead to backing up of chemicals and production fluids inside the umbilical, and the formation of hydrates. The water injection trees and upper wellbore are also considered to be high risk. Since the water injectors are completed into the oil zone (and can flow under their own pressure), gas can migrate past the SCSSV (towards the tree) during an extended shut-in and can form hydrates with the water in the well. This risk is greatest when waterflood has just started for the first time and the well is shut in within a few days of initial start-up. This is because the gas front in the reservoir would not have been pushed away from the wellbore and hence the gas will tend to migrate back into the wellbore. As more and more water is injected into the reservoir, the risk continues to decrease as the gas front in the reservoir is pushed away from the water injection wellbore (hence, the gas takes much longer to migrate back into the wellbore). This risk is difficult to quantify since the probability of gas migration into the wellbore is unknown. Any hydrates that are formed will be a result of these gas bubbles migrating up the wellbore. In the absence of any agitation, a column of hydrate bubbles will be established. With time, these hydrate bubbles will be pushed up to the tree where they will accumulate and may also form deposits. The fate of these bubbles is open to speculation, but could easily collapse during the shut-in or the restart to form a hydrate ‘slush’ with the excess water in the wellbore. Depending on the volume of gas, slushy hydrates and deposits in the wellbore upon restart, it may or may not be enough to stop flow and prevent the water from pushing this hydrate to below the SCSSV, where they will melt. Even excluding this scenario of gas leaking past the SCSSV, gas and oil migration will pose a risk at start-up as the valve is opened with oil or gas trapped beneath it unless downward flow is established in a timely fashion. It should be emphasised that any hydrate deposits that form in the tree or just beneath the tree in either case will likely not be quickly melted even if flow is established since the incoming water temperature is expected to be no higher than about 60°F and at pressures still within the hydrate region. Because of the uncertain nature of gas migration in the water injection system and the inability to inject any chemicals into the wellbore to prevent or remediate hydrates, then the risk is considered high for the water injection trees and upper wellbore. Since this is a problem that occurs during shut-in and can prevent start-up, this risk is included in both the ‘start-up’ and ‘shutdown’ categories for the sake of completeness.


OPRM-2003-0302D

Page 17 of 64

30-April-2006


Unrestricted

Medium Risk None. Low Risk Flowlines and risers have a low risk of hydrate formation during start-up since they will always be started up with hot-oiling (or will be started up before their cooldown time has expired). Hot-oiling ensures that the flowlines are always above the Hydrate Dissociation Temperature (HDT) and have a guaranteed 12-hour cooldown to HDT. Water injection flowlines and water injection tees have low risk of hydrate formation due to the low probability of any gas migration to these parts. Any oil or gas that migrates into the well is likely to accumulate at the tree. As per current Bonga procedures, the gas lift riser will be flushed with methanol (to displace potential hydrocarbons that might have backed into the gas lift valve sleds and portions downstream of it towards the flowline) during shutdown. Moreover, the gas injection side is higher in pressure so as to prevent backflow of hydrocarbons into the gas lift system (achieved by means of an orifice plate which ensures higher upstream pressure).

3.2

Steady-state High Risk None. Medium Risk None. Low Risk All sections of the Bonga subsea system are under low risk of hydrate formation due to the design of the Bonga system. By design, every part of the subsea production system is insulated to operate above hydrate dissociation temperatures and also to provide 8 to 12 hours of cooldown outside of the hydrate region in case of a shut-in (8 hours for wellbore and 12 hours for the rest of the system). Similarly, we do not expect any kind of backflow of hydrocarbons into umbilicals, water injection system or the gas lift system during steady-state.

3.3

Shutdown Shutdown risks are tricky to capture since any problem that occurs during shutdown will be manifested only when we try to restart the system. However, this paragraph attempts to capture possible hydrate problems that are solely the result of a shut-in and not necessarily problems that occur during start-up. High Risk The highest risks at shutdown are hydrate formation within chemical umbilical lines, gas lift umbilical lines and water injection trees due to backflow of hydrocarbons.


OPRM-2003-0302D

Page 18 of 64

30-April-2006


Unrestricted

In Bonga, the control system is designed such that the chemical valves at the trees are instructed to close as soon as the wells shut in. Similarly, the gas lift valves on the gas lift sled are also designed to be shut in as soon as the FPSO flowline boarding valves shut in. However, there has been past experience wherein hydrate formation has occurred due to backflow. An example of this is the Malampaya Project wherein methanol lines were plugged with hydrates due to backflow from the production system into the umbilical. Gas migration from the water injection reservoirs can also result in hydrate formation in the water injection trees. Gas migration can occur either during shut-in after a short period of operation (say within the first few days of start-up when the hydrocarbon front has not been pushed far enough into the reservoir) or during a long extended shut in (when gas eventually migrates back, as in the case of the Terra Nova Field in Canada)8. The gas from the reservoir can migrate past the SCSSV towards the trees and can result in plug formation. Medium Risk None. Low Risk All other parts of the subsea system have a low risk with respect to hydrate formation due to the fact that the Bonga production system is designed with a minimum 12-hour cooldown time after shut-in. Failure to protect the subsea sections against hydrate formation greatly increases the risk of forming a hydrate plug upon restart.

3.4

Aborted Start-up High Risk The aborted start-up has the highest risk of hydrate formation of any of the operations at Bonga. The wellbore, well jumper, production manifolds, gas lift sled, umbilicals and water injection trees are all particularly susceptible to hydrates during an aborted start-up. The risk is similar to that of the start-up case (ie wellbore, well-jumper, production manifold, chemical umbilicals and water injection tree are all at high risk). In addition, the gas lift umbilical also becomes a high-risk candidate with respect to hydrate formation. This is due to the fact that the water-heated portion of the Gas Lift Riser (GLR) system below the GLR valves takes some time to reheat once the system has cooled thus leaving the system vulnerable to hydrates. The water injection trees are at a high risk because any gas that may have migrated near the wellbore and/or accumulated beneath the SCSSV during the shut-in would not be moved very far during the initial start-up. This means that less time is required for the gas to migrate back to the wellbore or past the safety valve. Cold fluids have been moved further down into the wellbore where the pressure is higher and there is a greater likelihood of ‘seeing’ the gas as it bubbles up the wellbore.

8 At Terra Nova, this problem with gas migration only appeared during the first 6 weeks of water injection. However, Terra Nova has a different water injection strategy (involving alternating periods of injection followed by extended shut-in) and the properties of the reservoir are different than at Bonga.


OPRM-2003-0302D

Page 19 of 64

30-April-2006


Unrestricted

Medium Risk The flowlines are designed in such a way that they have a minimum of 12 hours cooldown after a shut-in from steady state. However, if a well has a warm start into a flowline during the cooldown period and this well has an aborted start-up, it is not very easy to estimate the resulting cooldown time. This uncertainty in the cooldown time necessitates identifying hydrate formation as a medium risk. The second scenario is starting up a cold well into a flowline, which has just been hot-oiled. The hot-oiling ensures that the flowline has a 12-hour cooldown prior to starting up the well and also that the cold fluids from the well do not reduce the flowline temperature to below the HDT. However, an aborted start-up on this well will result in not knowing the exact temperature of the flowline and thus result in an uncertain cooldown time (and hence a hydrate risk). Low Risk The risk in the water injection flowlines/tees is low due to the low probability of gas migration from the reservoir through the SCSSV and then past the tree.

4.0

HYDRATE PLUG DETECTION AND REMEDIATION As seen in Figure 3.8, every portion of the subsea system (except for waterflood flowlines and tees) is at risk with respect to hydrate formation either during start-up, shutdown or during an aborted start-up. Based on our current knowledge of hydrates with respect to Bonga, it is unlikely that hydrate formation will be detected in the system before a hydrate plug is formed. Therefore, this paragraph assumes that a hydrate plug has formed and provides guidance for determining approximately where the plug is in the subsea system and for remediating hydrate plugs. Lastly, there are examples for hydrate detection and remediation drawn from different fields from various parts of the world (Shell and non-Shell). In this document the pressure measurement is the crucial parameter in determining the location of the hydrate plug. However, it should be noted that the temperature measurement is important in determining if the system is in the hydrate formation region. The temperature and pressure must be in the hydrate region before hydrates can form. Figure 3.1 or the Bonga tool can be used to determine if the system temperature and pressure are in the hydrate region. If the system is operating outside of the hydrate region and a blockage is formed, then the cause is something other than hydrates. The location of hydrate plug can only be determined with limited accuracy. In the case of Bonga (and the guidelines in this document), the hydrate plug position can be determined to be between a particular pair of pressure sensors. Although not discussed in this document, there are other methods to determine a more exact plug location (eg ultrasonic sensors). The pressure sensors at Bonga can localise the plug to specific sections of the subsea system, including the flowline, riser, jumper/manifold or the wellbore. For the hydrate remediation guidelines presented in this document, this gross determination of plug location is sufficient. If the plug location needs to be determined more exactly, expert guidance is suggested. Information is also presented for hydrate plugs that form in umbilical lines and the waterflood lines. These lines are not equipped with instruments that help to locate and remediate a hydrate plug, as are the other portions of the subsea system. In these cases, more general information will be given.


OPRM-2003-0302D

Page 20 of 64

30-April-2006


4.1

Unrestricted

Flowlines/Risers Although the risk for hydrate formation in the flowline has been determined to be low, it is important to have guidelines in place to remediate a hydrate plug in the flowline. The loss of a flowline can affect production from multiple wells (all wells from that manifold), which has a significant impact on deferred production and revenue.

4.1.1

Hydrate Plug Formation Flowlines (Except PFL 03/04) The most likely time that a hydrate plug will occur (or be detected) in the flowline/riser section is during a restart following an aborted start-up. Plugs may occur during the shutdown period, but they will not be detected until the flowline is restarted. A hydrate plug during steady-state operation is considered unlikely since the flowing temperatures are outside of the hydrate region. However, the indications of having formed a hydrate plug will be the same for a restart and steady-state flow. Figure 3.9 gives a simplified schematic of the flowline with the hydrate plug along with the affected subsea components. For simplicity, this figure can be used for any flowline pair (except PFL 03/04). In all cases, Flowline 1 (not necessarily PFL 01) refers to the flowline with a hydrate plug and Flowline 2 (not necessarily PFL 02) refers to the second flowline in the pair that does not have a hydrate plug. During production, the Pigging Isolation Valve (PIV) is closed and a hydrate plug in Flowline 1 results in a pressure increase at the manifold (and at the tree of the wells flowing into Flowline 1) to the Shut-in Tubing Pressure (SITP). The pressure at the downstream end of the plug (measured at the riser base and topsides of Riser 1) decreases. The plug in Flowline 1 does not affect Flowline 2, since the flowlines are isolated when the PIV is closed. The lack of flow in the flowline also results in a decrease in flowline temperatures, however, due to the flowline insulation, this temperature decrease may be too slow to recognise over the short time-scale that the plug is expected to form. Thus pressure measurements will be the primary means of detecting a hydrate plug. Plug Formation in Flowlines (except PFL 03/04) Refer to schematic in Figure 3.9 for relevant locations of pressure gauges. Use Appendix 3A to determine the pressure tags for the relevant pressure sensors. •

Increase in pressure at manifold (Flowline 1 side (wells flowing to Flowline 1) to SITP –

[Pm-1])

and at the tree

Manifold pressure and tree pressure upstream and downstream of the choke read the same on the affected wells

•

Decrease in pressure at the base of Riser 1 (Prb-1) and topsides of Riser 1

•

Decrease in Riser 1 temperature topsides –

Due to flowline insulation, the temperature decrease may not readily observed


OPRM-2003-0302D

Page 21 of 64

30-April-2006


Unrestricted

Flowline PFL 03/04 The case of a hydrate plug in either PFL 03 or 04 is similar to the general case of hydrate plug in one of the flowlines except that the pressure downstream of the plug is determined at manifold PM3 instead of at the riser base (or topsides). There is an additional difficulty in this case since the downstream pressure will still see contributions from other wells flowing into PM3, but there should be a significant decrease in pressure (and temperature). However, the upstream pressure (PM4 manifold pressure in Flowline 1 and in all wells flowing to Flowline 1 at PM4) will increase to the SITP pressure. Plug Formation Between PM3 and PM4 (PFL 03/04) Refer to the schematic in Figure 3.10 for relevant locations of pressure gauges: •

Increase in pressure drop in Flowline 1 between PM3 and PM4 –

•

Pressure increase at PM4 (Flowline 1 side), pressure decrease at PM3 (Flowline 1 side)

Pressures at the tree of the wells flowing to Flowline 1 (through PM4) increase to SITP

Riser 2

Subsea Manifold

Pm-2

Topsides Pressure

Flowline 2 Prb-2 Riser Base Pressure

PIV Pm-1

Hydrate Plug

Riser 1

Prb-1

Flowline 1 Manifold Pressure OPRM20030302D_058.ai

Figure 3.9 – Schematic of Hydrate Plug in Flowline (Except PFL 03/04)


OPRM-2003-0302D

Page 22 of 64

30-April-2006


Unrestricted

Manifold (PM4)

Manifold (PM3)

Flowline 2

Flowlines to FPSO PFL 05/06

Manifold (PM3) Pressure

PIV Pm-1

Hydrate Plug Flowline 1

Manifold (PM4) Pressure OPRM20030302D_059.ai

Figure 3.10 – Schematic of Hydrate Plug in Flowline (PFL 03/04) Riser A hydrate plug that forms in the riser (Riser 1) will show the same indications as in the flowline except for the pressure reading at the base of Riser 1. In this case, both the manifold pressure (Flowline 1 side [Pm-1]) and the riser base pressure (Riser 1 [Prb-1]) increase to the SITP. There will still be a decrease in pressure at the downstream end of the plug (measured at Riser 1 topsides). The hydrate plug in Riser 1 will not affect Flowline 2, since the PIV is closed. Plug Formation in Riser Please refer to the schematic in Figure 3.11 for relevant locations of pressure gauges: •

Same indication as in the flowline –

Pressure also increases at the base of the Riser 1 to SITP

Riser 2 Topsides Pressure

Subsea Manifold

Pm-2

Flowline 2 Prb-2 Riser Base Pressure

PIV

Hydrate Plug

Prb-1

Pm-1

Riser 1

Flowline 1 Manifold Pressure OPRM20030302D_060.ai

Fig 3.11 Schematic of Hydrate Plug in Riser

Figure 3.11 – Schematic of Hydrate Plug in Riser


OPRM-2003-0302D

Page 23 of 64

30-April-2006


4.1.2

Unrestricted

Hydrate Plug Remediation This document is only meant to help provide guidelines for relatively simple cases. Cases involving hydrate plugs in both flowlines are much more complex in terms of safely performing a remediation and hence are not discussed in this document. In these cases additional support is recommended before any plug remediation procedures are attempted. Flowlines (Except PFL 03/04) The following discussion is based on Figure 3.9 and the convention that the hydrate blockage is in Flowline 1 (not necessarily PFL 01), and that Flowline 2 (not necessarily PFL 02) does not have a hydrate blockage. Refer to Figure 3.12 for a flowchart representation of the procedures presented in this section. Since the precise location of the blockage in Flowline 1 is not known, depressurisation of the flowline from both ends is the safest option. Four pressures are monitored during this process: •

Prb-1, the

pressure at the riser base of Flowline 1

•

Pm-1, the

pressure at the manifold of Flowline 1

•

Pm-2, the

pressure at the manifold of Flowline 2

•

Prb-2, the

pressure at the riser base of Flowline 2

Step 1 – Shut In Flowline 1 Once it has been determined that Flowline 1 has a blockage, the following steps should be followed as soon as possible: •

Shut in Flowline 1 by closing the topsides Flowline 1 shut-off valve

•

Shut in the wells feeding Flowline 19

•

Secure all wells flowing to Flowline 1 –

Displace wellbore and jumper with methanol

Step 2 – Set-up for Blowdown Configure topsides piping to allow blow down of Flowline 2 to the flare, and to allow blow down of platform end of Flowline 1 through the Low Pressure (LP) separator. Step 3 – Shut In Flowline 2 Wells Shut in Flowline 2 and the wells feeding Flowline 2 such that the pressure Pm-2 at the subsea manifold is within 14bar (200psi) of Pm-1, but the smaller the pressure drop the better. The pressure gradient across the manifold should be small in the event that the plug is near the manifold. A straightforward way to set a safe pressure at Pm-2 is as follows: (1)

Shut in Flowline 2 by closing a topsides valve.

(2)

Allow the pressure at the manifold, the wells feeding Flowline 2.

Pm-2,

to rise to close to

Pm-1,

and then shut in

9 After the wells feeding flowline 1 have been shut in, the pressure at the manifold, Pm-1, is expected to be 500psi to 3000psi greater than the pressure at the riser base, Prb-1. The difference between these two pressures (Pm-1 – Prb-1) is the pressure across the blockage, Pab. Pab times the pipeline internal diameter is the driving force on the blockage.


OPRM-2003-0302D

Page 24 of 64

30-April-2006


(3)

Unrestricted

Blow down Flowline 2 topsides to the flare until the target safe pressure at is reached.

Pm-2

Target safe pressure at Pm-2 is within ±14bar (200psi) of Pm-1. (4)

Close the topsides Flowline 2 valve.

Step 4 – Reduce Pressure in Flowlines Open PIV to equalise pressures between the two flowlines. Blow down Flowline 2 until Prb-2 reaches a pressure of 7bar to 14bar (100psi to 200psi) below Prb-1. When this occurs, begin simultaneous blowdown of both flowlines, making sure that Prb-2 remains 7bar to 14bar (100psi to 200psi) below Pm-1. Near the end of the pressure lowering, gas lift may be required to further lower the pressure. Step 5 – Reduce Pressure Outside of Hydrate Region Blow down the flowlines to pressures outside of the hydrate region so that the blockage will melt (dissociate). The lower the pressure, the faster the blockage will melt10. Blow down Flowline 1 and Flowline 2 to as low pressures as possible while maintaining Prb-1 7bar to 14bar above Pm-1. The pressure at the blockage must be less than the hydrate equilibrium pressure at ambient seafloor temperature11 in order for hydrates to melt (dissociate). If the pressure cannot be lowered enough to melt the hydrates in a reasonable amount of time (refer to Figures 3.6 and 3.7), then alternative means of hydrate remediation are necessary and will be recommended by the blockage response team. Step 6 – Hydrate Removal As soon as pressure communication is observed across the plug, methanol should be injected into the flowline (via one or more of the wells feeding that flowline). This will help ensure that residual hydrate in the flowline does not form another plug during displacement and will aid in melting the remaining hydrate. A total volume of about 50 barrels of methanol should be injected into the flowline prior to start-up. Step 7 – Dead-oiling Circulate dead-oil from Flowline 2 into Flowline 1. This moves residual hydrates closer to topsides, where blockages are easier to remediate if they form. Hot-oiling is preferred to dead-oiling, if it is available. Flowline PFL 03/04 This situation is similar to the general flowline plug case, except that there is an additional manifold (PM3) that needs to be taken into account. The flowlines should be configured as illustrated in Figure 3.10, in that Flowline 1 flows into either PFL 05 or 06 and Flowline 2 flows only into the other flowline (PFL 06 or 05). The Crossover Valve (XOV) at manifold PM3 is to remain closed during the remediation process. This will create a single large dual-flow loop that can be remediated using the procedures given for the other flowlines.

10

For instance, if the pressure is reduced to 7bar (100psi), it will take a hydrate plug (with the 702 fluid) about a day to melt; and if the pressure is only decreased to 14bar (200psi), it will take a hydrate plug about a month to melt. 11

15bar (215psi) for 702 or 9bar (130psi) for 710 Refer to Table 3.2 Figure 3.1.


OPRM-2003-0302D

Page 25 of 64

30-April-2006


Unrestricted

The case studies given in Appendix 3B Paragraph 2.0 give several examples where plugs have been safely remediated by depressurising the flowline at both ends. These examples are meant to illustrate the wide range of conditions that lead to hydrate formation and the various locations (within the flowline) where a plug can form. In all cases, the pressure in the flowline was safely reduced below the HDP and the plug melted. Particular note should be paid to the very well-documented case study of the ARCO hydrate plug. This plug formed in an insulated line and took 23 days to remediate once the pressure was reduced, and further reinforces that the removal of a hydrate plug is not a fast process and may take many days. Riser Remediation of a hydrate plug in the riser (Riser 1) can be handled in the same way as in the case of a hydrate plug in one of the flowlines. However, extra caution needs to be taken to ensure that the pressure at the base of Riser 1 is less than the pressure topsides at Riser 1. In this case, it may also be recommended to maintain a high pressure downstream (between hydrate plug and topsides) of the plug and to do a one-sided depressurisation by aggressively blowing down Flowline 2. This will ensure that any plug movement is away from the FPSO. Figure 3.13 presents a flowchart for plug removal in the riser. Note: Figure 3.13 states a pressure drop limitation across the plug of 28bar (400psi). This pressure drop may be exceeded if one-sided depressurisation is used.


OPRM-2003-0302D

Page 26 of 64

30-April-2006


Unrestricted

Close in all wells flowing to affected manifold Treat jumper and wellbore by displacing with methanol

If hydrate plug is in PFL 03/04, then: (1) Close in all wells flowing to PM 3. (2) Make sure the PIV at PM 3 is closed. (3) Use WSV to route flow from Flowline 1 to either PFL 05 or 06. (4) Route flow from Flowline 2 to another flowline (PFL 06 or 05).

Open PIV to equalise pressure in Flowlines 1 and 2

Begin blowdown of Flowline 2 (flowline without hydrate plug) to flare. Blow down Flowline 1 (flowline without hydrate plug) to LP separator

Make sure pressure gradient across plug does not exceed 14bar (200psig), measured as the difference between the manifold pressure (Pm-1) and the pressure at the base of the riser 1 (Prb-1)

Is the manifold pressure below HDP?

Use riser gas lift to further reduce the pressure in the flowline

Yes

Maintain pressure drop of 7 to 14bar (100 to 200psig) across the plug

No

Monitor pressure at the base of riser 1 (Prb-1) for signs of pressure communication (eg sudden pressure decrease)

Is there pressure communication across plug?

Yes

No

If possible, use gas lift to further reduce the pressure (whilst still maintaining an acceptable pressure drop across the plug)

Inject 50 barrels of methanol into the flowline through the manifold (via MIV 2). Start dead-oiling (or hot-oiling if available) the flowline from Flowline 2 to Flowline 1 OPRM20030302D_052.ai

Figure 3.12 – Remediation Procedure for Hydrate Plug in Flowline


OPRM-2003-0302D

Page 27 of 64

30-April-2006


Unrestricted

Close in all wells flowing to affected manifold Treat jumper and wellbore by displacing with methanol Open PIV to equalise pressure in Flowlines 1 and 2

Begin blowdown of Flowline 2 (flowline without hydrate plug) to flare. Blow down Flowline 1 (flowline without hydrate plug) to LP separator

Ensure pressure gradient across plug does not exceed 28bar (400psig), measured as the difference between topside of riser 1 and the base of riser 1

Use riser gas lift to further reduce the pressure. Note that this only applies to riser 2

Is the pressure at the base of riser 1 (Prb-1) below HDP?

Yes

Maintain pressure drop of 7 to 14bar (100 to 200psig) across the plug

No

Monitor pressure for signs of pressure communication across plug, either to decrease topsides (riser 1) or a sudden pressure increase (spike) at the base of riser 1 (Prb-1) Is there pressure communication across plug?

Yes

No

Inject 50 barrels of methanol into the flowline (via the gas lift riser). Start dead-oiling the flowline from Flowline 2 to Flowline 1 OPRM20030302D_053.ai

Figure 3.13 – Remediation Procedure for Hydrate Plug in Riser


OPRM-2003-0302D

Page 28 of 64

30-April-2006


Unrestricted

4.2

Wellbore Jumper and Manifold

4.2.1

Hydrate Plug Formation A hydrate plug in the jumper or manifold prevents flow from a particular well into the flowline. When a plug is formed in the jumper (refer to Figure 3.14) or manifold, the pressure at the tree increases to the SITP even though the choke is open. The pressures upstream (production pressure) and downstream (outlet pressure) of the choke are the same. The downhole pressure increases to the Shut-in Bottomhole Pressure (SBHP) and the pressure at the manifold begins to drop off. Temperature also begins to decrease at the tree, but this decrease may not be noticeable (since it will occur slowly). Note: The temperature must be in the hydrate formation region in order to form hydrates. When there is more than one well flowing to a single flowline, the same indications of hydrate formation are present, including the decrease in pressure at the manifold. The manifold pressure continues to see contributions from the other wells so the pressure does not decrease as low as it would with only one well flowing to the manifold, but the change in pressure is significant enough to be detected. Plug Formation in Jumper/Manifold •

Pressure at tree (production and outlet pressure) increases to SITP –

Pressure upstream (production pressure) and downstream (outlet pressure) of the choke equalise

•

Downhole pressure increases to the SBHP

•

Reduction in manifold pressure and temperature –

The magnitude of these decreases depends on the number of wells flowing into the flowline MIV 1

Methanol Line

XOV

MIV 2

ASV

PSV

Production Pressure Choke

AWV

Annulus

PWV PWV

SWV

Manifold

Outlet Pressure

PIV Hydrate Plug

WSV

WSV

SCSSV

Manifold Pressure

Downhole Pressure

Flowlines to FPSO

OPRM20030302D_066.ai

Figure 3.14 – Schematic of Hydrate Plate in Jumper


OPRM-2003-0302D

Page 29 of 64

30-April-2006


4.2.2

Unrestricted

Hydrate Plug Remediation If a plug is formed in the jumper, two-sided depressurisation is not possible. One-sided depressurisation may be used to melt the plug, but due to safety considerations, it should not be attempted unless the methanol remediation strategy fails. In order for methanol to melt the plug, the methanol must contact the hydrate. Due to the shape of the jumper section, it may not be possible to get methanol to the hydrate surface. In order to have a reasonable chance of getting methanol to the hydrate surface, the methanol should be ‘rocked’ into the jumper. This strategy has proved successful in the wellbore (refer to the Popeye case study), but has not been tried in a jumper. The flowchart shown in Figure 3.15 gives the steps to follow in order to use this methanol ‘rocking’ procedure. The first step is to isolate the affected jumper from the manifold by closing the WSV. Production from the wells flowing to the affected manifold does not need to be stopped. If the methanol strategy does not work, then the production from the other wells will need to be stopped. Methanol should first be bullheaded into the wellbore to protect that area against hydrates. Once the well is protected, all valves should then be closed except for the SCSSV, Production Master Valve (PMV) and Sacrificial Wing Valve (SWV) and the choke. Use MIV2 to inject methanol into the jumper. Once the pressure (production and outlet) reaches a level that is 21bar (300psig) higher than the SITP, close MIV2. Before initiating ‘rocking’, ensure that the SWV is open. A ‘rock’ has three steps: (1)

Close the Production Wing Valve (PWV). Conduct a blockage-breach test every four rocks or if a blockage breach has been indicated in Step 3 (refer to the discussion below).

(2)

Inject methanol through MIV2. This should increase the outlet pressure. Stop injecting methanol (close MIV2) when the outlet pressure is greater than the SITP by 300psi.

(3)

For 60 to 90 minutes, monitor for blockage breach. Blockage breach may be indicated in several ways (refer to the discussion below). If the blockage is not breached, then open PWV (and drop the outlet pressure to the downhole pressure).

‘Rocking’ the methanol into the jumper is achieved by repeating these three steps many times. After every four ‘rocks’, a blockage-breach test (discussed below) should be conducted. The success of this method depends on the proximity of the hydrate plug to the tree. Since the pressure increase in the line will be small in relation to the SITP, the volume of methanol that is injected during each pressure cycle will be small. Only for cases when the plug is reasonably close to the tree will this method work.


