Text For RBA Drilling and Completions
Text For RBA Drilling and Completions
Text For RBA Drilling and Completions
PHASE 2 DEVELOPMENT
OF THE
OSEIL OIL FIELD
APPENDIX
DRILLING PROGRAMME
TABLE OF CONTENT
1. Summary ...................................................................................................................................... 4
2. Introduction ................................................................................................................................. 8
3. Discussion ................................................................................................................................. 10
Figure 1Estimated Cost Difference between Oseil-4 & Oseil-3 the First Phase 2
Development Well...........................................................................................................5
Figure 2 Oseil Phase 2 Development Planning, Approvals, Procurement, Drilling &
Completions Schedule....................................................................................................7
Figure 3 Determination of the Technical Limit .....................................................................11
Figure 4 Well Time Prediction..............................................................................................12
Figure 5 Oseil Landslip Zonation .........................................................................................15
Figure 6 Multi-Well Pad Design ...........................................................................................16
Figure 7 Bore Hole Instability...............................................................................................18
Figure 8 Conventional & Slim Casing Design ......................................................................18
Figure 9 General Operating Envelope for API & Premium Connections..............................21
Figure 10 Conceptual Completion Schematic for Oseil Ph.2 Development.........................24
Figure 11 Material Selection for Tubulars ............................................................................32
Figure. 12 Oseil Annulus Velocity Profile for PI = 700 ........................................................38
Figure 13 Oseil Inflow Performance for Point Source @ 8440 ft MD (6800 ft TVD) ............38
Figure 14 Generic Well Design Chart .................................................................................41
1. Summary
Oseil Phase 2 Development includes drilling and completion of up to 15 additional wells into
the structures located around the existing Oseil Field and a new prospect area called Oseil
Tenggara to the southeast. These additional wells will be drilled directionally and with
underbalanced drilling from the existing Oseil well sites and will have 300-500 ft horizontal
sections in the Manusela reservoir. The wells will be completed open hole with electric
submersible pumps (ESPs) or gas gift. A schematic of the completion design is shown in
Fig. 10. Note, however, that well and completion designs presented in this document are
conceptual based on work completed to date and may change.
Significant efforts have been made to reduce the cost of these wells. Current cost estimates
for the first oil well are US$ 5.647 million with technical limits improvements leading to costs
for the final development well of US$ 5.54 million. Cost reductions relative to past wells are
due to reduced hole size, fewer casing strings, directional wells from existing locations,
batch drilling/completions, and economies of scale from more wells. An added cost
reduction will be from use of a better drilling rig: with more wells will be an increased number
of drilling contractors will participate in the bidding.
Operations will commence with mobilisation of the drilling rig and all materials and services
in late 2003 and completion of the first two wells on the Oseil 4 well pad in 1st qtr 2004. A
schedule is shown in Figure 2.
Attachments 1 & 2 contain detailed cost estimates for Oseil-3 and for Oseil-13 (the last well
on the Oseil-4 multi-well pad). For reference, the lowest well cost to date was for Oseil-4 at
US$11.626 million. The cost difference between Oseil-4 and the estimated cost of
US$5.647 million for Oseil-3, the first well in the Phase 2 development, is illustrated in Fig. 1.
The costs and differences between these wells is summarised as:
Difference Percent of
Item Oseil 4 Actual Oseil 3 Forecast
US $ Total $
Time 83 days 40 days 3,228,750 54%
Tangibles $ 1,413,451 $ 538,732 874,719 18%
Mob/Demob $ 1,683,142 $ 463,780 1,219,362 25%
Site $ 715,000 $ 30,000 436,588 9%
Others $ 490,789 $ 270,807 219,982 4%
Total 5,979,401 100%
The 54% difference in time prediction for the first well as compared to Oseil 4 is because of
the large amount of Removable Time (see section 3.1 for further explanation). The
difference is tangibles cost is a function of the simplified slim-hole casing and completion
design and expected cost savings when purchasing materials for 16 wells. The Mob/Demob
difference is also a function of the aggregate cost spread over 16 wells as compared to the
two well programme in 1998. The site upgrade costs per well for the Oseil-4 multi-well pad
are small as compared to the original site cost. The Others is the anticipated lower
services cost for a multi-well programme.
Figure 1Estimated Cost Difference between Oseil-4 & Oseil-3 the First Phase 2
Development Well
14,000,000
12,000,000
10,000,000
8,000,000
US$
6,000,000
4,000,000
2,000,000
0
Oseil-4 Oseil-3
The estimated well costs for the 16 well programme are shown in Table 14.
1.2 Uncertainties
The following uncertainties are present in the planning, costs, design and execution
assumptions made in this Appendix.
Schedule:
Approval to commence tendering before internal and BP MIGAS AFE approvals have
been obtained.
Longer than expected delivery times for tangibles.
Rig availability after 1Q04
Well Design:
Geomechanical issues restrict directional drilling plans such that targets cannot be
reached within current drilling programmes.
Tectonically stressed Kola Shale may cause serious hole instability and stuck pipe.
Significant over-pressure in overburden (e.g. Lola Kecil)
Inability to drill entire horizontal section through the reservoir in an underbalanced
condition due to frequency and permeability of fracture zones.
Inadequate hole cleaning in horizontal sections due to restrictions imposed by UBD.
Unable to fully isolate gas cap to allow drilling and production in the oil leg without
excessive gas production.
The actual level of corrosion is not currently known and this may seriously impact the
selection of materials.
Costs:
Well cost estimates assume the benefits of a continuous 16 well programme a
lesser well programme or an interrupted, discontinuous programme will cost more
per well.
Drilling Performance:
Ability of lead contractors, particularly drilling rig and directional drilling, to
significantly improve upon previous performance.
Improvement in average rates-of-penetration.
1.3 Schedule
The detailed planning and procurement process for this programme will commence
immediately internal approval is given. Drilling operations in the field are planned to
commence early in the first quarter 2004 and to continue through until the fourth quarter
2005. This aggressive programme will depend upon obtaining timely internal and external
approvals, procurement of goods and services and a significant improvement in performance
from previous drilling operations on the Oseil Field. An outline schedule is shown in Fig. 2
Figure 2 Oseil Phase 2 Development Planning, Approvals, Procurement, Drilling & Completions Schedule
2. Introduction
The Oseil Phase 2 Development Team (ODT) has the following main objectives:
Increase production.
planning and execution phases, their impact and mitigating plans or actions. The Risk
Assessment is used to communicate the risk picture to Management. A detailed Drilling
Operational Risk Assessment will be performed and subsequently communicated to the well
site team. This will guide the Drilling Team during the preparation of detailed operational
procedures.
FREQUENCY / NON-
MINOR SEVERE CRITICAL CATASTROPHIC
CONSEQUENCE CRITICAL
HIGH
MODERATE
LOW
REMOTE
NEGLIGIBLE
Key
ACCEPTABLE RISK ALARP ZONE UNACCEPTABLE RISK
Severity Definition
Non-Critical Event that causes no hazard to personnel, facilities or the environment.
Economic Value < US$10,000
Minor Event that causes light personnel injuries, local damage to facilities or the
environment
Economic Value < US$50,000
Severe Event that causes severe personnel injuries, severe damage to facilities or
the environment
Economic Value < US$500,000
Critical Event that causes personnel fatalities in the vicinity, critical damage that
threatens the integrity of facilities or extensive environment impact
Economic Value < US$10,000,000
Catastrophic Event that causes large numbers of personnel fatalities in the vicinity and
outside, total loss of facilities or long term extensive environment impact
Economic Value > US$10,000,000
Water
Losses/Contamination into High Low High
reservoir
(References 2 & 3)
3. Discussion
The following section discusses in detail the past drilling performance analysis, well
locations for Phase 2, the on-going well bore stability and geomechanical modelling work,
casing and completion design proposals, corrosion issues, the proposed drilling plan and
procurement of goods and services.
