Gas-Lift Best Practices

Potential Gas-Lift Development Ideas

Gas-Lift Value Engineering Workshop

Version: 11 January, 2023

This document lists the gas-lift best practices and potential development ideas that were identified, discussed, and supported at the Gas-Lift Value
Engineering Workshop. This workshop was held in the Shell Leeuwenhorst Training Center in Noordwijkerhout, The Netherlands, on 26 - 29 June,
2000. In the workshop, 58 "raw" best practice statements were recorded. Some of these were duplicates and have been combined. Thirty-four
potential development ideas were supported by the Operating Units attending the workshop. A few of these ideas overlap with associated best practices
but they have been listed anyway for completeness.

The best practices have been grouped into nine categories. These categories and the best practices in each category are listed below in the table of
contents. The best practices themselves, with discussion and recommendations, are presented in the sections that follow the table.

The potential development ideas have been grouped into seven of the same categories and are presented in their own table of contents and sections.

For the most part, the best practices can be implemented immediately and directly, as needed. Most (not all) of the potential development ideas will
require at least some development, and in some cases, will require research to bring them to a potentially usable state.

Table of Contents – Gas-Lift Best Practices

Category / Best Practice Page

Gas-Lift Best Practices

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

Field Development and/or Re-Development

1 Design up-front for gas-lift......................................................................................................................................... 8
2 Use gas-well gas for gas-lift....................................................................................................................................... 9
3 Use export gas pressure for gas-lift.......................................................................................................................... 10
4 Use remote booster compressors............................................................................................................................ 10
5 Optimize tubular sizes.............................................................................................................................................. 12
Gas-Lift Best Practices and Potential Development Ideas Page 2

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

6 Consider riser gas-lift for sub-sea wells.................................................................................................................... 13

Gas-Lift Best Practices and Potential Development Ideas Page 3

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Category / Best Practice Page

Production Equipment Design, Selection, and Use

1 Minimize well-head back pressure........................................................................................................................... 15
2 Maintain temperature above hydrate formation temperature...................................................................................16
3 Use waste heat to warm lift gas................................................................................................................................ 17

Gas-Lift Well and/or System Implementation

1 Convert poor continuous gas-lift wells to intermittent gas-lift...................................................................................18
2 Convert gas-lift wells to pumping.............................................................................................................................. 20
3 Use tandem gas-lift mandrels................................................................................................................................... 20

Gas-Lift Automation
1 Make all wells smart - make some wells brilliant..................................................................................................... 22
2 Fully automate all gas-lift wells................................................................................................................................. 24
3 Use automation to assure proper calibration of gas-lift models...............................................................................24
4 Use well test data and modelling to reduce need for pressure surveys and downhole instruments ........................25
5 Use a permanent down-hole pressure gauge with feedback control to the surface choke.......................................26

Gas-Lift Valve Design, Selection, and Use

1 Select gas-lift valves that have been manufactured to API and/or ISO standards...................................................29
2 Select gas-lift valves that have been performance tested.......................................................................................30
3 Select gas-lift valves that have a proven performance model.................................................................................31
4 Use ceramic seats in gas-lift valves......................................................................................................................... 32
5 Use gas-lift valves that don't need a bellows............................................................................................................ 33
6 Use nozzle-venturi gas-lift orifices............................................................................................................................ 33
7 Use pilot-operated gas-lift valves............................................................................................................................ 34

Gas-Lift Unloading and/or Kick-Off

1 Purge annulus liquid with nitrogen, diesel, or other light fluid..................................................................................36
2 Use foam for an unloading fluid................................................................................................................................ 37
3 Unload gas-lift wells with a swab unit...................................................................................................................... 37
4 Use shut-in wells for gas storage............................................................................................................................ 38
5 Unload a dual gas-lift well with one side only.......................................................................................................... 39
Gas-Lift Best Practices and Potential Development Ideas Page 4

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)
Gas-Lift Best Practices and Potential Development Ideas Page 5

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Category / Best Practice Page

Gas-Lift Well and/or System Operation

1 Test each gas-lift valve before and after use............................................................................................................ 40
2 Use gas-lift valve erosion test data........................................................................................................................... 41
3 Design how you plan to operate, operate the way you are designed.......................................................................42
4 Capture and analyze unloading data....................................................................................................................... 42
5 Know as soon as possible when a gas-lift well's design or operation is no longer OK and change it.......................43
6 Know actual optimum point for each well - maintain optimum operation, don’t exceed optimum point....................44
7 Use dedicated slick-line crews................................................................................................................................. 44
8 Use downhole shut-in tools when conducting bottom-hole pressure surveys..........................................................45

Gas-Lift Training
1 Develop the well analysis discipline........................................................................................................................ 46
2 Conduct well analysis workshops............................................................................................................................ 46
3 Provide "on site" training......................................................................................................................................... 47
4 Train on the job - no classrooms - use mentors....................................................................................................... 48
5 Use the "master/apprentice" training technique........................................................................................................ 49
6 Provide more frequent training................................................................................................................................. 49
7 Use job rotation for training/broadening of key gas-lift staff......................................................................................50

Gas-Lift Knowledge Acquisition

1 Implement a gas-lift valve tear-down data base.................................................................................................... 51
2 Host an Artificial Lift industry workshop.................................................................................................................. 51
3 Look outside Shell for best practice ideas................................................................................................................ 52
4 Use feedback from operating experience................................................................................................................. 52
Gas-Lift Best Practices and Potential Development Ideas Page 6

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)
Gas-Lift Best Practices and Potential Development Ideas Page 7

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Table of Contents – Potential Gas-Lift Development Ideas

Category / Potential Development Idea Page

Gas-Lift Well and/or System Implementation
1 Use small bubbles to enhance gas-lift, create with sintered injection point or ultrasonics........................................54
2 Use electric tubing to create vapor bubbles with arcs along the tubing, control on pressure...................................54
3 Use rail gun arcs, which are switched up hole.......................................................................................................... 54
4 Use a metal or plastic tubular that is porous to gas at one point..............................................................................55
5 Use an orifice that changes size downhole, combine with electric gas-lift valve......................................................55
6 Use a moveable injection point or slot...................................................................................................................... 55
7 Use reducible tubulars.............................................................................................................................................. 56
8 Use gas-lift and ESP simultaneously to reduce gas-lift requirements......................................................................57
9 Use gas to power a downhole pump........................................................................................................................ 57

Gas-Lift Automation
1 Use an automatic control system to optimize production (Delta P) of each well.......................................................59
2 Link CAO and modeling tools in real time (e.g. IDUN)............................................................................................. 60
3 Use automatic gas-lift valve retrieval/replacement................................................................................................... 60

Gas-Lift Valve Design, Selection, and Use

1 Implement an independent gas-lift equipment Quality Assurance process and use spot checks.............................62
2 Use a tracking system for gas-lift equipment – design, performance, failures, costs, deferment.............................62
3 Have new gas-lift designs reviewed by a panel/team or by a “peer” review.............................................................63
4 Sponsor a gas-lift design competition for NG gas-lift valves....................................................................................63

Gas-Lift Unloading and/or Kick-Off

1 Use advanced gas\foam sticks to help kick-off wells................................................................................................ 65
2 Punch holes and use E2 TECH patches to assist in gas-lift kick-off.........................................................................65
3 Splice the gas-lift conduit to the long string for unloading dual gas-lift wells............................................................66
Gas-Lift Best Practices and Potential Development Ideas Page 8

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)
Gas-Lift Best Practices and Potential Development Ideas Page 9

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Category / Potential Development Idea Page

Gas-Lift Well and/or System Operation

1 Use tracer technology to detect the operating gas-lift valve, if legal issues can be overcome.................................67
2 Install whistles in gas-lift valves with different pitches to determine operating gas-lift valve.....................................67
3 Use distributed temperature for rate measurements................................................................................................ 68
4 Use an inexpensive meter on each well and allocate gas via a single system meter...............................................68
5 Use a flapper check valve instead of normal check valves......................................................................................69
6 Use a handheld troubleshooting box with knowledge and historical data.................................................................69
7 Use surface controlled valves for deviated and horizontal wells...............................................................................70

Gas-Lift Training
1 Link the gas-lift demonstration program to CAO/SCADA – seeing is believing........................................................71
2 Develop and use a portable gas-lift model/simulator................................................................................................71
3 Build a gas-lift start-up simulator – require operators to qualify on it before undertaking operations........................72
4 Use a simulated environment to understand how to recognize and fix broken wells................................................72
5 Use a simulator to feed a neural net to learn failure symptoms for “smart” diagnosis..............................................73
6 Build a full-fledged gas-lift demonstration program.................................................................................................. 73

Gas-Lift Knowledge Acquisition

1 Publish gas-lift success and failure stories............................................................................................................... 74
2 Create a Shell “value of gas-lift” video...................................................................................................................... 74
Gas-Lift Best Practices and Potential Development Ideas Page 10

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)
Gas-Lift Best Practices and Potential Development Ideas Page 11

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Introduction
Starting in the next section, each gas-lift Best Practice is listed, by category. A brief description, appropriate discussion, and one or more recommended
actions are provided with each Best Practice.

During the gas-lift Value Engineering workshop, a complete outline was developed for presenting a detailed “way forward” plan for Best Practices. This
outline is shown below. It has been used, as an example, for one of the best practices in the Gas-Lift Automation category. If it is desired by the
Operating Unit sponsors, this more complete outline could be used to further develop other of the Best Practices or Potential Development ideas.

Here is the more complete outline.

Description: A more complete description of the best practice.

Risks/Barriers: A discussion of any risks or barriers that may confront use of the best practice.

Enablers: A discussion of people or processes that may help in the implementation of the best practice.

Where Does It Apply: A discussion of where the best practice applies, and if appropriate where it may not apply.

When Does It Apply: A discussion of when the best practice applies, and if appropriate when it may not apply.

Consequences (If Do): A discussion of the likely consequences of applying the best practice.

Consequences (If Don't Do): A discussion of the likely consequences of not applying the best practice.

References/Resources: A listing of references and/or resources, both within and outside of Shell, where more may be learned about this
best practice.

Others Experiences: A compilation of the experiences of others in using this best practice.

Value: The value or benefits that may be expected from using this best practice.

Cost: The expected cost of using this best practice.

Status/Timing: Some best practices can be applied straight away and some need to be further developed or supported before they can be used.

SepTAR Role: The potential role of SepTAR in helping to implement this best practice.
Gas-Lift Best Practices and Potential Development Ideas Page 12

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

OU Role: The role of the Operating Unit in implementing this best practice.

Potential Customers: A list of Operating Units that may be potential users or customers of this best practice.

OU Buy-In/Comments: Comments from OU's concerning this best practice. Do they plan to use it? Do they perceive problems in using it?

Feedback: Feedback (good or bad) from OU's that have used this best practice.
Gas-Lift Best Practices and Potential Development Ideas Page 13

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Best Practices

Field Development and/or Re-Development
The following gas-lift best practices should be considered when developing a new (green) field or re-developing an existing (brown) field that is expected
to (or already has) gas-lift as a means of artificial lift.

1 Design up-front for gas-lift Design and install initial (green field) production equipment based on the ultimate (brown field) requirements for
gas-lift. Incur any additional capital costs for gas-lift at the beginning of the project when the funds are available
to avoid needing to "beg" for funds later on when otherwise a retrofit is required.

Discussion: In principle, any "green" field development plan should consider the longer-term requirements for
artificial lift. If it is very likely that gas-lift will be required to recover the reserves that remain after primary
(natural flowing) production, provisions should be made for gas-lift initially. Such provisions may include:

 Gas-Lift System Pressure. Determine the most likely gas-lift system operating pressure. This pressure is
a function of the depths of the wells, the anticipated production rates, and the anticipated wellhead
backpressures. Various scenarios can be run using WinGLUE or WePS. If necessary, the Artificial Lift
team in SepTAR can assist in running these scenarios. If possible, it will be most economic to use the
same pressure for gas-lift that is required for gas export. (See Best Practice #3 below.) However,
depending on the well conditions, a higher pressure may be justified for gas lift. This pressure is needed to
be able to design the depths of the gas-lift mandrels.

 Gas-Lift Mandrels. Design and install gas-lift mandrels in the wells when they are initially completed. This
avoids the need to work the wells over to install gas-lift mandrels later when gas-lift is needed. There is a
"design gas-lift mandrel spacing" option in WinGLUE to address this. The recommended Best Practice
here is to err on the side of installing too many rather than too few gas-lift mandrels. It is always possible to
leave a "dummy" in any mandrels that are not needed. However, it can be very difficult to effectively gas lift
a well is there aren't enough mandrels to permit effective unloading.

 Gas-Lift Compression. Depending on when it is expected that gas-lift will be needed, it may or may not
be practical to install gas-lift compression initially. However, this should at least be considered so that
facilities can be designed to accommodate gas-lift compression when it is needed, without requiring any
facility re-work.
Gas-Lift Best Practices and Potential Development Ideas Page 14

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

 Gas-Lift Distribution. While actual gas-lift distribution facilities (pipelines, manifolds, measurement and
control, etc.) may not be required initially, it is good to consider these requirements and have a plan to
provide theses facilities. This can avoid a need to re-work other facilities later on when these facilities must
be added. A few Best Practices here are:
 Provide a way to measure the total amount of gas in the gas-lift system that is available for gas-lift.
Typically, this is the total gas available less gas that is used for other purposes such as fuel, export, etc.
 Provide a way to measure the pressure of the gas-lift system. This is an important variable in overall
gas-lift system and well control and optimization.
 Provide a way to measure and control the gas-lift gas to each well.

Recommended Action(s): Planning for artificial lift, and for gas-lift in particular if it is the most likely form of
artificial lift, should be a required part of every new (green) field development and every (brown) field re-
development plan.

2 Use gas-well gas for gas-lift Where it is available, design the gas-lift system to use gas from high-pressure gas well(s) to reduce or eliminate
the need for gas-lift compression. Or, use "connector" wells to connect high-pressure gas reservoirs to oil
reservoirs.

Discussion: If there is ample high-pressure gas-well gas available in a field, this can potentially be used to
partially (or possibly fully) offset the need for gas-lift compression. Several issues should be considered:

Some Compression is Normally Needed. Typically, a gas-lift system is essentially a "closed loop" system.
That is, the gas that is injected for gas-lift is produced back and has to be recovered and handled.
Normally, this requires that the gas be compressed, at least to the required export pressure, and that it be
conditioned for export (drying, etc.). So, the availability of gas-well gas cannot eliminate the need for
compression. But, it may reduce or eliminate the need for compressing to a higher pressure than the export
pressure.

Most Gas-Lift Gas is Re-Cycled. Typically, approximately 90% of the gas that is injected for gas-lift is re-
cycled and can be re-used for gas-lift. Typically, some small amount of gas is "lost" as low-pressure
vapors, etc. These "losses" must be "made up" with some additional source of gas. This additional source
can come from gas wells, even if the "produced" gas is re-compressed for gas-lift use.

Gas-Lift System Start-Up. A key issue is the initial start-up of a gas-lift system, or the re-start of the system
after any total field shutdown. In a start-up, the system must be supplied with enough gas to get the gas-lift
process started. It may be that gas-well gas can be used for this purpose.
Gas-Lift Best Practices and Potential Development Ideas Page 15

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Well Gas Issues. If gas-well gas is used for gas-lift, there are a few issues to consider:
 The gas must be controlled to a specified, desired pressure.
 The gas must be conditioned (dried, etc.) to a specified quality.
 The volume of gas must be sufficient to provide for gas-lift requirements, recognizing that these
requirements will likely grow over time.

Recommended Action(s): If gas-well gas is available in a field and if it can potentially be used to meet or
supplement the gas-lift requirements, this should be carefully studied before a decision is made to depend on
the gas-well gas and not provide the required gas compression facilities.

3 Use export gas pressure Where it is feasible and economic to do so, design the gas-lift system pressure to match the gas export (or
for gas-lift sales) pressure. Eliminate the need for booster compressor(s) for gas-lift.

Discussion: Typically, the operating pressure of most gas-lift systems starts out at the export pressure. This
avoids the need to "boost" the pressure to a higher level for gas-lift. However, this decision should not be taken
lightly. If the wells in a field are deep, or if the desired production rates are high, it may be very economically
attractive to provide a gas-lift system operating pressure that is higher than the field gas export pressure. Or,
saying this the other way around, there may be a very large economic penalty associated with limiting the gas-
lift system pressure to the gas export pressure. The lower pressure may limit the depth to which the wells can
be lifted. The largest single factor in gas-lift effectiveness is depth of injection.

Recommended Action(s): The recommended action, for any new (green) field development, or any (brown)
field re-development or conversion to gas-lift, is to carefully evaluate the required gas-lift system pressure. This
can be done by running various scenarios with WinGLUE or WePS. If necessary, the Artificial Lift team in
SepTAR can assist in running these scenarios.