OPRM-2003-0302D

Page 30 of 64

30-April-2006


Unrestricted

Blockage breach may be indicated by a rapid pressure reduction during methanol injection or during the monitoring period; by a slow but significant pressure reduction during the monitoring period; or by a significant increase in the amount of methanol that is injected during Step 2 (over the amount injected in earlier ‘rocks’ to the same pressure). After every four ‘rocks’ or after indication of a blockage breach, a blockage-breach test should be conducted in Step 1 of a ‘rock’. The blockage-breach test is: (1)

After completion of Step 3 of a ‘rock’ and closing of PWV, open WSV and observe the well outlet and manifold pressures for 15 minutes.

(2)

If the well outlet pressure remains constant and above the manifold pressure, then the blockage has not been breached – end of test.

(3)

If the well outlet pressure drops significantly and drops to the manifold pressure then blockage breach is indicated. Proceed to blockage-breach confirmation (Step 5).

(4)

If the well outlet pressure is about equal to the manifold pressure prior to opening WSV in Step 1, then it may not be clear whether the system behaves as described in Step 2 or Step 3. If this is the case, or it is not clear as to whether or not the blockage is breached for whatever reason, then MIV2 should be opened very briefly. If opening MIV2 does not raise the outlet pressure, or the outlet pressure rises and then decays back down (within 15 minutes) too near the manifold pressure, then blockage breaching is indicated – proceed to Step 5. Otherwise, this is the end of the test.

(5)

Confirm blockage breach by opening MIV2 (allowing methanol to flow into the jumper) and confirming that the outlet pressure does not increase. Continue until one jumper volume of methanol has been injected and until there is no evidence of flow restriction in the jumper – once these conditions are met, the blockage has been cleared sufficiently to restart the well.

Once the blockage has been cleared, open the WSV and push any remaining hydrate debris into the manifold with (additional methanol and) well production. If production from the other wells was never stopped, the manifold should be warm and help to quickly melt the remaining hydrate. The total time to remove a plug in this manner should be on the order of 1 day (24 hours). If the plug has not released within 3 days, then it is time to use an alternative remediation method. The alternative methods include: •

One-sided pressure reduction

•

Replacement of the jumper section

During execution of the methanol ‘rocking’ method, alternative remediation methods should be evaluated to determine the most feasible method for the current blockage. Both alternative methods listed require that production from the other wells to be stopped, so preparations should be made to shut in the other wells.


OPRM-2003-0302D

Page 31 of 64

30-April-2006


Unrestricted

Treat wellbore by displacing with methanol (should be done prior to the cooldown time)

Close the following valves on the affected tree: PWV, MIV1, MIV2, AWV, XOV, ASV, PSV Close WSV to isolate well from mainfold Open: PMV, SCSSV, SWV, choke

Open MIV2 and inject methanol injo jumper to 21bar (300psig) above the SITP

Close MIV2

Monitor pressure at tree (outlet pressure) for any change for 60 to 90 minutes

Conduct a blockage-breach test every fourth cycle by opening the WSV

Did pressure decrease at tree?

Close PWV

Yes

Inject 20bbls of methanol into wellbore and jumper, then proceed to start-up procedures

No

Open PWV and relieve pressure in the jumper into wellbore OPRM20030302D_054.ai

Figure 3.15 – Remediation Procedure for Hydrate Plug in Jumper/Manifold


OPRM-2003-0302D

Page 32 of 64

30-April-2006


Unrestricted

4.3

Wellbore/Tree (Upstream of Inhibitor Injection Point)

4.3.1

Hydrate Plug Formation The formation of a plug in the wellbore is similar to the case of a plug formed in the jumper section. When a wellbore hydrate plug forms, the downhole pressure gauge increases to the SBHP, and the production and outlet pressures approach the manifold pressure regardless of the open choke setting. Plug Formation in Wellbore/Tree •

Downhole gauge increases to SBHP

•

Pressure at tree is the same as the manifold pressure (in spite of choke being not fully open)

•

No change in production pressure as the choke opening is changed MIV 1

Methanol Line

MIV 2

XOV

ASV

PSV


AWV

Annulus

PWV

SWV Outlet Pressure

PWV

Jumper to Subsea Manifold Hydrate Plug

SCSSV Downhole Pressure

OPRM20030302D_067.ai

Figure 3.16 – Schematic of Hydrate Plug in Wellbore


OPRM-2003-0302D

Page 33 of 64

30-April-2006


4.3.2

Unrestricted

Hydrate Plug Remediation If a plug is formed in the wellbore, two-sided depressurisation is not possible. Generally, one-sided pressure reduction is not recommended for wellbore blockages due to the difficulty of safe execution in the wellbore environment. ‘Rocking’ methanol into the well is recommended as the first means of removing the hydrate blockage. The flowchart in Figure 3.15 shows the ‘rocking’ procedure steps. Once the plug is detected, close the WSV to isolate the affected well from the manifold. Production from the other wells does not need to be stopped immediately, but preparations should be made in case the hydrate plug cannot be easily removed. Open the SCSSV, PMV and the choke, and close all other valves (refer to Figure 3.16 for relevant valves). Use MIV2 to inject methanol into the well by opening PWV. Increase the production pressure until the maximum pressure is achieved (~345bar (5000psi) the rating of the tree) and then close MIV2 and monitor for blockage breach for approximately 60 to 90 minutes. Then slowly open the SWV to relieve the wellbore pressure into the jumper. Allow the production pressure to decrease to 21bar (300psig) less than the last known SITP, then close SWV. Repeat this cycle until the plug is melted and pressure communication is established between the downhole pressure and the production pressure. Initially, the top of the wellbore may be filled with gas. The methanol can be used to remove the gas and fill the top section of the wellbore with liquid. This will prevent the plug from breaking free and having enough momentum to cause any damage at the tree. Therefore, during the first few pressure cycles, the pressure should only be decreased to 7bar (100psig) less than the last known SITP. When the wellbore is liquid filled, the pressure should begin to increase very quickly with the addition of only a small amount of methanol. Pressure communication can be detected by a sudden change in the production pressure, which may be either a sudden increase or decrease depending on when in the pressure cycle communication is established. The downhole pressure sensor should also fluctuate, but this may be less noticeable than the production pressure. The volume of methanol that can be injected into the wellbore will also increase once the hydrate blockage is breached. Once the blockage is breached and pressure communication is established between the downhole pressure sensor and production pressure sensor, close the SWV (the PWV should still be open) and inject ~50 barrels of methanol into well. As methanol is injected, there should be some indication (pressure increase) noted on the downhole pressure sensor. If this is all successful, production can be restarted. This methanol ‘rocking’ has been successfully applied in the past (refer to the Popeye case study). At Popeye, the plug took roughly a day to remove once the methanol ‘rocking’ procedure commenced. In other cases, such as at Auger, a coiled tubing unit had to be brought in to melt a hydrate blockage which was much larger than the one experienced at Popeye.


OPRM-2003-0302D

Page 34 of 64

30-April-2006


Unrestricted

Select affected well and close the WSV to isolate well from manifold

All valves should be closed, except the SCSSV, PMV and the choke

Open PWV and use MIV2 to inject methanol into the well

Inject methanol into the well until maximum flowline pressure is achieved

Close MIV2

Close SWV

Monitor wellbore pressure and watch for pressure communication for 60 to 90 minutes

Pressure communication?

Yes

Inject 50bbls of methanol into wellbore and proceed to well start-up procedures

No

Open SWV to relieve wellbore pressure into jumper, pressure should be decreased to 21bar (300psig) below last known SITP. (During the first few cycles, the pressure should only be decreased to 7bar (100psig) below the last know SITP)

OPRM20030302D_055.ai

Figure 3.17 – Remediation Procedure for a Hydrate Plug in the Wellbore


OPRM-2003-0302D

Page 35 of 64

30-April-2006


4.4

Umbilicals

4.4.1

Hydrate Plug Formation

Unrestricted

The risk of hydrate formation in the umbilical lines is highest during times when there are large pressure fluctuations in the flowline. Many control system interlocks are put in place to prevent pressure fluctuation-induced hydrate formation in the umbilical lines. Examples of this include: Chemical Injection Valves (CIVs) have to be closed before reducing the choke and shutting in a well; CIVs automatically close when a well gets shut in etc). It should be noted that hydrate plugs in the umbilical might occur during steady-state operation due to pressure fluctuations created by slugging. These pressure fluctuations have the potential to push production fluids into the umbilical. Once production fluids are in the umbilical, it is relatively easy to form a hydrate plug due to the small diameter of the umbilical lines. Based on GoM experience, the formation of hydrate plugs in umbilical lines is fairly common. It should be noted that most of the cases of hydrate plug in the umbilical line are due to manual operation during transients (ie the correct operating logic was not followed and valves were opened/closed in the wrong order). A plug in one of the umbilical lines will be detected as an increase in the injection pressure of the affected chemical line and loss of flow of that particular chemical. The example in Figure 3.18 shows a hydrate plug formed in the methanol line, but there is the potential to form a hydrate plug in any of the umbilical lines. The detection and remediation process is the same for a plug formed in any of the umbilical lines. Hydrate Plug

MIV 1

Methanol Line

MIV 2


Annulus

PWV

SWV Outlet Pressure

PWV

Jumper to Subsea Manifold

SCSSV Downhole Pressure

OPRM20030302D_068.ai

Figure 3.18 – Schematic of Hydrate Plug in Umbilical Line


OPRM-2003-0302D

Page 36 of 64

30-April-2006


4.4.2

Unrestricted

Hydrate Plug Remediation Since hydrate plugs formed in umbilicals are typically much smaller hydrate plugs, it may be possible to push them out by increasing the pressure on the upstream end of the plug. Similar to the procedure of methanol rocking to remove a hydrate plug in the jumper or wellbore, the chemical in the umbilical can be used to cycle the pressure up and down in an effort to remove the plug. Note: This strategy attempts to mechanically remove the hydrate plug and not to physically melt it as in all the other remediation strategies. The flowchart in Figure 3.19 shows the blockage clearing procedure steps. The pressure in the umbilical should be increased to 70bar (~1000psig) above the normal hydrostatic pressure in the umbilical. Then the pressure can be released from the chemical line but still kept above the flowline pressure. Since the chemical in the line is not likely to actually melt the hydrate, the time between cycles can be short (about 15 to 30 minutes). This process may be repeated several times. The likelihood of success using this method is small, but is much easier than other remediation options. If this strategy does not work, then the system must be depressurised. This will involve stopping production from all wells flowing to the affected manifold and blowing down the entire flowline. The first step is to assess how critical this particular chemical is to maintain the current production. If clearing the blockage can wait until the next planned shut-in, then that would be recommended. Once this is done, the plug in the umbilical can be melted using one-sided depressurisation. In this case there is much less of a safety concern associated with one-sided depressurisation due to the small size of the hydrate plug.

Relieve pressure in umbilical to hydrostatic pressure

Increase pressure in umbilical to maximum pressure and leave for 15 to 30 minutes

Did this clear blockage?

Yes

Inject chemical through the affected chemical line to fully clear blockage

No

If chemical line is critical to operating system, begin preparations to shut in and blow down flowlines

OPRM20030302D_056.ai

Figure 3.19 – Remediation Procedure for a Hydrate Plug in an Umbilical


OPRM-2003-0302D

Page 37 of 64

30-April-2006


4.5

Gas Lift Riser

4.5.1


Unrestricted

The risk of forming a hydrate plug in the gas lift riser is highest during a shutdown and an aborted start-up. A plug in this section is detected by lack of gas flow in the riser and an increase in the gas lift riser topside pressure. Determining which pressure sensors are in communication with one another can help to localise the hydrate plug. Upstream of the plug all pressures should read the same, and downstream of the plug all pressures should read the same as the riser base pressure sensor in the flowline. The hydrate plug is located between the two adjacent pressure sensors not in pressure communication. If the gas lift riser pressure and the gas lift riser topside pressure are equal, then the plug is between the gas lift riser pressure sensor and the flowline (refer to Figure 3.20). If the riser base pressure and the gas lift riser pressure are equal, then the plug is between the gas lift riser and the gas lift riser pressure sensor (refer to Figure 3.21). Note: These figures refer to the plug being either upstream or downstream of the methanol injection point. There is a small risk that the plug is between the methanol injection point and the pressure sensor, but due to the small volume between these sections, this is highly unlikely. Plug Formation in Gas Lift Riser •

Increase in gas lift riser topside pressure

•

No flow in gas lift riser

Unlike other portions of the Bonga system, a plug in this riser is much easier to remediate should it form12. Methanol Line

Gas Lift Riser

GLR Topside Pressure

To Production Riser

GLIV1 Gas Lift Riser Pressure GLIV2

Hydrate Plug

To Subsea Manifold

Flowline Riser Base Pressure

OPRM20030302D_069.ai

Figure 3.20 – Schematic of Hydrate Plug in Riser Gas Lift System (Between Methanol Line and Flowline) 12

Email from Sada Iyer, March 2003.


OPRM-2003-0302D

Page 38 of 64

30-April-2006


Unrestricted

Methanol Line

Gas Lift Riser

GLR Topside Pressure

GLIV1 Hydrate Plug

To Production Riser

Gas Lift Riser Pressure

GLIV2

To Subsea Manifold

Flowline Riser Base Pressure

OPRM20030302D_070.ai

Figure 3.21 – Schematic of Hydrate Plug in Riser Gas Lift System (Between Methanol Line and GLR Topsides) Since the gas lift riser is close to the FPSO and since methanol can easily be delivered to the riser, the flowline does not need to be blown down to remove a hydrate plug in the gas lift riser. Flow in the flowline will actually help to remove the plug since it will warm up the lower portion of the gas lift riser. Note: There are many possible scenarios regarding the plug location in the gas lift riser. It is assumed that the plug does not form between the gas lift riser pressure sensor and Gas Lift Injection Valve (GLIV) 1 or between the gas lift riser pressure sensor and the methanol injection point. If either of these cases occur, the following remediation methods will still work, but the pressure gradient in the riser will tend to move the plug in the wrong direction (away from the flowline). However, due to the small volumes in these sections, there is not enough energy to move the plug any significant distance. If the plug is located downstream of the gas lift riser pressure sensor in the gas lift riser, refer to Figure 3.20 for a description of the remediation process. The first step should be to try and push the hydrate into the flowline. Close GLIV2, open GLIV1 and then use the methanol line to pressurise the gas lift riser to the maximum pressure (pressure measured at the gas lift riser pressure). Maintain this pressure on the upstream end of the plug and monitor the rate at which methanol is being injected into the gas lift riser. If the hydrate plug is solid, then the methanol volume will be very near zero. If the plug is moving or is porous enough to allow methanol to flow through, then some finite volume of methanol is needed to maintain the pressure in the gas lift riser. If after 6 hours, the pressure in the riser section has not changed and the volume of methanol injected is zero, then this method is not likely to work. Conversely, if methanol continually needs to be injected into the gas lift riser, then eventually the plug will either be melted or pushed into the flowline.


OPRM-2003-0302D

Page 39 of 64

30-April-2006


Unrestricted

If increasing the pressure with methanol does not remove the hydrate plug, then the gas lift riser will need to be blown down in order to remove the hydrate plug. GLIV1 should be closed and then the gas lift riser can be fully depressurised (topside gas lift riser and gas lift riser pressure are as low as possible). The valve upstream of the methanol injection point (GLIV2) is then closed and methanol is injected to pressurise the section between GLIV1 and GLIV2. The valve between the hydrate plug and the methanol-filled section (GLIV1) is then opened. Methanol can then be used to pressurise the section of the gas lift riser between the hydrate plug and GLIV2. At this point, the pressure on the methanol side of the plug should be greater than the flowline so that when the plug releases, it will move towards the flowline. If the plug does not release within 60 to 90 minutes, then the above process should be repeated. The time expected to remove a plug in the gas lift riser using the above method should be less than 1 day (24 hours). For the case when the plug is upstream of the methanol injection point, refer to Figure 3.21. The figure shows that the plug is between GLIV2 and the flowline, but this may not necessarily be true. Therefore, the pressure in the gas lift riser should be increased (or decreased) to a pressure that is 14bar (200psig) greater than the riser base pressure, leaving GLIV2 open. Open GLIV1 to relieve the pressure downstream of the plug to the flowline pressure, and then close GLIV1. Inject methanol into the gas lift riser and increase pressure until it is slightly less than the gas lift riser topsides pressure. Repeat this process every 60 to 90 minutes until plug releases. The time required to remove a plug using this method should be in the order of a day. However, this case is less likely to succeed than the other gas lift riser scenario (refer to Figure 3.20) and other remediation techniques may be necessary.


OPRM-2003-0302D

Page 40 of 64

30-April-2006


Unrestricted

Close GLIV2 and open GLIV1

Close GLIV1 and GLIV2

Use methanol to maintain the maximum pressure in the gas lift riser

Open GLIV2 and depressurise the gas lift riser

Monitor the gas lift pressure for 6 hours for any sudden pressure decrease and/or communication with riser base pressure, monitor amount of methanol injected into gas lift riser

Close GLIV2

Inject methanol into gas lift riser to maximum pressure

Open GLIV1

Use methanol to increase gas lift riser pressure to maximum

Is the injected volume of methanol greater than zero?

No

Monitor the gas lift pressure for any sudden pressure decrease and/or communication with riser base pressure

Yes

Pressure equalisation within 60 to 90 minutes?

Yes

Start methanol injection and begin gas lift

No

Close GLIV1

Continue injecting methanol until plug releases

Open GLIV2 and depressurise the gas lift riser

OPRM20030302D_057.ai

Figure 3.22 – Remediation Procedure for a Hydrate Plug in the Gas Lift Riser (Between Methanol Line and the Flowline)


OPRM-2003-0302D

Page 41 of 64

30-April-2006


Unrestricted


Increase/decrease gas lift riser topside pressure to 14bar (200psig) above riser base pressure

Inject methanol into gas lift riser until pressure equals the gas lift riser topside pressure

Open GLIV1 to reduce pressure at gas lift riser to riser base pressure

Open GLIV2

Monitor gas lift pressure for any sudden pressure increase and/or communication with gas lift riser topside pressure

Pressure equalisation?

Yes

Start methanol injection and begin gas lift

No


OPRM20030302D_061.ai

Figure 3.23 – Remediation Procedure for a Hydrate Plug in the Gas Lift Riser (Between Methanol Line and Topsides)


OPRM-2003-0302D

Page 42 of 64

30-April-2006


4.6

Water Injection Wells

4.6.1


Unrestricted

The most likely scenario in which a plug may form in one of the water injector wells is during a shut-in, which is performed within a few days of initial start-up. Since the Bonga water injection wells are completed into the oil zone, it is possible for gas to migrate back into the well and accumulate where the temperature and pressure are in the hydrate stable region. This involves the occurrence of two situations, gas migration into the well (from the reservoir) and a leaking SCSSV that allows gas to migrate up to the tree. Although hydrates would form during shut-in, they would not be noticed until start-up. During water injection, there is no hydrate risk. The risk also decreases with time as more water is injected into the reservoir and the gas front is pushed further away from the wellbore, which will make migration less likely within the duration of shut-in. Based on experience at Petro-Canada, this problem of gas migration was no longer a concern after about 6 weeks of water injection. A hydrate plug in the water injection wells will be indicated by lack of flow into the well (measured using the venturi meter). The pressure at the tree will also increase. If the plug is in the wellbore, both the injection and inlet pressure will be the same. If this is not the case, then the plug may be located in the tree or jumper instead of the wellbore. 4.6.2

Hydrate Plug Remediation Due to the lack of remediation options for the water injection wells, every attempt should be made to minimise hydrates from forming. The best means for this is to ensure that the SCSSV is closed during shut-in. Unfortunately, the SCSSV is not gastight and may still leak in the closed position. During shut-in, the plug may not have formed a solid mass, so if there is any reason to suspect gas migration into the well, the water injection should be started up as quickly as possible in an attempt to push any hydrate back down into the well. Once it is verified that there is a plug in the water injection wells, the water pressure can be increased in an attempt to move the plug below the SCSSV. In order to make sure that the plug is pushed back below the SCSSV, make sure that at least 50 to 150 barrels of water are flowed into the well. However, this is unlikely to be effective and may only create a more solid hydrate plug. If a plug is detected during start-up, then every attempt should be made to localise the plug and shut in the appropriate valves to prevent any further gas or hydrate from moving back through the flowline even though this is a small risk. At this point preparations should be made to intervene at the well to remediate the plug.


OPRM-2003-0302D

Page 43 of 64

30-April-2006


Unrestricted

Figure 3.24 – Schematic of Hydrate Plug in Water Injection Line


OPRM-2003-0302D

Page 44 of 64

30-April-2006


Unrestricted

Appendix 3A Pressure Tags Table of Contents TABLES Table 3A.1 – Pressure Tags for Production Wells ................................................................46 Table 3A.2 – Pressure Tags for Production Flowlines ..........................................................47 Table 3A.3 – Pressure Tags for Water Injection Wells..........................................................48

Section 3 Appendix 3A Pressure Tags OPRM-2003-0302D

Page 45 of 64

Issue 1.0 31-December-2004

Printed copies of this document may be obsolete. ‘Business Control Documents’ are online in SNEPCO Livelink.


Unrestricted

Well No

Downhole Pressure

Production Pressure (Upstream of Choke)

Outlet Pressure (Downstream of Choke)

690p1

01-PT-800

01-PT-801

01-PT-802

690p2

01-PT-804

01-PT-805

01-PT-806

702p14

01-PT-808

01-PT-809

01-PT-810

702p10

01-PT-812

01-PT-813

01-PT-814

702p15

01-PT-816

01-PT-817

01-PT-818

702p2

01-PT-820

01-PT-821

01-PT-822

702p4

01-PT-824

01-PT-825

01-PT-826

702p5

01-PT-828

01-PT-829

01-PT-830

702p9

01-PT-832

01-PT-833

01-PT-834

710p1

01-PT-836

01-PT-837

01-PT-838

710p2

01-PT-840

01-PT-841

01-PT-842

710p3

01-PT-844

01-PT-845

01-PT-846

710p4/803p1

01-PT-848

01-PT-849

01-PT-850

803p2

01-PT-852

01-PT-853

01-PT-854

803p3

01-PT-856

01-PT-857

01-PT-858

S690p3

01-PT-860

01-PT-861

01-PT-862

S690p4

01-PT-864

01-PT-865

01-PT-866

S702p3

01-PT-868

01-PT-869

01-PT-870

S702p6

01-PT-872

01-PT-873

01-PT-874

S702p7

01-PT-876

01-PT-877

01-PT-878

Table 3A.1 – Pressure Tags for Production Wells Section 3 Appendix 3A Pressure Tags OPRM-2003-0302D

Page 46 of 64




Unrestricted

Production Flowlines

Manifold Pressure

Riser Base Pressure

Topside Pressure

Gas Lift Riser Pressure (between GLIV1 and GLIV2)

PFL 08

03-PT-800

04-PT-812

04-PIT-304

04-PT-804

31-PIT-013

PFL 09

03-PT-801

04-PT-813

04-PIT-324

04-PT-805

31-PIT-023

PFL 11

03-PT-802

04-PT-814

04-PIT-344

04-PT-806

31-PIT-033

PFL 12

03-PT-803

04-PT-815

04-PIT-364

04-PT-807

31-PIT-043

PFL 05

03-PT-804

04-PT-810

04-PIT-404

04-PT-802

31-PIT-063

PFL 06

03-PT-805

04-PT-811

04-PIT-384

04-PT-803

31-PIT-053

PFL 03

03-PT-806

N/A

N/A

N/A

N/A

702p9

PFL 04

03-PT-807

N/A

N/A

N/A

N/A

690p1

PFL 01

03-PT-808

04-PT-808

04-PIT-444

04-PT-800

31-PIT-083

PFL 02

03-PT-809

04-PT-809

04-PIT-424

04-PT-801

31-PIT-073

Well No

Production Manifold

710p1 710p2

PM1

Gas Lift Riser Topside Pressure

710p3 702p2 702p15 PM2 710p4 803p3 702p10 PM3 702p14 702p5 PM4

690p2

PM5

702p4

Table 3A.2 – Pressure Tags for Production Flowlines Section 3 Appendix 3A Pressure Tags OPRM-2003-0302D

Page 47 of 64




Unrestricted

Well No

Injection Pressure

Inlet Pressure

690w1

01-PT-900

01-PT-901

690w4

01-PT-903

01-PT-904

702w1

01-PT-906

01-PT-907

702w2

01-PT-909

01-PT-910

702w4

01-PT-912

01-PT-913

702w5

01-PT-915

01-PT-916

702w6

01-PT-918

01-PT-919

702w9

01-PT-921

01-PT-922

710w1

01-PT-924

01-PT-925

702w8

01-PT-927

01-PT-928

710w3

01-PT-930

01-PT-931

803w2

01-PT-933

01-PT-934

803w4

01-PT-936

01-PT-937

R690w2

01-PT-939

01-PT-940

690w3

01-PT-942

01-PT-943

702w10

01-PT-945

01-PT-946

Table 3A.3 – Pressure Tags for Water Injection Wells

Section 3 Appendix 3A Pressure Tags OPRM-2003-0302D

Page 48 of 64




Unrestricted

Appendix 3B Case Studies Table of Contents 1.0

HYDRATE FORMATION IN FLOWLINE ....................................................................50 1.1

2.0

3.0

4.0

Case Study: Popeye ........................................................................................50

HYDRATE REMOVAL IN FLOWLINE ........................................................................51 2.1

Case Study: Tahoe ..........................................................................................51

2.2

Case Study: Petrobras.....................................................................................52

2.3

Case Study: Statoil ..........................................................................................52

2.4

Case Study: ARCO ..........................................................................................53

HYDRATE REMOVAL IN WELL.................................................................................58 3.1

Case Study: Popeye ........................................................................................58

3.2

Case Study: Auger...........................................................................................59

HYDRATE REMOVAL IN A CHEMICAL INJECTION LINE........................................60 4.1

Oregano...........................................................................................................60

Section 3 Appendix 3B Case Studies

OPRM-2003-0302D

Page 49 of 64 .

30-April-2006


Unrestricted

1.0

HYDRATE FORMATION IN FLOWLINE

1.1

Case Study: Popeye Taken from: AP Mehta et al, ‘Fulfilling the Promise of Low-dosage Hydrate Inhibitors: Journey from Academic Curiosity to Successful Field Implementation’, SPE Production and Facilities, February 2003, p73. During steady-state operation, the flowline was being treated with methanol. However, the volume of produced water was too large to be protected with the methanol and hence the flowline was operating in the hydrate region. What was observed during steady-state operation was that there was a slow gradual increase in the pressure drop along the flowline, which is attributed to the formation and accumulation of hydrates in the flowline. The figure below indicates variables that would typically be measured and that show an observable indication of hydrate formation. The pressure drop in the flowline shows a steady increase with the gas rate showing a steady decline. Since the formation of hydrates in this system was observable, actions could be taken to remove the hydrates before they accumulated sufficiently to form a hydrate plug.


OPRM-2003-0302D

Page 50 of 64 .