Technical Limit is a term used to describe a level of performance defined as the best
possible for a given set of drilling parameters. It is the theoretically flawless or perfect time
for performing each of the steps in well construction i.e. the entire process from rig move,
preparation to drill, drilling and completing the well. The perfect time can be described as
the actual well time, from past performance, less the removable time. This removable time
includes conventional lost time (e.g. equipment failures, downhole drilling problems, waiting
on equipment or services, stuck pipe, well control etc.) and lost time for less-than-optimum
performance (use of sub-optimal parameters or equipment e.g. bit performance, slow
tripping times) and time spent on unnecessary work (use of non-applicable procedures e.g.
frequent wiper trips, conditioning mud, surveys). The analysis of the removable time
required defining the problems and causes and seeking solutions in order to eliminate this
lost time from future operations (Ref 5).
CONVENTIONAL
THEORETICAL PERFECT INVISIBLE LOST
LOST OR DOWN
WELL TIME TIME
TIME
REMOVABLE TIME
TIME
TECHNICAL LIMIT
A summary of the Technical Limit Time estimate for a typical Phase 2 Development well is
given below:
Estimating the time required to drill and complete the Oseil Phase 2 Development wells is an
important part of the project planning and determination of project economics. In addition,
the analysis provides a scale against which performance can be measured. Equations
derived from statistical analysis of wells were used to predict the well times for the
development programme (Ref. 6). Because of the small amount of local data available, the
equations were modified to provide a best fit using engineering experience and judgement.
The analysis of the Oseil database shows a steady improvement in well times between
1993 and 1998. This is attributed to learning from the problems in the earlier wells,
improvement in performance of equipment and crews and major changes in the drilling
programmes made after experience on the first and second wells. The Oseil Development
wells are planned to be drilled directionally and whilst it may appear that the vertical wells in
the data base are not applicable when predicting drilling times, other well time analyses
(Ref. 7) have indicated that hole angle has very little effect on well times since the measured
depth of the well overrides this effect. The Geomechanical Modelling as referred to earlier
will help to verify if this is likely to be a valid statement for the Oseil wells.
The equations used to predict the Oseil Ph.2 Development well times are as follows:
Where,
The Learning Curve Factor LCF = 1.0 + C x exp (0.693(Drilling Order 1)/D)
And,
The Spud Date Factor SDF = (1.0 E) x exp (Well n Spud Date Well 1 Spud Date)
50.0
40.0
30.0
20.0
10.0
0.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Well Number
consequence of this has been an escalation in costs and continual upgrading. The cost of
the Oseil well sites and roads is between US$1 to 2.5 million each (about 15 20% of the
total well cost) and require constant maintenance to keep in a useable condition. The
principal reasons for the high costs are:
Topography
Propensity for landslides
Weather conditions very wet season October to April
Remoteness of location from civil engineering infrastructure
The landslides aggravated by the building of the Oseil-4 access road and well site has been
continually monitored since late 2000. The slides continue to slip further and have
completely blocked the original lower section of the access road to Oseil-2 & 4 and now
impinge upon the Bula Air River. The main reasons for the landslides are:
The topography around the Oseil Field is mountainous terrain up to 700m above sea level,
with steep and distinct ridges and crests with numerous valleys along which landslides, both
natural and induced, have occurred. The surface geology is Middle Triassic to Early
Jurassic rocks of the Kanikeh Formation consisting of interbedded mudstones, shales,
sandstones and siltstone. Tertiary Salas deposits consisting of older boulders or blocks
(sometimes kilometres in size) occur on the flanks of the Oseil Anticline. The Salas and
Kanikeh Formations are often covered by residual soil about 2 4 m in thickness. The
coastal plains are relatively flat low rolling hills that are often waterlogged after heavy rain.
Considering the degree of damage already done to the environment and the long-term
impact on the use of current well sites and continued access to these sites, further well site
construction work would be foolhardy. The physical conditions on the Oseil Anticline are
apparently not suitable for such large-scale earthworks. A preferable solution is to try to
contain the current damage and to avoid building work unless necessary. Drilling the
development wells from the existing well sites would save 85-90% of the typical cost of a
new well site, and would not aggravate the current land slides further, if conducted properly
and with appropriate geotechnical input into the construction design. Geotechnical
investigations have been carried out at the Oseil-1 well site. This work must be extended to
cover the Oseil-2 and 4 well sites.
However some of these cost savings by using existing well sites at Oseil-2 and 4 as the two
main multi-well pads from which to drill directional wells to reach the desired targets in the
reservoir are offset by the requirement to drill directionally and the increased risk involved in
this drilling work. The geomechanical study underway will assist in proper well planning to
mitigate these risks. Fig. 5 overleaf shows the landslide zonation on the north side of the
Oseil mountain ridge. Both Oseil-2 and 4 well sites are situated on stable, if narrow and
restricted, areas. There are severely unstable areas to the north (down-slope) sides of both
well sites. The Oseil-1 site is stable but has active slides upslope and down slope with
movement rates of 2 3 m per year. The geotechnical study of the Oseil-1 site indicted a
high risk of reactivation of the slides under the well site platform. For this reason, the Oseil-1
site has been avoided for the Phase 2 development drilling. It intention is to use the
currently large Oseil-4 well site as a multi-well pad by constructing 10 additional well cellars
either in a line, as shown in Fig. 6 or in a 5 x 2 grid with a 5 m spacing between centres.
The rig derrick, drill floor and substructure would be skidded between well cellars to access
each new well cellar in turn whilst all the other rig equipment and mud pits would remain
stationary. The Oseil-2 well site will require two additional well cellars to be constructed.
This site is considerably more restricted, particularly to the north (down slope) and south
(rock wall) sides. The plan will be to extend the site to the west to place all three cellars in
one line running E W. Construction cost for these site modifications have not been
included in the well costs presented in section 3.8 but are expected to be in the order of
US$150,000 to 500,000 per well pad site.
Typical output from a 3D linear elastic borehole instability risk analysis showing the
equivalent circulating density required to prevent hole collapse as a function of well
inclination and azimuth. Areas with red shading indicate well trajectories that require
overbalanced bottom hole pressure.
The benefits of the slim casing design are a 48% reduction in tangible costs for casing and
accessories and a 20-35% improvement in the drilling and casing installation time.
Casing Design Process: The Oseil Development well casing design was performed using a
combination of manual calculations and Landmarks StressCheck v. 3.0. It was based on
the deepest prognosed formation tops and lithologies (including a +100ft margin of error for
possible underestimation of stacking velocities), and the estimated formation pore and
fracture pressure gradients as further described below.
The objective of the design was to allow the Jurassic Manusela Formation reservoir to be
safely developed for production using 3-1/2 tubing inside 7 casing. The design also takes
into consideration the expected hazards i.e.
Pressure transitions at the top of the Imbricate Zone (Late Miocene to Late Triassic)
Upper Nief Fm. (Middle Miocene to Late Paleocene) with possible fresh water and
associated gas flows in the Imbricate Zone.
Brittle, tectonically stressed Kola Shale (Late Jurassic)
Sub-normal Manusela Fm. reservoir pressure
Hydrogen Sulphide in the Manusela Fm. reservoir.
A 5 liner has been included as a contingency in the event the 7 casing is required to be set
prior to reaching the base of the Kola Formation.
Kick Tolerance: This is the maximum volume of a (usually gas) influx, or kick, into the well
bore that can be tolerated based on a 0.5 ppg influx intensity i.e. the pore pressure at the
point of influx is 0.5 ppg higher than the hydrostatic pressure of the mud weight in the hole.