4 Use remote booster In those cases where some wells are far from the "main" compressor station, install booster compressor(s) in
compressors outlying areas to reduce the need to compress all gas to higher pressure and to provide adequate pressure for
far-away wells.

Discussion: If some gas-lift wells that are physically located far from the source of gas-lift gas, there may be
significant pressure drops in the gas-lift distribution system between the source location and the wells. This can
result in several "downside" gas-lift issues including:
 Inadequate lifting depth.
Gas-Lift Best Practices and Potential Development Ideas Page 16

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

 Incompatibility with the gas-lift design, if the design was made assuming a different (higher) pressure.
 Instability that may result from interference from other wells or groups of wells on the long distribution
system.

To address this problem, there are basically five choices:

Compress All Gas to High Pressure. Compress all of the gas to a higher pressure, so at least adequate
pressure can be delivered to the remote wells. Clearly, this has a high cost associated with it.

Install Multiple Compressor Stations. Install multiple primary gas compression facilities, each one located
relatively close to the wells it serves. This may also have a high cost associated with it - to provide the
necessary gas sources, facility infrastructures, etc.

 Install Large Distribution Lines. Install large gas-lift distribution lines so the pressure losses in the lines
will be minimized. This is clearly a matter of economics. If the added cost of the bigger lines, and the
improvement in pressure at the remote wells is sufficient to justify the cost, then this may be a good option.

Live with Lower Pressure. Live with an inadequate pressure at the remote wells. Normally, this is not a
satisfactory solution.

Use Booster Compressors. Provide remote, booster compressors to boost the gas that is required by the
remote wells. Presumably, this is only a portion of the total gas-lift gas that must be boosted, and therefore
the total compression costs will be less than with the first option. And, this avoids some of the problems
associated with the second option. This option, does, however have some issues associated with it:
 There may be a significant cost associated with operating the remote booster compressors.
 In some cases, there may be issues (security, safety, etc.) associated with operating remote booster
compressors.

Recommended Action(s): If a field is large, with some wells located far from the available source(s) of gas-lift
gas, it may be necessary to consider this option. Unfortunately, no standard guidelines can be given since each
case is (will be) different. As a general rule, if the gas pressure loss between the compressor station (or other
source of high-pressure gas-lift gas) and the wells where the gas must be used is more than 50 - 100 psi (3 - 7
bar), consideration should be made to boost the pressure to the remote wells.
Gas-Lift Best Practices and Potential Development Ideas Page 17

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

5 Optimize tubular sizes Where feasible, optimize the flow regime to maximize productivity. Use a variable or tapered tubing string with
larger size up hole and smaller size down hole. Install the tubing close to the perforated interval.

Discussion: The concept that underlies this best practice is that, as liquid and gas are produced up the well
bore, the pressure is reduced. This causes some gas to come out of solution, and both this gas and any "free"
gas (from the formation or from the gas-lift injection) to expand. This expanded volume will result in more
pressure losses due to increased friction between the flowing fluid and the walls of the production conduit.

There is a "flip" side to this. For any tubular size, there is a minimum fluid production rate that must exist to
sustain stable flow. If the rate falls below this minimum, flow will become unstable due to slippage and fallback
and may cease altogether. In some wells, the production rate declines with time, due to reservoir pressure
depletion, inflow problems, etc. In some of these wells, it becomes necessary to "down size" the production
conduit by installing smaller tubing, or by installing an insert "velocity" string.

In the normal case, the size(s) of the production conduit is chosen initially to accommodate the expected initial
production rates of the wells. Of course, these may be limited by the available size of the wellbore. So, it is
often the case that, later in a well's life, the tubulars are too large. It is difficult to design tubulars that can
effectively accommodate a wide range of production rates.

One potential way to address the issue of gas expansion and associated increased friction is to use a tapered
tubing string, with smaller tubing on the bottom and larger tubing high in the wellbore. If, later in the life of the
well, it becomes necessary to downsize the tubing to sustain continuous flow, this might be accomplished with a
shorter insert string that only needed to reach to the depth of the smaller tubing string.

One other comment has to do with the flow that must occur from the well's inflow point(s), that is from the
formation interface into the wellbore up to the bottom of the tubing. Typically, except perhaps in the case of
dual wells, this flow will occur in the casing. If this is a long distance, the flow may not be efficient. Therefore,
for effective flow, and for effective gas-lift operation, the end of the tubing should be installed as close as
practical to the top of the perforated interval and the gas-lift injection point should be as close as possible to the
end of the tubing.

Recommended Action(s): The effects of different tubular configurations, including the effects of tapered tubing
strings, can be investigated by running various scenarios with WinGLUE or WePS. If assistance with this is
required, the Artificial Lift team in SepTAR can help.
Gas-Lift Best Practices and Potential Development Ideas Page 18

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

6 Consider riser gas-lift For sub-sea gas-lift installations, consider the use of "riser" gas-lift to assist in the production of the fluid to the
for sub-sea wells surface and to help stabilize the flow and pressure in the riser.

Discussion: With the significant growth in the number of sub-sea wells in the Shell Group, the issues concerned
with flow and pressure drop between the wellhead and the host platform are becoming very important. If a well
has a long sub-sea flowline, and if it is several thousand feet from the seabed to the platform, the pressure
required to produce the well's fluids from the wellhead to the separator on the platform can be substantial.

Another important issue can in instability in the riser, due to liquid/gas slugging. Slugs that are produced from
the well may become worse in the sub-sea flowline and worse still in the riser. Significant slugs (large, rapid
changes in both flow rate and pressure) can have very negative impacts on the production facilities on the
platform.

For many wells, there are a limited number of choices to meet this pressure requirement and to address this
slugging issue.

 Enhance Artificial Lift System in the Well. Provide sufficient lifting from downhole in the well to produce
the well's fluids to the wellhead, to flow the production from the wellhead to the base of the platform, and to
lift the fluids up the riser to the platform. The requirement to "produce" the well's fluids from the wellhead to
the platform may increase the artificial lift "lifting" requirements by a very significant amount. If the well is
being gas-lifted, it may be very difficult to deliver either the required pressure or the required gas injection
rate to the well, at the desired depth. Also, if a large amount of gas is added downhole in the wellbore, all of
this gas must also be produced up the well, through the flowline, and up the riser. The pressure losses
associated with producing the additional gas may also be substantial. (Thus, in some and perhaps many
cases, this points to the benefits of pumping rather than gas-lift to produce these sub-sea wells.)

 Provide Wellhead Boosting. Provide some mechanism to "boost" the pressure at the wellhead. This
could be in the form of multi-phase pumping, progressive cavity pumping, electrical submersible pumping (if
the fluid isn't too gassy), or gas-lift. This option can, in principle, be applied either at or close to the
wellhead, or at the base of the riser. Clearly, any of these options would require a very significant
engineering design effort. The Artificial Lift team in SepTAR can assist with analyzing this option.
Gas-Lift Best Practices and Potential Development Ideas Page 19

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

 Use Riser Gas-Lift. Another option is to use riser gas-lift. This is a process whereby gas is injected into
the base of the riser, near the sea floor, and the production is "gas lifted" up the riser. There are several
issues with this approach:
 Advantages
 The gas (at least the gas for riser gas-lift) only needs to be supplied to the base of the riser, and not
to the remote wellhead, or to downhole in the well.
 In many cases, it may be possible to inject gas directly into the base (or near to the base) of the
riser, without the need for unloading gas-lift valves in the riser, or at least with only a minimum
number of unloading valves.
 This method provides a high degree of flexibility. The rate of gas injection can easily be controlled
from the surface to "optimize" the lift in the riser as the well's production conditions change.
 Since gas-lift uses gas, the presence of other gas in the produced fluid, which might cause a
problem for pumping, actually helps the gas-lift process.
 Challenges
 A separate gas injection line may be required.
 It may be difficult to design and operate this if more than one injection depth is needed.
 It may be difficult to produce the desired pressure reduction. The wellhead backpressure will be the
sum of the producing pressure at the bottom of the riser, plus the pressure drop in the seabed
flowline. Gas-lift is not ideal for creating minimum "bottom hole" (in this case riser bottom)
pressures.
 This may or may not produce a favorable impact on the issue of pressure and rate slugging or
surging. If the flow coming into the riser from the flowline is already slugging, it may not be possible
for the gas-lift system to reduce the slugging, and it may actually make it worse.

Recommended Action(s): Design and implementation of any of these options is a major undertaking. Before
any of these options is seriously considered, it is recommended that the advice of the Artificial Lift team, and
likely also others such as multi-phase pumping experts, be obtained.
Gas-Lift Best Practices and Potential Development Ideas Page 20

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Production Equipment Design/Selection/Use

The following gas-lift best practices should be considered when designing, selecting, and/or using production equipment. Proper design can help to
enhance gas-lift system and well performance.

1 Minimize wellhead Design wellheads, flowlines, manifolds, separators, etc. to minimize the wellhead backpressure on gas-lift
back pressure wells, to maximize their production rate.

Discussion: For a typical gas-lift system, every extra psi or kPa of pressure at the wellhead translates into an
extra (higher) psi or kPa of pressure downhole. This translates into a lower draw down between the reservoir
pressure and the flowing bottom-hole pressure, and therefore a lower production inflow rate. So, it is clearly a
good idea to do all that is reasonably possible to minimize the backpressure at the wellhead that the well’s
production system must overcome.

In simple terms, the backpressure at the wellhead is the sum of wellhead pressure losses, plus flowline
pressure losses, plus production facility back pressure requirements. Several steps can be considered and
several procedures can be used to achieve minimum wellhead backpressure. These include:

 Minimize Wellhead Pressure Losses. Pressure losses at the wellhead can be minimized by eliminating
restrictions to flow such as unnecessary piping restrictions, un-needed chokes or choke bodies, etc. There
are some cases where some restrictions are needed, but they should be designed to minimize the impact
on production.
 In some cases, it may be necessary to have a choke body, so the production can be restricted (choked)
is necessary.
 In other cases, it may be justified to create a small pressure drop to obtain an important measurement.
An example is use of the differential pressure (Delta P) method to estimate the well’s production rate
based on measured differential pressure across a large (small pressure drop) restriction at the
wellhead.

 Minimize Flowline Pressure Losses. Pressure losses in the flowline can be minimized by observing the
following steps:
 Choose an optimum flowline size to optimize (balance) capital cost vs. pressure losses that will result if
the flowline diameter is too small.
 Try to avoid using one flowline to carry the production of multiple wells. In addition to creating extra
pressure losses, this also creates other problems, for example with well testing.
Gas-Lift Best Practices and Potential Development Ideas Page 21

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

 If paraffin may be deposited in the flowline, establish a routine “pigging” or “hot oil” program to keep the
flowline clean.
 If sand may be deposited in the flowline, establish a routine process to flush the line to keep it clean.

 Minimize Production Facility Back Pressures. Production facility back pressures can be minimized by:
 Designing the inlet manifold to minimize pressure losses due to excessive pipe bends, un-necessary
restrictions, or too-small piping.
 Designing the production separation system to minimize pressure losses.
 Designing the equipment that is downstream of the production separator (bulk treating equipment, gas
compression equipment, etc.) to maintain a minimum backpressure on the production separation
equipment.

Recommended Action(s): In every new (green) field design or (brown) field re-design, take time to review and
evaluate all of the areas that may create back pressure on the well-heads and design and/or select the well-
head, flowline, and production facility equipment to keep the well-head back pressure to a practical minimum.

2 Maintain temperature Insulate gas injection lines, and/or install them to slope down, to help prevent the formation of hydrates in the
above hydrate formation injection lines at points where there may be a significant temperature drop due to a pressure drop.
temperature
Discussion: Hydrates can form in a gas line when water vapor is present in the gas and the temperature falls
low enough. When the pressure of the gas is reduced, as it often is across a choke or control valve, a
temperature drop will occur as well. Hydrates can block the flow of gas to the well and, as a result, cause the
well to stop producing.

The Best Practice ways to prevent hydrate formation are:

 Dry Gas. If the gas is sufficiently dry, hydrates will not form. Dry gas will also make it easier to accurately
measure the gas, and will help to prevent erosion or other problems with the gas-lift valves. A Best
Practice is to provide sufficient dehydration to assure that the gas is dry enough to prevent hydrate
formation. This can be done with a dehydration system, or possibly by injection of a hydrate suppression
chemical such as glycol.

 Minimize Pressure Drops. Another important Best Practice to minimize hydrate formation is to minimize
pressure drops. Clearly, any excessive pressure drop in the gas-lift delivery system means that less
pressure is being delivered to the well. This may have a negative impact on depth of lift. There must be
Gas-Lift Best Practices and Potential Development Ideas Page 22

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

some pressure drop associated with the process of controlling the gas injection rate into each well, but this
should be kept to a minimum.

 Insulate Gas Injection Lines. In those cases where some pressure drop across a control device is
unavoidable, it can be a Best Practice to insulate the gas-lift line near the pressure drop, and also insulate
the control choke or valve. This can help to prevent “freezing” or hydrate formation, especially in cold or
rainy weather.

 Avoid Low Spots. Another Best Practice is to install gas piping so that there are no low spots where
liquid can accumulate. If there must be some low spots, or long runs of essentially horizontal pipe, it can
help to provide a place where liquid can be drawn from the line, with the pipe sloping down to this point.

Recommended Action(s): The recommended approach is to provide a source of dry gas so that hydrate
formation will be minimized. If this can not be done, then one or more of the other steps should be used to
deliver relatively dry gas, without freezing or hydrate formation problems.

3 Use waste heat to warm Where feasible and economical, use waste heat from gas compressors or other process equipment to warm
lift gas gas-lift gas to help prevent formation of hydrates.

Discussion: Another way to help prevent the formation of hydrates is to warm the gas-lift gas in a heat
exchanger with waste heat from the gas compressor(s) or other process equipment. If gas freezing is or may
be a problem, this may make economic sense, especially if the warming effects will last until the gas is injected
into the gas-lift wells. This could be the case on an offshore platform, but might not be the case in a field with a
large distribution area.

Recommended Action(s): If hydrate formation is or may be a problem, evaluate the costs of installing a heat
exchanger vs. the likely benefits from warming the gas. The primary considerations will be the cost of the heat
exchanger and the likely lasting value of the heating, given the amount of warming that can occur, the external
temperatures to which the gas pipeline(s) will be exposed, and the distances between the source(s) of heat and
the injection of the gas.
Gas-Lift Best Practices and Potential Development Ideas Page 23

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Well/System Implementation

The following gas-lift best practices should be considered when implementing gas-lift in a field, or when working to find the optimum form of artificial lift in
a field.

1 Convert poor continuous Consider converting poor, inefficient continuous gas-lift wells to intermittent gas-lift to achieve lower overall
gas-lift wells to intermittent average producing bottom-hole pressure, greater overall average inflow, and higher overall lift efficiency.
gas-lift
Discussion: In many fields, the "de facto" method of artificial lift is continuous gas-lift. In many cases, it has
been considered poor practice to mix continuous and intermittent gas-lift on the same gas-lift system. The
primary reasons for this are:

 Intermittent gas-lift cycles, and the associated pressure and rate fluctuations, may cause instability in the
gas-lift system and thus interfere with the efficiency of the continuous gas-lift wells.

 Intermittent gas-lift is a very different process than is continuous gas-lift. It is not nearly as well understood,
it requires different design procedures, and it requires very different operating strategies.

However, intermittent gas-lift can be much more effective than continuous gas-lift for some types of wells. If
a well has a low bottom-hole pressure, a poor inflow performance (low productivity index), or a tubing size
that is too large to sustain continuous flow, it may be very inefficient in continuous gas-lift. Often these
wells are easily recognized. Often, they will produce with large pressure and rate "heads" and may actually
"self intermittent." That is, they will actually die (stop to produce) for significant periods of time in between
"self intermittent" production cycles.

In these cases, even though intermittent gas-lift is inherently inefficient, if it is properly planned, designed,
and operated, it may be more efficient than a poor continuous gas-lift operation. A significant advantage of
intermittent gas-lift in these cases, rather than some other alternative such as pumping, is that all of the
necessary infrastructure for gas-lift is already in place.

There are some Best Practices when considering converting poor continuous gas-lift wells to intermittent
operation.

 Be Selective. It is not necessary, nor desirable, to covert all of the wells to intermittent gas-lift. If some
wells can be more effectively lifted with continuous gas-lift, they should not be converted. If these best
practices are followed, it is only necessary, and desired, to convert the poor continuous gas-lift wells to
intermittent.
Gas-Lift Best Practices and Potential Development Ideas Page 24

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

 Use Constant Injection Rate. If possible, design the intermittent gas-lift wells to continue to use a
constant injection rate. This will minimize upsets in the gas-lift distribution system, and thus interference
with other wells on the system. This may be accomplished if the wells use a pilot-operated gas-lift valve.
This process works as follows:

 The gas is injected at a constant rate at the surface.