30-April-2006


Unrestricted

2.0

HYDRATE REMOVAL IN FLOWLINE

2.1

Case Study: Tahoe Taken from: AP Mehta, ‘Hydrate Plug Blockage and Remediation: Case Studies from Operations in the Gulf of Mexico’, EP 2001-3019. Where Tahoe Well #A4 experienced hydrate plugs in a 6in uninsulated flowline connecting the well to the Bud Platform at Main Pass 252 over a distance of 12 miles. This was part of the Tahoe Phase I development; in Tahoe Phase II, three additional gas wells and one oil well were added as subsea tiebacks to newly built facilities at Bud Lite, also at MP252. These new wells were also 6in, 12-mile, dual-uninsulated flowlines and were Shell’s first foray into deepwater subsea tieback development. The oil well Tahoe A3 has its own dedicated 4in x 8in pipe-in-pipe insulated flowlines. Hydrates have formed in all Tahoe gas lines at one point or the other, but each has been remediated quickly due to the ability of carrying out a two-sided depressurisation. A hydrate plug was also reported in March 1999 in the Tahoe oil line. How Methanol is injected on a continuous basis in the Tahoe gas lines to prevent hydrates. The expected sea-bottom temperature is 46°F (at a water depth of 1500ft), with pipeline pressures varying from 2500 to 3300psig. The subcoolings at Tahoe are on the order of 25 to 30°F, under normal flowing conditions. Hydrates are believed to have formed in Tahoe A4 due to failure of the methanol pump. Gradual pressure build-ups were observed in the gas line, but went unnoticed until a plug had formed. Alarms on the pressure monitors and pump were also not geared to pick up on the pump failure. In the case of the Tahoe A3 oil well, hydrate formation was more unexpected. Tahoe A3 was operating with a water cut of less than 1%. This was based upon a Base Sediment and Water (BS&W) of <1% obtained in simple shake-out tests. During these tests, it also appeared that the water was present as a tight emulsion, one that did not easily break out. During a shut-in at Tahoe A3, the low water cut led field personnel to assume that hydrates would not form in this oil line. Upon restart, after a 36-hour shut-in, a hydrate blockage was found. Presumably, the hydrates formed from the water that settled out over the duration of the shut-in and came into contact with the hydrate formers abundantly dissolved in the oil phase. This hydrate blockage was another reminder that hydrates can and do form in oil systems, and that a low water cut is a poor indicator of the hydrate-forming potential of an oil. Plug Remediation The hydrate plug in both Tahoe lines (gas and oil) was remediated by two-sided depressurisation. In the gas line, plug dissolution was relatively rapid. Methanol was also being injected into the gas line, which may have helped in dissolving the hydrate plug. The oil line was recommissioned by injecting methanol continuously upon restart until the pipeline had warmed up to a temperature well above the hydrate formation temperature.


OPRM-2003-0302D

Page 51 of 64 .

30-April-2006


Unrestricted

Key Learnings Monitoring the efficiency of methanol injection pumps is critical in operating gas flowlines requiring continuous inhibition. A surveillance plan for early detection of pump malfunctioning or failure by the addition of alarms is required for smooth operations. Hydrates can and do form in oil lines even at low water cuts. The water cut is a poor indicator in determining whether a line can form a hydrate blockage. For planned shut-ins, methanol should be injected in sufficient quantities to ensure a cold restart. If methanol could not be injected prior to a shut-in, blowing down the line to ensure that it stays outside the hydrate region is recommended.

2.2

Case Study: Petrobras Taken from: ED Sloan, ‘Offshore Hydrate Engineering Handbook’, Case Study C20. Petrobras reported a hydrate blockage in a subsea manifold, located at around 2000ft water depth. The manifold was initially loaded with water, and was not drained and loaded with ethanol prior to production start-up, as is normal practice. Consequently, a hydrate plug formed in the manifold, blocking valves in a production line. However, production was maintained through a test production line. Two methods were attempted to dissociate the pipeline. First, ethanol was injected into the manifold to begin dissociation. Some dissociation did occur (indicated by pressure increases), but the hydrate plug was still present after 2 days. Depressurisation of the manifold was then used to dissociate the plug. Depressurisation was carried out on both sides of the plug, dissociating the plug in 12 hours. Start-up of the pipeline was carried out by filling the manifold with ethanol and then resuming production. Overall, the hydrate plug was in the manifold for 60 days, but production was maintained throughout that time via a test production line. During depressurisation, all production from the wells flowing into the manifold had to be shut down. The total economic loss due to the hydrate was 31,500bbl oil and the wages of two engineers (1 week) and two technicians (3 days).

2.3

Case Study: Statoil Taken from: Gjertsen et al, ‘Removal of a Gas Hydrate Plug from a Subsea Multiphase Pipeline in the North Sea’ Proc BHR Group 1997 Multiphase ‘97 Conference. In January 1996, Statoil (Gjertsen et al, 1997) depressurised a hydrate plug in a North Sea line, which was alternatively used as a black oil producer and a gas injector to maintain reservoir pressure. The oil and water production rates were 18,000ft3/day and 16,242ft3/day respectively, and the gas to oil ratio was usually 100 to 360scf/ft3, a fairly low value. Since the plug was about midway along the 1.6 mile pipeline, there was not an option of using an inhibitor because pipeline topology prevented inhibitor contact with the plug. Since there were no connections at the well the plug had to be depressurised from the platform side only. By considering the hydrate formation curve it was determined that the plug equilibrium pressure was 261psia but that ice would form when the pressure was below 115psia.


OPRM-2003-0302D

Page 52 of 64 .

30-April-2006


Unrestricted

During dissociation the pressure was decreased in steps, and a slow bleedthrough was observed from 0 to 73 hours, from 73 to 90, 95 to 105 hours, and from 105 through 120 hours. During the time prior to 120 hours, the pressure was above the hydrate equilibrium pressure, and while the upstream pressure decreased steadily, it never decreased to the downstream pressure, indicating that the plug was not very permeable to black oil. A second mechanism was that the light oil ends may have been flashing to maintain a constant pressure upstream. However, the increase in downstream pressure occurred much more rapidly as the downstream pressure was lowered, indicating that the plug was porous, even to the black oil. After about 120 hours the line pressure was maintained between 145 to 261psia downstream of the plug. The plug dissociated about 50 to 60 hours after the downstream pressure had been reduced sufficiently for melting by heat influx from the ocean. This was indicated by a sudden upstream pressure decrease from 1890psig to 1160psig, while the downstream pressure increased from 218psig to 1015psig during the same period. The pressure was decreased to 145psig and kept there for over 30 hours to melt the remainder of the hydrates. Restart of the well was accomplished 2 weeks after the original plug developed. This case is another indication of the long times required to remediate a hydrate plug. In 1996 a Statoil black oil pipeline plug occurred in the Norwegian sector of the North Sea. After several precautions, the pipeline was depressurised from one side of the plug, and when the plug had melted the line was maintained at atmospheric pressure for over 1 day to eliminate the light components, which might form hydrates. Before start-up, methanol was injected in the amount of 530 gallons in the 6in ID, 1.6 mile line from the platform. The pipeline was then pressurised with diesel from the platform to the subsea valve in an amount which indicated that the pipeline was nearly empty of liquid after the previous depressurisation to atmospheric conditions. A further injection of diesel corresponding to two pipeline volumes was pumped into the pipeline and well. Subsequently the well and the pipeline were put into production without any hydrate problems.

2.4

Case Study: ARCO Taken from: ED Sloan, ‘Offshore Hydrate Engineering Handbook’, Case Study 14: Plug Formation Setting The gas field is located in the southern North Sea and consists of three subsea wells, flowing into a subsea manifold with a capacity of four well inputs. This well’s gas compositions, temperature and pressure promote hydrate formation, consequently Monoethylene Glycol (MEG) is injected into the manifold and wellheads to thermodynamically inhibit hydrates. The inhibited water, gas and condensate is then pumped through a 22 mile, trenched, insulated export pipeline to a processing platform where water is removed from the condensate. The MEG in the pipeline is recycled and piped back to the manifold via a 3in pipeline piggybacked to the export line.


OPRM-2003-0302D

Page 53 of 64 .

30-April-2006


Unrestricted

Blockage On April 14th, 1996 an unusually large liquid slug overran the platform primary separator causing a temporary shutdown. The liquid slug was remediated, but complete blockage of the pipeline had occurred during shutdown. It was hypothesised that the blockage was a result of a hydrate plug. The reasons were: •

The pipeline-free water, recovered during depressurisation at the platform, did not contain MEG inhibitor. The 3in MEG inhibitor line had ruptured

•

Through backpressurisation the blockage was found to be 150m away from the platform. At this location, the pipeline exited the mud line allowing contents to be rapidly cooled by ocean currents, causing hydrate formation

•

Slight decreases in pressure determined that the blockage had some porosity. This had also been observed for several Statoil hydrate plugs. (In contrast however, two DeepStar Field trials formed low-porosity, low-permeability plugs, which would transmit pressure very slowly and withstand high pressure drops)

•

A liquid slug, which shut down the compressors, was probably caused by a partial hydrate plug pushing a fluid front down the pipeline as it moved

The blockages proximity to the platform posed serious safety concerns. Pipeline depressurisation was necessary to dissociate the hydrate, however it had to be done on both sides of the hydrate plug. If only the blockages platform-side was depressurised, the pressure differential would cause a projectile to form which could destroy the riser piping and damage the platform. The projectile would be life threatening to workers on the platform and result in costly damages to the platform itself. Consequently, depressurisation had to be done through both the platform and the subsea manifold to ensure safety. Projectiles could form due to dissociation, if gas became trapped within multiple plugs. Slow depressurisation was required to remove pressure build-ups in the hydrate plug(s). Several methods were considered. Depressurisation Method Initial Ideas Three questions were raised to determine a proper depressurisation method: (1)

Will the remediation process effectively depressurise the pipeline?

(2)

What is the cost of equipment and modifications?

(3)

How much time is needed to complete the remediation?

Based on these questions, process engineers, consultants, safety management and diving specialists proposed three potential depressurisation methods. They were: (1)

Jack-up Rig Method:

Tow a jack-up rig to the site. From the rig, attach a high pressure riser to the manifold’s subsea tree and flare exiting gas via the rig’s flare stack.

Modification:

A spool piece would have to replace a non-return valve on the manifold’s fourth well intake.


OPRM-2003-0302D

Page 54 of 64 .

30-April-2006


(2)

(3)

Unrestricted

Time Required:

A drilling rig was not currently available, consequently a delay of approximately 8 weeks was needed to locate a suitable rig. The time required for hydrate removal could be 12 weeks.

Estimated Cost:

$1,980,000.

Feasibility:

The large amount of time required to locate a jack-up rig made this an ineffective remediation method, useful in the absence of other methods.

MEG Injection Line Method:

Connect the subsea manifolds spare fourth flange to the 3in MEG pipeline and flare gas at the platform.

Modification:

Subsea work would require a spoolpiece installed between the two pipelines. Secondly, a method of injecting methanol was needed to prevent future hydrate growth. The platform (while in operation) required significant modification to connect the MEG pipeline to its flare stack. To further complicate the matter, all of the MEG currently in the pipeline would need to be stored on the platform, which had limited storage space.

Time Required:

6 to 8 weeks.

Estimated Cost:

Unknown, expected to be higher than the other methods based on the large amount of modifications that were required.

Feasibility:

Substantial modifications to the platform made this remediation method costly and impractical. It was deemed unusable in any circumstance.

FPSO Method:

Connect an FPSO with a processing plant and flare to the subsea manifolds fourth flow loop and process the exiting gas. The connection between the manifold and FPSO would be made through a high-pressure, flexible riser.

Modification:

The platform required no modifications. A diving rig was required to do the subsea work. A valve skid containing both Emergency Shutdown Valves (ESDVs) and a MEG injection valve was also needed. The flexible riser and the manifold would be connected with a spoolpiece.

Time Frame:

A FPSO was available for immediate use, consequently the required time was expected to be 6 to 8 weeks.

Estimated Cost:

$1,906,000.

Feasibility:

This method proved to be the most feasible. The immediate availability of a FPSO and diving rig allowed modifications to begin. It was estimated that the FPSO could be at the site and begin within 2 weeks.


OPRM-2003-0302D

Page 55 of 64 .

30-April-2006


Unrestricted

Establishing Procedures/Permits It took approximately 2 weeks to develop potential remediation processes. Procedures were then written to firmly establish the processes required for the pipeline depressurisation. Procedures considered the safety, process and co-ordination requirements between the diving rig and the FPSO. All parties were educated about the tasks involved. Government permits were applied for at the Health and Safety Executive (HSE) Pipeline Inspectorate and the Department of Trade and Industry (DTI) Oil and Gas Office for additional gas flaring and well modification. The permits were expedited by local agencies to prevent delay in hydrate removal. 2 weeks were required to prepare procedures and permits for depressurisation. In the meantime, the FPSO and diving rig were being equipped for the operation and moving to the field. Depressurisation of the Pipeline The divers first task was to manually locate the subsea manifolds fourth intake and to isolate it from any trees or flow loops. The fourth well intake was then modified with a spoolpiece for connection with the high-pressure riser. The valve skid was now ready to be put in place. Due to the sandy ocean bottom, it became necessary to provide a foundation for the valve sled. The valve skid was placed on a concrete mattress and then stabilised with gravel bag supports coupled with Tirfors, chain blocks and ground anchors. This insured that no movement would transfer from the flexible riser to the valve skid. The valve skid contained ESDVs and a MEG injection system for the pipeline. The diving rig then inspected the flexible riser route to ensure that is was clear of debris. It proceeded to deploy 920ft of the high pressure riser via a tugger rigged with a dead-man’s anchor. The MEG in the riser provided some buoyancy, consequently the line was anchored through concrete mattresses. A 5 tonne clump weight was placed at the bottom of the riser with a buoyancy collar attached to the surface. The FPSO could only process gas at 600psig, consequently it required some modification to process the 1300psig pipeline gas. Additionally, a quick-release valve (QVD) was needed to enable the FPSO to escape from the riser in case of an emergency. This complicated the design because current quick-release valves could not withstand pressures of 1300psig. Initial design placed choke valves in the riser to reduce pressure for the quick-release valve, however this caused control problems and was deemed impractical. An innovative new quick-release valve was developed with a standard valve weak link with three additional hydraulic jacks for manual release. This valve could withstand 1500psig of pressure, allowing choke valves to be placed on the ship’s deck, which simplified control issues. This design enabled a safe, simplified, control of gas pressures from the deck of the FPSO. The buoyancy of the riser prohibited pipeline intake through the FPSO’s moonpool. Spoolpieces were used to allow riser intake from the side of the ship deck. The riser was also steam traced with 1000ft of 1in piping to maintain the minimum process temperature required by the FPSO. All valves and risers were tested and shown to be in working order. Overall the modification and instalment procedures required 1 week before pipeline depressurisation could begin.


OPRM-2003-0302D

Page 56 of 64 .

30-April-2006


Unrestricted

Determining the Pipeline Minimum Pressure Reducing pipeline pressure too much could result in ice formation. This causes significant problems because ice melting might have required significantly more time, than hydrate dissociation. Ice formation was prevented through use of the hydrate equilibrium curve for the field. At constant low pressure, hydrates will continually dissociate, maintaining the equilibrium temperature at that given pressure. The equilibrium pressure at 32°F was 200psig. To prevent ice formation, the pipeline pressure could not drop below 175psig. Consequently, the FPSO reduced the pipeline pressure to 185psig to maximise hydrate dissociation without ice formation. Depressurisation 23 days were required to completely dissociate the pipeline hydrate. Heat transfer between the ocean and the pipeline was slow because the line was trenched and insulated in the sea floor. Dissociation was slightly facilitated by occasional backpressuring which drew methanol into the plug. Backpressuring also proved beneficial in determining the location of the plug. The pressure was monitored for 12 hours after the hydrate was thought to be dissociated. No pressure variation was noticed so the flexible riser was recovered and the depressurisation apparatus dismantled. Throughout the whole operation, no equipment failure occurred and the operation progressed smoothly. Recommissioning the Pipeline After the hydrate was dissociated, there remained significant amounts of free water in the pipeline. The pipeline had to be recommissioned carefully to prevent reformation of hydrates. Above normal amounts of MEG were added to the system before pipeline start-up. One gas well was opened and the platform flow was high to maintain low pressure, preventing hydrate formation. The high intake caused a high gas velocity, which facilitated rapid water removal. The first 12-hour night shift reported 7000ft3 of water received from the separator, the water which would result from a 1.25 mile long (non-porous) hydrate plug. The high flow rate of gas was maintained until the water contained 40% MEG, ensuring that the line was fully inhibited. The pressures and intakes were then returned to normal operating levels. Conclusions The remediation team removed the hydrate plug efficiently. They achieved a monumental task in a very short period of time, preventing more severe economic losses. The procedure and methodology followed could be applied to many different situations. Communication, clear objectives and excellent resources helped in removing the hydrate plug. Despite the efficient remediation effort, the economic impact of the hydrate plug was substantial. The cost of depressurising the pipeline was almost 3 million dollars, without counting lost production. On top of this, relations between the buyers and producers were tested, due to lack of production. Fortunately, good initial relations between the two reduced the impact of the disruption. This case study shows the potential financial loss that can result from hydrate plugs. Hydrate prevention is key in preventing significant economic and production losses.


OPRM-2003-0302D

Page 57 of 64 .

30-April-2006


Unrestricted

3.0

HYDRATE REMOVAL IN WELL

3.1

Case Study: Popeye Where A hydrate plug was formed in the Popeye A4 well. The plug was detected upon initial start-up after completion of the well. The well contained some residual completion fluids including an NH4Cl solution. How The well was shut-in with the residual completion fluids and a mixture of CaCl2/MEG was dumped into the well. It was believed that the CaCl2/MEG mixture was sufficient to inhibit the formation of hydrates. However, later analysis revealed that the combination of the CaCl2/MEG mixture and the residual NH4Cl in the well did not provide sufficient hydrate inhibition and hydrates were formed. Plug Analysis The pressure at the wellhead varied between 3300psig (manifold pressure) and 5500psig (reservoir pressure). Based on the amount of methanol that could be injected into the well, the hydrate plug was determined to be near the top of the well. Plug Remediation The plug was removed by cycling methanol into the well. Methanol was pumped into the well up to 5500psig. This was then left at that pressure for about 1 to 2 hours. Then the wellhead pressure was relieved to the manifold pressure of 3300psig. The wellhead pressure was then monitored for signs of a pressure increase, which was due to either the hydrate plug melting or gas leaking through the hydrate plug. Once the pressure stopped increasing, methanol was again pumped into the well up to 5500psig. With each successive pressure cycle, more methanol was able to be pumped (indicating that the plug was either melting or being pushed down the well) and the amount of time it took for the pressure to stabilise after opening to the manifold increased. During the final pressure cycle, as the methanol was being pumped into the well, the plug broke free and the wellhead pressure did not increase as more methanol was pumped. During all of the pressure cycles, the SCSSV remained open. Key Learnings In this case, the well was thought to contain fluids that were inhibited against hydrate formation. If the well is to be shut in for any period of time, it is necessary to verify exactly what fluids are currently present in the well and what needs to be injected to ensure that hydrates will not form.


OPRM-2003-0302D

Page 58 of 64 .

30-April-2006


3.2

Unrestricted

Case Study: Auger Taken from: AP Mehta, ‘Hydrate Plug Blockage and Remediation: Case Studies from Operations in the Gulf of Mexico’, EP 2001-3019. Where A hydrate plug had formed in the Auger A1 well soon after completion and commissioning in July 1994. The well was producing oil with no detectable water based upon shake-out tests. The well had been shut-in soon after start-up due to a hurricane in the Gulf of Mexico. How The Auger hydrate plug was yet another reminder that an undetectable water cut, based upon shake-out tests, is not a good indicator of a fluids potential to form hydrates. Auger well A4 had been in production for at least 14 hours prior to being shut in (by closing the wing valve). The subsurface safety valve was shut in 2 hours later. Some completion fluids were expected to remain in the well even though it had been flowing for 14 hours. Upon shut-in, it is believed that water droplets dispersed in the oil (present as BS&W, not free water) would have quickly settled out. In addition any slugs of completion fluids would also be settling out upon shut-in. These water droplets and completion fluids would possibly be forming a thin layer of water along the tubing wall (since the well flowed ‘liquid full’) and initiate hydrate formation. Further contact with falling water upon shut-in could easily result in a hydrate plug since no methanol was injected into the well prior to shut-in. Plug Analysis Auger Well A4 had a shut-in pressure of about 7200psig and the crude bubble point is around 5500psig. The sea-bottom temperature is expected to be approximately 40°F. At 7200psig, the hydrate dissociation temperature is ~85°F, assuming that the water is fresh. This provides a tremendous driving force of over 45°F for the hydrates to form when the oil/gas comes into contact with any free water. The hydrate blockage at Auger formed during restart of the well. The SCSSV at a depth of 6000ft below the mudline could not be opened and it was initially suspected to be the location of the hydrate plug. The well was pressured up to as high as 10,000psig and bled down several times to open up the SCSSV. These attempts were unsuccessful. A wireline tagged the plug at 1290ft. It was decided by the remediation task force to use glycol or methanol to dissolve the hydrate plug by injection of these inhibitors from the wellhead. Plug Remediation Initially, hot water was circulated down a 0.5in tubing string in the annulus to warm up and melt the hydrate, but no movement was observed. Glycol was then pumped from the surface and methanol was injected about 100ft above the SCSSV, but this too did not help. At this stage a coiled tubing unit was mobilised. A 1.25in coiled tubing unit was rigged up and a mixture of 50% methanol (by volume) in water was circulated through four 0.25in nozzled from the base of the tubing. The circulation rate was set at 0.5bbls/min. The unit first encountered hydrates at 2136ft and these were easily washed down to 2970ft. Several hard plugs were observed at depths down to 3270ft, where the unit broke through washed down to 4700ft. According to Steve Norton of Shell Offshore Incorporated (SOI), the coiled tubing unit does not easily ‘feel’ the presence of hydrates but recognised their presence only after breaking through them.


OPRM-2003-0302D

Page 59 of 64 .

30-April-2006


Unrestricted

During the period of plug remediation, Veet Kruka and Fred Trascher at Westhollow conducted some experiments with Triethylene Glycol (TEG) and methanol to determine the rate at which they can melt a solid plug of hydrates. Tetrahydrofuran (THF) was selected as the hydrate former since it is completely miscible in water and forms hydrates easily at 4°C. The tests indicated that TEG could dissolve and disperse a hydrate plug in a stagnant column at least an order of magnitude higher than methanol. Based upon this work it was estimated that glycol sitting on top of ice in a wellbore would penetrate it at about 4in/hr and a 10ft/sec jet would penetrate at 3ft/hr. Key Learnings A major learning from the Auger plug incident was the need to inhibit any wet oil prior to a cold well start-up. If the water that accumulated in the wellbore had been dosed with methanol prior to shut-in it would have prevented hydrates from forming. Continuous methanol injection for a limited amount of time until the well reaches a safe operating condition is essential to prevent hydrates upon an unexpected shut-in. The safe condition time for a well is typically the time required for the fluid to attain a temperature high enough to allow a minimum of 8 hours of cooldown time prior the fluid entering into the hydrate region. The optimal inhibitor for dissolving a hydrate plug depends on the nature and location of the plug. Laboratory tests indicate that glycol may be more effective than methanol in melting a plug in a static column since it can keep penetrating the plug due to its higher density. Methanol may form a thin pad above the dissolved hydrate with the liberated water, which would prevent its further progress down the plug, thus decreasing its ability to melt it effectively.

4.0

HYDRATE REMOVAL IN A CHEMICAL INJECTION LINE

4.1

Oregano Oregano MeOH Umbilical Plugging Event Cause and Plan for Remediation From: Kevin Cooper Date: 01/21/2003 Summary The primary umbilical tubes supplying Methanol (MeOH) to the Oregano subsea field have been plugged. Indications are that the event that caused the plugging occurred on 12 December, 2002. The plugging event occurred due to an instantaneous pressure spike at the open CIT2 methanol injection valve causing backflow of well fluids into the methanol service umbilical. The valve is located downstream of the subsea choke and is often used for treatment of the subsea tree during shut-ins.


OPRM-2003-0302D

Page 60 of 64 .

30-April-2006


Unrestricted

Cause of Plugging Event While in preparation for normal monthly well testing, the Oregano #1 well was shut in for preparation to test the Oregano #2 well. MeOH injection was started into the OR1 tree to treat the wellbore, tree and jumper against hydrate formation. Pressure was bled off of the portion of the tree where the methanol was to be injected through the CIT2 valve by closing the choke and opening the Pipeline Shutdown Valve (PSDV) downstream of the choke. At this point, pressure was bled to the flowline pressure, which was approximately 3000psi. The chart below contains the PI data recorded during the event. When the CIT2 valve was opened, the topsides methanol pump pressure was approximately 2000psi with a corresponding pressure at the CIT2 valve of 3200psi including the hydrostatic head of methanol. At this point there was sufficient topsides methanol pump pressure to inject chemical into the tree. During the operation, the choke was commanded open and the PSDV was commanded shut. This is normal operating procedure for treating the tree and wellbore for shut-in as long as equalisation across the tree valves is performed first. However, when the choke was opened and the PSDV was closed, the shut-in tubing pressure was allowed into the tree downstream of the choke at the opened CIT2 valve and the topsides pump pressure plus methanol hydrostatic head was not sufficient to prevent the backflow of hydrocarbons into the methanol injection line. At the time the choke was commanded open and the PSDV was commanded shut, the shut-in tubing pressure was approximately 5000psi. At this time, the topsides pump pressure was 2800psi. Including the head of methanol and friction losses in the line, the pressure differential across the valve was on the order of 2000psi. It appears that backflow and plugging extended through the OR1 steel flying lead to the Umbilical Termination Head (UTH) and possibly up the main umbilical where the methanol circuit is common for both of the Oregano wells. It is not clear at this time whether the plug is a hydrate, emulsion or floc. As a result of the backflow into the common part of the umbilical, injection into both wells through the normal means is not currently possible. System Redundancy Methanol injection into the system is currently available through the annulus vent umbilical line. However, this line is the only means of annulus service for both Oregano and Serrano because the line is common to both fields. Therefore, extreme caution must be taken during operations involving this line such as annulus bleeds or methanol injection so that the ability to inject into the line is not jeopardised. Operations guidelines must be strictly adhered to at all times. The Oregano system does have additional umbilical tubes that are not currently in use that could potentially be used for methanol service. Specifically, the asphaltene solvent and the scale inhibitor lines appear to be feasible substitutes. Plans are currently being developed to perform the work required to commission these lines for use if remediation attempts prove to be unsuccessful.


OPRM-2003-0302D

Page 61 of 64 .

30-April-2006


Unrestricted

Plan for Remediation The current strategy to remediate the plugged umbilical is to perform a double-sided depressurisation of the Oregano system. This will facilitate pumping against and moving the plug with as large a pressure differential as possible. If the plug is a hydrate, the depressurisation will also promote faster melting of the plug. If this operation is not successful, it is envisaged that the unused asphaltene solvent tube and/or the unused scale inhibitor tube will be commissioned for methanol service or a minor intervention chosen to clear the methanol circuit. Remediation procedures are currently under development and will be added as appendices to this document when they are completed. Prevention of Future Plugging Events In order to prevent such events, it is essential that individuals who are tasked with performing routine operations on the subsea systems have adequate training and are well aware of the system’s response to valve movements and large pressure differentials. It is also essential that the subsea operating guidelines be precise and provide clear guidance for such operations. Though the Oregano operating guideline does include procedures for pressure equalisation across valves, it does not completely address the instantaneous pressure increase due to valve movements that caused the backflow and line blockage. The guideline is currently under review and will be updated to ensure adequate procedures.


OPRM-2003-0302D

Page 62 of 64 .