Kick Tolerance is usually expressed as a volume in barrels because this is the measure
drillers use most commonly when referring to kicks. Some areas use a kick tolerance
expressed as a mud weight e.g. a 12.6 ppg kick tolerance means the calculated mud weight
required to balance pore pressure and incorporates various sizes and lengths of BHAs. The
minimum acceptable tolerance should be >25 bbl and preferably at least 50 bbl.
Tubular Connections
The selection of suitable connections needs to reflect the intended service. The following
flow chart was used to assist with the selection of casing and tubing connections for the
Oseil Phase 2 Development wells:
A P I o r P r e m iu m C o n n e c t io n ?
Is M a t e r ia l Yes
C o r r o s io n R e s is t a n t A llo y ?
No
Is C o n n e c t io n S t r e n g t h R e q u ir e d Yes
t o b e 1 0 0 % o f P ip e B o d y
No
P r o d u c t io n C a s in g P r o d u c t io n T u b in g
o Yes
Tem p > 210 F
No
Type of Type of
F lu id F lu id
L iq u id G as L iq u id G as
A P I C o n n e c t io n A P I C o n n e c t io n
P r e m iu m C o n n e c t io n
Casing Design Philosophy: The choice of casing sizes and setting depths was based on
the following:
7 production casing (or 5 liner) production liner set at the base of the Kola Shale to
isolate the overburden from the sub-normally pressured reservoir.
A contingency 5 liner will be available if unexpected hole conditions force setting the 7
casing prior to reaching the planned section target depth and also for isolating the possible
gas cap on some fault blocks.
The most critical part of this casing design is the setting depth of the 9-5/8 casing. The 8-
1/2 hole section through the Imbricate Zone, Upper and Lower Nief and Kola Shale is
probably the most challenging section of hole to be drilled. Fresh water and gas flows have
occurred in the abnormally pressured Imbricate Zone requiring kill mud weights up to 11.7
ppg (Oseil-2) and minor to moderate lost circulation rates of 10 50 bbl/hour occur in all the
formations. There are formation dips of 20- 45 degrees that have a severe influence on the
well trajectory. Stuck pipe has occurred when drilling the Kola Shale on Oseil-2 and appears
to be related to a thinner, more highly stressed shale section over the crest of the reservoir
in this fault block.
There is possibility that the 7 casing may need to be set prior to reaching the planned
section target depth if the leak-off test at the 9-5/8 shoe is < 14.8 ppg and more severe
over-pressure is encountered in the 8-1/2 hole section, as occurred in well Lola Kecil-1 and
Bolifar Utara-1. To mitigate these risks the Rotating BOP will be installed on to the 13-5/8
or 11 5,000 psi WP BOP after setting the 9-5/8 casing and installing the 9-5/8 wellhead.
The 5 contingency liner may then be required to isolate the overburden from the sub-
normally pressured reservoir.
Section 9 of this programme deals with the issue of corrosion. It is recommended that either
the exposed section of 7 casing below the ESP packer utilises a L-80 13Cr steel or plastic
coated pipe to mitigate the problems of corrosion.
Calculated Results: The results of the Oseil Ph.2 Development casing design are
summarised below:
Table 7 Calculated Design Safety Factors
The Oseil Phase 1 Development completion design (Ref. 10) currently installed on the 3
wells had the following main objectives:
To perform extensive pressure build-up (PBU) testing of all three Oseil wells to obtain
essential data on reservoir size and properties.
Evaluations of the selected Artificial Lift System (ALS) for Phase 2 when enough
production history records and ESP failure rate data are available.
Evaluation of the corrosive environment to assist in the selection of suitable
materials.
At the time of writing no workovers have been performed on the producing Oseil wells
therefore evaluation of the ALS using Electric Submersible Pumps (ESP) system is
incomplete as is the issue of corrosion (see below). However, the current high completion
cost must be addressed for economic Phase 2 development. The current Oseil well
completion tangible costs are over US$1 million per well because the design incorporates
features required for reservoir evaluation as mentioned above such as surface controlled,
deep downhole shut-in to reduce wellbore storage effects and permanent, high accuracy
surface read-out pressure gauges. Many of these features can be eliminated when the
completion design objectives are long-term, trouble free production using low-cost, low-
maintenance equipment.
The proposed completion design will now feature 3-1/2, J-55, 9.2 ppf, EUE tubing with slim
hole, multi-stage ESPs with possibly tandem motors or tandem ESPs, variable speed drive
(VSD) or Soft-start, Fixed Speed Drive surface controller. Gas lift completions will be used if
sufficient gas is proven. A schematic of the conceptual completion design is shown in Fig.
10.
2
2. 3-1/2 Tubing
4 4. Swivel Joint
5 5. Adjustable Union
6 6. Selective Nipple
Artificial Lift
In late 2000, KUFPEC commissioned a study of the most appropriate Artificial Lift System
(ALS) for the Phase 1 Development. The study reviewed the most likely methods as
summarised below:
From Table 8, the systems that are considered most suitable for the Oseil Field taking into
consideration reservoir & fluid properties, depth of reservoir, required production rates are:
None of the above systems has any restrictions at the 6,600 ft (2,012 m) maximum setting
depth in Oseil Field. The report concluded:
Gas Lift has a number of attractive features for Oseil Field. The only big drawback in
the Oseil Field is the uncertainty on the amount of gas. Therefore, it is important to
prove sufficient gas to produce the wells at the required rate by drilling a well to
evaluate the possible gas cap in the fault block penetrated by well Oseil-1 & 4.
ESP is the most reliable system for producing the wells to maximum potential. Pump
life is the biggest concern. If ESP life is between 6 month and one year, then OPEX
will be extremely high.
The well completion design should incorporate the possibility of producing the wells
by Natural Flow.
Due to the highly corrosive nature of the produced fluids, the installation of a
corrosion probe with remote real-time monitoring and surface sampling programs
should be implemented.
Selection of the equipment material should take into account workover rig frequency
and the possibility of changing equipment during workover jobs.
Wellhead
A 13-3/8 X 9-5/8 X 7 X 3-1/2 X 3M H2S / CO2 service SpeedLock/Unihead Thru-Bore
Well Head system will be used for the Oseil Phase 2 Well Development Program. This will
consist of a 13-3/8 Slip-On-Lock Casing Head and a 9-5/8 x 7 x 3-1/2 Unihead. This
combination system has been chosen over the conventional flange type because the 13-3/8
SOL casing head is installed after the surface pipe has been set and does not require any
welding. Because of this, the BOP stack will be installed only once and will remain in place
throughout the drilling and completion process. All the casing hangers and pack-offs will be
landed and pressure tested through the BOP without removing the BOPs. Well control will
be maintained throughout each program. The SpeedLock/Unihead wellhead system has
many advantages:
Reduces BOP nipple-up/down time as much as 3.8 days/well when compared to the
Conventional Flange System (Operational Time Comparison of Oseil 1, 2, and 4
wells versus the Conoco Philips Ramba Field wells - refer to tables below).
Less equipment i.e. no additional spools and risers to connect to the bell nipple (one
connection needed only).
Less BOP pressure testing time required.
Reduced manpower needed to complete operation (one man operation using
pneumatic impact wrench).
Installed using a standard set of tools.
Isolates personnel and the environment from dynamic well conditions when setting
and cementing casing and pack-off annular spaces on all strings except the
conductor casing.
Initial capital cost is higher then conventional flanged wellhead systems because of the tool
requirements but this cost is quickly recovered when amortised over a multi-well program.
