This preview shows page 1. Sign up to view the full content.
Unformatted text preview: SPE 137319
Deploying the World’s First Commercial Dual Gradient Drilling System
J. David Dowell/Chevron Copyright 2010, Society of Petroleum Engineers
This paper was prepared for presentation at the SPE Deepwater Drilling and Completions Conference held in Galveston, Texas, USA, 5–6 October 2010.
This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been
reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its
officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to
reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright. Abstract
Chevron is working with AGR Subsea, Pacific Drilling, GE Oil and Gas, and others to deploy the world’s first commercial
Dual Gradient Drilling (DGD) System. The DGD system will be installed on a Pacific Drilling Samsung 12,000 Class rig
currently being built in Korea. It is expected to begin operations in the Gulf of Mexico in 4th Qtr 2011. This paper will
detail the design, fabrication, testing and preparations that are being made for that deployment.
Not only is significantly new hardware being added to the rig, but a new set of processes and concepts are being
embedded in the drilling organization to take full advantage of the hardware. This has required the simultaneous development
of two different, but linked, work streams: hardware and personnel. The “hardware” side of the work has concentrated on the
integration of completely new equipment into the rig to make it DGD ready. This includes a subsea pump, a completely new
riser, several riser “specialty” joints and all the rig modifications required to handle, run and support this new drilling system.
The “people” work stream includes the development of drilling and well control procedures, well planning and the extensive
personnel training needed to realize all the benefits of the technology.
This effort has impacted every facet of the Chevron deepwater drilling community. Chevron has reexamined the way rigs
and drilling assets have historically been deployed. This has resulted in the development of a “Value Driven Drilling
Schedule” to optimize utilization of the technology.
The DGD system could revolutionize how deepwater wells are drilled and how deepwater assets are explored and
This technology’s origins can be traced back to a Joint Industry Project (JIP) that was started in the mid 1990s. The industry
was moving to deep water and many drilling problems caused by the long column of mud in the riser were becoming evident.
Several JIPs were started at this time to investigate removing the mud from the riser and replacing it with the equivalent of
seawater. This would allow the deepwater wells to see the same pressure gradients that existed as the formations were
deposited and would solve many drilling problems. These methods were generally referred to as dual gradient solutions
because of the two different gradients maintained in the well: one in the riser and the other in the well below the mudline.
One of these JIPs was the “SubSea MudLift Drilling” project (Fig. 1). This JIP solved the dual gradient dilemma by
using positive displacement pumps at the seabed to pump the well mud to the surface through a dedicated line, separate from
the annulus of the riser. This JIP culminated in a successful field test of the equipment by drilling a well in the Gulf of
Mexico (GOM) in late 2001.
Numerous SPE papers (and other industry publications) detail the Subsea MudLift JIP progress and the results of the field
test. These papers can be used to gain a clear understanding of the efforts expended on what was the industry’s largest JIP at
this time (Smith et al 2001; Schumacher et al 2001; Eggemeyer et al 2001).
Although the Subsea MudLift JIP was a technical success and more than 90% of the project objectives were completed, it
was a commercial failure. Several factors contributed to this commercial failure:
1. There was a downturn in the economy (and hence the industry) when the JIP ended in December of 2001. The
actual field test occurred in September 2001 and was ongoing during an economically turbulent period.
2. The needed rig modifications and equipment purchase were too cost prohibitive for commercial use.
3. The largest hindrance was that no single operator had the deepwater portfolio of prospects or developments required
to support the financial and resource commitment needed to make the dual gradient system a reality.
Forward to today – the deepwater landscape has completely changed with many discoveries and developments in the GOM. 2 SPE 137319 Deployment Concepts
In late 2006, Chevron’s Deepwater Drilling organization began a systematic examination of its operations in deep water,
determined to improve safety, predictability and economics. Many advancements had been made in tubulars, rigs, and
methods but no large step-change technology had risen to the task. A small group of original JIP participants within
Chevron proposed revisiting the DGD landscape to determine if the technology now made economic sense.
technical/commercial/practical review of the technology was commissioned and by mid-2007, several efforts were underway
to detail the economics and some of the more daunting technical tasks.
A review of the potential pumping systems available in the industry revealed that the original concept MudLiftTM pumps
(MLP) were still the most technically qualified pumps for the job. With the pumps chosen, the bigger issue of how to deploy
them became the focus of an intense study.
Chevron commissioned two different companies to conduct detailed reviews of deployment options and their effects on
the riser configuration, operational costs and overall practicability. The following options were evaluated (Fig. 2):
1. Option 1A was a single riser with the MLP run in-line with the riser.
2. Option 1B was a single riser with the MLP set on a stump beside the well and connected to the well/riser through a
series of jumpers (suction/power fluid/mud return).