 The casing pressure gradually rises since gas is entering the casing at the surface but not leaving the
casing and being injected into the tubing.
 When the casing pressure rises high enough, the pilot-operated gas-lift valve opens and admits a large
amount of gas into the tubing beneath the liquid slug.
 When the casing pressure falls, the pilot-operated gas-lift valve closes and the process repeats.

 Coordinate Injection Cycles. If the wells cannot be designed to operate effectively with this method, the
next best thing is to coordinate the intermittent gas-lift cycles so that no two wells are injecting at the same
time. This can also minimize the overall effects on the gas-lift system and the other wells in the system.

 Automate the Gas-Lift Process. For either of these methods to work well, in the case where both
continuous and intermittent gas-lift is mixed on the same system, it is a Best Practice to monitor and
control the system with a gas-lift automation system. This system can assure that, regardless of the
intermittent process being used, it can be kept in "most efficient" operation at all times.

 Use a Standing Valve. In any intermittent gas-lift operation, it is a Best Practice to install a standing valve
below the bottom gas-lift valve. This will prevent back flow into the formation during the injection cycle.

Recommended Action(s): Continuous gas-lift wells that are performing poorly due to low bottom-hole pressure,
poor productivity, or too large tubing, can be very inefficient, can waste large amounts gas lift gas, and can
result in very ineffective production. They should be evaluated for conversion to intermittent gas-lift.

When a gas-lift system will contain both continuous and intermittent gas-lift, it should be automated to allow
continuous surveillance and continuous, effective monitoring and control of the injection and production
processes in both types of wells. There is good precedence for this type of operation in the Sirikit Field in Thai
Shell.
Gas-Lift Best Practices and Potential Development Ideas Page 25

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

2 Convert gas-lift wells Where appropriate and economically justified, consider converting gas-lift wells to pumping to increase current
to pumping production rates and ultimate recoveries.

Discussion: There may be cases where gas-lift has been used in a field, but the field and/or well conditions are
no longer favorable for gas-lift. This could be the case if:

 The productivity of some wells is higher than can be effectively produced with gas-lift. More oil can be
produced by pumping than by gas-lift.

 The reservoir pressure has declined below the level where continuous gas-lift can be sustained but the
inflow productivity is still good and the wells can produce much more than can be effectively lifted with
intermittent gas-lift.

 The wells are beginning to produce with high water cuts. This may result in overall rates that are too high to
be effectively gas-lifted and it may also result in a shortage of gas for gas-lift.

Recommended Action(s): In these cases, conversion of the wells to some form of pumping should be
considered. This is not a Best Practice document on pumping. However, there are many matters that should
be considered in choosing the best form of pumping. If conversion of (former) gas-lift wells to pumping appears
to be a reasonable option, the advice of the Artificial Lift team should be obtained.

3 Use tandem gas-lift When appropriate, for example in high-production-rate wells that need a high rate of gas injection, consider
mandrels using tandem gas-lift mandrels (that is, two mandrels at each depth) so that two gas-lift valves can be installed
at each depth to accommodate the high rate of gas injection.

Discussion: The largest conventional gas-lift valves are 1.5 inches in diameter. The maximum gas-lift injection
rates that can be effectively sustained through these valves are on the order of 3.0 - 4.0 MMCF / Day,
depending on the pressures available. In some gas-lift wells, that have very large tubing and very large
(potential) production rates, it may be required (desired) to inject much more than this amount of gas. In these
wells, use of tandem (or twin) gas-lift valve can be considered. A tandem or twin gas-lift installation is usually
referred to as a gas-lift injection "station."
Gas-Lift Best Practices and Potential Development Ideas Page 26

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

In these cases, there are some Best Practices to consider:

 If possible, unload the well with only one gas-lift mandrel and valve at each depth and use the tandem or
"twin" gas-lift injection station only at the intended gas-lift injection depth. And at this depth, use gas-lift
"orifice" valves for injection.

 If tandem unloading valves are required, minimize the number of different unloading "stations." This may
be achieved by providing a special, high pressure unloading system.

Recommended Action(s): For very large rate wells, special design approaches may be required to obtain the
optimum production rates. If more gas is needed for effective gas-lift than can be injected through one gas-lift
valve and/or orifice, consider use of tandem (or twin) gas-lift mandrels and valves.
Gas-Lift Best Practices and Potential Development Ideas Page 27

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Automation

The following gas-lift best practices should be considered when designing and implementing the oil-field instrumentation, control, and automation system
(CAO or computer assisted operations) for a gas-lift field.

1 Make all wells smart - Equip each gas-lift system and all wells with enough surface instrumentation and control to be classified as
make some wells brilliant "smart." That is, so that they can continuously "tell" the gas-lift staff their status and their control needs. It will
be justified to make some wells "brilliant" with downhole instrumentation and control.

Description: Each gas-lift system should be equipped with enough instrumentation to be able to continuously
determine the amount of gas available for gas-lift and its pressure. Each gas-lift well should be equipped with a
minimum set of instrumentation and control to allow it to be continuously monitored and controlled, and to
support continuous, routine surveillance. The minimum per well instrumentation and control should include:

 Injection pressure measurement

 Injection rate measurement
 Injection rate control
 Production pressure measurement
 Production rate measurement (or estimation based on differential pressure).

For some gas-lift wells, downhole instrumentation and control can be justified. This may be the case for sub-
sea wells, very high rate wells, important (high production rate) dual gas-lift wells, etc.

Risks/Barriers: There are no risks for the "base" case. The only barrier is the cost to install the needed
instrumentation and control. In some cases, where there are no existing gas injection measurement and control
facilities, this cost can be substantial.

For the "advanced" case (downhole instrumentation and/or control), the risks and barriers are being addressed
with the current development of this technology by the Shell Intelligent Well (iWELL) Team and the OU's that
are associated with it.

Enablers: The benefits from gas-lift surveillance and optimization are very well documented. There is already
much experience in the Shell Group for this. The benefits of the "advanced" case are being developed by the
iWELL team.
Gas-Lift Best Practices and Potential Development Ideas Page 28

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Where Does It Apply: The "base" case applies for all gas-lift systems and wells in all Shell OU's. The cost of
the minimum level of instrumentation and control is small when compared with the rewards. The application
opportunities for the "advanced" case are being developed by the iWELL team.

When Does It Apply: This can be applied now. It already is being used in PDO and Thai Shell, and it is being
implemented in SSB, SPDC, and BSP.

Consequences (If Do): Installation of appropriate gas-lift instrumentation and control provides access to the
wide array of gas-lift surveillance and optimization capabilities that are available in Shell. One of the "follow on"
consequences is that the required people (with the appropriate skills, etc.) must be in place to make effective
use of this system.

Consequences (If Don't Do): If this is not done, the opportunities for gas-lift surveillance and optimization are
seriously compromised. Basically, this is the "old way" of operating and leads to inefficient and ineffective gas-
lift.

References/Resources: To be determined.

Others' Experiences: To be determined.

Value: The value of gas-lift surveillance and optimization has been documented by PDO to be at least 5 - 10%
more oil with 5 - 10% less gas. Most gas-lift "experts" feel that the overall benefits may be much higher,
especially for typical "problem" gas-lift wells such as chronically unstable gas-lift wells, dual gas-lift wells, etc.

Cost: The cost of "base" case instrumentation and control will vary depending on the location, the existing
infrastructure, etc. A good "rule of thumb" would be to assume a cost of $25,000 per well for the needed
instrumentation and control equipment, including the necessary remote terminal unit (RTU) or connection to a
local DCS (distributed control system), facility RTU, or facility PLC (programmable logic controller).

Status/Timing: The "base" case can be implemented today. The required measurement and control
instrumentation exists and is readily available. The associated CAO (SCADA) systems are readily available.

SepTAR Role: For the "base" case, SepTAR can advise on specific configurations of instrumentation and
control hardware and associated equipment. It can advise on various CAO/SCADA system options. It can
provide associated gas-lift training and skills development.
Gas-Lift Best Practices and Potential Development Ideas Page 29

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

For the "advanced" case, the iWELL team can assist in all aspects of the project from candidate selection,
equipment selection and design, installation, commissioning, operation, and evaluation.

OU Role: The OU role must be to take the lead in all aspects of the process from initial planning, budgeting,
project management, installation, commissioning, operating, and follow-up evaluation.

Potential Customers: All Shell OU's that use gas-lift are potential customers for this best practice. This should
be the way for working for all gas-lift systems and wells.

OU Buy-In/Comments: This space is reserved for comments from OU's concerning this best practice. Do they
plan to use it? Do they perceive problems in using it?

Feedback: This space is reserved for feedback (good or bad) from OU's that have used this best practice.

2 Fully automate all gas- Provide full automatic monitoring and control of all gas-lift wells.
lift wells
Discussion: This best practice is very similar to Item #1. All gas-lift wells should be automated at least to the
"base" case level as described above. Selected wells should be further automated to the "advanced" case level
where the additional expense is justified by the additional benefits to be realized from downhole instrumentation
and control.

Recommended Action(s): Every gas-lift system should be evaluated for automation. In essence, every gas-lift
system that is worth operating is worth operating effectively. The question is not, "should the gas-lift system be
automated?" The question is, "to what degree should it be automated? If desired, the advice of the Artificial Lift
team may be obtained to help with this evaluation.

3 Use automation to assure Use automation to gather, store, process, and validate gas-lift information. Use this information to calibrate the
proper calibration of gas- various gas-lift models to assure optimum operation and surveillance of the gas-lift process.
lift models
Discussion: For effective gas-lift surveillance and optimization, models are used for several purposes,
including:
 Estimating the injection pressure at various depths in the well.
 Estimating the production pressure at various depths in the well.
 Calculating which gas-lift valve(s) are open.
 Estimating the amount of gas flow through each open (or partially open) gas-lift valve or orifice.
Gas-Lift Best Practices and Potential Development Ideas Page 30

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

 Estimating the flowing (or producing) bottom-hole pressure.

 Calculating the well's inflow performance.
 Calculating the well's gas-lift response curve (relationship of Production Rate to Injection Rate).
 Calculating the optimum distribution of gas to each well.
 Diagnosing various gas-lift problems, including both stable and unstable operations.
 Diagnosing various well inflow and/or outflow problems.

In the (near?) future, models will also be used to:

 Determine the optimum unloading process.
 Determine the optimum kick-off (restart) process.
 Diagnose gas-lift problems and determine potential causes of failure.

When all of the wells in a gas-lift system have been automated as discussed above in Item #1, real-time data
can be used to help validate and calibrate these various models. Once each model is properly calibrated to the
operating conditions of the gas-lift system and/or well it serves, it can provide much more accurate and effective
gas-lift surveillance and optimization.

Recommended Action: Clearly, it is recommended to use the most effective models to gain (and maintain) the
best understanding possible of the operation of each gas-lift well and its equipment. To do this, an automation
system is required to effectively collect the necessary information, "feed" it to the models, and use it to help
calibrate the models.

A recommended Best Practice is to obtain, read, and study the American Petroleum Institute (API)
Recommended Practice 11V8 - "Recommended Practice for Gas-Lift System Design and Performance
Prediction." This document contains valuable insights and recommendations for using and calibrating gas-lift
models for effective gas-lift surveillance and operations.

4 Use well test data and Where possible, use accurate well test data and well models (e.g. vertical pressure profile models) to determine
modeling to reduce need (estimate) producing down-hole pressure, thus reducing the need for physical pressure surveys or permanent
for pressure surveys and downhole pressure measurement devices.
downhole instruments
Discussion: In some cases it can be very expensive (and/or risky) to run pressure surveys. Also, it can be
expensive to install down-hole instrumentation, especially if the well is already completed without such
instrumentation. So, if the down-hole information needed for effective well and gas-lift surveillance can be
calculated (estimated) based on "accurate" well test data and calibrated models, this can be a very good thing.
Gas-Lift Best Practices and Potential Development Ideas Page 31

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Also, even when it is feasible to conduct pressure surveys, these are usually not run very often. In most fields,
the most frequent that pressure surveys are run is once per year or even less frequently. So, even when they
can be run, it is a very good operating practice to obtain "accurate" well tests and to use this information in
conjunction with calibrated models to perform continuous (or at least very frequent) surveillance on each gas-lift
well.

As discussed in Best Practice No. 3 above, the process of calibrating gas-lift models is made much more
effective by using the right type and level of automation to assure the collection and processing of accurate,
reliable, repeatable information. An effective automation system can also greatly assist in the process of
collecting, validating, and evaluating well test data. More information on automation systems for effective well
testing can be obtained from the Production Technology team in SepTAR.

Recommended Action(s): It is always recommended to use accurate well test and calibrated gas-lift models to
assist in day-to-day gas-lift surveillance. If this information can be calibrated with measured pressure surveys,
or with data from down-hole instrumentation, clearly this is advantageous. However, whether or not surveys
can be run or down-hole instrumentation exists, it is always recommended to obtain the most accurate possible
well test information, to use this and other information to calibrate the gas-lift models, and to use this
information as part of a routine gas-lift surveillance process.

It is not recommended to depend solely on well test data and models for gas-lift surveillance, unless there is no
alternative. If it is possible to run good pressure surveys, at least occasionally, or to install down-hole
instrumentation to continuously measure down-hole pressure(s), this is certainly a recommended Best
Practice.

5 Use a permanent down- Consider installing a downhole pressure gauge with a feedback control capability to permit optimization of well's
hole pressure gauge with production (e.g. producing bottom-hole pressure) through automatic adjustment of the surface backpressure
feedback control to the choke.
surface choke
Discussion: This "best practice" actually speaks to two separate issues. The first is the use of down-hole
instrumentation to continuously measure down-hole pressure, and perhaps other variables. The other is to use
this information as part of an on-line, real-time control system to optimize well production.

Clearly, the first point is a Best Practice. If the gas-lift staff can always know the producing down-hole
pressure, at the inflow point, and/or at the depth of the operating gas-lift valve, and/or at the depth of the
unloading gas-lift valves, this can be a tremendous aid in gas-lift surveillance.
Gas-Lift Best Practices and Potential Development Ideas Page 32

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

The second point sounds good, but is actually a little less clear. On the surface, it sounds right to be able to
continuously optimize a well's production by controlling the surface choke (or back-pressure control valve) to
optimize its bottom-hole pressure and thus its production rate.

However, there are some issues that need to be considered:

 In general, it is always desired to keep the wellhead backpressure to as low a value as possible. Any
additional backpressure at the surface will result in higher inflow pressure that will result in lower production
rate.

 Having said this, a small amount of backpressure is, in most cases, preferable to an unstable well. So, if
instability can be "cured" by applying a small amount of backpressure, this may be a very good thing.

 In general, if a well is unstable, there is some problem with its design or operation. The recommended Best
Practice, for any unstable well, is to determine the cause of the instability and if possible to fix the problem.

 Wellhead backpressure should only be applied to correct an instability problem if it cannot be corrected by
proper gas-lift design or operation.

 Furthermore, it is not clear that down-hole pressure is the best information to use for real-time wellhead
backpressure control. It may actually be much more effective to use surface pressure for this, since there is
a time delay between control actions taken on the surface and the results of these actions being realized
down hole.

 There is another issue. If down-hole pressure can be measured, and if it can be used for wellhead control,
a more logical use may be to help optimize the gas-lift injection rate.

 However, this is also not totally clear. Most gas-lift wells exist as part of a gas-lift system. In this system, it
is necessary to optimally distribute (or allocate) the available gas to the wells in the system. If there were
an on-line, real-time system that was attempting to continuously optimize each well, this system could end
up "fighting" with the system control logic.

 Therefore, a logical approach may be to use a real-time system for control of the overall gas-lift system and
the optimum distribution (allocation) of gas to the wells in the system. Measured down-hole pressure could
help tremendously in determining the actual gas-lift response curve (curve of production rate vs. injection
rate) for each well. This could then be used to help determine the optimum injection strategy for each well.
Gas-Lift Best Practices and Potential Development Ideas Page 33

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

 Remember, the most important goals of continuous gas-lift are: (1) lift from as deep as possible, and (2) lift
as consistently (stably) as possible. If these two goals can be achieved and maintained on a consistent
basis, gas-lift production can be optimized. It is not desired to be continuously changing the gas-lift
injection rate in an attempt to "fine tune" the production of each well. Rather, it is desired to determine the
actual gas-lift response for each well and use this to determine the optimum gas-lift distribution strategy for
each well.

Recommended Action(s): By all means, if possible, consider using down-hole measured pressure as a key
ingredient of gas-lift surveillance and as a key parameter to help determine gas-lift response. If it is desired to
use this for on-line, real-time control, either of the wellhead back pressure or of the gas-lift injection rate, consult
with the Artificial Lift team for advice in devising the best strategy.
Gas-Lift Best Practices and Potential Development Ideas Page 34

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Valve Design/Selection/Use

The following gas-lift best practices should be considered when designing, selecting, and using gas-lift valves, orifices, and related equipment.