K U F P E C - O S E IL C o n o c o P h ilip s -
C o n v e n t io n a l F la n g e c o n n e c t io n R am ba
C o n v . F lg e Speed
C A S IN G S IZ E 1 2 4 A ve. Conn. Loc S a v in g s
20" 1 6 .5 0 0 .0 0 0 .0 0 1 6 .5 0
1 8 - 5 /8 " 0 .0 0 2 7 .5 0 1 4 .5 0 2 1 .0 0 1 4 .0 0 2 .5 0 1 8 .5 0
1 3 - 3 /8 " 2 0 .5 0 3 0 .0 0 1 6 .5 0 2 2 .3 3 1 5 .0 0 1 .1 5 2 1 .1 8
9 - 5 /8 " 4 6 .0 0 2 7 .0 0 3 9 .0 0 3 7 .3 3 1 7 .0 0 1 .5 0 3 5 .8 3
T re e 3 1 .0 0 1 4 .0 0 9 .5 0 1 8 .1 7 1 0 .0 0 1 .5 0 1 6 .6 7
T o ta l H o u rs 9 2 .1 8
T o ta l D a y s 3 .8
CONVENTIONAL FLG.
ITEM UNIHEAD WELLHEAD
WELLHEAD
The Speedlock / Unihead Wellhead System allow the ability of each casing string to be run
and cemented and annulus isolated in the wellhead with dramatically reduced operational
nipple up and or down time when compared to the Conventional Flange Systems. However,
the running tools i.e. casing hanger running and setting tool, casing pack-off running and
setting tool, pressure test plug running tool and bore protector running and setting tool must
be purchased. The initial capital cost is greater for the Speedlock / Unihead Wellhead
System because of these running tools but this disadvantage in cost be recovered in the
operational time saved. The daily operating cost is expected to be in the order of
US$125,000 per day, and as shown above the time saving is about 3.8 days per well.
Therefore, the cost of the tools is recovered during the first well and provides operational
savings of potentially 57 days and US$7.125 million over the development drilling
programme.
Xmas Tree
The selected design and specification of the Xmas Tree will be a H2S/CO2 service stack-up
flanged gate valve type incorporating dual master valves (one hydraulically actuated), flow-
tee, production wing valve, manual adjustable choke and swab valve. The size of the tree
will be 3-1/8 bore and will be rated to 3,000 psi WP.
3.6 Corrosion
The Oseil reservoir fluids contain hydrogen sulphide, carbon dioxide and water containing
choride ions as shown on Table 11. As such, it may be classified as a wet, sour, acidic
environment. It is possible that the high viscosity, low API crude oil may provide some
protection, but this cannot be confirmed until there is an opportunity to check the effects, if
any, of the produced fluids on the completion during the first workover. However, it is
important to understand the risks and the cost of mitigation. To do this requires an
understanding of the corrosion and hydrogen embrittlement processes as briefly presented
below.
Metal corrosion in sour gas environments is an electrochemical process, which consists of
anodes and cathodes, and can be divided into two simple reactions: oxidation and reduction.
Oxidation is the process that results in corrosion of the metal at the anode. The oxidation
and reduction reactions are dependant on each other. The behaviour of iron in acidic
solutions containing Hydrogen Sulphide or H2S is oxidation. The corrosion of steel in H2S, a
mixture of iron, carbon and other elements to form an alloy, is dependant upon the
concentration of H2S in solution, which in turn is dependant on the partial pressure of the
H2S and the solution pH. This is an inverse measure of the concentration of hydrogen ions:
the greater the quantity of hydrogen ions the lower the pH value and the more acidic the
solution.
Carbon Dioxide or CO2 dissolves in water to form a weak acid with a low pH value. A low
solution pH accelerates corrosion. The primary factors that affect CO2 corrosion are the
partial pressure of CO2, temperature and chloride content. The corrosion will take the form
of uniform surface corrosion, otherwise known as uniform weight-loss corrosion, and
localised pitting and crevice corrosion. Another commonly used expression is General
Corrosion, combination of uniform, crevice and pitting corrosion.
The combination of H2S, CO2 and chlorides in solution creates complex reactions that are
difficult to predict. Under certain environments, in addition to corrosion of the steel, there is
a susceptibility to premature failure by fracturing often below the yield strength of the steel
by a process known as hydrogen embrittlement. This process is the charging of the metal
by interstitial hydrogen atoms, causing a weakening of the crystal lattice structure of the
metal. Normally the reduction process at the cathode will result in hydrogen gas, H2, being
produced. However, H2S hinders the re-combination of the two hydrogen atoms and allows
them to enter the metal, since hydrogen atoms are smaller than the spacing between the
metal atoms. Hydrogen embrittlement cause by H2S is known as suphide stress cracking
(SSC). The principal factors that contribute to SSC are environment (H2S concentration,
pH, temperature and applied or residual stress) and materials (yield strength, microstructure,
heat treatment and alloying elements). In addition to SSC there is also a phenomenon
called stress corrosion cracking (SCC) which is brittle fracture of the steel from the combined
effects of corrosion and applied or residual stress. A third mode of failure common to steels
is chloride stress cracking (CSC) that affects some austenitic microstructure steels at
temperatures above 200F.
In summary, there are six common modes of steel failure that are interrelated to different
degrees:
The factors that contribute to these failure modes have been given above. The physical
properties of the steel can be controlled by proper materials selection; the environmental
PHYSICAL PROPERTY UNITS OSEIL- 1 (1) OSEIL- 2 (2) OSEIL- 2 (3) OSEIL4 (4) OSEIL 4 (5)
% CO2 % mol / ppm 1.14 / 11,400 7.45 / 74,500 2.25 / 22,500 6.71 / 67,100 2.82 / 28,200
% H2S % mol / ppm 0.01 / 100 0.64 / 6,400 0.29 / 2,900 0.66 / 6,600 0.15 / 1,500
All corrosion rates are based on theoretical calculations since no empirical data is
available for this field. Corrosion rates have been estimated using a Helix software
package, Helicorr. This uses 3 phase oil/water/gas corrosion and erosion prediction
models to predict corrosion and erosion tendencies at any specified location in the
production system.
Experience in the Ninian Field, UK Sector suggests that under similar conditions
(even lower CO2 and H2S partial pressures), carbon steel tubing in ESP wells can
corrode severely within 9 months. The expected life of a L80 13Cr completion is in
excess of 2 years assuming that the water composition is valid.
Sulphide stress cracking strongly depends on the pH and therefore on the quality of
the water analysis of the Oseil produced fluids, it is strongly advised to do further
detailed analysis on the water samples to decide whether L80 13Cr is acceptable for
all conditions.
Casing, whether flow exposed or not, can be carbon steel if a flow wetted period of
less than 10 years is expected (water break through is only expected in 4 years). In
future wells carbon steel with adequate H2S resistance should be selected. It should
also be noted that significant erosion/ corrosion has been seen in casings where
ESP pump intakes have been located. This is due to the effect of turbulence and the
restricted flow area between the casing and the pump. This can result in premature
collapse of the casing (e.g. Forties, UK Sector). It is recommended to install a
Finally, it should be noted, that if H2S levels in the gas phase should increase, for instance
due to a mechanism such as reservoir souring, a metallurgy review will be required.
Corrosion Rate Calculations: The de Waard & Milliams 1995 corrected formula for
predicting corrosion due to carbon dioxide was used with a range of CO2 concentrations
(max. 7.45 mol %) in the produced gas phase and using water cuts of 1%, 50% and 90%.
The following worst-case corrosion results were obtained:
Table 12
Carbon Steel Well Components
Corrosion Rate
Tubular Size
Mils/yr
3-1/2 Tubing 64
7 Casing 42
9-5/8 Casing 32
Corrosion Rate
Tubular Size
Mils/yr
3-1/2 Tubing
4.1
Sulphide stress cracking will occur in N80 carbon steels and 13Cr N80 CRA steels. L80
13Cr will be adequate under current producing conditions and theoretically calculated pHs.
There is a low risk of pitting and crevice corrosion for L80 13Cr under Oseil anticipated
conditions. It is recommended that downhole completion components, including pump
bodies should be L80 13Cr material.