30-April-2006


Unrestricted

Appendix 3C Nomenclature AWV

Annulus Wing Valve

bbls blpd bpd

Barrels Barrels of liquid (oil and water) per day Barrels per day

FPSO

Floating Production, Storage and Offloading Vessel

GLIV GLR GOM

Gas Lift Injection Valve Gas Lift Riser Gulf of Mexico

HDP


KHI

Kinetic Hydrate Inhibitor

LDHI LP

Low Dosage Hydrate Inhibitor Low Pressure

MIV MMSCF MMSCFD

Methanol Injection Valve Million Standard Cubic Feet Million Standard Cubic Feet per Day

PFL PIV PM PMV PWV

Production Flowline Pigging Isolation Valve Production Manifold Production Master Valve Production Wing Valve

SCSSV SITP SWV

Surface Controlled Subsurface Safety Valve Shut-in Tubing Pressure Sacrificial Wing Valve

VIT

Vacuum Insulation Tubing

WC WSV

Water Cut Well Switching Valve

XOV

Crossover Valve

Section 3 Appendix 3C Nomenclature

OPRM-2003-0302D

Page 63 of 64

31-December-2004


Unrestricted

ADMINISTRATION AND DISTRIBUTION LIST Reporting No:

OG-03.80112

Title:

Bonga Hydrate Remediation Guidelines

Subtitle: Date of issue:

October 2003

Author(s):

DJ Peters, ER Cadena, AP Mehta, SD Iyer

Contributor(s): Reviewed by:

GJ Hatton

OGUS-OGUA

Manager/(designated alternative)

AD Leitko (Responsible for approving contents and distribution of report)

OGUS-OGUA

Project Number:

52006219

ECCN Number:

EAR99

Activity Code:

52006219

Sponsor/Customer:

Shell International Exploration and Production, Inc

Keywords:

hydrate, Bonga, flow assurance, plug, remediation

Electronic file:

OG.03.80112.pdf

Issuing Library:

Westhollow Technology Center (WTC)

Distribution:

(Please add recipient info as complete as possible ie company name, name, reference indicator; company name and location in bold)

Number of Copies

Shell Research & Technology Centre, Amsterdam (SRTCA)

1

Degree of Confidentiality:

Restriction on Distribution: Additional Distribution:

Shell Canada Limited, DMS Library

1

Petro-Chemical Knowledge Center Westhollow Technology Center

4

DJ Peters, OGUS-OGUA, HOU-WTC E-1467B

1

ER Cadena, OGUS-OGUA, HOU-WTC E-1460

1

AP Mehta, OGUS-OGUA, HOU-WTC E-1450

1

SD Iyer, SIEP-EPT-PDS, HOU-WCK REMOTE

1

AA Kaczmarski, SIEP-EPT-PDS, HOU-WCK 3340

1

M Bosha, SIEP-EPT-PDS, HOU-WCK 2316

1

KA Stevens, SIEP-EPT-PDS, HOU-WCK 3376

1

AR Green, SIEP-EPT-SDW, HOU-WCK 2114

1

W Schoppa, OGUS-OGUA, HOU-WTC E-1300

1

GJ Hatton, OGUS-OGUA, HOU-WTC E-1270

1

AD Leitko, OGUS-OGUA, HOU-WTC E-1420

1

Unclassified ⌧ Shell Confidential

Restricted to Shell Personnel Only (Standard)

⌧

Shell Confidential

Shell Most Confidential Unshared

Any additional distribution (outside the above mentioned distribution list) can only be effected with special permission of owner/custodian (see above).

Section 3 Appendix 3C Nomenclature

OPRM-2003-0302D

Page 64 of 64

31-December-2004


Unrestricted

Section 4 Production Flowline Wax Assessment


2.0

EXECUTIVE SUMMARY...............................................................................................4 1.1

Wax Deposition..................................................................................................4

1.2

Summary of Results...........................................................................................5

1.3

Recommendations .............................................................................................5

BACKGROUND............................................................................................................6 2.1

Sample Selection for the Present Study.............................................................6

2.2

Scope of Work ...................................................................................................7

3.0

PRODUCTION FLOWLINE LAYOUT AND PIPE CHARACTERISTICS.......................8

4.0

WAX-RELATED FLUID PROPERTIES.........................................................................9

5.0

6.0

7.0

8.0

4.1

Measurement Techniques..................................................................................9

4.2

Sampling and Basic Fluid Properties..................................................................9

4.3

Normal Paraffin Distributions ...........................................................................10

4.4

Critical Wax Deposition Temperatures .............................................................10

4.5

Kinetic Wax Deposition Rates..........................................................................12

WAX DEPOSITION SIMULATION RESULTS ............................................................13 5.1

East 10in Production Flowline Line ..................................................................14

5.2

East 12in Production Flowline Line ..................................................................17

5.3

West 10in Production Flowline Line .................................................................20

POUR POINT AND RESTART EVALUATION............................................................25 6.1

Dead Oil Pour Point .........................................................................................25

6.2

Live Oil Pour Point ...........................................................................................27

6.3

Gel Strength Measurement ..............................................................................27

6.4

Effect of Bonga Fluid Blending .........................................................................28

6.5

Impact on Chemical Treatment ........................................................................30

WAX RISKS AND WAX MANAGEMENT STRATEGY ...............................................30 7.1

Risks and Basic Management Strategy............................................................30

7.2

Surveillance and Adjustments to Management Strategy ..................................30

HEALTH, SAFETY AND ENVIRONMENT (HSE) .......................................................31


OPRM-2003-0302D

Page 1 of 41

30-April-2006


Unrestricted

Table of Contents (cont’d) TABLES Table 4.1 – List of Bonga Fluids Whose Wax Properties Were Measured ..............................6 Table 4.2 – Production Flowline Data .....................................................................................8 Table 4.3 – Basic Properties of B1 803 Oil .............................................................................9 Table 4.4 – Comparison of Bonga Pour and Cloud Points....................................................10 Table 4.5 – Measured Pour Points for Bonga B1 803 Sample NIG-O-129A .........................26 Table 4.6 – Required Restart Pressures for Bonga Production Flowlines (PFL) ...................29 FIGURES Figure 4.1 – Production Flowline Layout.................................................................................8 Figure 4.2 – Normal Paraffin Distributions for Various 803 Bonga Oils .................................11 Figure 4.3 – Critical Wax Deposition Temperatures for Various Bonga Oils .........................11 Figure 4.4 – Comparison of Kinetic Wax Deposition Rates for Bonga 803 Oils ....................12 Figure 4.5 – FPSO Arrival Temperatures – East 10in PFL ...................................................14 Figure 4.6 – Deposition Onset Location – East 10in PFL......................................................15 Figure 4.7 – Deposit Growth Rate (in Maximum Thickness) – East 10in PFL .......................15 Figure 4.8 – Deposit Growth Rate (in Volume) – East 10in PFL ...........................................16 Figure 4.9 – Estimated Pigging Frequency – East 10in PFL.................................................16 Figure 4.10 – FPSO Arrival Temperatures – East 12in PFL .................................................17 Figure 4.11 – Deposition Onset Location – East 12in PFL....................................................18 Figure 4.12 – Deposit Growth Rate (in Maximum Thickness) – East 12in PFL .....................18 Figure 4.13 – Deposit Growth Rate (in Volume) – East 12in PFL .........................................19 Figure 4.14 – Estimated Pigging Frequency – East 12in PFL...............................................20 Figure 4.15 – FPSO Arrival Temperatures – West 10in PFL ................................................21 Figure 4.16 – Deposition Onset Location – West 10in PFL...................................................22 Figure 4.17 – Deposit Growth Rate (in Maximum Thickness) – West 10in PFL ...................22 Figure 4.18 – Deposit Growth Rate (in Volume) – West 10in PFL ........................................23 Figure 4.19 – Estimated Pigging Frequency – West 10in PFL..............................................24 Figure 4.20 – Pour Points of Bonga B1 702 and B1 803 Blends...........................................29


OPRM-2003-0302D

Page 2 of 41

30-April-2006


Unrestricted

Table of Contents (cont’d) APPENDICES Appendix 4A – Sensitivity Analysis of CWDTS and Deposition Rates...................................32 Appendix 4B – Pour Point Measurement Techniques and Uncertainties ...............................37 Appendix 4C – Tables from Westrich (1999) Report (SIEP.99.6096)....................................38


OPRM-2003-0302D

Page 3 of 41

30-April-2006


1.0

Unrestricted

EXECUTIVE SUMMARY The objective of this study is to assess the wax-related risks in the Bonga production system. These risks include wax deposition under flowing conditions and oil gelling (pour point) during shut-in conditions. The Bonga flowline system is quite complex, comprising four flowline systems, five production manifolds and produced fluids from four reservoirs. Adding to the complexity, the production at each manifold is not always from a single reservoir, and some of the reservoirs are compositionally graded (ie variations of wax-related properties within the reservoir). Further, manifold temperatures are functions of the flow rates from each of the individual wells. The resulting matrix of possible conditions is quite large.

1.1

Wax Deposition Controlling Factors Our approach is to examine the base-case risks and to study sensitivities to the key variables. The key factors that determine wax deposition are: •

•

Fluid composition and wax-related fluid properties: –

Critical wax deposition temperature – varies by fluid

–

Kinetic wax deposition rate – varies by fluid

Flowline thermal profile, controlled by: –

Flowrate – production variable; expected Qfl > 10MBPD/flowline

–

Manifold temperature Tman – controlled by well flowrate; expected Tman > 49°C (120°F) in most cases

–

Flowline insulation – fixed in system selection

–

Line size – fixed in system selection

Base Case Based on this information, the base case was chosen to be: •

Qfl = 10MBPD

•

Tman = 49°C

•

Worst wax-related properties (B1 Well 803 sand deposition rates and B2ST3 well 702 sand Critical Wax Deposition Temperatures (CWDTs)

•

All flowlines studied

Sensitivity Studies •

Qfl = 5 and 20MBPD

•

Tman = 38 and 60°C

•

Low wax-related properties found in the partially biodegraded crudes (B1 Well 803 sand and B2ST3 Well 803 sand CWDTs as well as B2ST3 Well 803 sand deposition rates)


OPRM-2003-0302D

Page 4 of 41

30-April-2006


1.2

Unrestricted

Summary of Results Wax Deposition There is no wax risk in the short (1 mile) flowlines, west of the FPSO. Recommended pigging frequency is once per year for maintenance and surveillance. The wax risks in the long flowlines (east of the FPSO) are minimal if the conditions are base case or better. Using Qfl > 10MBPD, Tman > 49°C, and worst-case wax properties, recommended pigging frequencies range from 2 to 3 times per year. Wax risks increase substantially if either Qfl or Tman fall below base-case conditions. Wax risks decrease substantially if the produced fluid has low wax deposition rates (found in the partially biodegraded oils). Pour Point and Gel Strength A study has been made of the B1 Well 803 sand fluid, which has the highest pour point of any Bonga fluid yet measured in our labs (maximum 4°C, minimum -7°C). We have determined that this fluid is unlikely to exhibit a yield stress/gel strength under shut-in pipeline conditions. We expect that no pour point depressant will be required. These findings will be compared with those of chemical vendors when the chemical tender results become available.

1.3

Recommendations The wax deposition study has used Tman as a variable rather than connecting specific well production functions to manifold temperatures. For this reason, we recommend that a comparison be made of critical wax deposition temperatures to case-specific manifold and arrival temperatures based on production functions and wellbore thermal/hydraulic simulations. Frequent surveillance is recommended for the produced fluid wax properties to ensure that: •

The fluid is arriving above the critical wax deposition temperatures in each flowline

•

The produced fluid pour points have not increased


OPRM-2003-0302D

Page 5 of 41

30-April-2006


2.0

Unrestricted

BACKGROUND The wax-related properties of Bonga fluids have been assessed in several previous studies (Refs 12, 13 and 25). The testing was not comprehensive across all wells/sands, and some of the measurements were deemed to have high uncertainties (Ref 25). Table 4.1 lists the fluids that had some wax properties measured. Additional information is provided in Paragraph 4.0. Well

Sand

Comments

B1

670

Low wax content and pour point

B1

702

Low pour point; cloud point similar to B1 803; CWDT higher

B1

710

Pour point, wax content slightly below B1 803

B1

803

Highest wax properties (except CWDT) from B1, B2ST3 and B3ST samples

B2 ST3

702

Termed ‘kinetically inhibited’ but highest CWDT

B2 ST3

803

Termed ‘kinetically inhibited’ but high CWDT

B3 ST

690

Low wax content and pour point; cloud points lower than B2ST3 702 and 803

B3 ST

702

Limited data; low wax content and pour point

702 W6

709

Highest wax properties, based on limited volume of questionable sample

Table 4.1 – List of Bonga Fluids Whose Wax Properties Were Measured

2.1

Sample Selection for the Present Study Wax-related properties show a wide variation for Bonga fluids (Ref 25) and we therefore recommended that an expected worst-case fluid be identified and tested for wax deposition and pour point problems. As a general statement, the Bonga crude oils appear to be from the same oil family (private communications from Erik Tegelaar (SIEP-EPT-DE) and Nancy Utech (OGUS-OGUA)), and fluid property variations are largely caused by secondary alteration processes (eg biodegradation, water washing or gas stripping). Biodegradation has had a noticeable impact on the Bonga wax-related fluid properties (lowering wax content, pour points and cloud points). Therefore, the goal of the sample selection was to find a substantial volume of a non-biodegraded Bonga oil. At the time of the initial reports (Refs 12 and 25), the preferred fluid was expected to be B1 Well 803 oil, classified as ‘W0 – no biodegradation’. Subsequent to those reports, sample was obtained from 702W6 Well 709 sand (Ref 17). Tests on this showed low pour points (7 to 10°C) and higher cloud points (29 to 33°C) than the B1 Well 803 oil (21 to 24°C). However, the 709 oil was received as an emulsion with a rag layer; emulsion breaking and decanting left solids in the container that had to be removed with solvent rinses. Both the sample and the High Temperature Gas Chromatography (HTGC) analyses were reconstructed from the various sub-samples. In addition, only limited sample was available (blowdown of one 600cc SSB downhole sampling chamber). Owing to limited sample and questionable quality, no additional analyses of 702W6 Well 709 oil are currently possible.


OPRM-2003-0302D

Page 6 of 41

30-April-2006


Unrestricted

A sample of B1-Well 803 oil was found in storage in Nigeria and provided to Shell Global Solutions for testing and evaluation; it is used in the present study. While it cannot be guaranteed to have the absolute worst wax-related properties at Bonga, it is certainly among the worst and should be adequate for wax flow assurance measurements and models.

2.2

Scope of Work The key points in the scope of work are listed below: •

•

•

Wax Deposition –

Measure wax-related fluid properties of a non-biodegraded, high pour point fluid (B1 803): HTGC, pour point, cloud point, kinetic wax deposition rate

–

Verify wax deposition strategy by comparing range of CWDTs to expected arrival temperatures of production flowlines; run HYSYS Wax Deposition for selected cases if necessary

–

Follow-up/validate pigging frequency results given in oil offloading report; report issued separately

Gelling/Pour Point –

Challenge and assess if current strategy for treating high Pour Point (PP) wells is valid; develop detailed procedures if necessary

–

Test gel strengths of high pour point fluid to determine level of concern; also do pipeline restart tests if/when sample volumes become available (possibly get samples during unloading)

–

Model restart of high pour point fluid if required

–

Investigate effect of mixing oils at manifolds and topsides; determine if export oil should be treated for high pour point

Surveillance/Analysis –

Specify requirements for sample analysis during development drilling and as a part of surveillance after first oil


OPRM-2003-0302D

Page 7 of 41

30-April-2006


3.0

Unrestricted

PRODUCTION FLOWLINE LAYOUT AND PIPE CHARACTERISTICS The Bonga production concept brings all produced fluids from five Production Manifolds (PMs) to an FPSO located in the middle of the field. Each PM combines multiple wells with fluids from various pay sands. The West Field (ie west of the FPSO) consists of two Production Flowline Loops (PFLs), namely, PFL 8/9 and PFL 11/12 and the East Field consists of three PFLs, namely, PFL 1/2, PFL 3/4 and PFL 5/6. Since PFLs 3/4 and 5/6 are connected together, they are considered as one PFL for realistic wax deposition simulation. According to contractual specifications in the basis of design (Bonga Field Development Plan, 2001), all PFLs are of Pipe-in-pipe (PIP) configuration with an overall heat transfer coefficient (UOD) of 0.187 to 0.194 Btu/hr-ft2-°F (1.063 to 1.101W/m2-°C) depending on pipe size. Water depth is approximately 3400ft (1000m). Typical riser length is about 1700m. A simplified schematic of the production flowline layout is shown in Figure 4.1.

West

East PFL 3/4

PFL 8/9

PFL 5/6

FPSO PFL 11/12 PFL 1/2

Figure 4.1 – Production Flowline Layout The West PFLs are about a mile long (1.8km) and the pipe size is 10in with an ID of 8.876in (22.6cm). The East PFLs range from 3.6 to 5.5 miles (5.8 to 8.8km) and the pipe sizes vary from 10in to 12in with an ID of 8.876in to 10.62in (22.6 to 27cm). Table 4.2 lists the flowline characteristics used in this study. East 10in PFL Parr (bar)

WC (%)

PIP ID (cm)

Gas Lifted

PIP U Factor 2 (W/m -C)

Flowline Length (km)

21

0

22.6

No

1.063

8.8

Parr (bar)

WC (%)

PIP ID (cm)

Gas Lifted



21

0

22.6

No

1.063

1.8

Parr (bar)

WC (%)

PIP ID (cm)

Gas Lifted



21

0

27

No

1.101

5.8

West 10in PFL

East 12in PFL

Table 4.2 – Production Flowline Data Section 4 Production Flowline Wax Assessment

OPRM-2003-0302D

Page 8 of 41

30-April-2006


4.0

Unrestricted

WAX-RELATED FLUID PROPERTIES One objective of the current study was to validate the B1 803 oil sample (NIG-O-129A) newly obtained from Nigeria. The sample was validated using cloud point and HTGC measurements, after which kinetic wax deposition rates were measured as input for the deposition studies. Appendix 4C contains all wax properties obtained from Bonga Main fluids prior to the current report. The 670, 690, 702 and 710 sands show lower wax-related properties than 803 oil. (As noted previously, the 709 sand in well 702W6 looks as bad or slightly worse than B1 803 oil, but the 709 sample is questionable.) These data show that B1 803 oil is expected to be the worst case for wax-related fluid properties.

4.1

Measurement Techniques Wax-related properties are determined from several in-house measurements. These data are used as inputs and consistency checks to our thermodynamic and transport models for wax deposition in flowlines and wells. Both the measurements and models are described in Refs 11 and 12.

4.2

Sampling and Basic Fluid Properties Data about samples and trends across the Bonga Main fields are available in Refs 11 to 13 and 17. Major conclusions in those studies are that wax-related properties vary both between and within reservoirs. The primary cause of this variation is biodegradation, which metabolises paraffins and reduces cloud and pour points. From those studies, the B1-well 803-sand fluid was identified as a primary oil (in the geochemical sense; not biodegraded or otherwise altered) with the highest known cloud and pour points of the available Bonga fluids. At the time in 1999, this fluid was not available to Shell, but a sample has since been made available. This sample has been analysed and the basic results are given in Table 4.3. Cloud point was measured by the cold-finger technique, and maximum/minimum pour points were measured according to the ASTM D5853-95 protocol. Cloud Point (°C) Well

B1

Sand

803

SAM ID

NIG-O-129A

WTC ID

6140

Gravity (API)

33.9

Cold Finger

HTGC

35.6

37.0

Pour Points (°C)

4/-7

Table 4.3 – Basic Properties of B1 803 Oil Table 4.4 compares cloud points for Bonga Main fluids measured or derived by Shell Global Solutions (and, previously, SEPTAR Flow Assurance) using consistent cold-finger and HTGC methods. B1 803 oil has the highest pour and cloud point values of any of the fluids excluding the questionable 702W6 709 oil. A brief comparison was done between B1 803 oil and Bonga South West fluids. The highest measured cloud points at Bonga SW were seen for the 803 and 812 oils. However, these values (29 to 32°C including both cold-finger and HTGC techniques) are lower than Bonga Main B1 803 oils. ASTM D97 pour points were 2 to 4°C, similar to B1 803 oil. HTGC data were uniformly lower for the Bonga SW 803 and 812 oils than for B1 803 oil over the full carbon number range. This is further evidence than B1 803 is a suitable choice as a representative end-member fluid.


OPRM-2003-0302D

Page 9 of 41

30-April-2006


Unrestricted

Pour Point (°C) Well

Pay Sand

SAM ID

B-3ST

690

NIG-O-88A

B-2ST3

702

NIG-O-85H

B-2ST3

803

NIG-O-84H

702W6

709

NIG-O-93X

B1

702

B1

803

ASTM D97

ASTM D5853

Cloud Point (°C) Cold Finger

HTGC

Thermo Model

15

24

18

<-45

22

32

31

<-45

20

32

31

10/7

29 to 33

38

34 to 35

NIG-O-128A

<-37/ <-37

22

28

NIG-O-129A

4/-7

36

37

Table 4.4 – Comparison of Bonga Pour and Cloud Points

4.3 Normal Paraffin Distributions Normal paraffin distributions were measured using quantitative high-temperature gas chromatography. Figure 4.2 shows distributions from the 702 and 803 sands sampled in the B1 and B2ST3 Wells. The B1 803 sample has the highest concentrations over most of the carbon number range, which typically implies worse wax-related properties. Interpretation of the chromatogram did not find indications of biodegradation for this oil, indicating that is likely to be the (worst-case) end member of the 803 fluids. The cloud point calculated from the HTGC correlation is 37°C, in excellent agreement with the measured value of 35.6°C. This agreement is quite different from the B2ST3 fluids (Ref 12), where the HTGC and lab values differed by more than 10°C. The large difference in those values was speculated to be possible kinetic inhibition of crystallisation from biodegradation products; the fact that the B1 803 fluid does not show this effect is indirect evidence that B1 803 should behave as a primary oil.

4.4

Critical Wax Deposition Temperatures Figure 4.3 shows the CWDTs calculated for three Bonga fluids: B2ST3 702, B2ST3 803 and B1 803. (CWDTs for B1 702 oil are not shown because the oil is biodegraded and its normal paraffin concentrations are uniformly lower than B1 803. Therefore, it cannot be a worst case.) Although the paraffin distributions for B2ST3 702 is lower than for B1 803, it has higher CWDTs owing to a lower API gravity (density effects). For this reason, some of the deposition simulations use mixed properties (ie B1 803 deposition rate and B2ST3 702 CWDTs) as a sensitivity study. Note: A recalibration of our thermodynamic model since the original studies (Ref 12) resulted in higher CWDTs for the B2ST3 fluids than in the original reports. This effect is not substantial at expected arrival pressures. The new results are used in this study.


OPRM-2003-0302D

Page 10 of 41

30-April-2006


Unrestricted

10000

Concentration, ppm wt

Bonga B2ST3 702 Bonga B2ST3 803

1000

Bonga B1 702 Bonga B1 803

100

10

1 20

30

40

50 60 Carbon Number

70

80

Figure 4.2 – Normal Paraffin Distributions for Various 803 Bonga Oils

46 B2ST3 803 Sand B2ST3 702 Sand

o

Temperature, C

44

B1 803 Sand

42

40

38

36 0

50

100

150

200

250

300

350

400

Pressure, bar

Figure 4.3 – Critical Wax Deposition Temperatures for Various Bonga Oils


OPRM-2003-0302D

Page 11 of 41

30-April-2006


4.5

Unrestricted

Kinetic Wax Deposition Rates Wax deposition rates were measured using the Shell cold-finger technique. Figure 4.4 shows a comparison of the whole-oil deposition rates (correlations rather than raw data are used to compare trends more easily). Looking at the B2ST3 803 data, the deposition rate at the HTGC cloud point of 32°C yielded an extremely low deposition rate. When measured at the laboratory cloud point of 21°C, the rate increased by more than an order of magnitude; however, the test temperature is well below the expected arrival temperatures. The B1 803 oil sample, tested at its lab cloud point of 36°C, shows deposition rates much higher than the B2ST3 oil at 32°C. Compared with the B2ST3 test at 21°C, the B1 oil is also higher, at low ∆Ts. (Since the flowlines have pipe-in-pipe insulation, the low-∆T range is the appropriate range for comparison.) Therefore, the B1 803 wax deposition rate is the worst case for the fluids we have tested. The B1 803 oil has a lower degree of biodegradation than the B2ST3 702 and 803 sands and, in general, less benign wax properties than other Bonga oils analysed (Ref 25) (B1 well 690, 702 and 710 oils, B2ST3 well 702 and 803 oils, B3ST well 690 and 702 oils). Therefore, we consider the B1 803 wax deposition rate to be a likely worst case for the Bonga oils seen to date.

1

Wax + Oil: B2ST3 @ 32 C Wax + Oil: B2ST3 @ 21 C

Deposition Rate, mg/cm 2-hr

Wax + Oil: B1

@ 36 C

0.1

0.01

0.001 0

2

4

6

8

10

o

Oil-Wall ∆ T, C

Figure 4.4 – Comparison of Kinetic Wax Deposition Rates for Bonga 803 Oils


OPRM-2003-0302D

Page 12 of 41

30-April-2006


5.0

Unrestricted

WAX DEPOSITION SIMULATION RESULTS We used HYSYS steady-state simulator (Version 2.4.2, Build 3874) to simulate wax deposition in three selected production flowlines (ie two east lines and one west line) that adequately represent the Bonga production scenario. A Shell proprietary wax model (SD-HYPR-Extensions_WAX-2.3-GS) and a multiphase hydraulic routine (GZM-NEWPRS) were also used. Pressure/Volume/Temperature (PVT) tuned fluid compositions and flowline models (with topography, pipe and insulation properties) were provided by Wade Schoppa. Fluid composition of 702 sand was used in all simulations since it is the primary pay sand with the highest CWDTs obtained from B2ST3 sample. We selected a matrix of three manifold temperatures, ie 100°F (38°C), 120°F (49°C) and 140°F (60°C), and three production rates, ie 5, 10 and 20MBLPD, to evaluate the severity of wax deposition based on new deposition rates measured and new CWDTs. According to previous studies, a realistic low manifold temperature is about 120°F at a low rate of 10MBLPD based on minimum turndown rates specified in the basis of design (Schoppa and Kaczmarski, 2001 and Schoppa, 2002). Similarly, a manifold temperature of 100°F can be expected at 5MBPD according to a previous study (Schoppa, 2002). Both flowing wellhead temperatures and pressures as a function of rate as well as steady-state arrival temperatures at the FPSO from each PFL are not available to us at the start of the study. All simulations were performed on a monthly basis (720 hours time period). No water cut (ie 0%) is included in the production rates considered; thus, the results are more conservative. According to production data provided by Sada Iyer, water cut in late life can be as high as 80% in most PFLs. Effect of gas lift at the riser base was also not included in this study. Possible cooling could result from lift-gas injected at/near the riser base. Since heated lift-gas is to be used (minimum specification of 90°F as it enters the flowline), its effect on temperature drop is expected to be smaller. However, substantial cooling by the injected gas may occur if the lift-gas in the injection line is allowed to cool down to seabed temperatures. Notice that this temperature reduction may result in higher wax deposition in the riser section or PFL downstream of the gas injection point. The effect is not quantifiable in this study. Topside arrival pressure at the FPSO is set at about 300psi or 21bar (Bonga Field Development Plan, 2001). In late life, the topside arrival pressures will be 170psi (12bar) resulting in 1°F increase in CWDT at topside conditions. A sensitivity check on the effect of this temperature increase at the topside in the presence of high water cut on current wax mitigation strategy is not indicated to be significant. As a sensitivity check, wax depositions were ran on all three sets of CWDTs from B2ST3 702 sand, B2ST3 803 sand and B1 803 sand with the same B1 803 sand deposition rates. The results indicate minimum variation on deposit volume and pigging frequency over the selected conditions (refer to Appendix 4A Paragraphs 1.0 to 3.0). Results based on B2ST3 702 sand CWDTs are presented in Paragraph 5.1. A sensitivity analysis was also performed on variation of the kinetic deposition rates (using B2ST3 803 sand and B1 803 sand data for comparison) on flowline deposition with the same B2ST3 702 sand CWDTs. The results indicate minimum variation on deposit volume and pigging frequency over the selected conditions using B2ST3 deposition rates (refer to Appendix 4A Paragraph 4.0). Therefore, the use of deposition rates from B1 803 sand represents the worst case, as stated in Paragraph 4.5. Tabulated data are listed in Appendix 4A.