1 Select gas-lift valves that Specify that all gas-lift equipment, including gas-lift mandrels, valves, and associated down-hole equipment, be
have been manufactured manufactured and tested according to the applicable API (API Standard 11V1) and/or ISO specifications.
to API and/or ISO Standards
Discussion: The American Petroleum Institute (API) has developed a specification for the manufacture of gas-
lift equipment. It is called API RP 11V1. This specification may be used worldwide, but is primarily accepted
and used in the United States. The International Organization of Standards (ISO) is developing three gas-lift
standards:

 ISO 17078.1 - Side Pocket Mandrels

 ISO 17078.2 - Gas-Lift Valves
 ISO 17078.3 - Gas-Lift Latches and Running/Pulling Tools

It is the intention that, when the ISO standards have been adopted by the worldwide ISO process, they will
super-cede the API specification. For the short term, the API specification is the operative document. It is
expected that the ISO standards will be approved sometime in the 2002 - 2004 time frame.

When a company follows the API specification, or the ISO standards, it agrees to manufacture and test its
equipment according to these standards, and it is subject to technical audit to verify compliance. The
specification (and standards) are prepared to assure that the resulting equipment will be fully "fit for purpose" in
typical gas-lift operations.

The ISO standards will allow the user to choose among various "grades" of equipment. Higher grades will have
higher quality, and will cost more.

The recommended Best Practice is to insist on gas-lift equipment that is manufactured and testing according to
either the API specification of the ISO standards. This is particularly pertinent when considering less expensive
gas-lift equipment that is offered by "new" companies that may be less experienced in producing gas-lift
business. Some companies may offer so-called "equivalent" equipment. This equipment should NOT be
purchased unless it has been approved and given the "stamp of approval" by either the API or the ISO. It may
be slightly more expensive to purchase certified API or ISO equipment, but the life-cycle costs will almost
certainly be less due to a higher service life and greater performance from the "approved" equipment.
Gas-Lift Best Practices and Potential Development Ideas Page 35

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Recommended Action(s): When ordering any gas-lift equipment, insist on accepting only equipment that has
been certified by either the API or the ISO. If a company wishes to sell non-certified equipment, insist that they
obtain certification before selling their equipment to any Shell Operating Unit.

If a potential gas-lift equipment supplier wishes to investigate the certification process, they may contact the
Artificial Lift team for an introduction to the appropriate API or ISO personnel.

2 Select gas-lift valves that Where possible, select and use only gas-lift valves that have been performance tested according to the API RP
have been performance 11V2 Recommended Practice.
tested
Discussion: The American Petroleum Institute has published a recommended practice for gas-lift valve testing
and modeling. It is called API RP 11V2. This document provides recommended practices for conducting
performance tests on gas-lift valves to determine:

 Valve opening pressure

 Valve closing pressure
 Valve load rate
 Valve stem travel distance
 Valve flow coefficient
 Valve gas throughput rate

A significant number of gas-lift valves have been tested. The primary testing organizations are the University of
Tulsa TUALP (Tulsa University Artificial Lift Project) and the Valve Performance Clearinghouse (VPC), a
Petroleum Industry Joint Industry Project (JIP).

At this time TUALP is no longer active in gas-lift valve testing. And, while their tests were extensive, they did
not fully follow the API RP 11V2 guidelines.

Shell was instrumental in helping to form the VPC. Its tests are done strictly in accordance with the API
guidelines. The VPC conducts an annual gas-lift valve testing program, with the member companies agreeing
on the valves to be tested. Shell is one of these member companies. At the instruction of the member
companies, the VPC has re-tested some of the valves that were originally tested by TUALP.

Some valve testing is also conducted by the gas-lift valve companies. However, they have not been
forthcoming with their test results. One company, Weatherford, is a member of the VPC. Schlumberger Camco
has said that they plan to join the VPC.
Gas-Lift Best Practices and Potential Development Ideas Page 36

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

This gas-lift valve testing has not only defined the performance criteria of the tested valves. It has also found
some valves that do NOT perform as advertised by the valve manufacturers.

The recommended Best Practice is to only use gas-lift valves that have been tested by the VPC, and that are
recommended by the Artificial Lift team in SepTAR. A full list "tested and approved" gas-lift valves can be
obtained from the Artificial Lift team.

Recommended Action(s): Only use gas-lift valves that have been tested by the VPC and "approved" by the
Artificial Lift team in SepTAR. If an Operating Company wishes to consider an "un-tested" valve, it can be
tested by the VPC, either as part of the normal program, or as a "once off" test paid for separately by Shell.

3 Select gas-lift valves that Where possible, select and use only gas-lift valves that have a proven gas-lift performance model that has been
have a proven performance developed according to the API RP 11V2 Recommended Practice.
model
Discussion: In addition to the testing discussed in the section above, the API RP 11V2 also specifies methods
by which the performance of gas-lift valves can be modeled. These models, which are developed based on the
test results, can be used to predict when the valves will open, when they will close, and how much gas they will
handle under different conditions of pressure and temperature.

When these models exist, Shell has them entered into the WinGLUE gas-lift design, surveillance, and
optimization program so they can be used in routine gas-lift design, surveillance, and optimization work.

Typically, there is a gas-lift performance model for every gas-lift valve that is tested by the VPC. Therefore, a
recommended Best Practice is to only use gas-lift valves that have been tested by the VPC, "approved" by the
Artificial Lift team, and for which there are performance models in WinGLUE.

There are several different gas-lift valve performance models that have been generated by the VPC. The "best"
model, called the VPC model, is based on a full set of test data. If a full set of test data is not available, there
may be other models of lesser accuracy. One of these is the so-called "TUALP" model. This model is based on
the API recommended modeling process and the data gathered by TUALP. It is less accurate than the full VPC
model since the TUALP tests were not complete in gathering all of the recommended test data.

WinGLUE was developed before gas-lift valve performance models were available in the industry. In the
original WinGLUE program, the performance of gas-lift valves was estimated based on the Thornhill-Craver
equation. This equation is acceptable for "square edged" orifices, but not for gas-lift valves.
Gas-Lift Best Practices and Potential Development Ideas Page 37

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

WinGLUE still has the Thornhill-Craver equation as part of its logic. The user of WinGLUE must choose to use
the proper gas-lift performance model for the selected valve, or WinGLUE will default to use Thornhill-Craver.
So, a recommended Best Practice is to always choose and select the VPC gas-lift valve performance model
for each gas-lift valve in each well.

Recommended Action(s): When using WinGLUE, be sure to choose and select the best VPC model that is
available for the gas-lift valve being used. Do not use Thornhill-Craver except for orifices.

4 Use ceramic seats in gas-lift Consider the use of ceramic seats in gas-lift valves to avoid valve erosion during unloading.
valves
Discussion: Both ceramics and tungsten carbide are used to provide hardened gas-lift valve balls and seats to
help prevent erosion. There are several considerations that must be understood and weighed.

 Both ceramic and tungsten carbide are more expensive than conventional steel materials.
 Since both ceramic and tungsten carbide seats are harder than conventional steel, there is (or may be)
some amount of native leakage. That is, the ball may not tightly seal on the seat, even when the valve is
new.
 Even though both ceramic and tungsten carbide are harder than conventional steel, they may still erode to
some degree.

Given these considerations, it is not fully clear as to when and where the use of ceramic or tungsten carbide
balls and seats would be preferred over the use of conventional steel balls and seats. To overcome this
situation, a recommendation would be to conduct a series of tests of these "enhanced material" balls and seats
to determine:

 What is their "native" leakage rate?

 What is their susceptibility to erosion?

When these two factors have been determined, a proper evaluation can be made of when these "enhanced
materials" should be specified for gas-lift valves. It may be that they should be used in all cases, but this has
not as yet been proven.

There is a current recommendation to authorize the Valve Performance Clearinghouse (VPC) to conduct these
tests as a follow on to the gas-lift valve erosion tests is it conducting in 2000. This would be a recommended
course of action.
Gas-Lift Best Practices and Potential Development Ideas Page 38

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Recommended Action(s): Authorize the Valve Performance Clearinghouse (VPC) to conduct erosion tests on
gas-lift valves with hardened balls and seats. If the "native" leakage rate is not excessive and if the erosional
characteristics are acceptable, it may become a Best Practice to routinely use one or the other of these
materials.

5 Use gas-lift valves that Where possible, consider using gas-lift valves that don't use bellows, to avoid potential problems with bellows
don't need a bellows failures, leaks, etc.

Discussion: Most unloading gas-lift valves use a nitrogen-charged bellows. In some cases, the bellows is
augmented with a spring. In a few cases, there is no bellows and the closing force is provide solely by the
spring.

Not many performance tests have been performed on gas-lift valves that do not use a bellows. Before this
could become a recommended practice, tests should be conducted to determine the performance
characteristics of these valves over an extended period of time.

Recommended Action(s): At this time, use of gas-lift valves without bellows is not recommended. If it is
desired to pursue this possibility, such valves should be fully tested by the Valve Performance Clearinghouse to
determine their performance characteristics.

6 Use nozzle-venturi gas-lift Where pertinent, e.g. dual gas-lift wells and some single gas-lift wells, consider using the nozzle-venturi (Trade
orifices name Nova) gas-lift orifice to better control the injection rate from the injection conduit to the production conduit.

Discussion: The nozzle-venturi orifice has some attractive characteristics. It reaches critical (sonic) flow with a
very small pressure drop across the orifice. This means that, for most cases, the flow through the orifice is not
affected by the downstream pressure.

This may make this orifice especially attractive for use in dual gas-lift wells, where the desire is to operate each
side of the dual in an essentially independent manner, without interference from the other side of the dual well.

The Artificial Lift team has written and issued an extensive report on the use of the nozzle-venturi orifice. Use
of this orifice may be a Best Practice, especially in dual gas-lift wells. However, as clearly indicated in the
Artificial Lift team report, several precautions must be exercised to assure that these orifices are used correctly.
Gas-Lift Best Practices and Potential Development Ideas Page 39

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Since critical flow is reached with very little pressure drop across the orifice, a relatively narrow range of gas
flow rates can be used.

If the gas-lift injection rate is too low and the flow through the orifice is not critical, the flow will be very unstable.
If the injection rate is too high, the upstream pressure will rise and this may cause an upper unloading gas-lift
valve to re-open.

There is another concern. The "NovaTM" nozzle-venturi orifice is manufactured only by Schlumberger Camco.
To date, the price of these orifices has been exorbitant. They have recently indicated that, to enhance sales,
they are willing to discuss a program to sell the orifice at a substantially lower cost. If use of this orifice appears
warranted, the Artificial Lift team should be asked to approach Schlumberger Camco about this.

Recommended Action(s): If an Operating Unit has dual gas-lift wells, it is recommended that they carefully read
and study the Artificial Lift team report on the nozzle-venturi orifice. If use of these orifices appears warranted,
it is recommended that the Artificial Lift team be consulted on the design and operation of the wells.

If the Operating Unit wishes to make significant use of the nozzle-venturi orifice, it is recommended that the
Artificial Lift team be asked to contact Schlumberger Camco and explore ways to reduce the purchase price of
these devices.

7 Use pilot-operated gas- Where appropriate, as in intermittent gas-lift wells, use pilot-operated gas-lift valves as the operating injection
lift valves point.

Discussion: For continuous gas-lift wells, the normal recommendation is to use an orifice for the "operating"
gas-lift injection point. For these wells, the objectives are to inject the gas as deep and as stable as possible.

For intermittent gas-lift wells, the objectives are very different. The objective here is to inject the gas-lift "slug"
as rapidly as possible and then to stop injection altogether until it is time for the next intermittent gas-lift cycle.
An excellent way to do this is to use a pilot-operated gas-lift valve. With such a valve, the pilot action of the
valve causes it to very rapidly move from the "full closed" to the "full open" position. It remains full open, to
allow a high gas injection rate, until the casing pressure decreases to the closing pressure. Then, the pilot
action causes the valve to "snap" closed. This rapid open, rapid close action is ideally suited for intermittent
gas-lift. Therefore, use of properly sized pilot-operated gas-lift valves is a Best Practice for intermittent gas-lift
wells.
Gas-Lift Best Practices and Potential Development Ideas Page 40

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Pilot-operated gas-lift valves can be used with either form of intermittent gas-lift. The two forms are: (a)
continuous injection with the intermittent action controlled by the pilot-operated gas-lift valve, and (b) intermittent
injection that is controlled by a timer or an on/off surface controller. Here too the actual injection from the
annulus into the tubing is controlled by the pilot-operated gas-lift valve.

Recommended Action(s): If intermittent gas-lift is being used, or considered, it is recommended to consider use
of pilot-operated gas-lift valves to control the injection cycle. These can be used with either continuous or
intermittent injection at the surface. Contact the Artificial Lift team in SepTAR is assistance with selection and
design of valves is needed.
Gas-Lift Best Practices and Potential Development Ideas Page 41

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Unloading and/or Kick-Off

The following gas-lift best practices should be considered when designing and conducting initial unloading or kick-off (restart) of gas-lift wells.

1 Purge annulus liquid with After a completion or workover of a potential gas-lift well, purge the annular fluid with nitrogen, diesel, or
nitrogen, diesel, and/or another light fluid, to assist in the gas-lift unloading process.
another light fluid
Discussion: The primary factor that affects the design spacing of unloading gas-lift valves is the density of the
fluid in the annulus that must be "u-tubed" out of the annulus during the unloading process. The normal process
is to leave a completion fluid of treated brine salt water in the annulus. Typically, the density of this fluid is on
the order of 1.1 or slightly higher and thus its gradient is on the order of 0.5 psi per foot.

This "heavy" completion fluid dictates that the gas-lift mandrels that are designed (spaced) for the unloading
gas-lift valves must be placed relatively close together. This, in turn, may limit the ability to design (space)
mandrels deep enough to reach the "bottom" of the well, just above the packer. Since the most important goal
in gas-lift is to inject the gas as deeply as possible, if there is an inability to place the mandrels as deep as
possible, this can result in lower than desired gas-lift efficiency for the life of the well.

If a lighter fluid can be placed in the annulus of the well when it is completed, when it is worked over, or shortly
before it is initially unloaded and placed on gas-lift, and if the gradient of this fluid is known when the spacing of
the unloading gas-lift mandrels and valves is done, it may be possible to space the bottom valve lower and thus
gain in gas-lift efficiency.

Before this step is taken, all consequences of using the lighter "completion" fluid must be considered. The
completion fluid must be safe to install and safe to recover when the well is placed on gas-lift. It must provide
adequate well integrity and protection, it must provide adequate protection against corrosion, and it must not be
subject to loss due to leakage.

Another consideration, on the plus side, is that a lighter fluid may be less likely to cause erosion of the gas-lift
valve seat during the unloading process. This is discussed further in another Best Practice on valve erosion
that is covered below.

Recommended Action(s): If potential gas-lift wells are deep enough that it may be difficult to space unloading
mandrels/valves as deep as desired, it may be a Best Practice to consider using a lighter fluid (e.g. diesel,
nitrogen, or some other lighter fluid or gas) to displace the annulus and make it easier to unload to the desired
depth during the unloading process.
Gas-Lift Best Practices and Potential Development Ideas Page 42

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

2 Use foam for an unloading After a workover, consider leaving foam in the annulus of potential gas-lift wells, so that it will be easier to
fluid unload the wells on gas-lift.

Discussion: This is actually the same story as the item above. Foam could potentially be another light fluid to
consider for use during completion or re-completion, or shortly before beginning the unloading process.

Recommended Action(s): The same considerations as listed above should be considered here.

3 Unload gas-lift wells with a Consider using a swab unit to unload the completion or workover fluid from a gas-lift well, to avoid the need for
swab unit upper unloading gas-lift valves.

Discussion: Potentially, a swab unit could be used to lift part of the completion fluid out of the well before
starting the gas-lift operation. Potentially, this could be used if the top gas-lift mandrel is placed too deep to
allow initial unloading with the available gas injection pressure.

There are a number of pre-cautions that must be carefully followed if this practice is to be considered:

 Swabbing must be conducted through a properly installed "swab" valve or lubricator to assure well integrity
at all times.
 Gas should not be injected into the well's annulus during the swabbing operation. The objective is to
produce some of the completion fluid. This can be done by u-tubing the fluid. It does not require gas
injection.
 The "normal" objective of swabbing is to unload part of the "completion" fluid so the gas-lift unloading
process can start. Care must be used to not unload too much fluid so that the well can begin to produce.
Production of hydrocarbon liquids and gas could be dangerous during a swabbing operation.
 Care should especially be exercised to see that the well is not rapidly drawn down. This could endanger
the well's completion, especially if a sand control system has been installed.