Screening of Materials: Published manufacturers test data under similar conditions of
partial pressures of CO2 & H2S show that the steels containing 13% Chrome, commonly
referred to as 13Cr, 15Cr and also the 22 to 25Cr duplex steels are susceptible to SSC at
the temperatures present in the Oseil wells. Specifically, according to NACE MR-01-75, the
above mentioned materials cannot be used if the partial pressure (pP) H2S > 1.5 psi, and
with any amount of H2S if the water pH < 3.5. Because of the potential for catastrophic
failure of the tubing, these steels can be eliminated from further consideration.
Steels that are resistant to SSC with certain caveats as mentioned below are:
Low-carbon, low maximum yield stress steels (e.g. AISI 10XX to 41XX class or API
5CT Spec. grades H-40 to L-80) with Rockwell C hardness HRC < 22.
9Cr-1Mo low-alloy steel.
At the other end of the cost scale there are high Nickel-Chrome alloys such as Incoloy 825
and 925 or Inconel 718 that are resistant to SSC but cost about 11 times more than low-
carbon steels. The low-carbon, low maximum yield stress steels and low-alloy 9Cr-1Mo are
generally immune to SSC if the water content pH is greater than 3.5 in the temperature
range 150-212 deg. F and the Chloride ion concentration is less than 2% (i.e. 20,000 ppm).
These conditions are met by the downhole conditions on all three Oseil wells. However,
these steels will suffer badly from corrosion; how badly is difficult to predict until the first
completion is recovered from a well. Empirical data from sour oil wells indicates that the
corrosion rate will be initially slow because of the oil film on the steel, but that as water cut
increases above 30 40% the steel becomes water-wet and the corrosion rate will increase
to the rates as shown above.
Service
Completion Brines/ Produced
Environm ent
Drilling Muds only Fluids
Has corrosion
resistant alloy Yes Consider using CRA's
production tbg
been specified
No
Yes
Consideration can be given
o o
to using N80 (Q+T) or C95 (150 F) (175 F)
What is the
o o
in addition to the standard >65 C minimum service >80 C Consideration can be given to using H40, N80, or
temperature
sour-resistant casing P110 grades in addition to the standard sour-resistant
materials in API 5CT and casing materials in API 5CT and NACE MR-175-91
NACE MR-0175-91
o
(150 F)
o
<65 C
Use only the standard
sour-resistant casing What is the
materials in API 5CT and <95 ksi required strength >95 ksi Consideration to be given to using proprietary sour-
grade
NACE MR-0175-91 resistant casing grades with up to 110 ksi YS
s (Refer to Table 3 of NACE - consult with a relevant specialists
MR-0175-91 "For All
Temperatures" Column)
Has a suitable
combination of Yes Specify materials according to (API 5CT)
strength and SSC together with (NACE MR-0175-91
resistance been
achieved
Tubing Life and Costs: the expected life of the tubing, string in the wells can be calculated
using autoclave derived corrosion rates for various materials or corrosion rates calculated
using one of many mathematical models. The de Waard & Milliams formula for predicting
corrosion due to carbon dioxide was used to calculate the general corrosion rates for both a
low-carbon steel and API 13% Chrome steel under the conditions determined from Oseil-2
and Oseil-4 well test results. Three different weights of 3-1/2 OD carbon steel and 13Cr
L80 tubing were analysed for their resistance to failure at the calculated corrosion rates.
The axial forces and tri-axial stresses on the tubing were calculated using Bakers
Tubemove software under the conditions of shut-in and production. These forces and
stresses with appropriate safety factors have been used as the maximum likely loadings on
the tubing. The corrosion rates were used to calculate the reduction in wall thickness and
thus remaining strength of the tubing, and these results were compared with the loadings
from the Tubemove program to estimate the time to failure for three different failure modes:
The calculations of tubing life to failure used the shut-in load as the worst condition likely to
be encountered during the middle to late life of the tubing when the tubing has undergone
significant corrosion. The time to failure is determined by calculation to be when the
remaining strength in the tubing is equal to the maximum applied load at shut-in multiplied by
a 1.6 safety factor. The summary of these calculations is shown in Table 13.
Table 13 Calculated Tubing Life for L-80 Carbon Steel & API 13 Cr L-80
From the above table it is apparent that general corrosion resulting in a loss of strength will
be the mode of failure for both L-80 carbon steel and API 13 Cr L-80 steel, and as would be
expected, larger wall thickness found in the higher weight tubing results in a longer life
before failure.
The cost of the workover services and lost production must be added to the costs of tubing
to arrive at the total cost of the workovers over a 10-year field life:
The higher price of 3-1/2, 12.95 ppf, L-80 tubing is easily offset by the real cost to replace
the 3-1/2, 9.3, L-80 more frequently during the well life. Should a well fail unexpectedly, the
shut-in lost time could be about 30 days to allow time to mobilise the Kalrez rig and required
support services. This favours a tubing string that would not fail during the expected life of
the well. Since the table above is based on the tubing life impact from CO2 corrosion, the
risk of other forms of corrosion and catastrophic failure are still present. As already stated
above the only materials which would meet this criteria are the exotic high chrome-high
nickel CRA steels such as Sumitomos SM2535 at eleven times the cost per meter than L-80
carbon steel. At $2.9 million for 15 x 8,000 ft strings of SM2535, the economics of this once
only cost compares very favourably with the potential cost of replacing corroded API low-
carbon steel tubing.
However there is another important factor to be considered is the completion design. The
initial choice of artificial lift system was Electric Submersible Pumps or ESPs. These pumps
can provide the output to match the expected well performance i.e. ~6,000 bopd and ESPs
can deliver higher flow rates of total liquids to maintain the oil rate should the water cut
increase significantly, albeit by replacing the ESP with a higher output unit.
UNFORTUNATELY, ESPs suffer from a number of failure modes including power cable and
cable splice failure (the most common). The frequency of ESP repair/replacement
workovers is difficult to predict but the experience of Indonesian operators such as Maxus
S.E.S., Santa-Fe/Devon in Irian Jaya and East Java and Kondur (formerly Hudbay and
Lasmo Malacca Straits) who have used ESPs extensively over the past decade would
suggest that 6 months might be optimistic.
If the life of wells between workovers were reduced to as little as 6 months, it would require
pulling and re-running the CRA tubing many times. These materials suffer from a common
weakness: they are easy to damage through poor handling and from galling of the threaded
connections simply by backing out and re-making up. Low-carbon steels are considerably
more robust in these respects and the cost impact of damage is significantly less. A cost
comparison between these two possible tubing materials for the Oseil wells must take into
consideration a back-up string of the CRA material to allow re-cutting connections and
handling damage to be repaired.
It is likely that the time to failure of the low-carbon steel tubing due to the worst form of
corrosion, pitting, would still be greater than the likely time to failure of the ESP. However, it
would be prudent to replace the partially corroded low-carbon steel tubing string at this time
rather than wait until the tubing has corroded to a point when failure is imminent. This
analysis then favours the 3-1/2 9.3 ppf, J-55, EUE API 8RD Carbon Steel tubing. The 7
casing exposed below the ESP packer should be made from L80 13 Cr steel or internally
plastic coated to provide some resistance to corrosion.
Directional Drilling
The 16 surface hole will be rotary drilled vertically to 160 ft RKB . A Totco survey will be
obtained once TD has been reached.
The 12.1/4 surface hole will be drilled vertically to 2,500 ft RKB. A PDM motor and MWD
BHA will be used to control direction and angle due to the tough drilling and high dip
formation angles found ranging from 20-30 degrees in the Kanikeh Formation.