3. Option 2A was a two-riser option with the MLP hanging beside the well on a stump guide. The MLP is connected
back to the rig with a separate parallel riser carrying seawater power fluid and mud return lines. This riser is hard
hung-off at the vessel. The connection to the well was a hinged suction jumper from the well.
4. Option 2B resembled Option 2A but with a compensation system at the rig for riser hang-off. The MLP was
connected back to the drilling riser with a flexible suction hose.
5. Option 3A had the MLP set on a stump at the mudline and the seawater power fluid and mud return lines as separate
stand alone risers with jumpers to the rig.
6. Option 3B had the MLP set on a stump at the mudline and the seawater power fluid and mud return lines combined
in a single stand alone riser with jumpers to the rig.
The two contracted companies conducted detailed riser analyses on each option that included the following modes:
• Normal operations
• Standby/Disconnect/Hang-off events
• Extreme conditions
This included all the running and retrieval issues, static and dynamic loading due to things like vortex-induced vibrations
(VIV), clashing, hurricane issues, etc. Each option was then compared to the others based on long list of appraisal categories:
• The technical feasibility of the option and the amount of developmental effort required to make it a reality.
• The initial cost of the system (CAPEX)
• The operational cost of the system each year (OPEX)
• Amount and type of rig modifications required.
• Impact on the normal drilling routine
In the end, Option 1A was determined to be the most feasible with optimal safety and operational outcomes and with the
lowest long-term cost. At this point the economic studies were showing the technology to be cost effective, so Chevron’s
deepwater management approved pressing ahead with the technology.
During the deployment studies, it became apparent that numerous rig modifications would be required to make the DGD
system efficient. Since deepwater rig scheduling indicated the need for an additional rig, the decision was made to issue a
Request for Proposal (RFP) to the drilling rig contractors. The RFP was issued with detailed specifications asking a large
number of rig contractors for a proposal on supplying a “DGD-ready” rig.
Many different responses were received and after much analysis and deliberation, a contractor was chosen. Pacific
Drilling was chosen to supply the “DGD-ready” rig for several reasons:
• Pacific Drilling is a new drilling company and enthusiastic about being identified with such a differentiating
• They demonstrated eagerness to become part of the “Team” that was being established to bring DGD to reality.
• A new-build rig was proposed that fit the time schedule and offered the chance of shipyard modifications.
The Pacific Santa Ana is being built and modified to be the first DGD vessel. It is the fourth vessel constructed for
Pacific in Samsung Heavy Industry’s yards in Korea (Fig. 3). The Pacific Santa Ana is a dual derrick rig and a Samsung
12,000 class vessel. It is due to arrive in the GOM in October of 2011 which fits well with the DGD equipment schedule. SPE 137319 3 Work Streams
Operating this specialized equipment is not a direct part of Chevron’s core business and does not fit in Chevron’s typical
business models. As the project moved forward, Chevron’s deepwater management realized Chevron would need to drive
this deployment process to make it a success but needed a partner company in the role of supporting the equipment once it
Numerous companies were considered. Chevron chose AGR Subsea as its partner because AGR had already established
itself as a high quality service provider of a top-hole DGD system which was being marketed and used throughout the world.
The Riserless Mud Recovery system (RMR®) uses electrical disk pumps to capture the top-hole mud at the wellhead and
return it to the vessel so it can be used again. This eliminates the need for dynamic kill drilling (DKD) fluid that is
commonly used in starting deepwater wells.
AGR Subsea agreed to partner with Chevron to deploy the DGD system on the Pacific Santa Ana in the GOM and help
make the DGD system a commercial success. As such they have become an integral part of the project team. Chevron and
AGR project personnel are co-located at the same project facility.
With a rig selected, the deployment option finalized, and the project team gathered, project work is now proceeding in
two major areas: equipment and people preparation.
The equipment work stream is like any other hardware-based project: design and fabricate the equipment, test the equipment,
make it reliable, make sure it interfaces with the other equipment and make it do its job. The DGD project is no different.
The major task is making all those things happen on a drilling rig in a marine environment at 10,000-ft water depth.
The DGD unique equipment components are described in the following sections.
Drill String Valve (DSV)
The drill string valve prevents the drillpipe from u-tubing into the well when circulation is stopped (Fig. 4). With a DSV the
drillpipe is full during connections and it makes most of the drilling process “look” very normal to the rig crews.
Several versions of the DSVs have been designed and built, but none are commercially available today. Two different
valves are being built for this project and each will be tested in top-hole operations.
During actual operations the DSV will be available is several drillpipe sizes and will generally be run just above the BHA.