In summery, the Artificial Lift team in SepTAR cannot recommend swabbing as a Best Practice except in very
special, very well supervised and controlled conditions.

Recommended Action(s): If swabbing must be considered to begin the unloading process, this process should
be approached with extreme caution. Normally, this process should only be performed enough to "uncover" the
top gas-lift mandrel so that normal gas-lift unloading can begin.
Gas-Lift Best Practices and Potential Development Ideas Page 43

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

4 Use shut-in wells for gas When shut-in wells exist in a field, consider storing high-pressure gas in their annuli so this gas can be
storage available for use when it is necessary to re-start a gas-lift field after a field shutdown.

Discussion: There are two primary reasons for the size of the gas-lift distribution system to be as large as
possible. These are:

 The larger volume can help to "absorb" pressure upsets. If the pressure in the gas-lift distribution system
tends to fluctuate due to changes or upsets in one or more wells on the system, a large volume can help to
"cushion" the fluctuation and prevent upsets in other wells.
 A large volume can "store" gas that can be helpful if the gas-lift system has to be shut down and restarted
at a later time. Most gas-lift systems are "closed" systems. That is, the gas is re-circulated, re-
compressed, and re-used. If a system is shut down, it may be difficult to obtain enough gas to restart the
system.

The idea here is to expand the volume of the gas-lift distribution system by connecting shut-in or temporarily
abandoned wells to the system and keeping their wellbores filled with high-pressure gas. Several
considerations are necessary before pursuing this option.

 If the primary objective is to "cushion" upsets, a much more effective way to do this is with a fast-acting,
automatic gas-lift control system that can rapidly react to system upsets (e.g. compressor stops/starts,
production station stops/starts, etc.) and adjust the gas-lift injection into the wells to maintain (nearly)
constant system pressure. This process has been proved to work very well in existing Shell Operating Unit
fields. It could require a very large distribution system volume to begin to accomplish the same level of
system pressure stability.
 If the primary objective is to provide sufficient gas to re-start the system after a shutdown, again a very large
distribution system volume would be needed. A far better approach is to, if possible, provide some other
source of "start up" gas, for example from a gas pipeline, gas wells in the same field, etc.
 If this approach is taken, care must be used in selecting and testing the "storage" wells. It must be assured
that there are no leaks, either at the surface or downhole that could result in gas leakage or losses.
 Typically, the volume of gas-lift distribution systems in very small fields, e.g. on offshore platforms, is very
small. This approach may be valid in this case. Typically, this approach would not be of much value in a
large distribution system such as is typically found in more land or inland marine oil fields.

In summary, this would only be a recommended Best Practice in very special cases. All of the conditions
surrounding the "storage" wells, their integrity and safety, and their suitability to achieve the objectives should
be carefully weighed.
Gas-Lift Best Practices and Potential Development Ideas Page 44

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Recommended Action(s): If there are shut-in or temporarily abandoned wells that have proven integrity, it may
be appropriate to consider using them to expand the volume of the gas-lift distribution system. However, this
should only be done with extreme caution, and the contribution of these wells should be carefully weighed
against the limitations of this approach. It may not provide sufficient additional volume or storage to be of much
value.

5 Unload a dual gas-lift For dual gas-lift wells, consider designing one side of the dual to unload the annulus and the other side to only
well with one side only have enough gas-lift mandrels and unloading valves for kick-off and normal operation.

Discussion: Unloading of dual gas-lift wells is particularly challenging. Often, one side is unloaded and takes
all or most of the gas and the other side is not unloaded and gas-lifts very ineffectively. The idea here is to
design one side of the dual well to perform the necessary unloading of the annular fluid and design the
"unloading" valves in the other side to only be used during kick-off and normal operations.

In a dual gas-lift well, the well only needs to be unloaded to the depth of the upper (dual) packer. If the higher
pressure (potentially more productive) side is used for unloading, it may be possible to use a minimum number
of unloading valves in the lower pressure (potentially poorer) side.

An important point to remember here is that the unloading issue is concerned with the number of unloading gas-
lift valves, not mandrels. Even is a well has several mandrels, dummy valves can be installed in the upper
mandrels to make them unavailable for any gas-lift operation.

The Artificial Lift team is not ready to endorse this practice as a Best Practice. It should be tried in a number of
dual wells and the results should be evaluated. The Artificial Lift team would be pleased to work with any dual
gas-lift Operating Unit to test and evaluate this process.

Recommended Action(s): If an Operating Unit has a number of dual (or potential dual) gas-lift wells to be
unloaded, it is recommended that this process be tested and evaluated. The Artificial Lift team will be pleased
to assist.
Gas-Lift Best Practices and Potential Development Ideas Page 45

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Well and/or System Operation

The following gas-lift best practices should be considered when conducting routine gas-lift well and/or system design and operation.

1 Test each gas-lift valve Test and provide quality assurance on each gas-lift valve, before it is run into the well and after it is pulled from
before and after use the well. Test it before to assure that it meets the design specifications and desired opening pressure, closing
pressure, and gas passage capabilities. Test it after it has been used to help in "cause of failure" analysis.

Discussion: Decker Technology is building a gas-lift valve verification device. The first unit is being built for
Brunei Shell Petroleum (BSP). The purpose of the device is to test each and every gas-lift valve before it is run,
and again when it is pulled. The device will test:

 The opening pressure of the valve.

 The closing pressure of the valve.
 The load rate of the valve's bellows and/or spring.
 The maximum travel of the valve stem.

With this information, it will be possible to tell if the valve meets the design specifications that are advertised for
the valve. It will also be possible to determine specific gas-lift performance coefficients for the gas-lift
performance model that has been developed for this class of valve.

It will be possible for the Operating Company to establish acceptance criteria. If a new or re-conditioned gas-lift
valve does not meet the agreed acceptance criteria, it can be rejected before it is run. This can potentially save
a failed or very ineffective gas-lift operation before it occurs.

By testing the valves both before they are installed and after they are pulled, it will be possible to evaluate the
effects of actual use on valve erosion, bellows life, etc.

When a new or re-conditioned valve is tested, and if it passes the test (i.e. meets the agreed acceptance
criteria), the actual performance model coefficients for this particular valve will be electronically transmitted to
WinGLUE where they will be used for gas-lift design, surveillance, and optimization calculations.

The current plan (Fall 2000) is to test this device in BSP. If it successfully provides the gas-lift valve testing and
screening that is planned, it will be a recommended Best Practice to use this testing device in every Shell
Operating Unit that uses gas-lift. The device can be owned and operated by the Operating Unit, or the OU can
instruct its service company to obtain and use the device on its behalf.
Gas-Lift Best Practices and Potential Development Ideas Page 46

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Recommended Action(s): The recommended action is to closely watch the initial testing of this device in BSP.
If it proves successful, it is recommended that each Operating Unit that uses gas-lift acquire or gain access to
this device so that all of its gas-lift valves can be tested both before they are installed and after they are pulled.

2 Use gas-lift valve erosion Use gas-lift valve erosion test data and the erosion model in designing/selecting gas-lift valves and in designing
test data and conducting the unloading process, to prevent erosion of the gas-lift valve seat, stem, and ball.

Discussion: Gas-lift valves are designed to handle gas injection from the gas injection conduit (normally the
annulus) to the production conduit (normally the tubing). However, during the initial unloading process, the
valves must transmit completion fluid from the injection to the production conduit. In most cases, this fluid will
be much heavier and therefore much more erosive than gas. If the flow rate of this fluid through the valve is too
high, it can erode the valve seat and/or ball and make it impossible for the valve to ever close tightly. This can
lead to perpetual gas leakage through the valve and perpetual poor performance.

In 2000, the Valve Performance Clearinghouse (VPC) is conducting erosion tests of gas-lift ports. The results
of these tests will be used to evaluate the flow rate at which erosion may occur, and to develop a model that
can be used to evaluate whether or not a specific unloading rate may lead to erosion.

The purpose of these tests, which Shell is helping to fund on behalf of the Shell gas-lift Operating Units, is to
develop Best Practices for how to unload gas-lift wells to avoid erosion. Another expected deliverable from
these tests will be a model that can be implemented in WinGLUE. With this model, the gas-lift engineer/analyst
can "design" the unloading program for a well to assure that no erosion occurs.

Another purpose of these tests may be to evaluate other gas-lift port materials such as tungsten carbide and
ceramics. If these materials are less susceptible to erosion, as expected, and if they do not suffer too high a
"native" leakage rate, they may be preferred over "normal" steel on a routine basis.

Recommended Action(s): Watch for the results of the VPC gas-lift port erosion tests. When these results are
available, use them to help evaluate the gas-lift unloading process. Insist that the model(s) that are developed
are incorporated into WinGLUE so they can be used to "design" an effective unloading process. Also insist that
these tests be run on other materials so that the use of such materials as tungsten carbide or ceramics can be
evaluated.
Gas-Lift Best Practices and Potential Development Ideas Page 47

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

3 Design how you plan to This is more a "maxim" than a best practice. It speaks to the need to be very careful in designing a gas-lift
operate, operate the way system or well to be certain that the design is fully in alignment with the way that it can be operated (in terms of
you are designed pressure, gas rate, etc.) and then that the well be operated the way that it was designed to be operated.
Anything else can lead to very inefficient and ineffective operation.

Discussion: As stated above, this is more a "maxim" or a "golden rule" than a best practice. However, the logic
is irrefutable. It a well is not designed in alignment with the way it can be operated, there is no way the design
can be successful. For instance, if the well is designed to operate at a higher pressure than can be delivered to
the well, or if it is designed for more gas than can be delivered, the well will not unload and/or operate
effectively.

On the other hand, if a well is designed to operate within a certain range of injection rates, and if the gas-lift
valves, orifices, chokes, etc. are selected accordingly, and if an attempt is made to injection either less than the
minimum design rate or more than the maximum design rate, the well will not be stable and will not lift
effectively.

So, it is an essential Best Practice that gas-lift designers (engineers, analysts, etc.) know and communicate
very closely with the people who must operate the well. And, it is essential that the people responsible for
operations fully understand and agree with the design assumptions.

Recommended Action(s): Gas-lift should be performed by a team. The team must consist of those who
perform gas-lift design, gas-lift surveillance, and gas-lift operations. These people must all understand the
entire gas-lift process, from initial design to final operation. They must closely communicate with one another to
assure that each and every gas-lift well is designed the way it can be operated and is operated the way it was
designed to operate.

4 Capture and analyze Capture and store data (CHP, THP, FBHP, etc.) during unloading and analyze this data to understand how well
unloading data (or poorly) the well was unloaded, and to improve the process for the future.

Discussion: All too often, a well is unloaded as soon as possible, to accelerate when it can be placed on
production or to minimize deferment if the well has died and needs to be unloaded on gas-lift to reinstate
production. This may be expedient, but is a shame; for the initial unloading process is one of the most critical
times in the life of a gas-lift well. If it is done correctly, the well should be able to operate from "bottom" for a
very long time. It is not done correctly, it may not be possible to ever lift the well effectively.
Gas-Lift Best Practices and Potential Development Ideas Page 48

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

The "key" questions during unloading are: (a) did the well successfully unload to the desired operating depth,
and (b) were any unloading gas-lift valves damaged during unloading? The best way to answer these
questions is to gather the appropriate data during the unloading process and evaluate this information. The key
information is the injection pressure, the injection rate, the production pressure, and if available, the flowing
bottom-hole pressure. This information should all be collected on a continuous basis (at least one sample per
minute) during the entire unloading process.

Clearly, it is an important Best Practice to insist that this information be collected during the unloading process.
This means, of course, that the data collection process must be in place and fully operational before the
unloading process starts.

In addition to collecting the information, it must be analyzed. It should be possible to use the erosion data and
model(s) discussed above to help determine if any unloading valves were damaged. It should be possible to
use the record of injection and production pressures to calculate the final operating depth.

Recommended Action(s): It is essential to the success of gas-lift that the well be successfully unloaded. It is
recommended that the key operating information be collected during every initial gas-lift unloading operation
and that this information be analyzed to determine if the well was successfully unloaded to the desired
operating depth and if this was done without damaging any unloading gas-lift valves.

5 Know as soon as possible Through an effective surveillance process, become aware as soon as possible when a gas-lift wells is no longer
when a gas-lift well's design producing optimally and change the design or operation to once again be optimal. This can include cases
or operation is no longer where a well is no longer lifting at its desired operating depth, where it has become unstable, or where it is no
OK and change it longer operating at its optimum injection/production rate.

Discussion: Many things can occur to cause a gas-lift well to no longer be effective or efficient. The "problem"
may be a problem with the gas-lift system in the well, with the surface gas-lift or production system, or with the
well itself. The key point is that no well should be allowed to operate inefficiently.

With modern computer-based surveillance (CAO) systems, it is possible to monitor every gas-lift well on a
continuous basis. It is possible to determine, continuously, if the well is stable or not, if it is producing from the
desired depth or not, if is operating at the desired operating rate or not. Any well that is producing "off
specification" is either costing too much to operate or producing too little income.
Gas-Lift Best Practices and Potential Development Ideas Page 49

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Therefore, with the availability of modern gas-lift surveillance systems, it is an essential Best Practice to always
know the status of each well, and to highlight to the field gas-lift Well Analysts or Programmers whenever any
well is not operating at its desired depth and stability.

And, it isn't enough to know that a gas-lift well is not operating effectively. The essential next step is to react
quickly to any problem and "fix" the problem so the well returns to effective operation. Any day (any hour) of
sub-standard operation looses money.

Recommended Action(s): Modern surveillance systems, combined with modern training, can equip any gas-lift
operation to continuously know how each well is operating, to know if it is not operating OK, and if it is not, to
correct the problem(s) quickly to minimize any deferment.

6 Know actual optimum Maintain a current, accurate gas-lift response curve (Production Rate vs. Injection Rate) for each well, know
point for each well - maintain each well's optimum gas-lift production vs. injection point, don't exceed this point, and attempt to control each
optimum operation, don’t well to produce as near to its optimum point as possible, as much of the time as possible.
exceed optimum point
Discussion: With the modern technologies of gas-lift surveillance (CAO), well testing, and calibrated models, it
is possible to maintain an up-to-date gas-lift response curve at all times. This curve can be used to determine
the (economic) optimum operating point (injection rate and production rate).

It may not be possible to operate each well at its optimum point at all times. There may not be enough gas to
always injection the optimum amount in all wells, all of the time. However, it is a Best Practice and it is
possible to know the optimum point, to never inject more than the optimum amount, and to keep the wells as
close to optimum as much of the time as possible.

Recommended Action(s): It is not good enough to assure that each gas-lift well is merely lifting. With modern
technologies, it is possible to know the optimum operating point of each well and to actually operate each well
as close as possible to this point most of the time.

7 Use dedicated slick-line Use dedicated wire line (slick line) crews that understand the special needs associated with installation and
crews retrieval of gas-lift equipment. Pay them accordingly.

Discussion: It takes special knowledge and skills to properly run, install, retrieve, and pull gas-lift equipment.
Gas-lift valves must be properly installed in the gas-lift mandrels and they must be properly seated. It is
Gas-Lift Best Practices and Potential Development Ideas Page 50

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

essential that they be properly seated without jarring them too hard to the point of damage. It is essential that
the correct gas-lift valves be installed at the correct depths.

In addition, there is another significant advantage to using a dedicated crew for gas-lift wire line work. It can
enable closing the gas-lift surveillance loop much more rapidly. Often the lag time between running a pressure
survey and the related change in the gas-lift design, if a change is indicated, can be weeks or months. A
dedicated wire line team, that is part of the overall gas-lift surveillance team, can help reduce this time to days,
with the critical path item being the actual gas-lift analysis and design, or the process of obtaining the gas-lift
valves themselves.

The recommended Best Practice is to use wire line (slick-line) crews that are properly trained and certified.
Not all crews have this experience and these skills.

Recommended Action(s): Establish a certification process for wire line (slick-line) crews. This should include
both training and awareness in the specific goals of gas-lift. It should contain a process for the crews to
demonstrate their knowledge and ability to perform the gas-lift processes successfully.

8 Use downhole shut-in tools When conducting a downhole pressure build-up survey, consider using a downhole shut-in tool to eliminate the
when conducting bottom- effects of after flow and well-bore storage from "masking" the effects of the pressure build-up.
hole pressure surveys
Discussion: When a well's production is stopped at the surface, the well's fluids (oil, water, and gas) will
continue to "produce" into the wellbore from the formation for some time. This can "mask" the pressure build-up
effects. A way to overcome this is to actually "shut in" the well at the bottom of the wellbore by using a
downhole shut-in tool.

There is some risk associated with this. When such as device is used, it can potentially become stuck or sand,
scale, or other materials may accumulate on top of it. So, before this process is tried, the following steps are
recommended:

 Don't use this technique if the well is known to produce some sand.
 Before this is done, make a "gauge" run to below the depth where the down-hole shut-in device will be set
to make sure that the tubing string is clear of any foreign materials.