The 8.1/2 hole section will be directionally drilled using a steerable PDM motor and MWD
BHA . The well profile is designed to achieve lateral displacements ranging from 500-4,000
ft.. Kick-off depths will be initiated below the 9.5/8 casing with angle and direction built up
Drilling Fluid
The 16 surface hole will be drilled using water in conjunction with high-viscous slugs (80-
100 cp) composed of pre-hydrated bentonite to ensure proper hole cleaning. The hole will
be displaced to gel mud prior to running the 13.3/8 casing.
The 12.1/4 and 8.1/2 hole sections will be drilled overbalance using a water base inhibitive
polymer based mud system at 9-11 ppg mud weight (average fracture gradients observed in
both Oseil-2 and 4 areas is 14.0-14.3 ppg EMW) in a closed circulating system. The mud
system will be designed to handle sloughing / swelling of the reactive clays found throughout
the Imbricate Zone (interbedded siltstone, sandstone, claystone and limestone) and partially
reactive and techtonically stressed shales in the Kola Formation.
Cementing
The 16 conductor hole X 13.3/8 conductor casing will be cemented to surface using a 5
drill pipe cement stinger stabbed into the 13.3/8 float shoe with Class G Cement + 3%
Accelerator at 14.5 ppg weight.
The 12.1/4 surface hole X 9.5/8 surface casing will be cemented to surface using dual plug
loaded cement head based upon following criteria:
Lead: Class G Cement + Extender at 13.5 ppg
Tail: Class G Cement + Accelerator at 16.0 ppg
For on-site cement volume requirements, a 50% excess applied to the hole volume will be
used. The cement will be displaced using water plus mud.
The 8.1/2 intermediate hole X 7 intermediate casing will be cemented to 2000 ft RKB using
dual plug loaded cement head based upon following criteria:
Lead: Class G Cement + Extender + Fluid loss at 13.5 ppg
Tail: Class G Cement + Fluid Loss at 16.0 ppg
The top of the cement will be extended 500 ft inside the 9.5/8 X 7 annulus and include a
anti-corrosion plus bacteria pre-flush.
The cement will be displaced using water plus mud.
If losses occur during drilling a two-stage cement job may be considered using a DV collar.
Bit Selection
16 Conductor Hole - Rotary Drilling (surface 160 ft RKB)
Formation Kanikeh Interbedded siltstone, claystone, and sandstone
Maximize bit hydraulics to increase ROP and proper hole cleaning.
Bit Type: Milled tooth centre jet - Y13C (IADC - 1-3-1)
PDC - S91 (IADC S123)
12.1/4 Surface Hole PDM motor + MWD BHA (160 2,500 ft RKB)
Formation Kanikeh and Imbricate Interbedded siltstone, claystone, sandstone and
limestone (coal stringers)
Maximize bit hydraulics to increase ROP while controlling direction and angle.
Bit Type: PDC - S91 (IADC S123)
Milled tooth centre jet - Y13C (IADC - 1-3-1)
Electric Logging
No open-hole electric logs are proposed.
Mud Logging
Gas detection sensors (degasser, total gas, chromatograph, hydrogen sulphide H2S,
cuttings gas, and carbon dioxide CO2) will be run, monitored and recorded continuously
from surface to TD. A mud and drilling real time monitoring system will be run continuously
and include well site parameter monitoring, trip monitoring, pore and fracture pressure
gradient as well as kick and kill monitoring. Daily reporting including lithology log, formation
pressure log, drilling parameter log and chromatograph ratio logs will also be provided.
Collection of cuttings samples will be from surface to TD.
Solids Control
To provide maximum Solids Removal Efficiency, the following should be adhered to:
The shale shaker must be capable of handling the entire drilling pump rate and allow the use
of 180-200 mesh screens. Screen up as flow rate dictates.
The surface fluid-transfer system (shaker box/tank and overflow lines) must be capably of
handling high flow rates up to 1000 gpm.
The twin centrifuge system should be able to process 100 to 120-gpm of whole mud while
providing a discharge of 0.75 to 1.25 gpm.. Recovery of barite and maintaining LGS less
than 5% is required.
A planned dilution schedule may be implemented.
QHSE
Safety is a main concern. No operation is so important or urgent that it cannot be
done safely.
Health and Safety are of the utmost importance during all phases of the operation. All
operational personnel must ensure that they are fully conversant with the relevant HSE
regulations.
Particular attention must also be given to rig specific operating procedures, which may
highlight additional HSE requirements.
The following will be implemented during the drilling operations:
All incidents, accidents and near misses must be reported and investigated.
Abandon rig drills and safety meetings must be held prior to spud.
Hold a pre-spud meeting and ensure that all personnel on location attend.
Safety meetings should be held on a weekly basis and include all on/off duty personnel.
A safety meeting should be held prior to any hazardous/key operation i.e. casing running
etc.
Ensure shift handovers are effective and information on status is transferred.
Ensure all personnel on site use the proper Personnel Protective Equipment (PPE). If in
doubt on anything, ask your direct Supervisor, good communication is encouraged.
Underbalanced Drilling
The reservoir section in well Oseil-1 was drilled conventionally over-balanced using mud.
Large volumes of mud, water, cement and lost circulation materials were lost to the
formation and the results of the drill stem tests had strong indications of formation damage.
The total cost in materials and lost time spent fighting the lost circulation was about US$1.5
million. Apart from the damage to the reservoir and was a serious risk of losing well control
and stuck pipe. The appraisal wells Oseil-2 and 4 were drilled using under-balanced drilling
techniques whereby the hydrostatic pressure provided by the drilling fluid is less than the
reservoir pressure. This results in formation fluids flowing to surface in a controlled manner
by means of adjusting the degree of under-balanced pressure and annulus backpressure.
This method has the advantages and disadvantages as follows:
Advantages: Non-damaging, able to drill reservoir effectively, closed circulating
system, real-time formation PI determination.
Disadvantages: Expensive, inherently hazardous, large equipment & number of
personnel
Because of the high cost of underbalanced drilling a number of alternatives have been
reviewed for drilling the reservoir section in the Oseil Phase 2 Development. These are
briefly presented below:
One of the unusual features of the Oseil reservoir is the frequency and distribution of
fracture zones (see Geological Section for further details). These fractures have recorded
permeabilities in the range of 1 to 30 Darcies and the Productivity Index of the fractures is in
the range of 50 to 700 bbl/psi. This presents a problem when under-balanced drilling
because of the high inflow from the reservoir at very low-pressure draw down. If only one
fracture zone is encountered, further underbalanced drilling through the reservoir is
technically feasible whilst maintaining adequate hydraulic power and circulation through the
drilling motor and bit and sufficient annular velocity to circulate rock cutting from the well
bore. However, if a second or third high permeability fracture is encountered the effect is to
lose the ability to drill underbalanced at the bit because of the frictional back-pressure
imposed on the system by the inflow into the annulus, and the very likely possibility that
cross-flow from one fracture zone to the next will occur.
UBD Flow Modelling was performed using Neotechs WellFlo 7 Software and the actual
drilling parameters from Oseil-2ST to determine the feasibility of UBD technology to drill a
1,500ft horizontal lateral section.
Minimum annulus velocity required for hole cleaning > 200 ft/min
Wellhead annulus pressure assumed constant @ 50 psi
Increased annulus backpressure required to reduce draw down & oil inflow because
of additional point sources & increasing PI (see fig. 13).