OPRM-2003-0302D

Page 13 of 41

30-April-2006


5.1

Unrestricted

East 10in Production Flowline Line A 10in (22.6cm ID) East PFL with a flowline distance of 5.5 mile (8.8km) was selected for simulation (representing PFL 1/2).

5.1.1

Arrival Temperatures at FPSO Figure 4.5 illustrates the arrival temperature range at various manifold temperatures and production rates. Range of CWDTs (39 to 43°C or 103 to 109°F) is provided to enhance comparison. As shown, the East 10in PFLs are in wax deposition range for most of the conditions simulated (below 60°C and 20MBLPD).

Bonga Main (East) - 8.8-km Flowline (22.6 cm ID) 60

FPSO Arrival Temperature (oC)

50

CWDT at 21 bar

40

30

20

38 C, Manifold Temp 10

49 C, Manifold Temp 60 C, Manifold Temp 0 0

5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.5 – FPSO Arrival Temperatures – East 10in PFL 5.1.2

Deposit Onset Location and Deposit Thickness Figure 4.6 shows the location of wax deposition onset from topside and Figure 4.7 shows the deposit growth rate (in maximum deposit thickness per month) at various conditions. As shown, onset location is a strong function of manifold temperature and production rate; whereas, maximum deposit thickness (near riser base in most of the cases) is relatively insensitive to changes in temperature and rate due to the very low kinetic deposit rates of Bonga fluids (lower than the oils of most Shell global assets we tested). The deposit growth rate (in maximum thickness) is about 0.006in (0.15mm) per month. Notice this thickness is the maximum deposit thickness possible in the entire PFL per month, not an averaged thickness over the deposited area. Therefore, maximum thickness does not always correlate with total deposit volume.


OPRM-2003-0302D

Page 14 of 41

30-April-2006


Unrestricted

At 140°F (60°C) manifold temperature, waxes are likely to deposit only in the riser except for very low rates (such as 5MBLPD and below). If the manifold temperatures happen to drop below 100°F (38°C), such as when the rates are low, wax deposition is predicted in the wellbore. Bonga Main (East) - 8.8-km Flowline (22.6 cm ID)

Deposition Onset Location from Topside (m)

12000

10000

8000

6000


49 C, Manifold Temp 60 C, Manifold Temp 2000

0 0

5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.6 – Deposition Onset Location – East 10in PFL

Bonga Main (East) - 8.8-km Flowline (22.6 cm ID) 0.5

38 C, Manifold Temp 49 C, Manifold Temp

Max Deposit Thickness (mm/month)

0.4

60 C, Manifold Temp

0.3

0.2

0.1

0.0 0

5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.7 – Deposit Growth Rate (in Maximum Thickness) – East 10in PFL Section 4 Production Flowline Wax Assessment

OPRM-2003-0302D

Page 15 of 41

30-April-2006


5.1.3

Unrestricted

Deposit Volume and Pigging Frequency Figure 4.8 illustrates the amount of wax deposit accumulated per month at various conditions and Figure 4.9 illustrates the pigging frequency (number per year) based on accumulated volume. As shown, the amount of waxes deposited and, therefore pigging frequency, are a strong function of manifold temperature and production rate. Also, deposit volume and pigging frequency increase rapidly when rates are below 10MBLPD. At a realistic low rate of 10MBLPD and 120°F (49°C) manifold temperature, less than 1bbl (about 0.1m3 or 100 litres) of waxes is deposited per month (ie deposit growth rate in volume). This corresponds to a pigging frequency of about three times per year in the 10in East PFLs at 10 MPLBD and 49°C at the PM. Bonga Main (East) - 8.8-km Flowline (22.6 cm ID) 400

38 C, Manifold Temp

350

Wax Deposit Volume (liter/month)


300

250

200

150

100

50

0 0

5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.8 – Deposit Growth Rate (in Volume) – East 10in PFL Bonga Main (East) - 8.8-km Flowline (22.6 cm ID) 12


49 C, Manifold Temp

Pigging Frequency (#/yr)


6

4

2

0 0

5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.9 – Estimated Pigging Frequency – East 10in PFL Section 4 Production Flowline Wax Assessment

OPRM-2003-0302D

Page 16 of 41

30-April-2006


5.2

Unrestricted

East 12in Production Flowline Line A 12in (27cm ID) East PFL with a flowline distance of 3.6 mile (5.8km) was selected for simulation (representing PFLs 3/4 and 5/6).

5.2.1

Arrival Temperatures at FPSO Figure 4.10 illustrates the arrival temperature range at various manifold temperatures and production rates. Range of CWDTs is provided to enhance comparison. As shown, the East 12in PFLs are in wax deposition range for most of the conditions simulated (as long as they are below 60°C and 20MBLPD). The arrival conditions are very similar to the East 10in PFL described in Paragraph 4.1.

Bonga Main (East) - 5.8-km Flowline (27 cm ID) 60

FPSO Arrival Temperature (oC)

50

CWDT at 21 bar

40

30

20



5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.10 – FPSO Arrival Temperatures – East 12in PFL 5.2.2

Deposit Onset Location and Deposit Thickness Figure 4.11 shows the location of wax deposition onset from topside and Figure 4.12 shows the deposit growth rate (in maximum deposit thickness per month) at various conditions. As shown, onset location is a strong function of manifold temperature and production rates; whereas, maximum deposit thickness (near riser base in most of the cases) does not show a large variation with respect to changes in temperature and rate due to the very low kinetic deposit rates of Bonga fluids. The deposit growth rate (in maximum deposit thickness) is about 0.01in (0.25mm) per month, slightly higher than the East 10in PFL case. Again, this is an indication of the maximum deposit thickness over the entire PFL per month of deposition. At 140°F (60°C) manifold temperature, waxes are likely to deposit only in the riser except for very low rates (such as below 5MBLPD). If manifold temperatures drop below 100°F (38°C), wax deposition is predicted in the wellbore regardless of production rates.


OPRM-2003-0302D

Page 17 of 41

30-April-2006


Unrestricted



7000

6000

5000

4000

3000


2000

60 C, Manifold Temp

1000

0 0

5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.11 – Deposition Onset Location – East 12in PFL

Bonga Main (East) - 5.8-km Flowline (27 cm ID) 0.5



0.4

60 C, Manifold Temp

0.3

0.2

0.1

0.0 0

5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.12 – Deposit Growth Rate (in Maximum Thickness) – East 12in PFL


OPRM-2003-0302D

Page 18 of 41

30-April-2006


5.2.3

Unrestricted

Deposit Volume and Pigging Frequency Figure 4.13 illustrates the amount of wax deposit accumulated per month at various conditions and Figure 4.14 illustrates the pigging frequency (number per year) based on accumulated volume. As shown, the amount of waxes deposited and, therefore pigging frequency, are a strong function of manifold temperature and production rate. Also, deposit volume and pigging frequency increase rapidly when rates are below 10MBLPD.


38 C, Manifold Temp

350



300

250

200

150

100

50

0 0

5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.13 – Deposit Growth Rate (in Volume) – East 12in PFL At a realistic low rate of 10MBLPD and 120°F (49°C) manifold temperature, about 0.5bbl (85 litres or less than 0.1m3) of waxes is deposited per month (ie deposit growth rate in volume). This corresponds to a pigging frequency of less than two times per year in the 12in East PFLs at 10MPLBD and 49°C at the PM.


OPRM-2003-0302D

Page 19 of 41

30-April-2006


Unrestricted


38 C, Manifold Temp

7



6

5

4

3

2

1

0 0

5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.14 – Estimated Pigging Frequency – East 12in PFL

5.3

West 10in Production Flowline Line A 10in (22.6cm ID) West PFL with a flowline distance of 1.1 mile (1.8km) was selected for simulation (representing PFLs 8/9 and 11/12).

5.3.1

Arrival Temperatures at FPSO Figure 4.15 illustrates the arrival temperature range at various manifold temperatures and production rates. Range of CWDTs is provided to enhance comparison. As shown, the West 10in PFLs are out of wax deposition range if manifold temperatures are above 49°C (120°F) and above 5MBLPD. The arrival conditions are with higher temperatures as compared to the East PFLs described in Paragraphs 5.1 and 5.2 due to much shorter length.


OPRM-2003-0302D

Page 20 of 41

30-April-2006


Unrestricted

Bonga Main (West) - 1.8-km Flowline (22.6 cm ID) 60

o

FPSO Arrival Temperature ( C)

50

CWDT at 21 bar

40

30

20



5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.15 – FPSO Arrival Temperatures – West 10in PFL

5.3.2

Deposition Onset Location and Deposit Thickness Figure 4.16 shows the location of wax deposition onset from topside and Figure 4.17 shows the deposit growth rate (in maximum deposit thickness per month) at various conditions. As shown, onset location is a strong function of manifold temperature and production rate; whereas, maximum deposit thickness (near riser base most of the cases) does not show a large variation with respect to changes in temperature and rate due to the very low kinetic deposit rates of Bonga fluids. The deposit growth rate (in maximum deposit thickness) is only about 0.006in (less than 0.15mm) per month, much lower than the East PFL cases. At 120°F (49°C) manifold temperature, waxes are likely to deposit only in the riser. If manifold temperatures drop below 100°F (38°C), wax deposition is predicted in the wellbore regardless of production rates.


OPRM-2003-0302D

Page 21 of 41

30-April-2006


Unrestricted



3500

3000

2500

2000

1500


1000

60 C, Manifold Temp

500

0 0

5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.16 – Deposition Onset Location – West 10in PFL

Bonga Main (West) - 1.8-km Flowline (22.6 cm ID) 0.20

38 C, Manifold Temp



0.15

0.10

0.05

0.00 0

5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.17 – Deposit Growth Rate (in Maximum Thickness) – West 10in PFL


OPRM-2003-0302D

Page 22 of 41

30-April-2006


5.3.3

Unrestricted

Deposit Volume and Pigging Frequency Figure 4.18 illustrates the amount of wax deposit accumulated per month at various conditions and Figure 4.19 illustrates the pigging frequency (number per year) based on accumulated volume. As shown, the amount of waxes deposited and, therefore, pigging frequency, are a strong function of manifold temperature and production rate. Also, deposit volume and pigging frequency increase rapidly when rates are below 10MBLPD. At a realistic low rate of 10MBLPD and 120°F (49°C) manifold temperature, less than 0.1bbl (0.01m3 or 8 litres) of waxes is deposited per month (ie deposit growth rate in volume). This corresponds to a pigging frequency of less than once per year in the 10in West PFLs at 10MPLBD and 49°C at the PM. This means, for all practical purposes that the West PFLs should be pigged once per year accordingly.


38 C, Manifold Temp

140



120

100

80

60

40

20

0 0

5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.18 – Deposit Growth Rate (in Volume) – West 10in PFL


OPRM-2003-0302D

Page 23 of 41

30-April-2006


Unrestricted



4


60 C, Manifold Temp

3

2

1

0 0

5

10

15

20

25

Liquid Rate (MBLPD)

Figure 4.19 – Estimated Pigging Frequency – West 10in PFL Section 4 Production Flowline Wax Assessment

OPRM-2003-0302D

Page 24 of 41

30-April-2006


6.0

Unrestricted

POUR POINT AND RESTART EVALUATION Production of waxy crude oils can be a challenge due to the possible tendency to gel at low temperatures. As these fluids are cooled, wax starts to precipitate and particles become suspended in the fluid. As the temperature decrease continues, the concentration of wax particles increases and can ultimately reach a level at which an interlocking structural network is formed which changes the crude oil into a gel with solid-like properties. A commonly used indicator for the gelling temperature is the pour point. It should be emphasized however, that the pour point is not a well-defined rheological property. It is well known that the pour point of a crude oil is strongly dependent on the thermal history of the sample, causing the ‘minimum-maximum’ pour point phenomenon. Depending on the thermal history, the measured pour point can vary over a temperature range. For some oils this range can be as large as 50°C, while other waxy crudes hardly exhibit any difference. The pour point is especially sensitive to the pre-conditioning temperature of the oil. The pre-conditioning temperature determines whether all the wax is entirely dissolved into the oil or if precipitates are already present in the fluid. If the cooling process is started with all the wax dissolved, the process will result in a minimum type pour point, while starting at a lower pre-conditioning temperature will result in an increase of pour point. Additionally, the pour point of a crude oil is dependent on the cooling rate. If the rate is sufficiently high, the cooling process can outrun the wax precipitation kinetics and thus lower the pour point. As a result, it is therefore important that a pour point test is properly conducted. The sensitivity of pour point measurements is recognised by the American Society for Testing and Materials (ASTM) and special test procedures for crude oils have been developed to address these effects (a description of the different pour point tests is given in Appendix 4B). Furthermore, it is also important to base design considerations on the most relevant data. The maximum pour point of an oil can be above the minimum ambient conditions, while the minimum pour point can be below the ambient temperature. Generally speaking, for wellstream flows the operating temperatures are high and the cooling process continuous, a minimum pour point is the most representative. However, in case of dead oil circulation or, for example, an offloading line, due to the lower initial temperature and intermittent cooling process the maximum pour point may be more appropriate. Finally, it is not only the gelling temperature but also the gel strength and fluid rheology that determines whether it will be possible to restart a system after shut-in. In order to include these effects into an operability study, a model pipeline restart test is necessary.

6.1

Dead Oil Pour Point As with other wax-related properties, Bonga pour points show large variability across the field (Ref 11). Reported measurements of pour points from eight Bonga samples ranged from less than -45°C to 15°C. The highest of the reliable results were obtained for Bonga B1 803 samples. The results were measured by SPDC Warri and indicate an upper pour point of 10°C. This is above the ambient seafloor temperature and was therefore a major driver for this study.


OPRM-2003-0302D

Page 25 of 41

30-April-2006


Unrestricted

In order to determine if a possible gelling problem exists, a stored sample of B1 803 oil (SAM ID NIG-O-129A; WTC ID 6140) was located and provided for further study. After it was validated that the sample had not been altered during storage/transfer, with use of High Temperature Gas Chromatography (HTGC), a series of pour point measurements were conducted at Westhollow Technology Center (WTC) and Oil-phase DBR. The results of the pour point measurements on the NIG-O-129A sample under stock tank conditions are shown in Table 4.5. The minimum pour point measurement conducted at WTC indicates a lower pour point of -7°C, which is well below the ambient seafloor temperature. This value was confirmed by Oilphase DBR, where a minimum pour point of -8°C was measured. The maximum pour points measured by WTC and Oil-phase DBR (via third-party lab) were found to be 4 and 3°C respectively, in good agreement with each other. These results are just at the seafloor temperature but are several degrees lower than the upper pour point of 10°C measured by the SPDC Warri lab. The observed difference is within the reproducibility of the measurement as reported by ASTM. This difference in pour points could potentially lead to wrong conclusions with regard to whether or not a restart problem should be anticipated due to this the issue was further studied by conducting a Model Pipeline Test (MPT). The results of this test are discussed in a later section. For a comparison of the Bonga 803 results to other Bonga sands, refer to Appendix 4C Paragraph 2.0.

Laboratory

Protocol

Upper Pour Point (°C)

Lower Pour Point (°C)

SPDC Warri

ASTM D5853-95

10

N/A

Shell WTC

ASTM D5853-95

4

-7

Oil-phase DBR

ASTM D97 (modified)

N/A

-8

Oil-phase DBR ASTM D5853-95 (via third-party Lab)

3

N/A

Oil-phase DBR ASTM D97 (via third-party Lab)

0

N/A

Table 4.5 – Measured Pour Points for Bonga B1 803 Sample NIG-O-129A


OPRM-2003-0302D

Page 26 of 41

30-April-2006


6.2

Unrestricted

Live Oil Pour Point Solution gas has an impact on the pour point of a crude oil, therefore Oil-phase DBR also measured Bonga 803 oil pour point under live conditions. The method used to measure the live oil pour point was similar to the modified ASTM D97 used by Oil-phase DBR for the minimum pour point under stock tank conditions. For the live oil test the oil is introduced into a sapphire cell initially filled with synthetic gas mixture at test pressure. After the cell is charged, the oil is then heated to the reservoir temperature under seal (to avoid loss of light ends and oxidation) for an extended period of time. The composition of the synthetic gas mixture used for the measurement was based on the expected composition calculated by HYSYS using a Bonga 803 oil model. The test pressure of the test was 300psig, which is representative for a blowndown flowline. The pour point of the Bonga 803 oil under the described test conditions was -6°C +/- 1°C (21°F +/- 2°F). Comparison of the live oil results with the dead oil results shows that the difference is small and within the uncertainties of the tests. It is therefore concluded that the effect of solutions gas is small to negligible at the test conditions.

6.3

Gel Strength Measurement Although that a pour point is a good indicator whether or not a gelling problem might exist, it is ultimately the gel strength that will determine if a pipeline can be restarted. Therefore in addition to the pour point measurements, Oil-phase DBR was also asked to conduct a gel strength measurement. The gel strength was measured by simulating a restart situation of Bonga 803 oil using an MPT. The MPT system consists of a 7mm ID × 6m long stainless steel coil submerged in a temperature-controlled bath. The test is initiated by filling the loop with oil and circulating it around to remove all entrapped air bubbles. Once this step is completed, the flow is stopped and the test is started by cooling the oil at a specified cooling rate. The cooling process is continued until the target temperature is reached, after which the oil/gel is allowed to age for 12 hours to let the gel develop its strength. Once the gel is allowed to age, the restart process is started by pressurising the coil with nitrogen gas. The nitrogen pressure is increased in steps of 1psi over a 0 to 10psi range, 2psi over a 10 to 30psi range and 5psi thereafter, with 5 minutes between each pressure step to allow the oil to ungel. The lowest pressure at which the oil is observed to have started is then deemed as the yield pressure. Finally, because the yield pressure is dependent on the diameter of the pipe, the yield/gel strength is calculated by using a force balance:

τ yield =

p yield D 4L

Where: D = Coil diameter, L = Coil length


OPRM-2003-0302D

Page 27 of 41

30-April-2006


Unrestricted

Similar to pour point measurements, MPT pre-conditioning temperature and cooling rates are important parameters. Generally, the pre-conditioning temperature, cooling rate and restart temperature are chosen such that they closely represent actual field conditions. For the present MPT test, however, a different approach was chosen. The pour point measurements conducted as part of current study showed an upper pour point for the Bonga 803 sample that was 6°C lower than the value previously measured by SPDC Warri. The seafloor temperature is 6°C below the Warri pour point and therefore we chose a restart temperature 6°C below the DBR-measured pour point of 3°C (Trestart = -3°C). The ASTM pre-conditioning and cooldown procedure for pour points was to assure gelling at this temperature. By doing this, we ensured that a gel would form and that the results would be conservative. Following the procedure described above, the MPT resulted in a gel yield strength of 3Pa at a restart temperature of -3°C for the Bonga 803 oil sample. In order to determine if this would lead to any problems during a restart, the required restart pressure was calculated for the Bonga flowlines, refer to Table 4.6. The calculations show that the required restart pressure for all the flowlines will be significantly lower than the maximum available pump pressure. The highest required restart pressure is expected for 10in East PFLs 1 and 2, as can be seen in the tables. However, even for this PFL the restart pressure is only 19% of the maximum available pump pressure. Based on these results, it is therefore concluded that no restart problems will occur.

6.4

Effect of Bonga Fluid Blending The pour point and restart analysis, as discussed in the previous paragraphs, focuses on pure Bonga 803 oil. However, the pipeline fluids will consist of a blend of fluids with a pour point that is a function of the blend. The pour points of the other Bonga fluids were found to be significantly lower and Bonga 803 oil will only be present in a relatively small percentage, therefore the blend of 803 with other Bonga oils is expected to have a lower pour point. This effect is illustrated by Figure 4.6, which shows the pour point of a blend of 803 with 702 oil.


OPRM-2003-0302D

Page 28 of 41

30-April-2006


West 10in PFL

Unrestricted

East 10in PFL 1/2

East 12in PFL 3/4 and 5/6

East 12in PFL 5/6 only

Pipe ID

0.226m

Pipe ID

0.226m

Pipe ID

0.27m

Pipe ID

0.27m

Pipe Length

7000m

Pipe Length

21000m

Pipe Length

15000m

Pipe Length

9800m

Yield Stress (Pa)

Restart Pressure (bar)

Yield Stress (Pa)


Yield Stress (Pa)


Yield Stress (Pa)


1

1.24

1

3.72

1

2.22

1

1.45

2

7.43

2

4.44

2

2.90

1

2.48

2

1

3

3.72

1

3

11.15

3

6.67

3

4.36

6

7.43

6

22.30

6

13.33

6

8.71

12

14.87

12

44.60

12

26.67

12

17.42

2

24

53.33

24

34.84

2

30

43.56

41.33

60.002

24

29.73

30

37.17

48.43

60.002

16.14

60.00

1

27.00

1

Yield stress measured for Bonga 803 oil.

2

Maximum Pump Pressure.

60.00

Table 4.6 – Required Restart Pressures for Bonga Production Flowlines (PFL)

Blend Pour Points of Bonga B1 702 and 803 Oils 10

Upper Pour Point, C

0

-10

-20

-30

Upper limit; actual value may be lower.

-40 0

20

40 60 803 Oil Fraction, % vol

80

100

Figure 4.20 – Pour Points of Bonga B1 702 and B1 803 Blends


OPRM-2003-0302D

Page 29 of 41

30-April-2006


6.5

Unrestricted

Impact on Chemical Treatment The maximum pour points were measured in the 4 to 10°C range, however a strong gel was not formed at these temperatures. Based on our results, we therefore do not see a need to treat the Bonga oils for pour point depression. However, we recommend that ongoing surveillance be performed to confirm this with produced fluids. Currently, vendors are preparing chemical tenders for Bonga. We will review their recommendations when they are submitted and compare to our results.

7.0

WAX RISKS AND WAX MANAGEMENT STRATEGY

7.1

Risks and Basic Management Strategy In the course of this study, questions regarding wax risks have been answered. The B1 803 oil has been established as a primary (ie non-biodegraded) oil and is the likely worst-case fluid for wax properties (excluding CWDT, which is slightly higher for the B2ST3 702 oil). The B1 803 kinetic wax deposition rate has been measured and used as the basis for wax deposition calculations for the production system. From these calculations, a base-case pigging interval is recommended to be three times yearly for the east PFLs and once yearly for the west PFLs. Although the B1 803 pour point is found to be at seafloor temperature, the oil only formed a weak gel. Therefore, based on these fluid properties no restart problems are anticipated and thus no pour point depressant is required.

7.2

Surveillance and Adjustments to Management Strategy The key to managing wax risks is production surveillance. The surveillance plan for Bonga fluids should include the following. Monitor arrival temperatures and pressures and compare to relevant CWDT curves. Compare to pigging tables in Appendix 4A and adjust pigging frequency if necessary. Monitor pigging returns to determine solid volumes. If volumes are large, more frequent pigging is recommended. Monitor pressure drop while pigging. Total pressure drop can be separated as follows: ∆Ptotal = ∆Pflow + ∆Ppig + ∆Pwax The pressure drop from flow resistance (∆Pflow) should be constant if oil circulation rate is constant. The pig itself will cause pressure drop (∆Ppig), but this should also be constant during the pig run. In addition, the wax removed will cause pressure drop (∆Pwax), which will increase proportionally with the amount of wax removed as the pig traverses the flowline loop. Unfortunately, random pressure fluctuations may make it impossible to estimate ∆Pwax from the pressure data. However, if ∆Pwax can be measured and it is greater than 50 to 100psi, pigging frequency should be increased. If ∆Pwax is less than 50psi, follow recommendations in the pigging table.


OPRM-2003-0302D

Page 30 of 41

30-April-2006


Unrestricted

Monitor produced fluid properties. It is recommended that the samples be taken on a per-well basis if possible. If not practical, samples should be taken from individual flowlines. Samples should be taken at least every 6 months or more often if significant production changes occur (eg production from new wells or reservoirs, major change in production fraction from a given well or reservoir). Cloud points should be measured and compared to the measured (cold-finger) values in Table 4.5. If higher-than-expected cloud points are measured and the system is operating in the deposition regime (arrival T below CWDT), operations should pig sooner than planned and check pigging returns for excessive volume. Pour points should be measured. If upper pour point exceeds 10°C, pour point depressant should be injected as soon as possible. If possible, flowline shut-in should be postponed until chemical injection has begun and at least one line fill of treated oil has been produced. Treated production oil should be sampled and tested to confirm pour point has been lowered.

