SPE-181275-MS

Successful Development and Implementation of 9 5/8 Subsurface Safety

Valve for High Pressure, High Rate Gas Wells

Ali Al-Muslim, Karam Al-Yateem, and Hamed Nafa, Saudi Aramco; Thomas Swan and Amr Abdelhamid, Halliburton

Copyright 2016, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Annual Technical Conference and Exhibition held in Dubai, UAE, 26-28 September 2016.

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

Abstract
Because the need for energy is increasing worldwide, new technologies must emerge to accommodate this
increase. In the Kingdom of Saudi Arabia, supplying natural gas to satisfy local energy demand presents
a large challenge. When offshore high pressure and high temperature (HPHT) gas fields, Hasbah and
Arabiyah, were under development, this local high demand and the means to fulfill it had to be considered.
One of the necessities associated with safety and involving the greatest possible gain was the idea of the 9
5/8-in. big bore completion. The combination of size and rating (9 5/8 in. completion equipment rated for
10,000 psi) was not available in the market at that time; consequently, it had to be developed specifically
for this project. This paper discusses the development, testing, and qualification for a very important part
in the well completion: the 9 5/8-in. 10,000 psi Tubing Retrievable Safety Valve (TRSV).

Field Development Drivers

To safely produce hydrocarbons from oil and gas wells, equipping the well with a Sub-Surface Safety Vavle
(SSSV; TRSV or Wireline Retreviable Safety Valve (WRSV) is essential overall, but especially for offshore
installations and onshore populated areas. The valve and its control system are designed to automatically
shut in the well if catastrophic events are detected. The valve development also had to be completed quickly
and efficiently, because offshore HPHT discoveries were being exploited rapidly and effectively. Saudi
Aramco, a leading oil and gas operator worldwide, was eager to accept this challenge and develop this
equipment, which is the world's first TRSV of this size and pressure rating.

Subsurface Safety Valve Overview and Project Challenges

The subsurface safety valve is a critical item in any well completion. It is typically the first completion tool
in the tubing string installed below the wellhead; it provides fail-safe isolation of the reservoir from the
surface in the event of a well control loss event.
Because of the challenges presented by the field development, the TRSV base platform selected for the
field development is a premium valve that is currently available from a global safety valve provider. This
type of premium TRSV has been used on many high pressure, high rate gas well installations, although not
in the size and pressure that was required in this application.
2 SPE-181275-MS

The safety valve selected is a rod piston type valve with 100% metal-to-metal well containment of the
reservoir when the valve is closed. Fig. 1 shows a diagram of a TRSV in the closed position. Metal-to-metal
well containment provides the most reliable sealing for critical applications.

Figure 1—TRSV with rod piston actuation and 100% metal-to-metal reservoir pressure control/isolation sealing.

The TRSV supplier was approached about a 10,000 psi rated 9 5/8-in. TRSV. At the time, the most highly
rated 9 5/8-in. TRSV available was 5,000 psi. Many of the 5,000 psi units had been deployed globally.
A feasibility study was performed by the supplier to understand the design envelope and challenges
associated with achieving the desired output. The following items were identified as challenges:

• The desire to minimize the outer diameter (OD) to work within the special 14-in. casing inside
diameter (ID) (with or without bypass lines) to set the OD limit of the TRSV and the design
parameters to define the resultant ID. The maximum OD was 12.125 in., which is only 0.105 in.
(2.7 mm) more than the 5,000 psi TRSV. The resulting ID was 8.25 in. for the lowermost seal bore
of the TRSV, which is only 0.187 in. (4.7 mm) less than the typical 5,000 psi version.
• To achieve the aggressive OD/ID ratio at the 10,000 psi rating vs. the 5,000 psi version would
require a very high strength (140 ksi or greater) metallic material for critical components.
• Gas slam capabilities up to 200 ft/sec velocities would require further analysis and validation for
critical components to confirm the expected performance in the slam shut scenario of high pressure,
high rate gas.
• Inclusion of a proven self-equalizing feature into a high-pressure, large bore TRSV application.

• The produced fluids were sour; consequently, the material selection and compatibility
understanding would be critical to success.

Metallic Material Selection

To achieve the OD and ID targets recommended from the feasibility study, extra high yield strength material
(140 ksi) was used for the analysis. A relatively new material, corrosion resistance Nickel alloy, was the
primary selection because it met the high yield strength requirements and was suitable for sour service
conditions. It was known that the wells in one of the two fields would include elemental sulfur, and
insufficient data or history existed to validate the new material for this environment.
To understand the performance of the new material in the presence of elemental sulfur, the TRSV supplier
performed slow strain rate (SSR) testing in this environment. Other considerations in this testing included
the temperature (275°F) and the chloride content. Two different formation water samples were available for
the wells with elemental sulfur; one sample included a lower chloride content (less than 60,000 mg/l), and
the other sample had a higher chloride content (greater than 120,000 mg/l).
Two separate SSR tests were performed with elemental sulfur. One test used the lower chloride content
environment, and the other test used the higher chloride content. The lower chloride test result was
acceptable, whereas the higher chloride test result was unacceptable.
Based on the SSR testing of the new material, the following selections were made for metallic material
selection:
SPE-181275-MS 3

• For the wells with no elemental sulfur, the new alloy met the requirements and would be used for
critical components that need the greater yield strength.
For the wells with elemental sulfur, the poor test results for the new alloy used in environments with higher
chloride content indicated that it would not be a good choice should the wells produce waters with the higher
chlorides. Consequently, a different high strength material was selected: Nickel Alloy UNS N07716. This
selection was based on the guidelines of National Association of Corrosion Engineers (NACE) MMR0175 in
which 625+ (precipitation-hardenable alloy) is specified as suitable for use in elemental sulfur environments
in temperatures up to 350°F.