Difficult to maintain underbalanced conditions at bit once fractures are penetrated
with high PIs
Slight changes in parameters causes overbalanced conditions at bit
Annulus flow slugging annulus velocity erratic difficult to achieve stable
conditions (see fig. 12 below)
High pressure booster will be required as N2 injection pressure likely to exceed
rating of standard medium booster ~ 1800 psi
Standard 1500 scf/min capacity Nitrogen Membrane Unit sufficient
800
Case 4
Annular Liquid Velocity (ft/min)
4 1/2 inch DP to
600 3 1/2 inch DP X-over
and casing shoe BHA
Case 2
500
400
No Production
300
Case 3
4 5/16 inch DP to
4 1/2 inch DP X-over
200
0 1000 2000 3000 4000 5000 6000 7000 8000 9000
Measured Depth (ft)
Figure 13 Oseil Inflow Performance for Point Source @ 8440 ft MD (6800 ft TVD)
2410
PI=200
PI=1.3 PI=150
P:\Kufpec Indonesia 2003\VDR-CD\Drilling\Text_for_RBA_Drilling_&_Completions .doc
PI=100
2390Simon
Shaw Page 38 12/8/2003
PI=50
KUFPEC (Indonesia) Ltd.
Drilling & Completion Programme for RBA Document
Directional Drilling
The 6 hole section will be drilled below the 7 production casing through the remaining Kola
Shale Formation whilst building up the hole angle until 90 degrees horizontal is achieved as
close as possible to the top of the Manusela Fm. using a PDC drill bit with a steerable
positive displacement motor (PDM) and MWD/LWD. Once the horizontal plane and desired
azimuth (direction) is achieved drilling will be continued to a minimum of +/- 350 ft. lateral
with the primary objective to penetrate the anticipated fracture network systems while
maintaining a horizontal plane and direction. As mentioned in the section on UBD above the
challenge will be to penetrate a sufficient lateral section and maintain adequate hole
cleaning, well bore stability and well pressure control. Once two-phase fluids are introduced
down the drill string with an equivalent liquid volume greater than 22-24% (the measure of
nitrogen-to-liquid ratio) normal positive pulse type MWD will not function. Therefore, it is
planned to use Electro-Magnetic MWD. This system transmits the data gathered at the bit
i.e. hole angle, azimuth, magnetic & gravity tool face, tool temperature, drill pipe & annulus
pressure, gamma ray. Dependant upon the resistivity of the overburden, an extended range
antenna may be required for adequate transmission of the electro-magnetic waves to the
receiver at surface.
Drilling Fluid
The 6 hole section will be drilled from the Kola Shale to within the top of the objective
reservoir Manusela Carbonate using the water based inhibitive polymer mud system and
once significant fractures have been encountered changed over to a nitrified-diesel/crude oil
drilling fluid by means of under-balanced drilling (UBD). Returns to surface will be achieved
and maintained as a result of under balance conditions until either an extensive fracture
system and or multiple fractures are drilled through causing loss of complete circulation
(under balance becoming overbalance with possible cross flow interference of fracture
systems). If this event occurs, drilling will be continued but the circulating fluid and
technique will be slightly changed. A mud cap drilling system (MCD) using nitrified water
instead of crude oil as the main drilling fluid and either nitrogen and or water will be pumped
down the annulus to maintain a mud cap to prevent H2S gas from flowing to surface. The
extent of the horizontal displacement using this modified UBD-MCD drilling technique will be
a function of the fracture(s) system permeability and differential pressure(s) and the ability to
take in the drilling fluid and solids generated without causing hole instability.
Cementing
No cementing requirements are needed for the 6 horizontal hole section.
Bit Selection
6 Production Hole Steerable PDM motor + MWD/LWD BHA
Formation - Kola Shale and Manusela Carbonate Shale, dolomite and limestone
Bit Type: PDC bit - S94PX (recommended)
Button - EHP-51/EHP-53
Electric Logging
No open-hole electric logs are proposed.
Mud Logging
Same operational requirements as carried out in the previous hole sections.
Solids Control
High Capacity Centrifuge: A centrifuge will be the key component to treat and isolate the
solids build up in the under balance drilling UBD circulating fluid. The centrifuge will be
rigged-up to the closed-loop surface circulation system to process the diesel/crude oil drilling
fluid. The high capacity centrifuge is to be complete with an adjustable feed pump and be
able to process in the order of 200 gpm.
Cuttings Drying Equipment: The cuttings from the UBD drilling fluid will have to be dried to
reduce the oil content to less then 80 gm/kg. The recovered oil will then be recycled back
into the closed loop circulating system. The unitised equipment arrangement consists of a
minimum of 2-10 de-sanders and 20-4 de-silters with the discharge flow to a shaker bed.
The shaker bed or high G force dryer is capable of producing a continual 6 G force and
utilizing pyramid type screen designs. The drying system will be closed with an exhaust
arrangement to vent hydrogen sulphide gas through a scrubbing system.
5000
Lead:
Cmnt Class 'G' + Extender Drill 8.1/2"
Intermediate Csg., Directional Hole &
+ Fluid Loss 13.5 ppg
7" , 26 lb/ft, L-80,
Tail:
6000 Set 7" Intermediate Csg.
BTC, Drill into Kola shale
Cmnt Class 'G' + Fluid NU UBD/RBOP & Test
8,000 ft RKB MD and then set 7" Csg
Loss 16 ppg
PDC drillable float eqpt Low er NIEF into the Kola shale
TOC ~ 500ft above 9.5/8"
shoe
KOLA Shale 7000
H 2 S gas
The well time and cost estimates were calculated using the technical limit time estimates as
a target base case with variable contingencies added for the expected learning curve,
measured depth and spud date factors as referred to in section 4. The cost model used the
actual prices for goods and services obtained during the 2002 Oseil Phase 1 tender process.
The actual prices from the forthcoming tenders for Phase 2 are expected to be lower
because of the economies-of-scale benefits derived from a much larger programme for
Phase 2 Development as compared to Phase 1. In addition, lower rates are likely from the
proposed changes to the GSA and Exhibits. These changes place a greater degree of the
manageable risks onto KUFPEC rather than the Contractors. This apparent over-estimation
of times and costs is considered to be off-set by the uncertainties with the drilling and
completion programme such as contractor performance, geological variations and market
forces e.g. the effects of oil price and regional drilling activity influence on prices. The
accuracy of these costs are to be considered as Order-of-Magnitude i.e. in the range 15%
to + 15%.
The following assumptions were used when developing the well cost model:
Initial mobilisation and rig up time and final demobilisation time is 15 days each.
The drilling rig is used to drill and complete wells Oseil-3, 5, 6, 7 and 8 then proceeds
to drill wells Oseil-9, 10, 11, 12 and 13 (Oseil Tenggara A appraisal well) all from the
Oseil-4 multi-well pad. The drilling rig is then moved to the Oseil-2 multi-well pad to
drill wells Oseil-14 and 15. The rig is moved to a new well site at Nief Utara A
location and after drilling this exploration well will be moved to a new multi-well pad to
drill and complete three Oseil Tenggara development wells.
Drilling Rig moves between cellars on the same pad require 2 days.
All wells, with the exception of Oseil-3 and Nief Utara will have a 500 ft long
horizontal section in the reservoir.
A completions rig will be used to batch complete wells Oseil-9, 10, 11, 12 and 13
followed by Oseil-14 and 15.
Completion Rig moves between cellars on the same pad requires 1 day.
No well site costs have been included in any of the well cost estimates. Well sites
and access roads will be included in the Opex costs.
The well design is given in this appendix: 13-3/8 conductor set at 165ft, 9-5/8
surface casing set at 2500ft, 7 production casing set at between 6500 to 8250ft.
ESP set at 4,000 ft MD with gas compressor, 4-sensor pump instrument., carbon-
steel materials, soft-start ESP controller.
ESP Packer set at 3800 ft and 3-1/2 low-carbon steel tubing.
Flanged gate valve Xmas Tree with 4 valves and manual choke, Trim FF, 3,000 psi
WP.
13-3/8 Speedlock / Unihead 3,000 psi well head.
Table 14 presented below includes the exploration well Nief Utara. This well is included to
meet KILs Seram Non-Bula PSC exploration commitment in the event that the Oseil
Tenggara 'A' well is successful and is classified as an Oseil Field appraisal well for cost
recovery purposes. All costs were calculated using real US Dollars as of 1 Jan 03 and
are Gross amounts (not KUFPEC net costs).