Recommended Action(s): When running pressure build-up surveys, consider using a down-hole shut-in device
to eliminate the effects of "after flow" from masking the pressure build-up data.
Gas-Lift Best Practices and Potential Development Ideas Page 51

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Training

The following gas-lift best practices should be considered when designing and implementing gas-lift training and staff development.

1 Develop the well analysis Develop the well analysis discipline to have a group of people who focus on well productivity and well
discipline surveillance.

Discussion: Several Operating Units are developing an organization or cadre of Well Analysts to help conduct
routine gas-lift surveillance and operations. PDO has taken a lead in this, followed by SPDC and Gabon. The
Artificial Lift team in SepTAR has written a suggested job description for Well Analysts and has offered Well
Analysis training programs.

In general, Well Analysts are people who work in or close to field operations. They understand field operations
and artificial lift. They understand gas-lift design principles as well as gas-lift surveillance and operational
issues. They focus on gas-lift surveillance so they are usually the ones to first know when a gas-lift well is not
performing optimally and often they are the ones to rapidly address problems to minimize deferment.

It is clearly a Best Practice to develop and use Well Analysts to help optimize artificial lift operations, including
gas-lift.

Recommended Action(s): If your Operating Unit already has Well Analysts, use this and the following best
practices to make the most of this valuable resource. If your OU does not have Well Analysts, discuss the
needs with the Artificial Lift team in SepTAR and with PDO or one of the other OU's that use them.

2 Conduct well analysis Hold regular workshops where Well Analysts, Programmers, Production Technologists, and others come
workshops together to share best practices, worst practices, receive update training, etc. These workshops can be held
intra-asset, cross asset, or cross OU. Perhaps intra-asset and/or cross-asset workshops can be held more
frequently, with cross-OU workshop annually or bi-annually.

Discussion: Effective gas-lift surveillance requires teamwork. The gas-lift team should include the Operator(s),
Well Analyst(s), Programmer(s), Production Technologist(s), and others that are involved in the gas-lift
operation. This is essential but not sufficient for full knowledge sharing.
Gas-Lift Best Practices and Potential Development Ideas Page 52

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Much can be learned by sharing experiences with others who are involved in the same or similar work. For this
to occur, it is a Best Practice to hold periodic Well Analysis Workshops or Seminars. There may be many
different ways to do this successfully, including:

 In large Production Assets, where there may be more than one Well Analyst, Programmer, etc. hold a Well
Analysis/Programming meeting weekly to discuss problems, insights, new ideas, etc.
 In a large Operating Unit, with more than one Production Asset, hold a cross-Asset Well Analysis Workshop
on a Quarterly basis to share problems, needs, learnings, ideas, etc.
 Hold cross-OU Well Analysis Workshop on an annual or bi-annual basis. These can be coordinated by the
Artificial Lift team in SepTAR.

To make Well Analysis seminars or workshops effective, it is recommended that they be facilitated by an
experienced Well Analyst, Programmer, or Production Technologist from the Operating Unit, from another
Operating Unit, from the Artificial Lift team in SepTAR, or by a consultant. In this way, the facilitator can draw
out the best information from each participant and can add value by bringing in outside experience and/or
training.

Recommended Action(s): To augment and strengthen the Well Analysis corps in an Operating Unit, it is
recommended to hold periodic Well Analysis meetings, seminars, or workshops to share experiences, discuss
problems, explore new ideas, and receive external input and training.

3 Provide "on site" training Provide "on site" training and/or field gas-lift seminars. Make this similar to the "produce the limit" process.
Focus on a process to: (1) Find the problem, (2) Provide training on how to identify and solve the problem, (3)
Solve the problem, and (4) Perform follow-up.

Discussion: Official classroom training, in the Shell Training Center in The Netherlands or at a Consulting
facility, can be valuable. However, often this training does not reach all of the necessary staff, and often it is not
provided at the optimum time, when it is most needed.

The recommended Best Practice here is to find ways to provide training "on site" in the Operating Unit and to
make it available to all staff in the OU that can benefit from it. To make this cost effective, it may be necessary
to offer this training on a regional basis, to one or more Operating Units in a common geographical area.

Recommended Action(s): Provide "on site" training on both the principles and the practices of effective gas-lift
on at least an annual basis in each Operating Unit.
Gas-Lift Best Practices and Potential Development Ideas Page 53

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

4 Train on the job - no Develop a process to train gas-lift staff "on the job" so they can learn while doing.
classrooms - use mentors
Discussion: Parts of this "best practice" are realistic and beneficial and one other part is merely an "over the top
statement" to make a point. The three parts of this are discussed separately below:

 Train on the Job. This can be a highly effective means of training, and perhaps the most effective. It is
often very beneficial to learn while actually working on real problems on ones own actual wells. The
"problem" with this is that there are usually not enough qualified trainers to be able to provide on-the-job
training to all who need it. Therefore, this method of training must be augmented by other means as well.

Often, it may not be necessary to have a qualified trainer actually on the job, side by side with the trainee. It
may be possible to provide on-the-job training with the trainer located at a remote, central site and actually
providing training to several people at the same time. Also, it may be possible to offer effective on-the-job
training by using other means such as training manuals, training programs, videos, CD's, computer-based
training systems, on-line training systems, training simulators, mentors, etc.

 No classrooms. This idea is not serious. However, it does make a serious point. Classroom training is
essential for some forms of education. However, it is never sufficient. That is, it is never good enough to
provide a few days of classroom training and then to expect the students to be fully trained and qualified.
Classroom training must be augmented by some form of local or on-the-job training to allow the students to
become fully qualified gas-lift analysts.

 Use mentors. The idea of using mentors, or highly skilled gas-lift staff to serve as mentors of younger or
less experienced staff is an excellent one. Typically, the mentors are not "official" trainers, but are
experienced gas-lift Well Analysts or Programmers. Part of their responsibility is to provide hands-on, one-
on-one, on-the-job training for younger or less experienced staff.

Some Operating Units may wish to establish a range of Well Analyst positions through which a person can
grow over time.

 Expert Well Analyst. An Expert Well Analyst might serve an entire Operating Unit. He/she might be the
"resident" gas-lift expert to help solve difficult problems, provide gas-lift training, and act as an interface
with Other Operating Units and with the Artificial Lift team in SepTAR.
 Senior Well Analyst. A Senior Well Analyst might be the lead Well Analyst in each Production Asset.
He/she might lead the gas-lift surveillance process, address the most pressing problems, and provide
mentorship for the junior staff.
Gas-Lift Best Practices and Potential Development Ideas Page 54

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

 Well Analyst. A Well Analyst might perform the majority of the actual gas-lift surveillance work.
 Junior Well Analyst. A Junior Well Analyst might be a Well Analyst in training, working under the
tutelage of his mentor.

Recommended Action(s): No one is born with knowledge of gas-lift. It must be learned by all. It is incumbent
on every Operating Unit that has gas-lift to develop a comprehensive gas-lift development program that
includes the appropriate blend of classroom training (whether in The Netherlands, in the region, in the
Operating Unit, or in the Asset) with other forms of local, on-the-job, one-on-few or one-on-one training.

5 Use the "master/apprentice" Develop and use gas-lift "masters" (local experts) to help train new gas-lift personnel (apprentices). This can go
training technique hand-in-hand with on-the-job training.

Discussion: This item is a re-statement of the item before. Clearly the "master/apprentice" or "mentor" program
is an excellent way to provide rapid, valuable gas-lift training. It is clearly a Best Practice.

Recommended Action(s): See the item above.

6 Provide more frequent Provide gas-lift training and/or awareness classes more frequently, at least annually and perhaps quarterly in
training stead of once every few years. Include assessment testing with the training to help assure that the trainees are
learning the material.

Discussion: The process of becoming effective in gas-lift must be on going. It cannot all be learned in one
classroom session, or one on-the-job experience. For this reason, training should be perpetual. There are
many effective ways to do this:

 Provide routine (once per year or at least once each two years) classroom training.
 Provide update courses in between the formal classroom sessions. Do not always present the same
material.
 Provide gas-lift seminars or workshops (discussed above) where gas-lift Well Analysts, Programmers,
Production Technologists, and others can come together to share best practices, discuss problems and
failures, learn about new tools and techniques, etc.
 Provide mentoring programs. This must start with a program to "train the trainers." Knowledge of gas-lift
does not make a person an effective trainer. This skill must be learned just like any other skill.
 Provide training aids such as training programs, computer-based training, videos, CD's, a gas-lift training
simulator, etc.
Gas-Lift Best Practices and Potential Development Ideas Page 55

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Clearly, not all people are equally suited to perform gas-lift work. There must be effective tools to evaluate gas-
lift staff and appropriately reward and promote those that learn and perform well. For this purpose, there must
be adequate gas-lift certification programs to evaluate each person's knowledge and skills.

Recommended Action(s): As stated above, each Operating Unit that is actively involved in gas-lift should
develop and maintain an active program for training and developing gas-lift staff.

7 Use job rotation for training, Use job rotation between the field and the office. Provide field staff with an understanding and appreciation of
broadening of key gas-lift gas-lift project considerations, design issues, etc. Provide office staff with an understanding and appreciation of
staff practical field operating issues and concerns. Consider expanding this to include rotating staff between Shell
and other operating companies, service companies, etc.

Discussion: This is one of those areas where it will be best to “walk before we run.” That is, it will be best to try
gas-lift job rotation on a local basis, in an Asset, before expand it to think of wider job rotation across assets,
across Operating Units, or even with other companies.

It would be very logical to start with short-term job rotations between Programming and Well Analysis. This
could potentially be expanded to include full, one or two-year rotational assignments. In any case, it is
recommended that each job rotation be associated with a form of mentoring process so that the person that is
new to an assignment area can learn as rapidly as possible.

The idea of expanding the concepts of job rotation beyond the bounds of an Operating Unit are intriguing. The
most logical next step could include job rotations, or cross postings, across Operating Units. An even more
interesting possibility could be to investigate a rotation between a Shell Operating Unit and another Operating
Company, or a Service Company. At this level, it would be recommended to work with the Artificial Lift team in
SepTAR.

Recommended Action(s): Short or even longer-term job rotations can be an excellent way to gain and share
learnings. It is recommended that this be considered in at least the major gas-lift Operating Units. If there is
interest in more extensive forms of job rotations, the Artificial Lift team in SepTAR would be happy to discuss
how this might be approached most effectively.
Gas-Lift Best Practices and Potential Development Ideas Page 56

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Knowledge Acquisition

The following gas-lift best practices should be considered acquiring additional knowledge about gas-lift and gas-lift best practices.

1 Implement a gas-lift valve Implement a data base of gas-lift equipment that contains both operation and failure (e.g. tear down)
teardown database information. Design this to be used to gain understanding of causes of failures and procedures to overcome
them.

Discussion: This idea is the same as one of the Potential Development ideas discussed below in the Gas-Lift
Valve Design, Selection, and Use category. The title of the idea is “Use a tracking system for gas-lift
equipment – design, performance, failures, costs, deferment.” Please refer to that section for a discussion of
the idea.

Recommended Action(s): See the recommended actions in the companion section below.

2 Host an Artificial Lift Hold an artificial lift (or gas-lift) industry workshop with attendees from other Operating Companies, Service
industry workshop Companies, Consultants, Universities, etc. to greatly expand the possibilities for generating new ideas and
sharing both best and worst practices.

Discussion: Originally, it was intended to invite representatives of Service Companies, Consultants, and
Universities to attend the gas-lift Value Engineering Workshop. This idea was disallowed by the Shell Legal
Department due to concerns about sharing information in the area of downhole gas-lift measurement, control,
and optimization.

This workshop did discuss this area to a limited degree. But much more of the focus was on gas-lift Best
Practices, and other areas that can advance the gas-lift technology and business.

It would seem feasible, and should be acceptable to Shell Legal, to fashion another gas-lift workshop with a
wider attendance. It might be necessary to limit the range of the discussion to avoid certain sensitive areas.
However, based on the large number of good Best Practices and Potential Development ideas that were
generated at this workshop, it seems very likely that many more worthwhile ideas could be generated from a
broader and more experienced audience.

Recommended Action(s): The recommended action is to evaluate the results and longer-term effects of this
current workshop after a year or two. If the results of this workshop have proven to be highly worthwhile and
Gas-Lift Best Practices and Potential Development Ideas Page 57

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

beneficial, as is expected, then it may be appropriate to consider holding another workshop, this time with a
broader attendance.

3 Look outside Shell for best Look to the nuclear industry, the space industry, and other "high tech" areas to find "best practices" that may
practice ideas apply to gas-lift.

Discussion: This idea is easy to say, but difficult to practice. A logical first step would be to start in our own
industry. There is already clear precedent to perform benchmarking and to share best practices with other
Operating Companies.

One way to do this is at the annual Gas-Lift Workshop that is sponsored by the American Society of Mechanical
Engineers (ASME) and the American Petroleum Institute (API). This meeting is held annually, usually in
Houston, Texas, in late January or early February.

Another way to do this is to organize a specific benchmarking exercise with one or a group of Operating
Companies. If such a session is carefully organized and facilitated, it can be an excellent learning experience
for all involved and can avoid any of the pitfalls that find disfavor with the Legal Department.

This discussion of “starting at home” in the Petroleum Industry is not made to discourage the possibility of
seeking input from other industries. This may be a fruitful exercise if one or more people can find the time and
resources to give it a try.

Recommended Action(s): Benchmarking and sharing best practices with other Operating Companies is clearly
a Best Practice and is recommended. It can be practiced on a regional basis with leadership by one or a local
group of Operating Units. Or, it can be practiced on a global basis with leadership by the Artificial Lift team in
SepTAR. In either case, it is certainly recommended as a good way to learn.

4 Use feedback from Use feedback from actual operating experience in the design and selection of gas-lift equipment and operating
operating experience procedures.

Discussion: This idea should be a given in every Operating Unit. It has already been discussed above.
Teamwork, with the team including members of Operations, Well Analysis, Programming, Production
Technology, Engineering, and others is essential for effective gas-lift. A key component of every team
experience must include sharing information on what works, what does not work, what is needed to improve
performance, etc.
Gas-Lift Best Practices and Potential Development Ideas Page 58

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Recommended Action(s): As stated elsewhere in this document, every gas-lift designer should fully
understand, appreciate, and accommodate the operational issues associated with the gas-lift system and wells.
And, every gas-lift operator and analyst should fully understand the gas-lift design assumptions and objectives
and strive to achieve them to optimize gas-lift performance.
Gas-Lift Best Practices and Potential Development Ideas Page 59

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Potential Gas-Lift Development Ideas

Gas-Lift Well and/or System Implementation
1 Use small bubbles to There is already a “small bubbles” project that is part of the 2000 BAA process. This project is clearly well
enhance gas-lift, create supported by the OU’s. This item and the next three items all speak to this issue. All of these OU’s
with sintered injection expressed interest in being willing to help test this concept when it is ready for field-testing.
point or ultrasonics
Sponsoring Operating Units: PDO, SPDC, Shell EXPRO, Woodside, Norske Shell

Discussion: There are several concepts for creating and/or injecting small bubbles in the production steam.
These include the idea of using a special gas-lift valve, of using an insert in the tubing string to create the small
bubbles, or of using some other means to create and/or sustain the bubbles.

Recommended Action(s): It is recommended that this investigation be carried out by SepTAR as part of the
BAA Process. It is recommended that the Operating Units not attempt to undertake this project on their own.
However, when it becomes time for field-testing, involvement of the OU’s will be very welcome.

2 Use electric tubing to This is an extension of the idea above.

create vapor bubbles
with arcs along the tubing,
control on pressure
Sponsoring Operating Units: BAA Process

Discussion: See the discussion above.

Recommended Action(s): See the recommended actions above.

3 Use rail gun arcs, which This is an extension of the idea above.
are switched up hole, to
create and sustain the
small bubbles
Sponsoring Operating Units: BAA Process
Gas-Lift Best Practices and Potential Development Ideas Page 60

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Discussion: See the discussion above.

Recommended Action(s): See the recommended actions above.

4 Use a metal or plastic This is an extension of the idea above.

tubular that is porous
to gas at one point to
generate small bubbles
Sponsoring Operating Units : BAA Process

Discussion: See the discussion above.

Recommended Action(s): See the recommended actions above.

5 Use an orifice that One of the primary objectives of the surface controlled gas-lift valve project is to be able to change the
changes size downhole, size of the downhole gas-lift valve (or orifice) to adjust the rate of gas injection. In reality, the adjustable
combine with electric device will be a valve (orifice) that can be adjusted over a wide range of openings to allow a wide
gas-lift valve range of injection rates.

Sponsoring Operating Units: SPDC

Discussion: A number of concepts are being pursued for surface-controlled gas-lift valves/orifices. These
include controlling these devices with electrical signals, hydraulic signals, and wire-less power and
communications. All of these thrusts are being steered and/or monitored by the iWELL team in SepTAR.