8.0

HEALTH, SAFETY AND ENVIRONMENT (HSE) Assessment of the potential risks associated with the interpretation, usage, field-implementation of the technical results of the present study was carried It was determined that such risks are very low. Additionally, it was determined the engineering predictions and recommendations of this study do not raise significant HSE issues or concerns. It is further advised that the users of technical report conduct their own HSE risk assessment of the usage implementation of the results and recommendations of the present report.

and out. that any this and


OPRM-2003-0302D

Page 31 of 41

30-April-2006


Unrestricted

Appendix 4A Sensitivity Analysis of CWDTs and Deposition Rates Table of Contents 1.0

EAST 10IN PFL DATA AND CWDT SENSITIVITY – 720 HOURS SIMULATION ......................................................................................33

2.0

EAST 12IN PFL DATA AND CWDT SENSITIVITY – 720 HOURS SIMULATION ......................................................................................34

3.0

WEST 10IN PFL DATA AND CWDT SENSITIVITY – 720 HOURS SIMULATION ......................................................................................35

4.0

EAST 10IN PFL – DEPOSITION RATE SENSITIVITY – 720 HOURS SIMULATION ......................................................................................36

Section 4 Appendix 4A Sensitivity Analysis of CWDTs and Deposition Rates

OPRM-2003-0302D

Page 32 of 41

30-April-2006


1.0

Unrestricted

EAST 10IN PFL DATA AND CWDT SENSITIVITY – 720 HOURS SIMULATION Basis: 803 B1 Deposition Rates 803 B2ST3 CWDT's 702 Fluid Composition

Basis: 803 B1 Deposition Rates 803 B1 CWDT's 702 Fluid Composition o

FPSO Arrival Temperature ( C) 803 B2ST3 Liquid Rate (MBLPD) 5 10 20

o

Manifold Temperature ( C) 38

49

60

17.8 23.7 27.6

23.8 32 37.3

29.9 40.4 47.1

Deposit Onset from Topside (m) 803 B2ST3 Liquid Rate (MBLPD) 5 10 20

o


49

60

10862 10862 10862

5913 1603 302

1788 0 0

Max Deposit Thickness (mm) 803 B2ST3 Liquid Rate (MBLPD) 5 10 20

o


49

60

0.11938 0.07443 0.04493

0.13100 0.05161 0.01264

0.09006 0 0

Wax Deposit Volume (liter) 803 B2ST3 Liquid Rate (MBLPD) 5 10 20

38

49

60

324.1 204.7 125.6

197.7 35.4 2.9

69.7 0 0

Liquid Rate (MBLPD) 5 10 20

o


49

60

17.8 23.7 27.6

23.8 32 37.3

29.9 40.4 47.1

Deposit Onset from Topside (m) 803 B1 Liquid Rate (MBLPD) 5 10 20

o


49

60

10862 10862 10862

6847 2838 535

2359 99 0

Max Deposit Thickness (mm) 803 B1 Liquid Rate (MBLPD) 5 10 20

o


49

60

0.11938 0.07443 0.04493

0.13230 0.08216 0.01927

0.14219 0.02038 0

803 B1 Liquid Rate (MBLPD) 5 10 20

o


38

49

60

35 55 90

57 317 3913

161 ---


Manifold Temperature ( C)


38

49

60

17.8 23.7 27.6

23.8 32 37.3

29.9 40.4 47.1


38

49

60

10.4 6.6 4.0

6.3 1.1 0.1

2.2 0 0

803 B1 Liquid Rate (MBLPD) 5.0 10.0 20.0

o


49

60

10862 10862 10862

7575 4439 1009

3123 240 0


o


49

60

0.11938 0.07443 0.04493

0.13324 0.08270 0.02513

0.14308 0.02451 0

Wax Deposit Volume (liter) o


49

60

324.1 204.7 125.6

230.2 68.6 5.6

98.1 1.6 0

702 B2ST3 Liquid Rate (MBLPD) 5 10 20

o


49

60

324.1 204.7 125.6

255.5 101.6 13.0

125.2 4.7 0

Pigging Interval (days) o


49

60

35 55 90

49 164 2004

115 7013 10000


Pigging Frequency (#/yr) o


o

702 B2ST3



Pigging Frequency (#/yr) 803 B2ST3 Liquid Rate (MBLPD) 5 10 20

o

FPSO Arrival Temperature ( C) 803 B1



Pigging Interval (days) 803 B2ST3 Liquid Rate (MBLPD) 5 10 20

Basis: 803 B1 Deposition Rates 702 B2ST3 CWDT's 702 Fluid Composition

o


49

60

35 55 90

44 111 866

90 2391 10000



49

60

10.4 6.6 4.0

7.4 2.2 0.2

3.1 0.1 0

702 B2ST3 Liquid Rate (MBLPD) 5.0 10.0 20.0

o


49

60

10.4 6.6 4.0

8.2 3.3 0.4

4.0 0.2 0


OPRM-2003-0302D

Page 33 of 41

30-April-2006


2.0

Unrestricted

EAST 12IN PFL DATA AND CWDT SENSITIVITY – 720 HOURS SIMULATION




o


49

60

19.7 25.0 28.6

26.2 33.6 38.4

32.8 42.3 48.1


o


49

60

7500 7500 7500

4218 895 117

923 0 0


o


49

60

0.24193 0.14635 0.08732

0.19401 0.06506 0.01799

0.10587 0 0


38

49

60

388.2 242.4 148.2

199.8 27.1 2.4

46.8 0 0


o


49

60

19.7 25.0 28.6

26.2 33.6 38.4

32.8 42.3 48.1


o


49

60

7500 7500 7500

5023 1339 364

1218 0 0

Max Deposit Thickness (mm) 803 B1 Liquid Rate (MBLPD) 5 10 20

o


49

60

0.24193 0.14635 0.08732

0.19555 0.07666 0.01972

0.12728 0 0


o


38

49

60

50 79 130

96 711 8124

412 ---




38

49

60

19.7 25.0 28.6

26.2 33.6 38.4

32.8 42.3 48.1

Deposit Onset from Topside (m)

38

49

60

7.3 4.5 2.8

3.7 0.5 0.0

0.9 0 0


o

702 B2ST3



38

49

60

7500 7500 7500

5665 2833 590

1466 64 0


o


49

60

0.24193 0.14635 0.08732

0.19681 0.11957 0.02921

0.13724 0.03111 0



49

60

388.2 242.4 148.2

250.3 53.1 6.3

75.2 0 0


o


49

60

388.2 242.4 148.2

285.3 85.3 11.6

105.6 1.6 0



49

60

50 79 130

77 363 3049

256 10000 --



o

702 B2ST3




o

FPSO Arrival Temperature ( C) 803 B1






o


49

60

50 79 130

68 226 1667

182 11830 --



49

60

7.3 4.5 2.8

4.7 1.0 0.1

1.4 0 0


o


49

60

7.3 4.5 2.8

5.3 1.6 0.2

2.0 0 0


OPRM-2003-0302D

Page 34 of 41

30-April-2006


3.0

Unrestricted

WEST 10IN PFL DATA AND CWDT SENSITIVITY – 720 HOURS SIMULATION




o


49

60

26.3 29.2 30.8

35.2 39.2 41.5

44 49.2 52.1


o


49

60

3552 3552 3552

535 64 0

0 0 0


38

49

60

0.13503 0.07903 0.04627

0.05478 0.01859 0

0 0 0

38

49

60

126.04 76.50 45.60

15.08 0.95 0

0 0 0



49

60

26.3 29.2 30.8

35.2 39.2 41.5

44 49.2 52.1


o


49

60

3552 3552 3552

811 240 0

0 0 0



o


38

49

60

89 147 247

746 11856 10000

----




38

49

60

26.3 29.2 30.8

35.2 39.2 41.5

44 49.2 52.1


38

49

60

4.0 2.4 1.5

0.5 0.0 0

0 0 0


o

702 B2ST3



38

49

60

3552 3552 3552

1035 480 116

0 0 0

Max Deposit Thickness (mm)

o


49

60

0.13503 0.07903 0.04627

0.07290 0.02337 0

0 0 0


o


49

60

0.13503 0.07903 0.04627

0.08300 0.03408 0

0 0 0



49

60

126.04 76.50 45.60

25.33 4.45 0

0 0 0


o


49

60

126.04 76.50 45.60

37.42 8.26 2

0 0 0



49

60

89 147 247

444 2528 10000

----




o

702 B2ST3




803 B1

o




o



o




o


49

60

89 147 247

301 1362 6981

10000 10000 10000



49

60

4.0 2.4 1.5

0.8 0.1 0

0 0 0


o


49

60

4.0 2.4 1.5

1.2 0.3 0.1

0 0 0


OPRM-2003-0302D

Page 35 of 41

30-April-2006


4.0

Unrestricted

EAST 10IN PFL – DEPOSITION RATE SENSITIVITY – 720 HOURS SIMULATION

Basis: 702 B2ST3 CWDT's Dep Rates: 803 B2ST3 Fluid Composition: 702

Basis: 702 B2ST3 CWDT's Dep Rates: 803 B1 Fluid Composition: 702 o


o


49

60

17.8 23.7 27.6

23.8 32 37.3

29.9 40.4 47.1


o


49

60

10862 10862 10862

5913 1603 302

1788 0 0


o


49

60

0.00616 0.00197 0.00057

0.00846 0 0

0.01062 0 0

o

FPSO Arrival Temperature ( C) 803 B1 Liquid Rate (MBLPD) 5 10 20



60

17.8 23.7 27.6

23.8 32 37.3

29.9 40.4 47.1

o


49

60

10862 10862 10862

7575 4439 1009

3123 240 0

o


49

60

0.11938 0.07443 0.04493

0.13324 0.08270 0.02513

0.14308 0.02451 0



49

60

10.7 3.4 1.0

10.6 0 0

7.5 0 0


o


49

60

324.1 204.7 125.6

255.5 101.6 13.0

125.2 4.7 0



49

60

1054 3353 11590

1062 ---

1504 ---



49



38



o


o


49

60

35 55 90

44 111 866

90 2391 --



49

60

0.3 0.1 0

0.3 0 0

0.2 0 0


o


49

60

10.4 6.6 4.0

8.2 3.3 0.4

4.0 0.2 0


OPRM-2003-0302D

Page 36 of 41

30-April-2006


Unrestricted

Appendix 4B Pour Point Measurement Techniques and Uncertainties There are currently at least five ASTM pour point protocols, most of which are designed for petroleum products rather than crude oils. The two most widely used are ASTM D97-96a and D5853-95. As implemented by Shell, these tests require 100 to 250ml of fluid. ASTM D97 is an older petroleum-products pour point protocol that requires heating to a prescribed temperature (60°C) and cooling in a series of baths until the oil gels or solidifies. Shell has devised a ‘mini’ D97 pour point that uses 30ml of sample. The ‘mini’ pour point has been calibrated to the ASTM D97 method. The repeatability (95% confidence limits, same lab) of D97 pour points (measured on fuel oils) is reported by ASTM to be 5°F; the reproducibility (95% confidence limits, different labs) is 12°F. In addition, chemical vendors use variations of ASTM D-97 to screen chemicals and choose concentrations. They may use very small volumes (1 to 2ml per sample), and results have higher uncertainty than other protocols. For that reason, validation after pre-screening is always recommended, using much closer adherence to standard protocols. ASTM D5853 is designed specifically for crude oils and recognises the potentially strong effect of thermal history on the oil gelling temperature. Two separate heating and cooling protocols are employed in order to see the effects of two substantially different thermal histories. The minimum pour point protocol requires heating to 105°C and cooling in air for 20 minutes at room temperature before entering a series of cooling baths. This protocol lowers the measured pour point in two ways: •

It ensures that all wax is in solution before cooling is started

•

It cools relatively quickly, potentially ‘outrunning’ the wax kinetics and reaching a lower temperature before the gel forms

The maximum pour point protocol requires heating to 60°C or less, cooling in air for 24 hours, a short reheat to 45°C followed by the cooling baths. This protocol raises the measured point by allowing a long time for ‘seed crystals’ to form at room temperature, which in turn decreases the time for gel formation. ASTM D5853 is difficult to adapt to small volumes; therefore, we have devised no ‘mini’ technique for ASTM D5853. The repeatability of D5853 using crude oils is reported by ASTM to be 6 to 12°F, but the reproducibility is as high as 32 to 40°F(!). Clearly, minor variations between labs could have large consequences. ASTM D5853 addresses the problems specific to the pour points of crude oils, therefore it is regarded as the better test method.

Section 4 Appendix 4B Pour Point Measurement Techniques and Uncertainties

OPRM-2003-0302D

Page 37 of 41

30-April-2006


Confidential

Appendix 4C Tables from Westrich (1999) Report (SIEP.99.6096) Table of Contents 1.0

FLUID PROPERTY DATA FOR BONGA OILS WAX MEASUREMENTS FOR STOCK-TANK (DEAD) CONDITIONS................................................................39

2.0

FLUID PROPERTY DATA FOR BONGA OILS: SELECTED GEOCHEMISTRY AND PVT PARAMETERS ..........................................................................................40

3.0

BEST ESTIMATES OF KEY WAX-RELATED PROPERTIES FOR BONGA OILS UNDER STOCK-TANK (DEAD) CONDITIONS .........................................................41

Section 4 Appendix 4C Tables from Westrich (1999) Report (SIEP.99.6096)

OPRM-2003-0302D

Page 38 of 41

30-April-2006


Confidential

FLUID PROPERTY DATA FOR BONGA OILS1 WAX MEASUREMENTS FOR STOCK-TANK (DEAD) CONDITIONS

1.0

Wax Content Pay Sand

Well

Lab 2

Lab 3

670

B-1

1.6

1.0

690

B-3ST

1.5

702

B-1

2.7

702

B-2ST3

2.3

702

B-3ST

–

710

B-1

7.3

803

B-1

9.3

803

B-2ST3

4.7

709*

702W6*

1.8-2.3

Lab 4

Pour Point Data (°C)

Lab 2

Lab 3

-33

-39

1.1

-51

3.9-4.2

-39

-42

Lab 4

Lab 5

-36

<-36

8.4-9.0

– 4.8

6

12-15

9

Lab 6b

Lab 3

(-41)

(-12, -15)

-30 1.0

Lab 6a

Cloud Point (°C) Lab 4

Lab 5

Lab 6c

-3

(-2)

15

(11)

22

Lab 6d

Lab 6e

Lab 6f

24

18

32

31

24 <-36

-45

<-45

(-25)

23

– (18)

12, 15

-33

12, (55)

21 <-36

<-45

(-11)

7-10*

24

(22) (10)

20 29-33*

38, (29) 20

32

31

38*

34-35*

(1)

It is important to note that different labs were dealing with specific stock-tank oil samples that they were handled and transferred differently. Issues relating to sample representativeness and data quality have downgraded the reliability of some measured values (these are enclosed in parentheses). Uncertainty in this regard for some of the samples is unknown. Other data were downgraded because either the analytical methods were considered less reliable or there were reported problems with the measurements (enclosed in parentheses and italics). Laboratories that were used to measure wax-related fluid properties at Bonga, referred to as Lab 2 to Lab 6 in the table, are listed below.

(2)

Geochemical Applications/Consulting Team (SEPTAR, Rijswijk) – Pour point by standard ASTM (D97); wax content by UOP46.

(3)

Core Lab (Aberdeen) – Methods not specified in report tables; cloud point referred to inwax appearance temperature in (WAT).

(4)

Oil Test – Data from PVT reports; methods not specified in report tables.

(5)

Shell Research and Technology Center, Amsterdam (SRTCA) – Pour point by both min and max-ASTM methods (identical results by both techniques); cloud point by AMS 259. SRTCA reported difficulties in making cloud point measurements, and the data reported above also may be unreliable due to the poorer quality dead oil samples that were made available.

(6)

Multiphase and Flow Assurance Team (SEPTAR, Houston): (a)

Pour point by min ASTM method – considered more reliable.

(b)

Pour point by automatic pour point analyser (APA) – has large errorbars, considered less reliable.

(c)

Cloud point – Houston standard cold-finger method.

(d)

Cloud point by Cross-polar Microscopy (CPM); data measured by DB Robinson (Calgary).

(e)

Predicted cloud point based on concentrations of C31-C60 normal paraffins determined by HTGC.

(f)

Predicted cloud point based on thermodynamic wax model, using HTGC-derived normal paraffin data as input.

* Inserted for comparison into Westrich’s (Ref 25) table from reference (Ref 17).


OPRM-2003-0302D

Page 39 of 41

30-April-2006

Depth Top (ss – ft)

1

API

Total Acid Number (TAN)

Stage of Biodeg (from HRGC)

API

GoR (scf/bbl)

Degree of Undersat (psi)

Reservoir Temp (°C)

670

B-1

OMC-7010

6756

21.2

1.42

W5 – Mod

21

325

750

40

671

B-2

OMC-8270

–

–

–

W5 – Mod

–

–

–

–

690

B-3ST

OMC-8074

9290

29.1

0.76

W4 – Mild

30

550

1204

70

702

B-1

OMC-7011

8040

27.9

0.76

W3/W4 – Mild

28

660

445

56

702

B-2ST3

OMC-8441

8925

29.3

0.74

W3 – Mild

29

500

–

63

702

B-3ST

–

9342

–

–

–

29

–

–

71

710

B-1

OMC-7012

9182

32.6

0.39

W1 – Onset

33

1090

472

71

803

B-1

OMC-7013

9920

34.9

0.35

W0 – None

35

1420

302

80

803

B-2ST3

OMC-8442

10,900

30.2

0.76

W2/W3 – Light

30

780

578

93

Stage of biodegradation is based on the criteria in Utech et al 1999, which relates the characteristics of whole oil/gas chromatograms to the extent of biodegradation. The High Resolution Gas Chromatography (HRGC) data for the Bonga crudes, on which this interpretation is based, can be found in Buiskool Toxopeus and van der Veen (1999).


2.0

1

Well

PVT Data (SIDS, Bonga Team)

FLUID PROPERTY DATA FOR BONGA OILS: SELECTED GEOCHEMISTRY AND PVT PARAMETERS

Page 40 of 41


OPRM-2003-0302D

Pay Sand

Gelchemistry (RTS)

Sample ID No (OMC # – RTS)

Confidential

30-April-2006

Well

670

B-1

671

B-2

690

B-3ST

< -35

702

B-1

702

Pour Point (°C)

Std Cloud Point (°C)

1

Stage of Biodegradation (from HRGC)

Pour Point (°C)

Std Cloud Point (°C)

W5 – Mod

< -35

10 to 15

W5 – Mod

< -35

10 to 15

-3 to 15

W4 – Mild

-33

10 to 15

< -35

24

W3/W4 – Mild

-18

18

B-2ST3

< -35

22 to 23

W3 – Mild

-18

18

702

B-3ST

< -35

–

–

–

710

B-1

6 to 10

W1 – Onset

12

24

803

B-1

9 to 15

21 to 24

W0 – None

12

24

803

B-2ST3

< -35

20

W2/W3 – Light

-10

21

709*

702W6*

7 to 10*

29 to 33*

W1 – Onset*

12*

21*

< -35

740

1

Stage of biodegradation is based on the criteria in Utech et al 1999, which relates the characteristics of whole oil/gas chromatograms to the extent of biodegradation. The HRGC data for the Bonga crudes, on which this interpretation is based, can be found in Buiskool Toxopeus and van der Veen 1999. * Inserted for comparison into Westrich’s (Ref 25) table from reference (Ref 17).


3.0

Pay Sand

Predicted Values Based on Biodegradation Model of Utech et al, 1999

Confidential

30-April-2006

BEST ESTIMATES OF KEY WAX-RELATED PROPERTIES FOR BONGA OILS UNDER STOCK-TANK (DEAD) CONDITIONS

Page 41 of 41


OPRM-2003-0302D

Based on Measured Data


Unrestricted

Section 5 Offloading Riser (Wax Assesement)


EXECUTIVE SUMMARY...............................................................................................3

2.0

BACKGROUND............................................................................................................3

3.0

OIL OFFLOADING RISER LAYOUT AND CHARACTERISTICS .................................4

4.0

WAX DEPOSITION SIMULATION RESULTS ..............................................................6

5.0

4.1

Simulation Basis ................................................................................................6

4.2

Wax Deposition and Growth Rate ......................................................................6

4.3

Pigging...............................................................................................................9

4.4

5-year Wax Deposition Profile..........................................................................11

4.5

Effect of Riser Buoy Insulation on Wax Deposition...........................................13

HEALTH, SAFETY AND ENVIRONMENT (HSE) .......................................................15

TABLES Table 5.1 – BOOR Data – Basis.............................................................................................5 Table 5.2 – Wax Properties Basis ..........................................................................................6 FIGURES Figure 5.1 – BOOR Layout .....................................................................................................4 Figure 5.2 – Ambient Temperatures as a Function of Water Depth ........................................5 Figure 5.3 – SPM Arrival Temperature ...................................................................................7 Figure 5.4 – Deposit Growth Rate – Maximum Thickness ......................................................8 Figure 5.5 – Effect of Offloading Rate and Inlet Temperature on Deposition Profile ...............8 Figure 5.6 – Deposit Growth Rate – Cumulative Deposit Volume...........................................9 Figure 5.7 – Estimated Pigging Frequency ...........................................................................10 Figure 5.8 – Estimated Pigging Interval ................................................................................11 Figure 5.9 – 5-year Wax Deposition Profile ..........................................................................12 Figure 5.10 – Total Deposit Volume Over a 5-year Period – 32°C Inlet and 600MBPD ........12 Figure 5.11 – Effect of Buoy Insulation on Fluid Temperature Profile – 600MBPD ..............13

Section 5 Offloading Riser Wax Assessment

OPRM-2003-0302D

Page 1 of 16

30-April-2006


Unrestricted

Table of Contents (cont’d) FIGURES Figure 5.12 – Effect of Buoy Insulation on Wax Deposition Profile – 600MBPD, 49°C Inlet .....................................................................................14 Figure 5.13 – Effect of Buoy Insulation on Wax Deposit Volume ..........................................14 Figure 5.14 – Effect of Buoy Insulation on Pigging Interval...................................................15

APPENDICES Appendix 5A – Raw Data – 720-hour Simulation ...........................................................................16


OPRM-2003-0302D

Page 2 of 16

30-April-2006


1.0

Unrestricted

EXECUTIVE SUMMARY The wax deposition risk and pigging requirement of the Bonga Oil Offloading Riser (BOOR) was re-evaluated using updated wax-related fluid properties. The key findings are: • Wax deposition/risk in BOOR is less severe than previously predicted using updated wax-related fluid properties and models • Pigging is recommended at least once per year upon startup. Pigging operation can be optimised by checking deposits in the pig return and monitoring arrival temperatures • Inlet temperature has a strong effect on deposit thickness, deposit volume and pigging frequency • Effect of insulation (assumed C-float) from the buoyancy elements near the mid-point of the riser on deposition profile is observed but not significant enough to alter the pigging recommendation. • If the BOOR is not pigged: –

Maximum wax deposit thickness is predicted to be 1mm (after 1 year) and 3mm (after 5 years) at 32°C (90°F) inlet temperature and 600MBPD offloading rate

–

Cumulative deposit volume is predicted to be 20bbl (after 1 year) and 60bbl (after 5 years) at 32°C (90°F) inlet temperature and 600MBPD offloading rate

We recommend pigs used in pigging BOOR be customised by pig manufacturers. Care should also be taken to design procedures for proper solid handling due to wax precipitates in the FPSO hull when offload and deposits from pig return.

2.0

BACKGROUND A wax deposition study of the BOOR was carried out in September 2000 (Ref 18). Since then, production functions have been updated with new fluid property information and Shell deposition models have been improved. A Bonga wax analysis (Ref 19) was performed to reassess the wax-related fluid properties and wax deposition risks in the Bonga production flowlines to the FPSO. As a follow-up of the Bonga production flowline study, wax deposition and pigging requirements were re-evaluated for the BOOR from the FPSO to the Single Point Mooring (SPM) offloading buoy (which connected to oil tanker) using updated wax properties and models.


OPRM-2003-0302D

Page 3 of 16

30-April-2006


3.0

Unrestricted

OIL OFFLOADING RISER LAYOUT AND CHARACTERISTICS The schematic of the BOOR layout is shown in Figure 5.1. There are two offloading risers with an OD of 22in. The two risers are of different overall length to allow for vertical separation between them. Syntactic foam buoyancy elements are arranged over certain length of each riser near the mid-point to provide the wave shape. The risers are suspended at each end (FPSO and SPM). The deeper riser with an overall length of approximately 2.3km (1.4 mile) was used for worst-case study due to its colder temperature profile. The riser characteristics are provided in Table 5.1 (Ref 20). An intermittent offload cycle was used – oils from the FPSO were offloaded for 20 hours every 5 days at a flow rate of 600MBPD (ie a total of 1mmbbl oil transferred per offload from both risers). The ambient sea temperatures around the risers vary from 7 to 26.5°C (45 to 80°F). Figure 5.2 shows the ocean temperature profile used for the Bonga Field.

Bonga 22in Oil Offloading Riser Shapes Initial 1000 900 800

Elevation (m)

700 600 500 400 300 200 100 0 0

200

400

600

800

1000

1200

1400

1600

1800

2000

Horizontal Distance (m) OPRM20030302D_062.ai

Figure 5.1 – BOOR Layout


OPRM-2003-0302D

Page 4 of 16

30-April-2006


Unrestricted

Pinlet

Riser Length (m)

Riser ID (in)

Riser Thickness (in)

Fusion-bond Epoxy Coating (in)

Uod Factor (W/m2-°C)

34.5

2265

20

1

0.02

150

Tin (°C)

No of Riser

Buoy Section (m)

Offload (hr/5 days)

Oil Volume per Offload (MMbbl)

Offload Rate (MBPD/line)

32-66

2

450

20

1

600

Table 5.1 – BOOR Data – Basis

Figure 5.2 – Ambient Temperatures as a Function of Water Depth


OPRM-2003-0302D

Page 5 of 16

30-April-2006


4.0

Unrestricted

WAX DEPOSITION SIMULATION RESULTS We used the HYSYS steady-state simulator (version 2.4.2, build 3874) to simulate wax deposition in the offloading riser. A Shell proprietary wax model (SD-HYPRExtensions WAX-2.3-GS) and a multiphase hydraulic routine (GZM-NEWPRS) were also used. Processed and stabilised stock oil fluid composition was used to represent the offloaded oils (predominantly 702 sand since it is the primary pay sand). Tabulated raw data are listed in Appendix 5A.

4.1

Simulation Basis The basis of simulation is listed below: • Three offloading riser inlet temperatures ie 32°C (90°F), 49°C (120°F) and 66°C (150°F), were used to illustrate the effect of inlet temperature on wax deposition. According to previous design (Ref 23) oil is to be cooled to 43°C (110°F) via crude cooler prior to entering the FPSO hull and the offloading temperature is about 38 to 43°C (100 to 110°F) • Two oil offload flow rates, ie 600MBPD (1mmbbl oil/offload) and 900MBPD (1.5mmbbl/offload), were used to illustrate the effect of flow rate on wax deposition • Simulations were performed on a basis of 20-hour offload operation every five days (eg 6 offloads over 30-day period) • No water cut (ie 0%) is assumed in the oils offloaded and the Gas/Oil Ratio (GOR) is negligible (ie processed and stabilised oil) • The same wax-related fluid properties used in the Bonga production flowline wax study were used, ie Critical Wax Deposition Temperatures (CWDTs) of B2ST3 702 sand and kinetic deposition rates of B1 803 sand listed in Table 5.2 (Ref 19) The CWDT is approximately 43°C (109°F) at 10 to 20bar range (< 300psi)

Wax Properties

CWDTs

Kinetic Disposition Rates

Sand

702 B2ST3

803 131

Table 5.2 – Wax Properties Basis

4.2

Wax Deposition and Growth Rate Figure 5.3 illustrates the effect of inlet temperature on arrival conditions at SPM (oil tanker loading point) over the range of flow rates studied. It is clear that there will be wax deposition in the risers over most of the conditions studied and in the FPSO hull. Cares should be taken to properly handle wax precipitates/solids as suggested in previous study (Ref 18). However, no wax deposition occurs when inlet temperatures are above 52 to 55°C range (126 to 130°F) depending on flow rates since the SPM arrival temperatures will be higher than the CWDTs. No information about the heating capacity of the FPSO is available to us.


OPRM-2003-0302D

Page 6 of 16

30-April-2006


Unrestricted

Figure 5.4 illustrates the maximum deposit growth rate. The maximum rate is estimated to be < 0.1mm/month (< 0.004in/month). Note: This is the maximum thickness change possible in the entire riser system. The deposit thickness is not uniformly distributed due to variations in the temperature profile. A higher rate of 900MBPD will result in thinner deposits owing to smaller oil-wall ∆T caused by higher heat-transfer coefficients. Figure 5.5 illustrates the BOOR deposition profile after a 30-day period (six offloads) at two flow rates and two inlet temperatures. At 49°C (120°F) inlet temperature, higher flow rates move the wax onset location further downstream and most of deposits are located near the SPM. However, 32°C (90°F) inlet temperature is below CWDT. Therefore, wax deposition begins at the riser inlet regardless of flowrate, and deposits are more uniformly distributed. Figure 5.6 illustrates the cumulative deposit volume over a 30-day period. The effect of inlet temperature and flow rates can be clearly seen. At 32°C (90°F) inlet temperature and 600MBPD, almost 1.6bbl/month of waxes are deposited.

Figure 5.3 – SPM Arrival Temperature


OPRM-2003-0302D

Page 7 of 16

30-April-2006


Unrestricted

Figure 5.4 – Deposit Growth Rate – Maximum Thickness

Figure 5.5 – Effect of Offloading Rate and Inlet Temperature on Deposition Profile


OPRM-2003-0302D

Page 8 of 16

30-April-2006


Unrestricted

Figure 5.6 – Deposit Growth Rate – Cumulative Deposit Volume

4.3

Pigging Figure 5.7 illustrates the pigging frequency (yearly) required to clean the offloading riser. We use a pigging criterion based on the pressure drop that would be caused if the pigged wax forms a plug in front of the pig, specifically OPPiu9 < 50psi (3.4bar). As shown in Figure 5.7, the pigging frequencies are small (ie intervals are long) and are relatively insensitive to variation in the conditions studied (where wax is deposited). Note: A pigging frequency less than once per year should be regarded as once for all practical purposes. On the other hand, the associated pigging interval at 32°C (90°F) inlet temperature and 600MBPD is in the magnitude of 600 days as shown in Figure 5.8. We recommend the offloading risers be pigged at least once per year upon start-up. The amount of deposits collected in the pig return can then be examined to help optimise pigging operations after startup, along with arrival temperature monitoring. In addition, an inlet temperature of 16°C (60°F) was studied as a sensitivity check (included in Appendix 5A). At this temperature, very large amounts of waxes would have already precipitated/deposited prior to entering offloading risers.