It will be capable of being used in wells in 10,000 ft of water, to 35,000 ft TVD with 18.5-ppg mud, and will be built to the
same specifications as any other BHA tool.
Subsea Rotating Device (SRD)
The subsea rotating device is very similar to a common drilling rotating head (Fig. 5). It is the uppermost piece of equipment
in the DGD system. It will be run approximately 60 ft above the MLP. The SRD serves to separate the 8.6-ppg fluid in the
riser from the higher weight mud in the well. It ensures that gas doesn’t enter the riser and provides a slight pressure on the
well (less than 50 psi) needed to feed the MLP.
The SRD sealing element with bearings and drillpipe seals is run on the drillstring and tripped to the surface during each
drillpipe trip for servicing. It can hold pressure from both below and from above.
The SRD riser joint will have a pressure rating equivalent to the riser . The sealing element will have a minimum static
working pressure differential of 1,000 psi from above and 2,000 psi below and a dynamic working pressure differential of
1000 psi from above and below.
Solids Processing Unit (SPU)
The solids processing unit is designed to ensure that solids reaching the MLP are no bigger than 1½” x ½” x ½”, dimensions
which the pump is designed to handle. Cutters in the SPU will shear anything larger. Cuttings smaller than the required
minimum will pass through the SPU without being affected. After passing through the SPU, mud and cuttings are fed to the
MLP and pumped to the surface through a riser-mounted mud return line.
The SPU riser joint will be about 30 feet above the MLP and will also have a pressure rating equivalent to the riser.
Several valves are available to control the flow into the MLP.
MudLift Pump (MLP)
The heart of the DGD system is the MLP (Fig. 6). As originally conceived and designed in the JIP, the MLP is a sixchamber (80-gallon) diaphragm pump powered by seawater pumped from the surface. It is a positive displacement type
pump with independently controlled suction and discharge valves. Because each chamber can be operated independently, the
MLP can operate as two triplex pumps, a quintaplex, a quadraplex, a triplex, a duplex or as a single chamber pump. This
ability results in redundancy when the pump is operating at less than maximum capacity.
The GE-built MLP has a maximum rated flowrate of 1800 gpm with all chambers working. Maximum discharge
pressure rating is 6,600 psi above seawater ambient. The pump has two major modes of operation: constant inlet pressure
mode, used for most operations, and constant rate mode used for certain well control operations. 4 SPE 137319 Riser Dump Joint (RDJ)
During the numerous riser analyses performed, it became obvious that having the SRD in the riser would impact emergency
riser disconnects. The SRD traps the fluid in the riser and does not allow for the free exchange of seawater in and out of the
riser after an emergency disconnect as would happen with conventional systems. Trapping the fluid in the riser increases the
loads on the riser during the rapid up and down movement of riser. During rough sea states, these loads might become too
great for the riser system. Solving this problem required either a much stronger riser or a way of allowing the riser
fluid/seawater to freely move in and out of the riser tube.
The solution was to install a modified riser fill joint in the bottom of the riser. This riser dump joint (RDJ) opens during
an emergency disconnect and allows the free movement of fluids to reduce riser loading.
The “People” Part of the Process
Building new equipment to specifications can sometimes be difficult but even more challenging is getting people to change
the way they work. With the implementation of any new technology, people need training. The training provides the basis
for understanding and empowers the employees with confidence in their own ability to adapt to the technology. If this
confidence issue is not addressed, the project could encounter resistance from the critical employees needed for success.
To help change this situation the organization must develop a business strategy that clearly shows the value of the new
technology for everyone involved. Within Chevron this change situation is recognized and a priority has been placed on
ensuring people are properly trained. The process stresses engagement, communications, education and involvement.
Planning a DGD well requires an understanding of the equipment, the new procedures and, as is always the case, the
limitations of the rig. To develop the new DGD well planning process, the DGD project team began embedding personnel
within asset teams to produce DGD well plans. These were produced in parallel with the standard single gradient plans even
though the well would be drilled with a standard single gradient rig. The idea was to work with the asset team to refine the
process and work the issues while at the same time educating the entire organization. It also brought the side benefit of
highlighting the types of wells that would greatly benefit from the DGD technology.
The DGD technology has also caused Chevron to rethink how rigs are deployed. DGD technology can only provide real
value during drilling operations and is of less value during completions. Chevron is working this issue and developing what
is sometimes called the “Value Driven Drilling Schedule”. The DGD rig will be kept busy drilling the wells where DGD
technology yields the most value.
To kick off the communications of DGD concepts, the DGD project team instituted as series of “Lunch and Learn” sessions
in 2009 (Fig. 7). Over a four-month period, five two-hour sessions were conducted to introduce all of deepwater operations
personnel and selected outsiders to DGD procedures. Topics included Drilling Ahead, Tripping, Running Casing, Cementing
and Well Control. These sessions were part lecture and part question and answer periods. They also served to plant the seeds
of curiosity so that knowledge and interest would begin to grow.