Table 14 Oseil Phase 2 Drilling & Completion Cost & Time Estimates
Total Time
WELL WELL PAD & TARGET Cumulative Total
REMARKS TOTAL US $ per Well
NO. LOCATION US $
Days
All purchasing and tenders will be conducted according to KUFPEC procedures and the
Indonesian Government / BP MIGAS Regulation BP 077 Rev III. Briefly, a minimum of three
valid quotes will be obtained wherever possible and the award will be based on the lowest
price of the technically qualified quotes. In the event that less than 3 quotes are obtained, a
re-tender will be conducted unless there are valid reasons to believe that a re-tender will not
result in obtaining the desired number of quotes for evaluation. The system of tendering will
invariably be using the 2-envelope method whereby the administrative and technical parts
will be submitted in one envelope and the commercial part in a second envelope at the same
time. The evaluation of the administrative and technical parts of the tenders will be
conducted first and only the qualified bidders commercial quotes will be considered for the
subsequent commercial evaluation.
According to the KUFPEC process, tendering may only commence after approval of the AFE
related to the goods and services required. This RBA will address the need for a waiver of
this requirement to order to commence the often-lengthy tender process after approval of the
RBA, not after approval of the internal AFE. However, in order to maintain sufficient control
according to the established delegation of authority, the award of contracts (i.e. commitment)
will not be made until approval of the AFE at the appropriate level. The same conditions
also apply under the Indonesian government and BP MIGAS system. In brief, the Oseil
Phase 2 Plan of Development (POD) is submitted for approval by BP MIGAS. Once the
POD is approved the Work Programme and Budget can be revised and submitted for
approval, and then in turn well AFEs are submitted for approval. Award of contracts is
predicated upon prior approval of sufficient funding in the AFE. However, in this case
dispensation must be obtained from BP MIGAS to award contracts prior to formal approval
of the AFEs. If this dispensation is not granted and award of contracts is delayed until the
Work Programme and Budget approval and AFE approval is obtained, a delay of 3 to 5
months in the commencement of operations will ensue.
The schedule shown in Fig. 4 is based on the assumption that dispensation from KUFPEC
to commence the tender process after approval of the Company AFEs is obtained and also
that BP MIGAS allows KUFPEC to award contracts at its sole risk, subject to later audit
approval, prior to approval by BP MIGAS of the AFEs.
It has been conservatively estimated that the cost-benefit to be realised by tendering for
goods and services for a full 16 well programme is in the order of US$250,000 to 500,000
per well.
DESCRIPTION DRY HOLE COMPLETION TOTAL PRIOR YEARS CURRENT YEAR TOTAL UNDER BUDGET UNDER BUDGET
1 2 3 4 5 6 7 8
1 TANGIBLE COSTS
2 CASING 305,919 0 305,919
3 CASING ACCESSORIES 15,634 0 15,634
4 TUBING 0 33,400 33,400
5 WELL EQUIPMENT - SURFACE 59,417 71,581 130,998
6 WELL EQUIPMENT - SUBSURFACE 0 295,893 295,893
7 OTHER TANGIBLE COSTS 0 0 0
8 0 0 0
9 TOTAL TANGIBLE COSTS 380,970 400,874 781,843
10 INTANGIBLE COSTS 0 0 0
11 PREPARATION AND TERMINATION 0 0 0
12 *SURVEYS 0 0 0
13 *LOCATION STAKING AND PREPARATION 0 0 0
14 *WELLSITE AND ACCESS ROAD PREPARATION 0 0 0
15 *SERVICE LINES & COMMUNICATIONS 0 0 0
16 *WATER SYSTEMS 0 0 0
17 *RIGGING UP / RIGGING DOWN 3,906 40,000 43,906
18 * 0 0 0
19 SUBTOTAL 3,906 40,000 43,906
20 DRILLING / WORKOVER OPERATIONS 0 0 0
21 *CONTRACT RIG 830,015 51,579 881,593
22 *DRILLING RIG CREW / CONTRACT RIG CREW 31,775 4,515 36,290
23 *MUD, CHEMICAL & ENGINEERING SERVICES 188,537 0 188,537
24 *WATER 0 0 0
25 *BITS, REAMERS AND CORE HEADS 170,592 3,675 174,267
26 *EQUIPMENT RENTALS 1,014,179 84,638 1,098,817
27 *DIRECTIONAL DRILLING AND SURVEYS 1,224,217 2,518 1,226,735
28 *DIVING SERVICES 0 0 0
29 *CASING INSTALLATION 75,236 0 75,236
30 *CEMENT, CEMENTING AND PUMP FEES 73,277 12,046 85,323
31 * 13,052 3,032 16,084
32 SUBTOTAL 3,620,880 162,002 3,782,882
33 FORMATION EVALUATION 0 0 0
34 *WATER (CORING) 0 0 0
35 *MUD LOGGING SERVICES 34,241 7,150 41,391
36 *DRILLSTEM TESTS 0 0 0
37 *OPEN HOLE ELECTRICAL LOGGING SERVICES 145,492 51,661 197,152
38 * 0 0 0
39 SUBTOTAL 179,733 58,811 238,543
40 COMPLETION 0 0 0
41 *CASING LINER AND TUBING INSTALLATION 34,373 19,666 54,039
42 *CEMENT, CEMENTING AND PUMP FEES 0 0 0
43 *CASED HOLE ELECTRICAL LOGGING SERVICES 0 0 0
44 *PERFORATING AND WIRELINE SERVICES 12,873 0 12,873
45 *STIMULATION TREATMENT 0 0 0
46 *PRODUCTION TESTS 0 0 0
47 SUBTOTAL 0 0 0
48 47,246 19,666 66,912
49 GENERAL 0 0 0
50 *SUPERVISION 162,183 34,544 196,727
51 *INSURANCE 0 0 0
52 *PERMITS AND FEES 5,000 0 5,000
53 *MARINE RENTAL AND CHARTERS 58,459 7,469 65,928
54 *HELICOPTERS AND AVIATION CHARGES 33,430 7,695 41,125
55 *LAND TRANSPORTATION 15,350 15,350 30,700
56 *OTHER TRANSPORTATION 1,116 0 1,116
57 *FUEL AND LUBRICANTS 83,274 12,895 96,168
58 *CAMP FACILITIES 65,560 18,487 84,047
59 * ALLOCATED OVERHEADS - FIELD OFFICE 10,661 1,117 11,778
60 * - JAKARTA OFFICE 0 0 0
61 * - OVERSEAS 0 0 0
62 *TECHNICAL SERVICES FROM ABROAD 0 0 0
63 * 0 0 0
64 SUBTOTAL 435,034 97,556 532,590
65 TOTAL INTANGIBLE COSTS 4,286,798 378,035 4,664,833
66 TOTAL COSTS 4,667,767 778,908 5,446,676
References 1
7. Drilling Time Analysis: A Total Quality Management Tool for Drilling in the
1990s A.G. Kadaster et al SPE 24559
10. Integrated Borehole Stability Studies: Key to Drilling at the Technical Limit
and Trouble Free Cost Reduction E.van Oort et al SPE/IADC 67763
12. Oseil Field Completion Design J.C. Mantecon, Helix Well Technologies,
KUFPEC 2035/002
13. Oseil Field Artificial Lift System Selection J.C. Mantecon, Helix Well
Technologies, KUFPEC 2035/001
14. Oseil Metallurgy Selection Review Heiko Morgenroth Helix Well Technologies
PSD427
15. Increasing the Run Life of ESPs in high H2S Wells, B.L. Wilson, T.P. Comeau,
Mobil, SPE 28527,