OPRM-2003-0302D

Page 9 of 16

30-April-2006


Unrestricted

Offloading risers are flexible lines. Based on Shell internal documents (Refs 21 and 22) flexible lines can be pigged as normal carbon steel pipelines, provided that the radius of the bends in the flexibles is controlled (to avoid sharp bends, in the BOOR case, controlled via buoyancy elements). However, care should be taken on the selection of materials for the pigs to avoid aggressive pigging behaviour that may cause carbon steel pick-up if a stainless steel carcass is applied in the flexible. This is because the oval flow path of the flexibles resulting from the slight change of internal ID as the carcass structure opens and closes when bending occurs. Also, pig brushes should be avoided if a plastic liner (typically acts as primary corrosion barrier) is applied to maintain pipe integrity. The SPM should allow flow path control elements for pig to pass through and come back via another riser to the FPSO. The loading hose in the last section between the SPM and the tanker should not be pigged. This is because the soft linings of the hoses are likely to be damaged if pigged and tankers are not usually capable of receiving a pig. We would recommend pigs used in pigging BOOR be customised by pig manufacturers.

Figure 5.7 – Estimated Pigging Frequency


OPRM-2003-0302D

Page 10 of 16

30-April-2006


Unrestricted

BOOR – 2.3km Flowline (20" ID)

Figure 5.8 – Estimated Pigging Interval

4.4

5-year Wax Deposition Profile In the absence of pigging, we simulated wax deposition over a 5-year period using 600MBPD flow rate. An inlet temperature of 32°C (90°F) was used to compare with previous study in 2000 (Ref 18). The results are shown in Figure 5.9. The maximum deposit thickness in the riser grows from almost 1mm (0.04in) at the end of the first year offloading operation to 3mm (0.12in) at the end of the fifth year. The total cumulative deposit volume in the riser is predicted to be 20bbl in the first year and 60bbl at the end of the 5-year period, as shown in Figure 5.10. As a result, the 5-year deposition profile and total deposit volume are predicted to be less than the results from previous studies. Note: The previous work is based on a different simulation basis eg wax-related fluid properties, simulation environment and simulator package.


OPRM-2003-0302D

Page 11 of 16

30-April-2006


Unrestricted

Figure 5.9 – 5-year Wax Deposition Profile

Figure 5.10 – Total Deposit Volume Over a 5-year Period – 32°C Inlet and 600MBPD


OPRM-2003-0302D

Page 12 of 16

30-April-2006


4.5

Unrestricted

Effect of Riser Buoy Insulation on Wax Deposition The possible insulation of the 450m buoy section near the mid point of the offloading riser was taken into account to evaluate its impact on wax deposition. Commercial C-Float syntactic foam buoyancy elements rated up to about 1200m (4000ft) water depth was assumed. The thermal properties of the C-Float foam used in this study are: thermal conductivity of 0.112W/m°C (0.065 Btu/hr-ft°F), density of 737kg/m3 (46lb/ft3) and specific heat capacity of 1674.7J/kg°C (0.4 Btu/Ib°F). This results in very small temperature decline over the continuous 450m buoyancy section (refer to Figure 5.11). The effect of this buoy insulation on wax deposition profile can be seen in Figure 5.12 (deposit thickness profile) using 600MBPD and 49°C (120°F). No deposition is seen in the buoyancy section due to the excellent insulating properties assumed. Translated this result to wax deposition volume and pigging, both the cumulative deposit volume and the pigging interval are reduced only slightly (shown in Figures 5.13 and 5.14 respectively). Therefore, the recommendation of once per year pigging frequency is not altered, taking into account of buoy insulation effect.

Figure 5.11 – Effect of Buoy Insulation on Fluid Temperature Profile – 600MBPD


OPRM-2003-0302D

Page 13 of 16

30-April-2006


Unrestricted

Figure 5.12 – Effect of Buoy Insulation on Wax Deposition Profile – 600MBPD, 49°C Inlet

Figure 5.13 – Effect of Buoy Insulation on Wax Deposit Volume


OPRM-2003-0302D

Page 14 of 16

30-April-2006


Unrestricted

BOOR - 2.3-km Flowline (20" ID)

Figure 5.14 – Effect of Buoy Insulation on Pigging Interval

5.0

HEALTH, SAFETY AND ENVIRONMENT (HSE) We have carried out the assessment of the potential risks associated with the interpretation, usage and field implementation of the technical results of the present study. It was determined that such risks are very low. Additionally, it was determined that the engineering predictions and recommendations of this study do not raise any significant HSE issues or concerns. It is further advised that the users of this technical report conduct their own HSE risk assessment of the usage and implementation of the results and recommendations of the present report.


OPRM-2003-0302D

Page 15 of 16

30-April-2006


Unrestricted

Appendix 5A Raw Data – 720-hour Simulation Oil Tanker Arrival Temperature (°C) Oil Rate (MBLPD)

BOOR Inlet Temperature (°C)

600

900

16

14.4

15.3

32

26.2

28.6

49

38.4

42.1

66

50.9

55.8

Max Deposit Thickness (mm) Oil Rate (MBLPD)


600

900

16

0.03744

0.02802

32

0.08012

0.05956

49

0.09653

0.05815

66

0

0

Wax Deposit Volume (bbl) Oil Rate (MBLPD)


600

900

16

0.6

0.5

32

1.5

1.2

49

1.1

0.3

66

0

0

Pigging Interval (days) Oil Rate (MBLPD)


600

900

600

900

600

1363

1759

1363

524

689

524

727

2752

727

Pigging Frequency (Number per Year) Oil Rate (MBLPD)


600

900

16

0.3

0.2

32

0.7

0.5

49

0.5

0.1

66

0

0

Section 5 Appendix 5A Raw Data – 720-hour Simulation

OPRM-2003-0302D

Page 16 of 16

30-April-2006


Unrestricted

Section 6 Pour Point Depressant Risk Assessment


RISK DOCUMENT TO VALIDATE NOT INJECTING POUR POINT DEPRESSANT FOR INITIAL BONGA WELLS ............................................................2

2.0

RECOMMENDATIONS .................................................................................................2

3.0

RESULTS AND DISCUSSION......................................................................................2

4.0

RISK ANALYSIS ..........................................................................................................5

5.0

4.1

Probability of Forming a Gel...............................................................................5

4.2

Probability of Removing a Gel............................................................................5

4.3

OPEX Related to Pour Point Depressant (PPD) ................................................5

4.4

Other Considerations .........................................................................................6

CONCLUSIONS............................................................................................................6

TABLES Table 6.1 – Measured Pour Points for Bonga B1 803 Sample NIG-O-129A ...........................3 Table 6.2 – Required Restart Pressures for Bonga Production Flowlines (PFL) .....................4 FIGURES Figure 6.1 – Pour Points of Bonga B1 702 and B1 803 Blends...............................................4


OPRM-2003-0302D

Page 1 of 6

30-April-2006


1.0

Unrestricted

RISK DOCUMENT TO VALIDATE NOT INJECTING POUR POINT DEPRESSANT FOR INITIAL BONGA WELLS Pour point is viewed as a risk for Bonga 803 and 710 sands. Based on analysis done so far, the Bonga 803 and 710 fluids can have pour points that range from 3°C to 10°C. Production of pour point fluids can be a challenge due to the possible tendency of the oil to form a gel when the flowline has cooled down to a temperature, which is below the measured pour point. Scenarios under which this can happen for Bonga include the following: •

A planned shutdown (under low water-cut scenario) wherein the flowline is protected against hydrates by treating it with methanol and yet the flowline could cool below the pour point

•

A situation under which the flowline is not circulated with dead oil after it has been blown down

During a shutdown, the fluid could gel-up and might potentially result in restart problems since gels typically have non-zero yield strength (ie some pressure is needed to dislodge a gel). This document summarises the analysis performed to assess pour point risk for Bonga. The Flow Assurance Sub-team (FAST) reviewed this work on 14 April 2003.

2.0

RECOMMENDATIONS Based on the analysis done thus far:

3.0

•

It is recommended not to inject a pour point depressant for the initial Bonga 803 and 710 wells. However, it may be prudent to wait for testing results from the vendors before making a final decision

•

It is also recommended to analyse samples obtained during the initial well unload to rig and to perform active surveillance during the initial months of production to verify pour point properties. This includes sampling of the appropriate sampling of the fluid (according to given procedures) and verification of fluid properties by a properly qualified laboratory

RESULTS AND DISCUSSION As with other wax-related properties, Bonga pour points show large variability across the field (Ref 24). Reported measurements of pour points from eight Bonga samples ranged from less than -45°C to 15°C. The highest of the reliable results were obtained for Bonga B1 803 samples. The results were measured by SPDC Warri and indicate an upper pour point of 10°C. This is above the ambient seafloor temperature and was therefore a major driver for this study. The B1 803 sample was identified as among the worst fluids from a wax perspective due to its primary nature (ie no waxes lost) and hence was determined to be adequate for wax flow assurance experimental measurements and modelling. Table 6.1 shows the results on the pour point measurement obtained from various labs on the B1-803 fluid. It was also determined that the presence of a small amount of gas in the samples (pour point measurement done with live fluid at 300psig) made the pour point to be slightly lower but did not substantially alter the pour point.


OPRM-2003-0302D

Page 2 of 6

30-April-2006


Laboratory

Unrestricted

Protocol

Upper Pour Point (°C)

Lower Pour Point (°C)

SPDC Warri

ASTM D5853-95

10

N/A

Shell WTC

ASTM D5853-95

4

-7

Oil-phase DBR

ASTM D97 (modified)

N/A

-8

Oil-phase DBR (via third-party Lab)

ASTM D5853-95

3

N/A

Oil-phase DBR (via third-party Lab)

ASTM D97

0

N/A

Table 6.1 – Measured Pour Points for Bonga B1 803 Sample NIG-O-129A Pour points are typically characterised by a maximum pour point and a minimum pour point. It is entirely likely (as suggested by the above Bonga data) that the maximum pour points can be above ambient conditions and minimum pour points can be below ambient conditions. Ref 24 indicates that a minimum pour point is more likely for production systems due to the presence of higher temperatures and turbulence while a maximum pour point is more likely during dead-oiling and oil offloading. The above data indicates that the minimum pour point for Bonga B1 803 is about -7/-8°C while the maximum pour point ranges from 3°C to 10°C. Although a pour point is a good indicator whether or not a gelling problem might exist, it is ultimately the gel strength that will determine if a pipeline can be restarted. Therefore, in addition to the pour point measurements, a gel strength measurement was also conducted. The gel strength measurement was done at DBR. The pour point measurements conducted as part of DBR study showed an upper pour point for the Bonga 803 sample that was 7°C lower than the value previously measured by SPDC Warri (3°C measured at DBR vs 10°C measured at Warri). The seafloor temperature is 6°C below the Warri pour point (10°C to 4°C = 6°C), therefore we chose a restart temperature 6°C below the DBR-measured pour point of 3°C (Trestart = -3°C). By doing this, we ensured that a gel would form and that the results would be conservative. More details can be found in Ref 24. The measured gel strength was found to be 3N/m2. In order to determine if this would lead to any problems during a restart, the required restart pressure was calculated for the Bonga flowlines, refer to Table 6.2. The calculations show that the required restart pressure for all the flowlines will be significantly lower than the maximum available pump pressure. The calculations in Table 6.2 are also made with the conservative assumption that the entire flowline loop is filled with dead oil that has gelled. All the initial high pour point wells flow into the west flowlines which are the shortest flowlines. Table 6.2 indicates that a maximum gel strength of 48N/m2 can be restarted with the topsides pump pressure in the West flowline loops. Also, if you assume that only leg of the flowline loop (ie one flowline instead of two) is filled with the gelled fluid then maximum gel strength of up to 98N/m2 can be restarted.


OPRM-2003-0302D

Page 3 of 6

30-April-2006


West 10in PFL

Unrestricted

East 10in PFL 1/2

East 12in PFL 3/4 and 5/6

East 12in PFL 5/6 only

Pipe ID

0.226m

Pipe ID

0.226m

Pipe ID

0.27m

Pipe ID

0.27m

Pipe Length

7000m

Pipe Length

21000m

Pipe Length

15000m

Pipe Length

9800m

Yield Stress (Pa)


Yield Stress (Pa)


Yield Stress (Pa)


Yield Stress (Pa)


1

1.24

1

3.72

1

2.22

1

1.45

2

7.43

2

4.44

2

2.90

1

2.48

2

1

3

3.72

1

3

11.15

3

6.67

3

4.36

6

7.43

6

22.30

6

13.33

6

8.71

12

14.87

12

44.60

12

26.67

12

17.42

2

24

53.33

24

34.84

2

30

43.56

41.33

60.002

24

29.73

30

37.17

48.43

60.002

16.14

60.00

1

27.00

1

Yield stress measured for Bonga 803 oil.

2

Maximum Pump Pressure.

60.00

Table 6.2 – Required Restart Pressures for Bonga Production Flowlines (PFL) Lastly, all the above discussion focuses on the pour point properties of the B1-803 Bonga fluid which is one of the worst fluids in Bonga in terms of wax properties. Mixing of this fluid with other benign fluids (eg 690 fluids or 720 fluids) will cause the pour point to reduce further. The amount of reduction that can be achieved is shown in Figure 6.1. As seen in Figure 6.2, a 70/30 mixture of 803 and 703 fluid reduces the maximum pour point to less than 0°C.

Blend Pour Points of Bonga B1 702 and 803 Oils 10

Upper Pour Point, C

0

-10

-20

-30

Upper limit; actual value may be lower.

-40 0

20

40 60 803 Oil Fraction, % vol

80

100

Figure 6.1 – Pour Points of Bonga B1 702 and B1 803 Blends Section 6 Pour Point Depressant Risk Assessment

OPRM-2003-0302D

Page 4 of 6

30-April-2006


4.0

Unrestricted

RISK ANALYSIS The following factors were considered while trying to make a decision to inject pour point depressant:

4.1

•

The probability of forming a gel in the first place (prevention)

•

The probability of removing the gel (remediation)

•

Operating Expenditure (OPEX) related to the pour point depressant

•

Other consideration (pour point system plugging up the methanol system and vendor analysis)

Probability of Forming a Gel This is directly related to the pour point. B1 803 has a pour point that ranges anywhere from 3°C to 10°C. In spite of having a pour point above ambient conditions the probability of forming a gel ranges from low to medium. This is because of the following reasons:

4.2

•

The above pour points are maximum pour points while production fluids tend to exhibit minimum pour points which are less than ambient for B1 803 (Ref 24).

•

Current procedures call for dead-oiling the flowline after a blowdown. This ensures that benign fluids replace ‘high-risk’ pour point fluids in the flowline during a long shutdown

•

It is possible to have a planned shutdown wherein the flowline is treated with methanol (and never blown down). In such a case, the flowline can cool down to below the pour point and form a gel. However, in this case the pour point would be slightly lower than that measured due to presence of more gas

•

Even a small amount of mixing with other benign oils can reduce the pour point quite substantially (refer to Figure 6.1). It is possible to ensure that high pour point oils are mixed with benign oils during production (except for PFL 08 and PFL 09, where all the production is from 710 fluids)

Probability of Removing a Gel This is related to the gel yield strength of the gel. It was not possible to form a gel for the B1 803 Bonga fluids at ambient conditions. Formation of a gel at -3°C indicated a weak gel structure corresponding to a gel strength of 3N/m2. The main drawback of these gel strength calculations is that they have not been correlated to field data. Calculations have indicated that a pressure of only 4bar is needed to break this gel structure even if the entire west flowline loop is filled with the B1-803 fluid. Calculations have shown that a gel strength of up to 96N/m2 can be dislodged with topsides pressure of 56bar even if an entire flowline is filled up with dead oil. Moreover, even if the topsides pump cannot dislodge the gel, there always exists the possibility of renting a low-rate high-pressure pump to dislodge the gel. Taken altogether, the probability of removing the gel (in case it forms) is high.

4.3

OPEX Related to Pour Point Depressant (PPD) Current OPEX projections for PPD range from 1 to 2MM$ a year. This is a substantial amount of OPEX that can be saved by not injecting the PPD.


OPRM-2003-0302D

Page 5 of 6

30-April-2006


4.4

Unrestricted

Other Considerations The Bonga production trees are designed such that the methanol and chemical injection have a common port. In spite of the presence of the insert to avoid mixing of the PPD and the methanol (which are incompatible chemicals) there still exists some risk of mixing and forming a plug, thereby jeopardising the methanol system. The probability of this occurrence ranges from low to medium. However, the impact of this is quite high considering that we could lose the methanol injection system to the tree. Preliminary testing results from the chemical vendors also indicate that they are unable to form a stable gel and thereby select an appropriate chemical. Hence, it may be prudent to defer injection of the chemical (if needed) until more representative production samples have been obtained either during well unloading or during actual production surveillance. All the above arguments are in favour of NOT INJECTING a PPD. However, the impact resulting from this event is quite high in the sense that we could potentially lose a flow line. Hence we need to carefully involve all the relevant parties while making this decision.

5.0

CONCLUSIONS Based on the analysis done thus far, the following conclusions are reached: •

The risks of a flowline plugging due to pour point problems ranges are low. This is based on the analysis done on one the worst fluids (from a geochemical perspective) at Bonga – the B1-803 fluid

•

The risks can be lowered further by putting operating procedures in place to mitigate the risk (dead-oil displacement after blowdown)

•

Active surveillance must be done during the initial months of production to verify pour point properties. This includes sampling of the appropriate sampling of the fluid (according to given procedures) and verification of fluid properties by a qualified laboratory

•

It is also recommended to collect samples and measure pour points on the 803 and 710 wells while the wells are being unloaded to the rig


OPRM-2003-0302D

Page 6 of 6

30-April-2006


Unrestricted

Section 7 Scale Review.


INTRODUCTION............................................................................................................. 3

2.0

SUMMARY OF CONCLUSIONS AND RECOMMENDATIONS..................................... 3 2.1

Bonga Formation Water Composition ................................................................. 5

2.2

Bonga Produced Gas Composition ..................................................................... 6

2.3

ScaleChem Predictions and Analysis.................................................................. 8

TABLES Table 7.1 – Matrix of Pressure, Temperature Used for Scale Analysis (Obtained from Bonga Subsea Systems Engineering Team).............................. 7 Table 7.2 – Critical Parameters for Severity of Uninhibited Scale (From SIEP Scaling Manuals) ........................................................................... 11 FIGURES Figure 7.1 – Scale Tendency as a Function of Temperature and Pressure for BaSO4 (0.1)................................................................................... 12 Figure 7.2 – mg/L as a Function of Temperature and Pressure for BaSO4 (0.1)....................................................................................................... 12 Figure 7.3 – Scale Tendency as a Function of Temperature and Pressure for BaSO4 (1)...................................................................................... 13 Figure 7.4 – mg/L as a Function of Temperature and Pressure for BaSO4 (1).......................................................................................................... 13 Figure 7.5 – Scale Tendency as a Function of Temperature and Pressure for BaSO4 (10) ................................................................................... 14 Figure 7.6 – mg/L as a Function of Temperature and Pressure for BaSO4 (10)........................................................................................................ 14 Figure 7.7 – Scale Tendency as a Function of Temperature and Pressure for BaSO4 (100) ................................................................................. 15 Figure 7.8 – mg/L as a Function of Temperature and Pressure for BaSO4 (100) ...................................................................................................... 15 Figure 7.9 – Scale Tendency as a Function of Temperature and Pressure for CaCO3 (0.1).................................................................................. 16

Section 7 Scale Review

OPRM-2003-0302D

Page 1 of 24

30-April-2006


Unrestricted

Table of Contents (cont’d) FIGURES Figure 7.10 – mg/L as a Function of Temperature and Pressure for CaCO3 (0.1) ................................................................................................ 16 Figure 7.11 – Scale Tendency as a Function of Temperature and Pressure for CaCO3 (1) ................................................................................... 17 Figure 7.12 – mg/L as a Function of Temperature and Pressure for CaCO3 (1) .................................................................................................. 17 Figure 7.13 – Scale Tendency as a Function of Temperature and Pressure for CaCO3 (10) ................................................................................. 18 Figure 7.14 – mg/L as a Function of Temperature and Pressure for CaCO3 (10) ................................................................................................ 18 Figure 7.15 – Scale Tendency as a Function of Temperature and Pressure for CaCO3 (100) ............................................................................... 19 Figure 7.16 – mg/L as a Function of Temperature and Pressure for CaCO3 (100)............................................................................................... 19 Figure 7.17 – Scale Tendency as a Function of Temperature and Pressure for CaCO3 (10) ................................................................................. 22 Figure 7.18 – mg/L as a Function of Temperature and Pressure for CaCO3 (10) ................................................................................................ 22 Figure 7.19 – Scale Tendency as a Function of Temperature and Pressure for CaCO3 (10) ................................................................................. 23 Figure 7.20 – mg/L as a Function of Temperature and Pressure for CaCO3 (10) ................................................................................................ 23 Figure 7.21 – Scale Tendency as a Function of Temperature and Pressure for CaCO3 (10) ................................................................................. 24 Figure 7.22 – mg/L as a Function of Temperature and Pressure for CaCO3 (10)................................................................................................. 24


OPRM-2003-0302D

Page 2 of 24

30-April-2006


1.0

Unrestricted

INTRODUCTION. Previous Bonga scale studies done by Frigo1 (June 1997), Morgenthaler et al2 (June 2000) and Lorimer et al3 (December 2000) have been reviewed. Our work was done primarily to evaluate scaling potential under operating conditions not considered in the previous studies (especially low pressure operating points) and to assess the current inhibition strategies in place for scale control at Bonga. The immediate focus was to address whether a subsea scale inhibitor is needed for Bonga during the first few years of operation. In the absence of data files used by Morgenthaler, the scaling calculations (self-scaling and mixing water) have been reworked using the latest information on production profiles and thermal hydraulic modelling obtained from the Bonga Subsea Systems Engineering Team. Exact reproduction of Morgenthaler’s results was not seen and this may be due to different strategies/philosophies used in the scale modelling work (eg brine reconciliation, ratios of brine and gas used etc) using SCALECHEM v2.2 simulation software. However, the trends of results and the major conclusions are similar to those of Morgenthaler et al. We have based our calculations on the brine and gas compositions as shown in Tables 7.1 and 7.2, which are similar to those used in the study performed by Morgenthaler et al2 (June 2000). Table 7.3 shows the various pressure temperature conditions evaluated in our study. The brine and gas flow rates were obtained via simulations (unavailable to past studies) from Bonga Subsea Systems Engineering Team (Susan Lindsey).

2.0

SUMMARY OF CONCLUSIONS AND RECOMMENDATIONS We would like to reiterate the recommendation of Frigo1 and Morgenthaler et al2 that we need more water samples from the Bonga region. The robustness of the water sample used in our analysis (and also that used by Morgenthaler) has been critiqued and evaluated by Morgenthaler et al in detail. The water analysis is indeed in question and needs to be verified via analysis on a fresh water sample from a Shell-certified laboratory. Whenever feasible, water samples should be obtained during developmental drilling in the region. The FEAST5 (Fluids Evaluation and Stability Testing) networks best practices should be applied to all sampling and subsequent analysis to ensure that the data we get are representative and under strict Shell Group guidelines. Efforts should be made to acquire or share data with other operators in the region or participate in studies aimed at understanding regional formation water chemistry. Our work is based on the water chemistry identified in the Bonga Development Basis of Design Document (Rev 5). The composition in Table 7.1 forms the design basis and hence has been used in our study. Basis our calculations and analysis, for the range of flow ratios of brine to gas and operating conditions that will exist at various stages of operation, the only two scaling minerals that could form scale would be barium sulphate and calcium carbonate. 1

Frigo, D: Scale review draft report sent via email to Bonga Project Team (1997).

2

Morgenthaler, L, Bell, F: ‘Water Compositions and Scaling Predictions for the Bonga Field: Sensitivity to Carbon Dioxide, Organic Acid, and Barium Concentrations’ (June 2000).

3

Lorimer, S, Wallace, C, Gibson, G: ‘ Review of Bonga Corrosion, Souring and Scaling (Rev 1)’ (Dec 2000).

5

Contact at [email protected] for FEAST sampling protocols and analysis.


OPRM-2003-0302D

Page 3 of 24

30-April-2006


Unrestricted

Basis information obtained through Bonga Subsea Systems Engineering Team (Susan Lindsey), the gas to water ratio (kscf per day of gas produced per bbl of water) over field life in Bonga would range from 0.1 to 100. Barium sulphate self-scale formation (from formation water) is likely for low temperatures (~40°F) at all pressures considered (150 to 5000psia) (scaling tendency >8). However, the amount of scale (mg/L) that would form in all these cases is minimal (~1 to 2mg/L) and below limits of those required for causing plugging/deposition problems4. The likelihood of barium sulphate scale to occur in the field is therefore minimal. No subsea treatment for barite scale is therefore recommended. This confirms the findings of Morgenthaler2. Surveillance and monitoring of produced water is strongly recommended, especially to verify the analysis of Bonga formation brine used in this study with respect to barium scale. Scale analysis should be redone when the new production water sample is analysed to scrutinise scaling risks and operating strategies. Calcite self-scale formation (from formations water) is likely at pressures lower than 1000psia (scaling tendency ~5, amount of scale ~300mg/L)) and at temperatures greater than 145°F. From the scaling manual4, the scaling risk is higher around 175°F and at pressures close to ~350psia, where scaling tendency ~10 and amount of scale ~600mg/L. The severity of calcite scaling causing production problems in these conditions will be low to moderate4. Subsea scale treatment is advised if Bonga’s operating conditions fall under these operating conditions (and produced water is formation water only). Topsides scale treatment is however advised, as separator pressures will be around 150psia where the scaling tendency increases to a value of ~ >10 and amount of scale is ~500mg/L. These conclusions are similar to those of Morgenthaler et al2. The impact of scaling calculations on specific Bonga conditions (anticipated thermal hydraulic conditions at various nodes and produced water compositions) during field life is explained in detail in Section 8. Surveillance and monitoring of produced water is strongly recommended. Scale analysis should be redone when the new production water sample is analysed to scrutinise scaling risks and operating strategies. Scaling risk for scenarios with production water having 25, 50 and 75% seawater content have been calculated. For these scenarios, calcite scaling is likely at conditions of high temperature (175°F) and low pressure (150psia). As these scenarios are not likely subsea, hence subsea-scaling risk is minimal for these ratios. Also, self-scaling risk from seawater production alone is negligible at all conditions tested.

4

Frigo, D: Scaling Manuals. SIEP 99-5679 and SIEP 99-5780.


OPRM-2003-0302D

Page 4 of 24

30-April-2006


2.1

Unrestricted

Bonga Formation Water Composition We have used the design basis formation water chemistry similar to the one used by Morgenthaler et al 2. Morgenthaler et al 2 have performed extensive simulations to analyse sensitivity of formation water to variables like carbon dioxide, organic acids and barium concentrations. The reconciliation of the brine to balance for electro-neutrality may be done by changing either the calcium or bicarbonate composition in the brine. However, we expect lesser errors during the analysis of calcium. Hence we chose to reconcile using the bicarbonate composition wherein the higher alkalinity (bicarbonate 2800mg/L) is attributed to unmeasured organic acids content (acetate as shown in Paragraph 2.1.2). Morgenthaler et al2 report a similar approach. This would need to be verified with analysis of new water samples from Bonga. The analysis of the new water sample should be carried out by a Shell lab (or a Shell-certified lab) using standard Shell certified protocols5. Paragraph 2.1.1 gives the original composition of water composition used by us as well as Morgenthaler et al2.

2.1.1

Original Brine Composition Before Reconciliation (Same as Reported by Morgenthaler et al2) •

Sodium (mg/L)

10440

•

Potassium (mg/L)

5140

•

Calcium (mg/L)

280

•

Magnesium (mg/L)

14

•

Barium (mg/L)

0.81

•

Strontium (mg/L)

2.1

•

Iron (mg/L)

15

•

Chloride (mg/L)

19305

•

Sulphate (mg/L)

500

•

Bicarbonate (mg/L)

2830

•

pH at 77°F

8.22

2


5

Contact at [email protected] for FEAST sampling protocols and analysis.