The DGD project team also began producing a newsletter called “The Return Line” that is issued regularly to keep
interested parties current on the project. It includes general updates on the various pieces of equipment, the status of the
procedures development and a frequent “Single Gradient vs. Dual Gradient” section to spark thought.
Co-location of key personnel has also helped improve the communication in the DGD project team. All Chevron and
AGR personnel, and some of the contractors, work together at AGR’s offices in West Houston.
The DGD process changes almost every drilling and well control procedure on the rig. For example, the well constantly
wants to u-tube to equilibrium. This requires the use of completely new equipment and processes for many of the routine
procedures conducted on the rig. It’s not that these new procedures are difficult, but they are different and they must be
learned and taken to heart with this system.
During the JIP, more that 15 man-years of effort were put into developing the procedures that were proven during the
field test. Most of these remain valid for Chevron’s new deployment, but some of the DGD equipment has changed and the
rigs have definitely become more sophisticated in the last 10 years. This has required that all the procedures be revised and
adapted to current operations.
To ensure that these procedures are efficient and safe, a series of HAZOPs was recently completed. All the drilling and
well control procedures were HAZOP’ed by a multi-discipline team of 15 or more engineers. The team included
representatives from Chevron, AGR, Pacific Drilling, GE, Stress Engineering, and others. The HAZOPs examined each
procedure in detail and asked “what if?” for huge variety of things that potentially could go wrong at the worst possible time.
Each was then added to a “Risk Assessment” spreadsheet. Additional work was required to mitigate and lower identified
risks. SPE 137319 5 Education
From the finished HAZOPs the final procedures are written and the training materials developed. All the people involved
with the DGD deployment will receive training based on their job requirements, including all the rig personnel. The training
is generally divided along the following lines:
1. Basic DGD Concepts
2. Drilling Procedures
3. Well Control Procedures
4. Well Planning Procedures
5. Maintenance Procedures for Specialized Equipment
Included in this training are DGD simulator exercises where the drilling and well control procedures are practiced. The
duration of this training for items 1, 2 and 3 will be approximately three weeks. It will be timed to coincide with the factory
acceptance test (FAT) of the DGD equipment to allow for hands-on training with the actual equipment and will be used to
reinforce the simulator exercises.
The DGD technology represents a new way of drilling. It requires rethinking processes and techniques that have been
commonly used in the industry for more than 30 years. Although the concept of DGD can be understood in only a few
minutes, experience has shown that it takes 3 to 6 months to begin to “think” dual gradient and understand the nuances and
capabilities of the system.
The DGD system is inherently safer than single gradient operations in almost all ways. The well has riser margin (is dead
without the fluid in the riser) which makes emergency disconnects safer and less environmentally hazardous. All the HAZOP
work conducted by Chevron has shown the system to be as-safe and generally safer than current operations.
Deploying the DGD technology requires a huge commitment from many departments other than drilling. Its effects on a
company cascade through many levels and changes the way we do business. It will forever change the way we drill in deep
Smith, K., Gault, A., Witt, D. and C. Weddle. 2001. SubSea MudLift Drilling Joint Industry Project: Delivering Dual Gradient Drilling
Technology to Industry SPE 71357 presented at the 2001 SPE Annual Technical Conference and Exhibition held in New Orleans,
Louisiana, 30 September–3 October.
Schumacher, J., Dowell, J.D., Ribbeck, L. and J. Eggemeyer. 2001. Subsea MudLift Drilling: Planning and Preparation for the First Subsea
Field Test of a Full-Scale Dual Gradient Drilling System at Green Canyon 136, Gulf of Mexico. SPE 71358 presented at the 2001
SPE Annual Technical Conference and Exhibition held in New Orleans, Louisiana, 30 September–3 October.
Eggemeyer, J., Akins, M., Brainard, R. Judge, R. Peterman, C. Theti, R. and L. Scavone. 2001. SubSea MudLift Drilling: Design and
Implementation of a Dual Gradient Drilling System. SPE 71359 presented at the 2001 SPE Annual Technical Conference and
Exhibition held in New Orleans, Louisiana, 30 September–3 October. 6 SPE 137319 Figure 1 Figure 2 SPE 137319 Figure 3 Figure 4 7 8 SPE 137319 Figure 5 SPE 137319 Figure 6 9 10 SPE 137319 Figure 7 ...
View Full Document
This note was uploaded on 11/23/2010 for the course PETE 4XX taught by Professor Mehmetcebeci during the Spring '10 term at Middle East Technical University.
- Spring '10