Recommended Action(s): Since this project is already well underway, under the direction of the iWELL team,
the Operating Units should follow their lead and not attempt to “steer” this project on their own. It is
recommended that the lead of the iWELL team be followed and that these devices be used, as appropriate,
when they have been developed and proven.
Gas-Lift Best Practices and Potential Development Ideas Page 61

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

6 Use a moveable injection The most important objective of gas-lift is to inject the gas as deeps as possible. This “desired” depth is a
point or slot function of the gas injection pressure and the gradient (weight) of the fluid being produced. The idea of
this initiative is to be able to move the point of injection to the optimum depth without having to “work
over” the well to move the depth of the “bottom” gas-lift mandrel.

Sponsoring Operating Units: SPDC, BAA Process

Discussion: On first glance, the idea of a moveable injection point would be extremely difficult and expensive to
achieve. However, there might be practical approaches, such as using multiple gas-lift mandrels in the
expected vicinity of the desired lift depth and having an easy way to move the point of injection by changing the
gas-lift mandrel/valve that is currently operational.

It would appear likely that some other technologies should be developed and proven before this idea is pursued
further. These might include the idea of surface controlled gas-lift valves, automatically run and retrieved gas-
lift valves, etc.

Recommended Action(s): This appears to be a revolutionary idea. It is recommended that this idea be held by
the Artificial Lift team in SepTAR for future consideration. It is recommended that an Operating Unit not attempt
to pursue this idea on its own.

7 Use reducible tubulars The idea here is to be able to adjust the size of the production conduit to the optimize size for the rate that
can be produced by the well. The reason for mentioning “reducible” tubulars is that often, in older wells,
the size of the production conduit is too large to sustain efficient continuous flow. Often, there is a desire
to reduce the size of the production conduit either with a smaller tubing string or by inserting a small string
(e.g. coiled tubing) inside the production conduit.

Sponsoring Operating Units: SPDC, BAA Process

Discussion: The idea of having an adjustable tubing size is great, but probably “out of reach” in the near term,
due to costs. One feasible idea might be to design a well with a permanent “insert” string, produce the well up
the annulus between the insert string and the production conduit, and use the new “expandable tubular”
concept to expand the size of the insert string, when and as needed, to achieve optimum annular flow.

Recommended Action(s): This idea would clearly require a significant amount of research and development
before it could become field worthy. Therefore, this should remain in the sphere of the BAA process and be
ranked with other projects before any funding by individual Operating Units.
Gas-Lift Best Practices and Potential Development Ideas Page 62

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

8 Use gas-lift and ESP This is not really a “new development” idea. There are already existing “combination” systems in use in
simultaneously to reduce the world where pumping and gas-lift are used in the same well.
gas-lift requirements
Sponsoring Operating Units: PDO, Shell EXPRO, Woodside, Norske Shell

Discussion: The issue here is not one of new development, but one of proper application. It is already feasible
to implement both pumping and gas-lift in the same well. In fact, many offshore (and in particular sub-sea) ESP
applications are already equipped with a back-up gas-lift system.

The issue here is one or appropriate application. Clearly, there is additional cost associated with implementing
both pumping and gas-lift in the same well. Presumably, the well would be pumped, with gas-lift used to assist
in lifting the fluid above (downstream of) the pump. The reason for this is that, since pumps don’t like gas, it
would be preferred to pump the well first and use gas-lift to assist this operation rather than the other way
around.

This raises several issues, including determining the depth at which gas could be injected. Since gas-lift is
most effective when the gas is injected deep in the well, and since the pressure immediately downstream of
(above) the pump is high, it may be difficult to inject the gas deep enough to obtain effective gas-lift.

Recommended Action(s): To summarize, this item does not require research and development, but it requires
very careful engineering and economic analysis to determine if it will be economically feasible and attractive in a
well. If desired, the Artificial Lift team in SepTAR can assist with evaluating candidate wells.

9 Use gas to power a This is also not a “new” idea that requires research and development. One example of this is the use of
downhole pump gas to “power” a jet pump. This has been investigated and reported by C-FER.

Sponsoring Operating Units: PDO, BAA Process

Discussion: C-FER has presented a paper and has proposed to conduct additional technical and economic
feasibility of this idea. The idea is to use high-pressure gas (e.g. gas-lift gas) to “drive” a downhole jet pump.
The jet pump would used to help produce the well.

Based on initial studies, this could potentially help produce wells with a low bottom-hole pressure and/or a poor
productivity. These are the same types of wells that are typically produced by intermittent gas-lift. According to
Gas-Lift Best Practices and Potential Development Ideas Page 63

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

the studies conducted to date, this technology might be more effective in lifting these wells than intermittent
gas-lift. For one thing, the injection and production processes would be continuous rather than intermittent, with
the associated continuous changes in injection pressure and rate and production pressure and rate.

Based on the results reported to date, the case for this idea is not compelling. Before it is pursued in Shell, the
results of further work being conducted by C-FER should be evaluated. Is this makes both technical and
economic sense, a good place to test this could be Thai Shell where intermittent gas-lift is currently being used.

Recommended Action(s): Review the results of the proposed C-FER studies before spending money on this in
Shell. If a field trial appears warranted in Shell, consider conducting a test in Thai Shell.
Gas-Lift Best Practices and Potential Development Ideas Page 64

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Automation

1 Use an automatic control This idea is very similar to Best Practice #5 in the Gas-Lift Automation area. In that case, the proposed
system to optimize pro- “input variable” is the downhole pressure. Here it is the production rate as determined by the Delta P
duction (Delta P) of each technique.
well
Sponsoring Operating Units: PDO, Shell EXPRO, Woodside, Norske Shell

Discussion: The same discussion as presented with the Best Practice is pertinent here. It may not be a good
idea to continuously adjust the injection rate (or the producing well-head back pressure) of a gas-lift well. This
control could “fight against” the higher-level control of the gas-lift system and result in a continuously unstable
situation.

However, as mentioned before, it should be a very good idea to use the estimated production rate of the well to
determine the “real time” gas-lift response curve (Production Rate vs. Injection Rate). This curve could be used
to determine the optimum response curve of the well and this could then, in turn, be used in the real-time
allocation of gas to each well.

And, as discussed in other documents, there are many reasons to implement the Delta-P technique, or some
similar means to estimate production rate on a continuous basis. These include;

 Determine each well’s production rate on a continuous basis.

 Determine immediately if a well is experiencing any deferment.
 Determine estimated daily production.
 Determine estimated production during a well test, for well test validation.

In summary, it is a very good idea to estimate the production rate of each well on a continuous, real-time basis.
And, it is a very good idea to use this information to determine the current (real time) gas-lift response curve.
And, it is a very good idea to use this curve to determine the optimum allocation of gas to each well.

However, before this information is “fed” into an on-line, real-time automation system to automatically adjust the
injection rate in real time, this should be carefully developed, tested, and evaluated.

This idea does not require new research and development. SPDC is already implementing a gas-lift automation
system in which they are using the Delta-P technique to estimate the production rate of each well on a
continuous basis. However, some new CAO software will be required to fully use this information to maximum
advantage in the gas-lift process.
Gas-Lift Best Practices and Potential Development Ideas Page 65

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Recommended Action(s): It is definitely recommended to pursue real-time estimated of each well’s production
rate with the Delta-P or some similar techniques. It is definitely recommended to use this information to help
determine the optimum gas-lift control strategy. It is not recommended to use this signal as an input to an on-
line, real-time gas-lift control logic until this logic has been very carefully developed, tested, and evaluated.

It is recommended to work with the Artificial Lift team in SepTAR to develop the optimum strategy to use real-
time production information in conjunction with a gas-lift automation system.

2 Link CAO and modeling This idea is very similar to the one above, but it may go a step further. An additional idea here is to link
tools in real time a real-time gas-lift automation (CAO) system to other models such as WinGLUE to help determine and
(e.g. IDUN) diagnose gas-lift problems on a real-time basis.

Sponsoring Operating Units: SPDC, Shell EXPRO, Woodside, Norske Shell

Discussion: No actual research or development is required to accomplish this goal. It can be done by making
the necessary enhancements to Shell’s gas-lift automation (CAO) system and its gas-lift models such as
WinGLUE. This process has already started. If desired, it can be continued with minimal risk and effort.

Recommended Action(s): The sponsoring Operating Units can make this item happen by requesting the
Artificial Lift and Production System Technology teams to define the specific links and having Shell Services
International (the company that provides CAO programming) and AppSmiths I(the company that provides
WinGLUE programming) make the necessary programming additions.

3 Use automatic gas-lift The idea to run and retrieve deep gas-lift valves with a tractor or some other automatic technique to be
valve retrieval/replacement able to reach depths or locations that can't be reached with wireline.

Sponsoring Operating Units: PDO

Discussion: There are cases, especially in highly deviated, horizontal, and/or multi-lateral wells where it is not
possible to run wireline tools to the desired deepest point of gas-lift. In these cases, the depth of gas-lift is
usually restricted to the depth that can be reached by wireline.

It could be very favorable in such cases to be able to install gas-lift valves and run other types of well operations
in depths where wireline can not be used.
Gas-Lift Best Practices and Potential Development Ideas Page 66

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Recommended Action(s): The SepTAR iWELL team is already exploring potential sources and uses of devices
such as tractors for performing wireless operations in wells. The work of the iWELL team should be closely
followed. They should be encouraged to make this one of their development objectives.
Gas-Lift Best Practices and Potential Development Ideas Page 67

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Valve Design, Selection, and Use

1 Implement an independent The idea here is to create and independent agency of some form to monitor gas-lift equipment design,
gas-lift equipment Quality manufacturing, testing, shipment, installation, etc. to assure that this equipment is developed according to
Assurance process and accepted industry standards such as those developed by the American Petroleum Institute (API) and the
use spot-checks International Standards Organization (ISO).

Sponsoring Operating Units: SPDC

Discussion: Currently, only one company (Schlumberger Camco) builds gas-lift equipment with the API
certification. Weatherford and Altec claim to follow API specifications but do not have the API certification.
Other companies do not even make this claim. (Note. Currently, work is underway to produce ISO standards.
The intent is to eventually replace the API specifications with the ISO standards.)

Even when a company does claim to produce gas-lift equipment according to API or ISO standards, there is no
independent agency to verify this claim. A company that claims API or ISO certification is subject to audit, but
this happens infrequently.

The idea here is to create an independent agency that can represent the gas-lift Operating Companies by
randomly checking gas-lift equipment, from any manufacturer, and test it against accepted API or ISO
standards.

Recommended Action(s): If pursuit of this idea is requested by the sponsoring Operating Unit, the Artificial Lift
team can work with one or more consultants in the gas-lift industry to create this independent agency. The idea
would be to have this agency be supported by the Operating Companies in the form of a Joint Industry Project
(JIP).

2 Use a tracking system for The idea with this is to develop a system to track the performance of gas-lift equipment before it is
gas-lift equipment – design, installed, as it is installed, while it is in use, when it is pulled, and after it is torn down. Experience with
performance, failures, other types of equipment has shown that there is much to be learned from the collection and evaluation
costs, deferment of this information.

Sponsoring Operating Units: SPDC, Shell EXPRO, Woodside, Norske Shell

Gas-Lift Best Practices and Potential Development Ideas Page 68

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Discussion: Currently, a tracking system is being developed for Electrical Submersible Pumps by an industry-
sponsored joint industry project (JIP). This system is called the ESP-Reliability Information and Failure
Tracking System or ESP-RIFTS.

The idea here would be to develop a similar system for gas-lift. Brunei Shell Petroleum is staring the
development of such a system. It may be possible to leverage from this development to have a system that can
be used across all Shell gas-lift OU’s. If this does not work out, it may be possible to develop an industry
system that is comparable to the ESP-RIFTS system.

Recommended Action(s): The recommended action here is to follow the work of Brunei Shell Petroleum and
evaluate their system. If this proves to be useful for the Shell OU community, then this may be the best
approach. If this does not work out, the sponsoring OU’s can ask the Artificial Lift team in SepTAR to pursue
the development of a Gas-Lift-RIFTS system that is comparable to the one for ESP’s.

3 Have new gas-lift designs A large number of people are involved in gas-lift design, surveillance, and operation work. A significant
reviewed by a panel/team portion of the gas-lift Best Practices deals with training these people. But, this will not be a quick process.
or by a “peer” review Therefore, this essence of this idea is to provide gas-lift teams with a system of “check s and balances” to
support them in their work.

Sponsoring Operating Units: PDO, SPDC

Discussion: If an Operating Unit would like to have their gas-lift designs reviewed by a panel of experts or
peers, this could be arranged through the Artificial Lift team in SepTAR. The team could come from with the
Operating Unit, from another OU, from SepTAR, or from a consultant.

Recommended Action(s): If the sponsoring OU’s wish to pursue this idea, they can contact the Artificial Lift
team in SepTAR.

4 Sponsor a gas-lift design There have not been very many advances in gas-lift equipment technology in the past several years. Yet,
competition for gas-lift there are several problem areas such as unloading, valve leakage, valve interference, etc. The idea here
valves is to sponsor a design competition with the objective of sparking some new, innovative ideas in the gas-lift
equipment business.

Sponsoring Operating Units: Shell EXPRO, Woodside, Norske Shell

Gas-Lift Best Practices and Potential Development Ideas Page 69

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Discussion: Other industries have been successful in making significant advances by coming up with new,
revolutionary (break through or game changer) ideas. Sometimes those on the “inside” in an industry can only
think of ways to do the same things, maybe a little better. Sometimes it takes others, with a fresh perspective,
to come up with new, fresh ideas.

The idea here is to sponsor a design competition among universities, think tanks, consultants, and others to
potentially address some of the long-standing problems in the gas-lift business. Some of these problems
include:

 Difficulty in successfully unloading to the desired operating depth.

 Difficulty in keeping unloading valves closed after the unloading process is finished.
 Valve leakage.
 Interference between valves in dual gas-lift wells.
 Valve bellows leakage and/or failure.

Recommended Action(s): If the sponsoring Operating Units wish, the Artificial Lift team in SepTAR could
sponsor a design competition among selected universities, think tanks, consultants, etc. To be effective, the
scope of the competition should be limited to a few objectives – for example, to design a gas-lift valve that will
fully close when it is not in use, with no possibility of leaking.
Gas-Lift Best Practices and Potential Development Ideas Page 70

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Unloading and/or Kick-Off

1 Use advanced gas\foam The idea here is to use foam sticks to restart flowing production in a naturally flowing well that has
sticks to help kick-off wells stopped production for some reason. This is only pertinent in those cases where the well is not (yet)
equipped for normal gas-lift operations.

Sponsoring Operating Units: SPDC

Discussion: It is sometimes (often?) the case that naturally flowing wells are not (yet) equipped for artificial lift.
If one of these wells stops producing, some process such as swabbing, temporary gas-lift, rocking, etc. must be
used to return it to natural flow. Often this is a short-term measure before an artificial lift system (gas-lift,
pumping, etc.) can be installed.

For several reasons, swabbing and rocking are not good practices. They can be expensive, and they can
potentially damage the well or the well’s completion.

Potentially, foam sticks can be administered quickly and easily and sometimes they can be effective. A foam
stick is dropped into the well. When it reaches bottom, it disintegrates, creating foam that is much lighter than
the water in the bottom of the well. This allows the natural pressure of the well to start the well flowing again.

Use of foam sticks should never be considered a long-term solution. If a well is subject to periodically loading
up and stopping production (dying), artificial lift should be installed. This is a good idea, even if the well might
continue to produce on natural flow for some time. It may be (is likely) that the production rate of the well can
be increased, with artificial lift, above the rate that can be achieved with natural flow.

Recommended Action(s): If an Operating Unit has a number of natural flowing wells that are not (yet) equipped
with artificial lift, it may be a good idea to have foaming sticks on hand and to try them to return a well to natural
flow. However, at the same time (or preferably even much earlier), plans should be make to install artificial lift
in the well to optimize the production and minimize any deferment.

2 Punch holes and use E2 Another possible way to restart naturally flowing wells that are not (yet) equipped for artificial lift is to
TECH patches to assist in punch a hole in the tubing, protect the hole with an insert or patch, and inject gas through this hole to
gas-lift kick-off restart production.

Sponsoring Operating Units: SPDC

Gas-Lift Best Practices and Potential Development Ideas Page 71

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Discussion: This method will only work if there is a source of gas available to the well. Even then, it should be
considered only as a temporary expedient, and not as a long-term solution. Any hole in the tubing, no matter
how well it is “protected” by an insert or patch, is a potential source of leakage and perhaps even worse (e.g.
casing damage.) So, if this must be done as an expedient to return a naturally flowing well to production, work
should begin immediately to install artificial lift in the well.