OPRM-2003-0302D

Page 5 of 24

30-April-2006


2.1.2

Unrestricted

Input Base Case Brine Composition After Reconciliation (Modified Water Composition) Paragraph 2.1.2 gives the composition used in our study after reconciliation (saturation at reservoir conditions for calcite and barite). •

Sodium (mg/L)

10592

•

Potassium (mg/L)

5140

•

Calcium (mg/L)

280

•

Magnesium (mg/L)

14

•

Barium (mg/L)

0.81

•

Strontium (mg/L)

2.1

•

Iron (mg/L)

15

•

Chloride (mg/L)

19000

•

Sulphate (mg/L)

500

•

Bicarbonate (mg/L)

2000

•

Acetate (mg/L)

1700

•

pH at 77F

7.76

Note: The water composition before and after reconciliation is almost the same except for the bicarbonate content. This water is super saturated in calcite when taken to reservoir conditions and has been used for scaling analysis to give conservative estimates.

2.2

Bonga Produced Gas Composition Paragraph 2.2.1 shows the composition of gas used in our analysis. We have used the same composition of Bonga produced gas as that used by Morgenthaler et al2. The moisture content (0.16%) was calculated at reservoir conditions (145°F, 4500psia) and then added to the gas stream composition.

2.2.1

Input Base Case Gas Composition (used by Morgenthaler et al (2000) •

CO2 (mole %)

2.5

•

H2O (mole %)

0.16

•

Hydrocarbon gas (mole %)

97.34

2



OPRM-2003-0302D

Page 6 of 24

30-April-2006


Unrestricted

CO2 content used herein (refer to Paragraph 2.2.1) is higher than that in the Bonga Basis of Design. This would provide conservative results and was chosen to match its value used in previous studies2. Table 7.1 shows the matrix of conditions used for analysis. They cover expected range of temperature (40 to 175°F) and pressures (150 to 5000psia) conditions at Bonga. 7

T (°F)

P (psia)

175

5000 4500 3000 1000 350 150

145

5000 4500 3000 1000 350 150

100

5000 4500 3000 1000 350 150

40

5000 4500 3000 1000 350 150

Table 7.1 – Matrix of Pressure, Temperature Used for Scale Analysis (Obtained from Bonga Subsea Systems Engineering Team) 2



OPRM-2003-0302D

Page 7 of 24

30-April-2006


Unrestricted

The main difference between this test matrix and the one used by Morgenthaler et al (2000) is the inclusion of the new separator (topsides) pressure conditions of 350 and 150psia. From the production data obtained from Bonga Subsea Systems Engineering Team, the various ratios of gas to brine that will exist in the various Bonga flowlines at various stages of the project life are: 0.1, 1, 10, 100kscf of gas per bbl of brine.

2.3

ScaleChem Predictions and Analysis ScaleChem has been used to evaluate scaling tendencies using the input data discussed earlier. Severity of the scaling and extent of problems anticipated are based on the scaling manual and instructions therein by Frigo3 (1997). The following cases have been run to evaluate scaling potential and associated risks.

2.3.1

Case #

Seawater

Formation Water

A (self-scaling)

0

100

B

25

75

C

50

50

D

75

25

E (seawater only)

100

0

Case A: Self-scaling Calculations (Produced Water is Formation-water Only) The methodology used in ScaleChem for performing self-scaling calculations is as follows: •

Reconcile the formation brine attributing the excess alkalinity to unmeasured acetate (salt of organic acid). This brine is then taken to reservoir conditions and used for scaling analysis)

•

This reservoir brine is then used with the gas phase in the scaling scenario option to calculate scaling tendency and amount of scale for the range of process conditions (refer to Table 7.1) and for the various gas-to-brine ratios

•

The gas composition used for simulations is in Paragraph 2.2.1

Basis our calculations and analysis, for the range of flow ratios of brine to gas that will exist at various stages of operation, the only two scaling minerals that could form scale would be barium sulphate and calcium carbonate

3

Lorimer, S, Wallace, C, Gibson, G: ‘ Review of Bonga Corrosion, Souring and Scaling (Rev 1)’ (Dec 2000).


OPRM-2003-0302D

Page 8 of 24

30-April-2006


Unrestricted

BaSO4 Scale Figures 7.1, 7.3, 7.5 and 7.7 show the scaling tendencies for barite scale as predicted by SCALECHEM for the four ratios 0.1, 1, 10 and 100 respectively. Figures 7.2, 7.4, 7.6 and 7.8 give the corresponding amounts of baryte scale (mg/L) for the four ratios. The trends in these graphs are very similar and consistent trends are also observed at the higher ratios (1000, 2000kscf gas per bbl of water). BaSO4 solubility is moderately affected by pressure, however it is strongly influenced by temperature. BaSO4 solubility increases sharply as a function of temperature. Therefore, scaling tendency increases as temperature drops, the highest value being at the lowest temperature (40°F). For a ratio of 0.1kscf gas per bbl of brine, Figure 7.1 shows the anticipated scaling tendency for the various conditions tested. The produced fluids will become saturated with baryte as temperature drops with saturation levels increasing as the fluids move to topsides. Production problems have been observed only when the degree of baryte super saturation becomes ~ 5 to 8, ie scaling tendency is ~ 5 to 83 (refer to Table 7.2). This threshold is reached when temperature reaches around 40°F. Hence, BaSO4 scale formation is very likely at conditions of lower temperature (~40°F). However, from Figure 7.2 the amount of scale (1 to 2mg/L) that precipitates out of solution is not significant to cause plugging or deposition problems (~ 50mg/L) based on guidelines provided in the SIEP Scaling Manual3. Increasing the ratio of gas to brine, ie 1, 10, 100, has no significant effect on either the scaling tendencies or the amount of scale as shown in Figures 7.3 to 7.8. Even though scaling tendencies are greater than threshold limits at lower pressures and temperatures, the amount of scale is not significant to cause plugging or deposition problems. Conclusions for Baryte Scale in Case A The baryte-scaling tendency reaches values above threshold for scale formation at lower temperatures (40°F). The amount of scale formed (mg/L) is not high enough to cause any concern regarding production problems (plugging/deposition). As such, the analysis shows a low risk of scale formation based on field experience. Based on data used and analysis done herein, no subsea baryte scale treatment is recommended. However, it is strongly recommended to obtain a good quality produced water sample and to perform scale analysis using the new data due to concerns with water analysis (eg Ca, Ba content) used for the simulations. We strongly recommend that topsides surveillance and monitoring strategies be put in place to assist operations in the management of possible scaling issues. CaCO3 Scale Figures 7.9, 7.11, 7.13 and 7.15 show the scaling tendencies for calcite scale as predicted by SCALECHEM for the ratios 0.1, 1, 10, 100 (kscf of gas per bbl of brine) respectively. Figures 7.10, 7.12, 7.14 and 7.16 give the corresponding amounts of calcite scale (mg/L) for these four ratios. For ratio of 0.1kscf of gas per bbl of brine, Figures 7.9 and 7.10 give scaling tendency and amount of scale that could potentially form. As suggested in the SIEP Scaling Manual3 (refer to Table 7.5), for calcite scaling to cause production problems, the threshold limit for scaling tendency is about 4 and corresponding amount is about 300mg/L. This threshold is reached in the system with operating conditions of temperatures greater than 145°F and pressures less than 1000psia. 3

Lorimer, S, Wallace, C, Gibson, G: ‘ Review of Bonga Corrosion, Souring and Scaling (Rev 1)’, Dec 2000.


OPRM-2003-0302D

Page 9 of 24

30-April-2006


Unrestricted

Under these conditions, low scale formation problems could appear. However, scale severity is increased at pressure around 350psia and temperature around 175°F. For higher ratios (ie 1, 10 and 100), Figures 7.11 to 7.16 show that the trend of results (scaling tendency and amount of scale formation). Based on our analysis, at these ratios, moderate problems could appear in the system with operating conditions having temperature around 175°F and pressure around 350psia. Conclusions for Calcite Scale in Case A Calcite self-scale formation is likely at pressures lower than 1000psia (scaling tendency ~5, amount of scale ~ 300mg/L)) and at temperatures greater than 145°F. From the scaling manual4, the scaling risk is higher around 175°F and at pressures close to ~350psia, where scaling tendency ~10 and amount of scale ~600mg/L. The severity of calcite scaling causing production problems in these conditions will be low to moderate4. At Bonga’s operating conditions, in the event of significant production of formation water only, subsea scale treatment is recommended. Topsides scale treatment is however advised for all scenarios. The impact of these scaling calculations on specific Bonga conditions (at anticipated thermal hydraulic conditions and at various nodes and under different produced water compositions) is explained in detail in Section 8. It is strongly recommended to obtain a good-quality produced water sample and to perform scale analysis using this data due to concerns with water analysis (eg Ca, Ba content) used for the simulations. Surveillance and monitoring of produced water is strongly recommended, as there are concerns about water analysis (eg very low barium content). This is especially crucial since the major assumption of our work and recommendations is the basis claim of reservoir engineers that produced water will be primarily seawater. Scale analysis should be redone when the new production water sample is analysed to scrutinise operating guidelines and strategies. We strongly recommend that topsides and subsea surveillance and monitoring strategies be put in place to assist operations in scale management. Note: Figures 7.1 to 7.16 indicate the scaling tendency and amount of scale possible at a specific water to gas ratio. The number in the brackets in the title for each figure, eg 0.1, indicates the specific ratio ie kscf of gas per barrel of water produced.

4

Frigo, D: Scaling Manuals. SIEP 99-5679 and SIEP 99-5780.


OPRM-2003-0302D

Page 10 of 24

30-April-2006


Scale Mineral Calcium Carbonate

Unrestricted

Ω

Expected Scenario

Range

Scale unlikely

1 to 4

Scale possible

≥4

Scale very likely

Comments

(mg/litre) Minor component of co-deposited scale possible

≥ 10

Few problems

< 300

Moderate

Limit may be lower for co-deposited CaCO3

300 to 700

Severe Barium Sulphate

χ Range

> 700

Scale possible

1 to 5

Scale possible

5 to 8

Scale likely

>0

NORM deposition possible Productivity problems begin

>8

Minor problems Moderate

< 50

Productivity problems begin

50 to 400

Severe

> 400

Table 7.2 – Critical Parameters for Severity of Uninhibited Scale (From SIEP Scaling Manuals)


OPRM-2003-0302D

Page 11 of 24

30-April-2006


Unrestricted

16.0

14.0

12.0 5000 10.0

4500 3000

Scale Tendency

1000

8.0

350 150

6.0 150

4.0

350 1000

2.0

3000

Pressure (Psi)

4500

0.0 175

5000

145

100

40

Temperature (F)

Figure 7.1 – Scale Tendency as a Function of Temperature and Pressure for BaSO4 (0.1)

1.4

1.2

1.0

5000 4500 3000

0.8

1000

mg/L

350 150

0.6

0.4

150 350 1000

0.2

3000

Pressure (Psi)

4500

0.0 175

145 Temprature (F)

5000 100

40

Figure 7.2 – mg/L as a Function of Temperature and Pressure for BaSO4 (0.1)


OPRM-2003-0302D

Page 12 of 24

30-April-2006


Unrestricted

16.0

14.0

12.0 5000

10.0 Scale Tendency

4500 3000 1000 350

8.0

150

6.0 150

4.0

350 1000

2.0

3000

Pressure (Psi)

4500

0.0 175

5000

145

100

40

Temperature (F)

Figure 7.3 – Scale Tendency as a Function of Temperature and Pressure for BaSO4 (1)

1.4

1.2

1.0 5000 4500 0.8

3000 1000

mg/L

350 0.6

150

0.4

150 350 1000

0.2

3000

Pressure (Psi)

4500

0.0 175

145 Temperature (F)

5000 100

40

Figure 7.4 – mg/L as a Function of Temperature and Pressure for BaSO4 (1)


OPRM-2003-0302D

Page 13 of 24

30-April-2006


Unrestricted

16.0

14.0

12.0 5000 10.0

4500 3000

Scale Tendency

1000

8.0

350 150

6.0 150

4.0

350 1000

2.0

3000

Pressure (Psi)

4500

0.0 175

5000

145 Temparture (F)

100

40


1.4

1.2

1.0 5000 4500 3000 1000 350 150

0.8 mg/L

0.6

0.4

150 350 1000

0.2

3000

Pressure (Psi)

4500

0.0 175

145 Temperature (F)

5000 100

40



OPRM-2003-0302D

Page 14 of 24

30-April-2006


Unrestricted

16.0

14.0

12.0 5000 4500 3000 1000 350 150

10.0 Scale Tendency

8.0

6.0 150

4.0

350 1000

2.0

3000

Pressure (Psi)

4500

0.0 175

5000

145

100

40

Temperature (F)


2.5

2.0

5000 4500 3000 1000 350 150

1.5 mg/L

1.0

150 350

0.5

1000 3000

Pressure (Psi)

4500

0.0 175

145 Temperature (F)

5000 100

40



OPRM-2003-0302D

Page 15 of 24

30-April-2006


Unrestricted

30.0

25.0

20.0

5000 4500 3000 1000

Scale Tendency 15.0

350 150

10.0 150 350 5.0

1000 Pressure (Psi)

3000 4500

0.0 175

5000

145

100

40

Temperature (F)

Figure 7.9 – Scale Tendency as a Function of Temperature and Pressure for CaCO3 (0.1)

700

600

500 5000 4500 400

3000 1000

mg/L

350 300

150

200

150 350 1000

100

3000

Pressure (Psi)

4500

0 175

145 Temperature (F)

5000 100

40

Figure 7.10 – mg/L as a Function of Temperature and Pressure for CaCO3 (0.1)


OPRM-2003-0302D

Page 16 of 24

30-April-2006


Unrestricted

25.0

20.0

5000 4500

15.0

3000 1000

Scale Tendency

350 150

10.0

150 350

5.0

1000 3000

Pressure (Psi)

4500

0.0 175

5000

145

100

40

Temperature (F)

Figure 7.11 – Scale Tendency as a Function of Temperature and Pressure for CaCO3 (1)

700

600

500 5000 4500 400

3000 1000

mg/L

350 300

150

200

150 350 1000

100

3000

Pressure (Psi)

4500

0 175

145

Temperature (F)

5000 100

40

Figure 7.12 – mg/L as a Function of Temperature and Pressure for CaCO3 (1)


OPRM-2003-0302D

Page 17 of 24

30-April-2006


Unrestricted

30.0

25.0

20.0

5000 4500 3000 1000

Scale Tendency 15.0

350 150

10.0 150 350 5.0

1000 3000

Pressure (Psi)

4500

0.0 175

5000

145

100

40

Temperature (F)


700

600

500 5000 4500 400

3000 1000

mg/L

350 300

150

200

150 350 1000

100

3000

Pressure (Psi)

4500

0 175

145 Temperature (F)

5000 100

40



OPRM-2003-0302D

Page 18 of 24

30-April-2006


Unrestricted

230 220 210 200 190 180 170 160 150 140 130 Scale Tendency 120 110 100 90 80 70 60 50 40 30 20 10 0

5000 4500 3000 1000 350 150

150 350 1000 Pressure (Psi)

3000 4500 175

5000

145

100

40

Temperature (F)


2000 1800 1600 1400 5000 1200

4500

mg/L 1000

1000

3000 350 150

800 600

150 350

400

1000 200

3000

Pressure (Psi)

4500

0 175

145 Temperature (F)

5000 100

40



OPRM-2003-0302D

Page 19 of 24

30-April-2006


2.3.2

Unrestricted

Case B-D: Mixed Water-scaling Analysis (Produced Water is a Combination of Formation Water and Seawater) The input seawater composition (from Grant Gibson – Bonga Subsea Systems Engineering Team – EPP) used in simulating scaling scenarios (Cases B-E) is as follows: •

Sodium (mg/L)

1968

•

Potassium (mg/L)

245

•

Calcium (mg/L)

262

•

Magnesium (mg/L)

1339.5

•

Barium (mg/L)

0.042

•

Strontium (mg/L)

5.66

•

Iron (mg/L)

0.025

•

Chloride (mg/L)

5075.7

•

Sulphate (mg/L)

3435.5

•

Bicarbonate (mg/L)

36.6

•

pH at 77°F

8.3

The seawater analysis above is electro-neutrally balanced (calculation checked using ScaleChem). However, it is different from a typical seawater composition available in literature7 (wherein eg Na+ is ~ 10400mg/L and Cl- is ~19400mg/L). The water analysis report6 mentions that the composition above is similar to other water samples collected in offshore Niger Delta regions. This could be true because of a dilution effect from the Niger Delta water. The Bonga seawater may need to be sampled again and re-analysed to confirm previous work. The analysis should be done by a Shell-certified laboratory or under Shell supervision. The scaling calculations were not affected significantly by accounting for increased salt content. We have used the above analysis as Bonga seawater. A quick simulation using standard seawater composition7 yields different magnitudes of scaling tendencies and amounts of scale, however the major conclusions are similar to those obtained from simulations using the above compositions.

7

rd

Handbook of Chemistry and Physics, 83 edition, Pages 14 to 17.


OPRM-2003-0302D

Page 20 of 24

30-April-2006


Unrestricted

The methodology used in SCALECHEM is as follows: •

Reconcile the seawater brine as above and saturate it with calcite at reservoir conditions (pressure 4500psia, temperature = 145°F)

•

This saturated seawater is then mixed with formation water (refer to Paragraph 2.1.2) in various ratios at reservoir conditions. In Case E, only seawater was used under reservoir conditions

•

The output brine resulting from this mixing calculation is then used as production brine for scaling calculations to assess scaling risk over the entire facility (bottom hole to topsides). Similar to the self-scaling analysis, the simulations have been run at ratios ranging from 0.1 to 100kscf of produced gas per barrel of produced water and for a matrix of thermal-hydraulic conditions (refer to Table 7.1)

•

The gas composition used for simulations is in Paragraph 2.2.1

It is worth mentioning that there is no appreciable change in the composition of the saturated seawater, indicating that it is almost saturated at surface conditions with this salt. We have performed simulations using two combinations: (1)

Saturated seawater + saturated formation water.

(2)

Saturated seawater + supersaturated produced water.

Results in case (2) above will give conservative estimates. The trends and values obtained for scaling tendency and amount of scale are similar and we will show the results in cases of simulation using 10kscf of gas in this report. In all these simulations, it is observed that calcium carbonate seems to be the only likely scale. At the ratios considered (refer to Paragraph 2.3.1), the mixing of the seawater and formation water yields a benign production water in terms of calcite scale problems. From these simulations, it is concluded that calcite scale will cause moderate problems around conditions of high temperature (> 145°F) and low pressure (~150psia). No appreciable scale is seen when seawater alone is produced and therefore self-scaling is not an issue when production water is seawater only (Case E). The results of the simulations for Cases B-D are shown in Figures 7.17 to 7.22. For the ratio of 10kscf per day of gas produced per barrel of water; Figures 7.17 and 7.18 give the scaling tendency and amount of scale respectively for Case B. The magnitude of scaling tendency and amount of scale obtained for remaining ratios are very similar. Similarly, Figures 7.19 and 7.20 highlight the results for this ratio for Case C. Case D results are shown in Figures 7.21 and 7.22. Note: The 350psia calculation node used in the self-scaling simulations (refer to Table 7.1) was replaced by 500psia to obtain a better midpoint for any extrapolation that might be necessary.


OPRM-2003-0302D

Page 21 of 24

30-April-2006


Unrestricted

4.5

4.0

3.5 3.0

5000 4500 3000

2.5

1000

Scale Tendency

500

2.0

150

1.5 150 1.0

500 1000

0.5

3000

Pressure (Psi)

4500

0.0 175

5000

145

100

40

Temperature (F)


250

200

5000

4500

150

3000

1000

mg/L

500

150

100

150 500

50

1000 3000

Pressure (Psi)

4500

0 175

145

Temperature (F)

5000 100

40



OPRM-2003-0302D

Page 22 of 24

30-April-2006


Unrestricted

2.5

2.0

5000 4500

1.5

3000 1000

Scale Tendency

500 150

1.0

150 500

0.5

1000 Pressure (Psi)

3000 4500

0.0 175

5000

145

100

40

Temperature (F)


160

1 40

1 20 500 0

1 00

450 0 300 0

m g /L

100 0

80

350 150

60

1 50

40

35 0 1 000

20

300 0

P re ssu re (P si)

4 500

0

175

145 T em p eratu re (F )

50 00 10 0

40



OPRM-2003-0302D

Page 23 of 24

30-April-2006


Unrestricted

0.9

0.8

0.7 0.6

5000 4500 3000

0.5

1000

Scale Tendency

350

0.4

150

0.3 150 0.2

350 1000

0.1

3000

Pressure (Psi)

4500

0.0 175.0

5000

145.0

100.0

40.0

Temperature (F)


1 1 1 1

5000 1

4500

m g /L 1

1000

3000

350

150

0 0

150 350

0

1000

0

3000

P re s s u re (P s i)

4500

0 1 7 5 .0

1 4 5 .0

T e m p e ra tu re (F )

5000 1 0 0 .0

4 0 .0



OPRM-2003-0302D

Page 24 of 24

30-April-2006

Shell Nigeria E & P Company Ltd

Unrestricted

Section 8 Risk-based Evaluation of Scaling Tendencies for the Subsea System


INTRODUCTION............................................................................................................. 2

2.0

SUMMARY ..................................................................................................................... 3

3.0

WAY FORWARD ............................................................................................................ 4

TABLES Table 8.1 – Scaling Risks for Wells During Early Life and Late Life ........................................ 3


OPRM-2003-0302D

Page 1 of 4

30-April-2006


1.0

Unrestricted

INTRODUCTION This section provides a risk-based evaluation of scaling tendencies for the Bonga Subsea System starting from the wellbore until the topsides. The following methodology was used to evaluate the scaling risks: (1)

The entire Bonga production was simulated on a well-by-well basis using PIPESIM (with the Field Planning Tool (FPT) feature).

(2)

The production function of every well was examined to quantify wellbore risk. The production function was split into early and late-life scenarios.

(3)

(a)

Early-life constitutes the dry phase of the well with a water cut less than 2% of the entire liquid production from that particular well. During this time, the pressure and temperature range over which the well spends most of its early life is evaluated for scaling tendency.

(b)

Late-life constitutes the wet phase of the well with substantial water cut (> 20%). Similarly, the pressure and temperature range over which the well spends most of its late life is evaluated for scaling tendency. It is important to note that almost all of the Bonga reservoirs (except for 803p2) produce a substantial amount of water and the water cut rises quite dramatically from 0% to greater than 20% within a very short period of time (2 to 6 months). This dramatic rise in water production is in sharp contrast to the average production life of most Bonga wells (range from 5 to 10 years) and hence an intermediate water cut case (mid-life) blends into the late-life cases. 20% was chosen because this is also the time at which the riser base gas lift is turned on, which causes a substantial lowering of manifold pressure. Although a water cut between 2 and 20% is not explicitly covered, it is included in the late-life scenario. This is justified because scaling problems do not occur when the well begins to make substantial water (refer to Paragraph 2.0).

(c)

As mentioned above, the pressure and temperature range over which the well spends most of its life was examined for scaling risks. This will account for the most likely risk for scaling. For example, well 702p5 has a pressure range that varies from 1000psi to 650psi (early-life, prior to water cut). However, the well spends most of its life between 600psi and 700psi (early-life) and hence we looked at a pressure of 700psi to evaluate scaling tendencies during early-life.

A similar approach is taken to evaluate risks for flowlines and risers. Production profiles in each flowline and riser are evaluated and split into early and late-life according to similar criteria as listed above. Scaling risks are then evaluated by looking at the pressure and temperature regions where the flowline and riser spends most of its lift.

The following assumptions are made while evaluating scaling tendencies: (1)

Scaling tendency has been evaluated only for the Bonga Subsea System starting from the wellbore up to the topsides. Scale has been identified as a risk at topsides and there is provision for the injection of scale-inhibitor at the topsides.


OPRM-2003-0302D

Page 2 of 4

30-April-2006


2.0

Unrestricted

(2)

Current scaling evaluations have been made with a water sample that was obtained from the original Bonga exploration wells (Bonga No 2 well, sample analysed in May 97). There is apparently a more credible water sample (stored in Nigeria from both Bonga-5 (702-w4) or Bonga-7 (702-p9)), but an analysis on this has not been performed. It is important to re-examine this entire report once analysis from the new water sample becomes available. In particular, attention must be paid to the barium content of the water sample (current sample shows that amount of Ba is less than 1ppm) since this can have substantial implications with respect of mixing with seawater.

(3)

Reservoir engineers have indicated that most of the produced water (water cuts > 20%) in Bonga will be seawater that has been injected. The percentage of seawater in the produced water is estimated to be greater than 70%. Although we have calculated scaling tendencies at a seawater percentage of as low as 25%, this assumption needs to be validated at the FPSO when the wells begin to produce water.

(4)

We have also assumed that any small amount of water that is produced during early life will be formation water). Moreover, we have evaluated scaling tendencies at a ratio of 10kscf/bbl and 100kscf/bbl of water produced. This is most consistent with the production GoRs and amount of water that might be produced. For example, a production of 20,000bbl per day of oil (GoR of 500) and 100 barrels per day of water results in an approximate ratio of 100kscf/bbl.

SUMMARY Based on the main report and the production functions, we came to the following conclusions: (1)

Baryte scale is not a problem for Bonga, all subsequent scaling tendencies refer only to calcite scales. However, it is critical that we verify the water composition for barium after the field starts cutting water.

(2)

Calcite scaling is not a problem when the field starts to produce water. This is mainly due to the fact that a mixture of seawater and formation water leads to highly reduced scaling tendencies.

(3)

Scaling is a potential problem only during early life where formation water will be produced. Table 8.1 describes the problem. Early Life (< 2% WC)

Late Life

Comments

Wells

5 wells out of 21 wells are problematic

No Problem

Well Nos 702p5, 710p4, 803p2, 803p3

Flowlines

1 out of 8 flowline loops are problematic

No Problem

PF-12

Risers

All risers are problematic

No Problem

–

Table 8.1 – Scaling Risks for Wells During Early Life and Late Life


OPRM-2003-0302D

Page 3 of 4

30-April-2006


Unrestricted

Although, the above regions of the subsea system have been identified as being problematic during early life, the actual risk of scale deposition is quite low due to the following reasons:

3.0

•

A small amount of water typically implies that the pipe is oil-wet (as opposed to being water-wet). This means that precipitated scale will not have a chance to deposit on the walls of the pipe. This shows that the likelihood of scale formation is small

•

Due to the small amount of water being produced, the actual amount of scale deposited will be quite low. This will give us a chance to analyse the water being produced and take remedial action if necessary. Moreover, the teleconference with Phil Webb1 indicated that calcite scale deposits over much longer timeframes than baryte scale. This shows that scale formation is not likely to be catastrophic

•

Calcite scale can be remediated with an acid job. Although this would be highly undesirable, a potential job could be arranged with the right concoction of acid and corrosion inhibitors. The timescale for such a remediation job would be the order of hours (as opposed to1 baryte scale, which is almost impossible to remediate). This shows that scale could be remediated.

WAY FORWARD In view of the above conclusions, we do not recommend scale inhibition subsea. However, it is critical to analyse produced water from the field as soon as the field begins to produce water. This is to verify our water analysis with respect to calcite scale and most importantly to verify our barium content. A higher barium content (say an increase from 1ppm to 12ppm) can result in substantial incompatibility with seawater as shown by Morgenthaler et al (July 2000) and some action needs to be taken.

1

Teleconference with Phil Webb, 24 October 2004.


OPRM-2003-0302D

Page 4 of 4

30-April-2006

Flow Assurance Guidelines

Recommend Documents