OTC 20820

Taking the Gas Lift Valves to a New Level of Reliability

Jørn Andre Carlsen, Statoil; Øyvind Stokka and Erling Kleppa, PTC

Copyright 2010, Offshore Technology Conference

This paper was prepared for presentation at the 2010 Offshore Technology Conference held in Houston, Texas, USA, 3–6 May 2010.

This paper was selected for presentation by an OTC program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been
reviewed by the Offshore Technology Conference and are subject to correction by the author(s). The material does not necessarily reflect any position of the Offshore Technology Conference, its
officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Offshore Technology Conference is prohibited. Permission to
reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract mus t contain conspicuous acknowledgment of OTC copyright.

Abstract commencement of an activity or operation by description

Gas lift valves are an integral part of the tubing in gas of the required well barrier elements to be in place and
lifted wells. Many operators use tubing with premium specific acceptance criteria.”
threads for these wells. Additional safety equipment,
including the packer and the downhole safety valve, has
been subjected to a program intended to prove its
capabilities as a safety device. This is not the case with
standard gas lift valves, developed and delivered according
to the governing standard, ISO 17078-2. On the contrary,
this standard states that the valves are only intended to be
a flow check and not a pressure safety device. This paper
discusses the development of a new validation standard,
where the intention is to prove the equipment’s safety
capacity, as well as new equipment developed to meet
these challenges.

Introduction
Most operating companies design their wells according to
external or internal regulations with respect to safety
equipment and safety levels. Each component needs to
meet material standards and specific equipment standards.
The standards will also describe production requirements
as well as acceptance testing procedures for the equipment
in question. New equipment typically has to be qualified to
a certain validation procedure described in a standard
issued by an external standards organization and often to
additional internal company requirements.

Barrier Philosophy
The safety philosophy of preventing uncontrolled release
of hydrocarbons is often referred to as barrier philosophy.
The barrier philosophy for a given well typically depends
on region, nationality and operating company. The Norsok
D-10 is an example of a standard describing elements Fig. 1 The barrier schematic for a Statoil subsea
preventing hydrocarbon or pressure release to the well. The inner pressure containing envelope is
atmosphere and it defines well barriers as; “Well barriers managed by the blue items, i.e. the production packer,
are envelopes of one or several dependent well barrier liner, tubing, gas lift valve and downhole safty valve.
elements preventing fluids or gases from flowing The outer barrier is managed by the production casing
unintentionally from the formation, into another formation and the subsea horizontal tree system, marked with
or to surface. The well barrier(s) shall be defined prior to red.
2 OTC 20820

A key element in this philosophy is that the barrier ISO 10432 Subsurface safety valve equipment. In
elements are accepted on the basis of acceptance criteria. particular, the focus was to investigate what elements were
The acceptance criteria are often described in standards for present in the standards that answered the requirement of
known equipment. Examples of standards describing the Norsok D10 standard of being a “description of a
acceptance criteria are; “ISO 10423/API 6A Drilling and specific acceptance criteria”.
production equipment –Wellhead and Christmas tree
equipment” or the “ISO 14310:2008 Petroleum and The main missing item in ISO 17078-2 (gas lift valves), is
natural gas industries -- Downhole equipment -- Packers the lack of leak acceptance criteria comparable to other
and bridge plugs”. The purposes of such standards are to safety elements in the well. Additionally, a validation
define the process of qualification and production of the program that substantiates the probability for the valve to
equipment. The qualification process will normally relate maintain its safety capability throughout the lifetime of the
to the required safety level of the equipment based on its well was also missing.
criticality and the expected lifetime of the equipment.
Statoil stated that “The gas lift valve(s) shall together with
As an example the ISO 10423 describes a series of tests the tubing be part of the primary barrier”. In light of this,
for one qualification level and also a leak criterion for this the same leak acceptance criteria as used for downhole
level, e.g. the PSL level 3G describes a leak criterion of packers and bridge plugs was selected. The current criteria
zero visible bubbles for 15 minutes. In a similar manner was the V1 criteria, describing the acceptable leak limit as
the “ISO 13679 Procedures for testing casing and tubing 20 cc gas per 10 minute hold period, taken from the ISO
connections”, has a leak criterion for gas of 9 cc/15 min. 14310 standard.