Recommended Action(s): If there is an existing source of injection gas at a naturally flowing well, if the well
must be restarted in some fashion, and if foam sticks or some other non-intrusive process will not work, this
method may be considered as a temporary expedient. It should never be used as a long-term solution for gas-
lifting a well.

3 Splice the gas-lift conduit One of the greatest problems in dual gas-lift wells is the interference that often occurs between the two
to the long string for sides of the dual. In many cases, one side “robs” most of the gas and the other side is lifted very
unloading dual gas-lift wells in effectively, if at all. The idea here is to isolate the two strings but injecting down the annulus for one
side and down a separate, small conduit for the other side.

Sponsoring Operating Units: SPDC

Discussion: This is a good idea IF it is feasible and practical. There are several considerations that must be
carefully weighed before this is attempted:

 First, there must be room in the wellbore for the additional string of pipe.
 Second, it must be possible to inject enough gas through the (normally very small) spliced conduit. When
gas is injected through a small pipe, there may be a very large pressure drop that may inhibit the desired
rate of gas injection.
 Third, It must be possible to “kick off” the side of the dual that is lifted with the spliced string. Potentially, the
well can be unloaded with the other side of the dual, but it must be possible to kick off each side
independently.

Recommended Action(s): If use of this method is desired, it is recommended that operation of the well be
carefully modeled beforehand to assure that the desired rate of gas can be injected down the spliced tube with
out excess pressure loss. Also, Well Services should be consulted to be certain that the extra (third) string of
pipe can be accommodated in the wellbore.
Gas-Lift Best Practices and Potential Development Ideas Page 72

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Well and/or System Operation

1 Use tracer technology to A technique has been developed to detect the operating gas-lift valve by injecting a tracer in the gas-lift
detect the operating gas- gas and measuring the time it takes for the tracer to be returned to the surface. This technique has been
lift valve, if legal issues can licensed to Halliburton, so it may not be used without their permission.
be overcome
Sponsoring Operating Units: PDO

Discussion: It has been shown that the operating gas-lift valve can, at least in some cases, be detected by this
method. There are several considerations that must be weighed before this is attempted on any significant
scale.

 First, this idea, which was developed by Shell Research in the USA, has been licensed to Halliburton. Their
permission must be obtained before it can be used.
 Second, it is not clear that this method can be used to detect multiple operating valves. Thus, its use may
be limited to those wells that are operating through only one valve or orifice. If a well is operating through
multiple valves and/or a valve and an orifice, the results may be non-conclusive.
 Third, this is obviously not a continuous process. If other continuous means can be used, such as with a
permanent downhole pressure measurement, or even with accurate surface pressure measurements and a
good wellbore model, this may be preferable to using a periodic technique.

Recommended Action(s): If there are gas-lift wells where the operating depth is in doubt, and where there are
not sufficient other measurements to estimate to gas-lift injection depth(s) with reasonable accuracy, this
method could be tried. If there is a desire to try it, the Artificial Lift team in SepTAR can make the necessary
arrangements with Halliburton for a trial. If it turns out to be a method for long-term use, the Artificial Lift team
can negotiate the appropriate arrangements with Halliburton.

2 Install whistles in gas-lift The idea here would be to listen to the sound of the gas as it flows through each gas-lift valve. If the
valves with different pitches were different for each valve, and if there were a way to detect this sound, it could be possible to use
to determine operating gas- this sound (or these sounds) to determine which valve(s) were open and transmitting gas into the
lift valve production conduit.

Sponsoring Operating Units: PDO, SPDC, BAA Process

Discussion: It would seem, with modern technology, that it might be preferable to determine the operating gas-
lift valve(s) with accurate downhole or even surface pressures and models. However, this does not work
Gas-Lift Best Practices and Potential Development Ideas Page 73

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

acceptably well in some cases, and if the previous idea of using a tracer either does not work or is not available
to Shell for some reason (e.g. unfavorable license arrangements with Halliburton), this method could be tried.

Several issues would need to be addressed:

 How to obtain the different pitches in a reliable manner without creating an unacceptable pressure drop.
 How to “hear” the various pitches.
 How to detect the situation where more than one valve would be open and there would be multiple pitches
at the time.

Recommended Action(s): It is recommended that accurate pressure measurements be the first method used to
determine the operating gas-lift depth(s). The reason is that the pressure data has so many other beneficial
uses as well. The second approach should be the tracer mechanism. If neither of these methods prove
satisfactory, then this method could be tried. It would require a significant amount of research, both to create
the multiple pitch whistles, and to develop the pitch detection device.

3 Use distributed temp- This idea is already under test in Shell EXPRO. It may have application in some wells, especially multi-
perature for rate lateral wells where it is desired to know where the flow is coming from.
measurements
Sponsoring Operating Units: : Shell EXPRO, Woodside, Norske Shell

Discussion: It may be possible to determine or estimate the flow rates, or at least where the flow is coming from
in a multi-lateral well, with the use of a fiber-optic temperature measurement. It may also be possible to
determine where gas is being injected into the production conduit by the same means.

Recommended Action(s): Current field trials of this method should be followed. If they prove successful, this
should be considered on wells where downhole instrumentation and/or control is being installed.

4 Use an inexpensive meter Normally it is desirable to install (reasonably) accurate meters to measure the rate of gas injection into
on each well and allocate each gas-lift well. This measurement is not only important to allocate gas to each well. It is also important
gas via a single system for gas-lift surveillance. However, there are some cases, e.g. sub sea completions, where it is not
meter possible to install and maintain accurate meters on each well. In these cases, a simple measurement on
each well, such as a simple differential pressure technique, may be acceptable.

Sponsoring Operating Units: : Shell EXPRO, Woodside, Norske Shell

Gas-Lift Best Practices and Potential Development Ideas Page 74

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Discussion: Typically, gas-lift gas is measured with an orifice meter or a turbine meter. However, in some
situations such as sub sea wells, this may not be feasible. In such cases, however, it may be feasible to use a
more simple measurement technique to give sufficient accuracy.

Recommended Action(s): Each case should be evaluated based on its own conditions and merits. If
reasonably accurate measurements can be used, they should be. However, if they cannot be used for some
reason, a more simple method should be considered. An approximate measurement is better than no
measurement at all.

5 Use a flapper check valve Gas-lift valves use a “back check” valve to prevent fluid from flowing from the production conduit to the
instead of normal check injection conduit during those times when the production pressure is higher than the injection pressure.
Valves Typically, ball valves are used. It is a common complaint that these valves leak.

Sponsoring Operating Units: : Shell EXPRO, Woodside, Norske Shell

Discussion: The idea here would be to use a “flapper” valve instead of a ball valve for the “back check” valve.
It isn’t clear if this would be an improvement or not. The flapper would have to be hinged, and it would have to
seal against a seat that could also erode.

In any case, if such an idea were to be considered, it should be developed by one of the gas-lift companies, not
by Shell.

Recommended Action(s): If the sponsoring Operating Units are interested in pursuing this idea, the Artificial Lift
team in SepTAR could contact the three primary gas-lift valve companies (Camco Schlumberger, Weatherford,
and Altec) and ask them to consider this idea and its feasibility. It is appears feasible, Shell should offer to test
gas-lift valves with “flapper” check valves if they are manufactured. The first tests should be conducted by the
Valve Performance Clearinghouse to determine if the valves will leak, and to determine their flow
characteristics.

6 Use a hand-held trouble- The idea here is to use a hand-held computer that the field Operator can carry with him when he/she is
shooting box with checking the wells. It should contain current and historical information on each well and should be able
knowledge and historical to help the Operator to troubleshoot problems encountered in the field.
data
Sponsoring Operating Units: SPDC
Gas-Lift Best Practices and Potential Development Ideas Page 75

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Discussion: The premise for this device is that the field Operator is the one to perform first line surveillance on
the gas-lift wells. This may or may not be the case. In many locations, it is the Well Analyst who performs first
line surveillance.

The underlying premise of this case may true for fields that do not have gas-lift automation (CAO). For
automated fields, the key gas-lift information should be available on the desktop of the Well Analyst and he/she
should be the best equipped to perform gas-lift surveillance.

However, if there is no gas-lift automation, and if all gas-lift data gathering is performed manually, it may be very
valid to equip the field Operator with enough information and tools to perform first line surveillance.

Such a hand-held device would need to be extremely reliable and user friendly. It would also need to be very
secure. In some fields, similar devices have been known to disappear on a frequent basis.

Recommended Action(s): As stated elsewhere, the recommended practice is to automate gas-lift operations
and to continuously monitor the key information needed for first line gas-lift surveillance. However, if a field is
not now (and can not be for some time) automated, it may be very beneficial to provide hand-held computers to
assist the field Operators. There is ample precedent for developing and using such devices. If the sponsoring
Operating Unit wishes to pursue this idea, the Artificial Lift and Production System Technology teams can
provide information on potential sources.

7 Use surface controlled This idea is already being pursued by the iWELL team in SepTAR.
valves for deviated and
horizontal wells
Sponsoring Operating Units: SPDC

Discussion: The iWELL team in SepTAR is already pursing this idea. There are a number of potential
approaches, including electrically operated, surface controlled gas-lift valves, electrically-operated valves with
the electrical signal being transmitted in a wireless manner via the tubing and casing, and hydraulically-
operated valves.

Recommended Action(s): The recommended approach here is to follow and support the work of the iWELL
team. When one or more of the approaches is ready for field test, they will welcome having one or more
Operating Units in which to test these valves.
Gas-Lift Best Practices and Potential Development Ideas Page 76

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Training
1 Link the gas-lift demon- The existing Shell gas-lift demonstrator is a PC program intended to demonstrate many typical gas-lift
stration program to operating conditions and problems. It is an excellent training tool. But, it is based on purely hypothetical
CAO/SCADA – seeing is situations. The idea here would be to line the demonstrator to CAO so it could be used to demonstrate
Believing “real” gas-lift situations.

Sponsoring Operating Units: PDO

Discussion: The gas-lift demonstrator is a wonderful tool for visualizing gas-lift operations and problems. If this
tool could be integrated with the gas-lift automation (CAO) system, it could potentially display actual operating
conditions. This could greatly enhance its training value and turn it into a powerful diagnostic tool.

Making this “link” would require several steps:

 The first step would be relatively easy. This would be to “feed” the actual surface pressures and rates (from
CAO) into the demonstrator and to display these values.
 The second, and more difficult, step would be to use these “real” pressures to simulate the downhole
operation of the well. This could be done by using the WinGLUE vertical pressure model .

Recommended Action(s): If the sponsoring Operating Unit wishes to pursue this item, the Artificial Lift team in
SepTAR could contact AppSmiths and Shell Services International. Working together, with AppSmiths working
on the demonstrator and SSI working on the CAO system, they should be able to perform this integration for a
reasonable amount.

2 Develop and use a portable The Artificial Lift team in SepTAR has already prepared a detailed specification for a portable gas-lift
gas-lift model/simulator model/simulator. They have already obtained quotations for this device. And, SPDC has already
expressed a high degree of interest in obtaining one or more simulators.

Sponsoring Operating Units: SPDC

Discussion: This model/simulator can be built. The cost for the first unit will be $100,000. The cost of
subsequent units will be about $50,000.
Gas-Lift Best Practices and Potential Development Ideas Page 77

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

The idea of the simulator, which will be approximately six feet tall, will be to [physically demonstrate many of the
important gas-lift operations and problems – unloading, kick off, normal operation, unstable operation, etc. It
should be an excellent training tool.

Recommended Action(s): If the sponsoring Operating Unit, and/or other OU’s, is interested, this item can be
pursued quickly by activating the quotation that the Artificial Lift team has already received. (Actually, the
quotation was received by Shell Services International at the request of the Artificial Lift team – contact Jim
Hall.)

3 Build a gas-lift start-up This idea can be based on either the gas-lift demonstrator (Item #1 above) or the physical gas-lift
simulator – require operators simulator (Item #2 above), or both. This training tool would simulate the gas-lift unloading process
to qualify on it before and kick-off process.
undertaking operations
Sponsoring Operating Units: SPDC

Discussion: Unloading is a critical time in the life of any gas-lift well. If it is done correctly, the well unloads to
the “bottom” gas-lift mandrel and begins lifting normally. If it is not done correctly, one or more of the unloading
valves may be damaged, and/or the well may not work down to the desired depth. Both conditions can lead to
very ineffective gas-lift, and especially if one or more of the unloading valves is damaged, it may never be
possible to work down, until the damaged valve(s) are replaced.

Recommended Action(s): If the sponsoring Operating Unit wishes to pursue this item, the Artificial Lift team in
SepTAR can contact AppSmiths. They can add start-up simulation logic to either the gas-lift demonstrator or
the gas-lift simulator, or both.

4 Use a simulated environ- This item can also be based on either the gas-lift demonstrator or the simulator. This training tool would
ment to understand how simulate various gas-lift problems and provide the trainees with an opportunity to learn how to diagnose
to recognize and fix broken and correct these problems.
wells
Sponsoring Operating Units: SPDC

Discussion: The idea here is to crate a learning environment to train gas-lift Operators, Well Analysts, and
others to recognize, diagnose, and fix gas-lift operating problems.
Gas-Lift Best Practices and Potential Development Ideas Page 78

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Recommended Action(s): If the sponsoring Operating Unit wishes to pursue this item, the Artificial Lift team in
SepTAR can contact AppSmiths. They can add simulation of gas-lift problems to either the gas-lift
demonstrator or the gas-lift simulator, or both.

5 Use a simulator to feed a Once the two items above have been completed, the resulting simulator (or demonstrator) can be linked
neural net to learn failure to a neural net program and “teach” it to recognize and diagnose gas-lift problems. This neural net
symptoms for “smart” program could then be implemented in the gas-lift automation (CAO) system for real-time gas-lift problem
diagnosis recognition, diagnosis, and development of solutions.

Sponsoring Operating Units: BAA Process

Discussion: This idea would be a logical outgrowth of the two items above. If they are successfully developed,
it should be very feasible to take this next step and complete the loop with a full gas-lift diagnosis system.

Recommended Action(s): If AppSmiths addresses the above two items under the direction of the Artificial Lift
team in SepTAR, they would be in an ideal position to address this item as a follow-on project.

6 Build a full-fledged gas-lift This item is a logical outgrowth of the above five items. When they have been developed and tested,
demonstration program there will be the basis for a full-fledged gas-lift demonstration program.

Sponsoring Operating Units: PDO

Discussion: This should be the final goal. As each of the above five items are undertaken, it should be with a
view to ending up with a full-fledged gas-lift demonstration program that can be used for training and
certification of all gas-lift staff.

Recommended Action(s): No action is recommended on this item at this time. This should wait for completion
of the above five items. Then it can be addressed directly.
Gas-Lift Best Practices and Potential Development Ideas Page 79

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Gas-Lift Knowledge Acquisition

1 Publish gas-lift success The idea here is to publish both gas-lift success stories, and failures, on a routine basis, in various
and failure stories Shell publications such as the EP Newsletter, the Artificial Lift Newsletter, and other appropriate
publications such as the Journal of Petroleum Technology, World Oil, etc.

Sponsoring Operating Units: SPDC, Shell EXPRO, Woodside, Norske Shell

Discussion: It was clear at the Gas-Lift Value Engineering Workshop that the importance of gas-lift is not fully
recognized and appreciated by Shell management. This has an impact on funding for gas-lift developments,
gas-lift research, gas-lift staffing, gas-lift training, etc.

It is important to keep the gas-lift story fresh and before the key decision makers in the Shell Operating Units
who are responsible for funding and staffing gas-lift operations.

Recommended Action(s): Publish at least one gas-lift success story and/or failure at least three times per year
in the EP Newsletter. Publish gas-lift stories in every issue of the Artificial Lift Newsletter. Publish at least one
gas-lift story each year in an industry publication.

2 Create a Shell “value of A well-made video can be another way to get the gas-lift story before key technical and management staff
gas-lift” video in Shell.

Sponsoring Operating Units: PDO, Shell EXPRO, Woodside, Norske Shell

Discussion: A well-made video should tell the convincing gas-lift story but it should be short enough that
management groups will be willing to view it at management meetings in the Operating Units and in SepTAR.
The total length should be no more than 15 or 20 minutes. The video should be professionally written,
produced, and acted.

Recommended Action(s): If the sponsoring Operating Units wish to pursue this item, the Artificial Lift team in
SepTAR should contact with a professional video company to produce a high quality product. Several people in
the Operating Units, SepTAR, and various consulting companies can help to provide information. But, the video
should be written, produced, and acted professionally to have maximum impact.
Gas-Lift Best Practices and Potential Development Ideas Page 80

Best Practice or
Num. Potential Development Short Discussion and Recommendation(s)

Prepared by:
Cleon Dunham
SepTAR, EPT-ASP
Artificial Lift and Production System Technology Teams

Reviewed by:
Jim Hall, SSI
Production Computer Assisted Operations