In summary, these standards identify the equipment to be a The developed testing program can be divided into four
part of the safety system for the well and describe phases. Three phases, including back check and water flow
processes and procedures in order for this equipment to be testing, were executed at the International Research
fit for its purpose. Institute of Stavanger (IRIS). The gas flow test phase was
done at K-Lab, a full scale laboratory owned by Statoil,
The ISO 17078-2, Flow control devices for side- located at the gas facility plant at Kårstø, Norway.
pocket mandrels
If one is not familiar with the ISO 17078-2 standard, one Phase 1. Open and close (back check) tests at ambient and
may be misled to believe that this standard is also referring elevated temperature, with and without spring installed.
to safety equipment in the same manner the above
mentioned standards do. This is not the case and the Phase 2. Unloading tests (water), 600 bbl, 1.5 bbl/min.
standard itself points out that “These devices are designed Back check leak tests at each 200 bbl with industrial water
and intended to prevent reverse flow through a flow and gas.
control device. They are not designed nor intended to be a
part of the safety system”. Phase 3. Full scale gas flow test at 1885 psi. 100 open and
close cycles with back check leak test after each 10th open
In trying to design a system based on a standard of and close cycle and a minimum of 140 hrs at maximum
qualified barrier elements the content of ISO 17078-2 gas flow circulation 250,000 m3/d. The flow period was
presents a problem. Other safety elements within the well for future tests set to 24 hrs.
construction, like the packer, tubing and wellhead, are
establishing procedures to qualify safety equipment, while Phase 4. This final function test with water and gas is a
the ISO 17078-2 is stating that the equipment is not “to reptetition of the tests performed in phase 1.
provide a tight shut-off pressure safety seal”.
Qualification of equipment for the new validation
A chain is no stronger than its weakest link, and in this criteria
case a weak link will be the gas lift valves qualified to ISO The qualification testing started in 2005 by testing eight
17078-2. One will also be in breach with Norsok D-10 or current designs from various suppliers. It soon proved
similar standards, as the valves do not offer a “tight shut- difficult to achieve a successful test result, even with some
off pressure safety seal”. modifications to the valves. In 2007 Statoil evaluated the
test program in order to see if the test program was too
Development of a new qualification requirement ambiguous. The conclusion from this evaluation was to
for gas lift valves continue the same test program with no modifications.
By having the above facts in mind Statoil decided to
establish a validation procedure that is in line with It was decided to invite new companies to present possible
common practice for other safety devices. The current valve designs. These designes were evaluated and it was
standard for gas lift valves, ISO 17078-2 was combined decided to go further with a design from Petroleum
together with the ISO 14310 Packers and bridge plugs and Technology Company (PTC).
OTC 20820 3

The first step was to review what designs needed to be The design was divided into the following
implemented for the gas lift valve to be a safety barrier for activities
the intended lifetime. The lifetime of a gas lift valve had
been set to 12 years, and several common failure modes Fluid calculations utilizing Computational Fluid
were noted. The top ranked failure modes were: Dynamics (CFD) analysis, including particle tracking
analysis. The potential for flow cutting was investigated
A. Flow cutting in sealing seats due to unloading using a CFD program. It was found that the flow capacity
the well. Flow cutting is mainly a problem when for the valve could be increased both for gas and liquid
unloading the annulus of completion fluids. The rates. Later tests confirmed the increased capacity and that
valves will have an upper limit for tolerable fluid the valve allows for more gas flow than the Thornhill-
velocities that does not cut the metal. The API RP Craver equation estimates. The test program has also
11V5 describes a method of unloading a well. If verified that the model’s prediction of seal surfaces
this method is followed, the valves will most remains intact after flowing of gas or liquids, even with
likely not be damaged. A case by case evaluation liquids containing a certain amount of solids.
has to be done to compare the actual fluids and
control measures against the procedure. It has not
been possible to identify if the damaged valves
have been subjected to unloading outside of this
recommendation. However, the more resistance a
valve has against fluid erosion, the lower
probability it should have for damage, thereby
having a potential to reduce the time for
unloading the well. Based on this information, the
design basis was a desire for
increased unloading rates and more important, an
increased resistance against erosion.

Fig. 2 The Gas Lift valve developed under this

B. Spring fatigue failure. The spring needs to be
project can be retrofitted in all side pocket
able to lift the dart in a vertical position. It should
mandrels made according to ISO 17078-1
also have some strength in order to overcome
potential smaller debris deposits that increase the
required closing force. There was also a need to
reduce the utilization level of the spring in order
to minimize fatigue problems.

C. Damage during installation. A number of

valves were reported as being damaged during
installation. The conclusion was that an improved
external seal system was required.

D. Multi phase injection or wet gas erosion

problems. The gas was in some cases wet, and it
seemed that a potential for micro particle
movement was an issue in some wells, perhaps
due to insufficient well cleaning. As a result, the
design should allow for a level of multiphase
flow and particles

E. Damage to valve seat due to chattering in low

delta pressure applications. Most spring closing
valves will have a potential for valve chattering in
a closing or opening mode. This chattering has a Fig. 3 Velocity calculations done by Computational
potential to damage the seals. A system to reduce Fluid Dynamics
the effect of valve chattering was implemented.
4 OTC 20820

Detailed risk analysis. The Failure Mode, Effect and After completion of the Statoil program, PTC decided to
Criticality Analysis (FMECA) were the preferred pump an additional 600 bbl of water mixed with 4 kg/m3.
qualitative risk analysis for the detail design. The objective The valve also passed this erosion test at a zero bubble per
with FMECA is to identify potential component failures 10 minutes gas test criterion.
and reduce the risk to an acceptable level through risk
reducing actions. This is achieved by identifying The valve developed under this program has also been
component functionality, component failure modes, subjected to additional testing initiated by other operating
causes, consequences, risks (criticality) and need for risk companies. One test focused on pumping water at and
reducing actions. back check testing at a zero bubble criterion. The other test
focused on pumping a completion fluid based on salt
Prototyping. Prototype valves were made to perform according to a recipe leaving salt particles in the brine.
qualification testing and to optimize manufacturing. A set The valve was the first to pass the tests for both
of quality control requirements is built into the companies.
development and delivery of the gas lift valves. Valves in
various types of materials were made and were taken Results from the use of barrier qualified gas lift
through initial testing. Ultimately, one gas lift valve type valves
was selected for the main test. The valves have been installed on a large number of wells
with very good results. It is difficult to measure gas
1 ½” OD orifice valve (orifice range: 1/8” to ½”) backflow in the field due to time requirements,
 True metal-to-metal seals (no soft seals) temperature effects and volumes involved. However, it has
 Valve design is optimized for seal area protection been possible to measure a distinct improvement in wells
(flow erosion) and fields known for having considerable problems with
 Spring activated valve with positive sealing leaking gas lift valves:
 Spring is protected from flow
 Vibrations are eliminated by design One company had wells in a field where a fluid was
 25-180°C working temperature envelope (V0 injected with the injection gas for production reasons. This
tested at +2 to +121C) fluid formed droplets that moved with very high velocity
 Designed for 10 000 psi working pressure through the gas lift valves. This eroded the valves and the
 Spring can be pre-set to various opening wells experienced a lifetime of only 3 months for the gas
differentials lift valves. When the valves developed under this program,
the SafeLift valves, were installed, the leak problems
stopped and the first valve stayed in the well for over 2.5
FAT testing years before it was pulled due to orifice changes. The
valve did not show any wear signs and it is also worth
The program has also developed a process for Factory
noting that this was the prototype validation valve that had
Acceptance Testing to ensure that all valves delivered to
been subjected to the testing at IRIS and Kårstø prior to
the field are capable of delivering safety barrier features
the installation.
when leaving the factory. The FAT program is stricter as
compared to the requirements in ISO 17078-2, and each
Another company had issues with wells involving
valve has to pass a gas test criterion of 20 cc/10 min at a
injection of fluid in the gas. Undetermined problems with
pressure of 100 psi and 10 000 psi.
the gas lift valves occurred. The lifetime of the valves in
these wells was also about 3 to 4 months and by changing
Extended erosion testing. After completing the new
to SafeLift valves the lifetime was increased to more than
validation program, the question of how well the valve
1.5 years.
would performe in an erosive liquid media remained. A
separate erosion testing program was created. The erosive
Conclusions
fluid was made by mixing water with Barite. The program
The current “ISO 17078-2, Flow control devices for side-
was divided into two steps where the Barite concentration
pocket mandrels” is not an appropriate standard for wells
was increased for each step from 0.5 kg/m3 to 2 kg/m3. A
needing the gas lift valve to be a part of the barrier
total of 1,200 bbl were pumped at a rate of 1 bbl/min and
envelope and to be a pressure safety valve. The standard
polymers were added to keep the Barite particles in
deviates in philosophy from standards validating
suspension. The program was intended so simulate a clean
comparable safety devices.
well before unloading where the cleanliness of the well
was defined to be less than 150 Nephelometric Turbidity
A new validation program for gas lift valves has been
Units (NTU ).
created. The test is more aligned with validation programs
for comparable safety equipment.
The valve being subjected to the erosion fluid performed
at a zero bubble per 10 minutes gas test criterion
Strong evidence from both the laboratory and the field
throughout the test.
suggest that the valve developed to the validation program
OTC 20820 5

is considerably more robust and has an extended lifetime

in harsh conditions. The increased robustness and lifetime
will increase the safety level of the well and reduce
intervention time and cost.

Acknowledgements
The authors will like to express gratitude to Statoil and
Petroleum Technology Company for accepting publication
of this material. Gratitude is also offered to the laboratory
personnel at International Research Institute of Stavanger,
that has spent months of testing of the equipment, as well
as the personnel at K-Lab (Statoil full scale gas test
laboratory).

References
ISO 10423/API 6A Drilling and production equipment –
Wellhead and Christmas tree equipment.

ISO 13679 Procedures for testing casing and tubing

connections

ISO 14310:2008 Petroleum and natural gas industries --

Downhole equipment -- Packers and bridge plugs

ISO 17078-2, Flow control devices for side-pocket

mandrels

Norsok D-10 Well integrity in drilling and well operations