Design Considerations
A key area of any safety valve is the flapper and seat closure mechanism; this is the primary feature
that provides reservoir isolation when needed. For this reason, a high level of design effort is expended
to understand the interaction and forces/stresses that are present. Because of the complexity of this
particular flapper-seat design, traditional engineering calculations could not be used to adequately verify
its strength and sealing capabilities. Finite element analysis (FEA), provides engineers with a more
complete understanding of the stress patterns and intensity, and enables them to make design adjustments.
Consequently, the flapper and seat were modeled during the design process, and FEA was run at the
maximum (10,000 psi) pressure differential (Fig. 2) and at low (200 psi) pressure differential (Fig. 3) to
confirm suitability at both high and low pressure extremes.

Figure 2—FEA results of the flapper at 10,000 psi loading.

4 SPE-181275-MS

Figure 3—FEA results at 200 psi loading.

The optimization of the flapper and seat design through the FEA techniques provided a high level of
assurance that the components would work as desired during validation testing.
The flow dynamics encountered during a high rate gas slam closure event provided another area of
analysis for the flapper-seat. In extreme high rate events, flow stream effects can create pressure differentials
across the flapper body that can defeat the flapper closure spring force; if these effects are sufficiently high,
they may prevent the flapper from closing. Computational fluid dynamics (CFD) techniques were used to
analyze these effects to enable them to be understood before validation testing. Modeling the flapper and
surrounding housings in the flow environment enabled the optimization of the surrounding bore contours
to ensure a positive closing bias at the maximum conditions. Fig. 4 shows the flow velocity contours
determined from CFD modeling analysis; Fig. 5 shows the flow velocity vectors around the flapper that
were determined from CFD modeling.

Figure 4—Flow velocity contours from CFD modeling analysis.

SPE-181275-MS 5

Figure 5—Flow velocity vectors around flapper from CFD modeling.

After the flapper and seat analysis was completed and the full valve design was finalized to meet
functional specifications, a prototype valve was built for validation testing. Fig. 6 shows the flapper and
seat subassembly.

Figure 6—Flapper and seat subassembly (pen used to show relative scale).
6 SPE-181275-MS

Validation Testing
A key area to validate through testing is the capability of the valve to function properly in high rate gas slam
conditions. In this condition, the well is on production and producing a high rate of gas, and the safety valve
is called on to "slam" closed, to provide reservoir isolation in an emergency well control loss scenario. To
validate this functionality, gas slam testing was performed at two different flow facilities.
The first facility used was Southwest Research Institute (SWrI) in Texas, USA (Fig. 7). The SWrI facility
has a gas system pressure of approximately 2,000 psi and flows nitrogen from pressurized tanks. The
velocity that can be achieved at SWrI is near that of the gas velocity during the early well life when reservoir
pressures are high.

Figure 7—9 5/8 in. TRSV rigged up in the vertical position for gas testing at SWrI.

Two slam closure tests were performed at SWrI, as shown in Table 1; the valve successfully closed off
the flow without sustaining damage to the valve.
SPE-181275-MS 7

Table 1—High rate gas closure parameters at SWrI.

Run Target Tubing Actual Time to Post Gas Flapper Leak

# Flow Velocity Flow Close (sec) Slam Flapper Test Press (psig)
(ft/sec) (MMscfd) Leakage (SCFM)

1 Max 34 192 8 0.3 1217

2 Max 43 224 8 0.6 1212

The second flow facility used was GL Noble Denton, located in the UK. It has gas system pressure of
approximately 800 psi and uses a continuous supply of natural gas from the national gas grid in England.
The velocity that could be achieved at GL Noble Denton would exceed the gas velocity rates that occur
later in the life of the well when reservoir pressures are lower. Fig. 8 shows the vertical position rig-up for
TRSV testing at the GL Noble Denton facility.

Figure 8—9 5/8 in. TRSV rigged up in the vertical position for gas testing at GL Noble Denton.

In the initial gas slam tests in the UK, the hydraulic control line pressures during the valve closures
exhibited a pressure spike at the moment of flapper closure. The TRSV was modified to reduce the
magnitude of the pressure spikes to remain within the system control pressure rating limits. After the TRSV
was modified, the testing was repeated.
In addition, to more accurately record and understand the nature of the pressure pulse, three pressure
transducers were used in the test set-up to record control line pressures. One pressure transducer was located
at the valve (Fig. 9); the other two other transducers were located 5 and 10 ft from the valve. Data showed
that the pressure pulse was very short in duration and that it reduced in magnitude away from the valve.
8 SPE-181275-MS

Figure 9—Transducer located at the control line fitting immediately next to the TRSV.

During the gas slam testing in the UK, the gas velocity achieved exceeded the gas velocity of 160 ft/sec
expected later in the life of the well when reservoir pressure is depleted. The maximum velocity tested was
207 ft/sec. Table 2 provides the results of the gas slam testing from the GL Noble Denton facility.

Table 2—Gas slam testing results from the GL Noble Denton facility.

Run Target Tubing Actual Time to Post Gas Flapper Leak

# Flow Velocity Flow Close (sec) Slam Flapper Test Press (psig)
(ft/sec) (ft/sec) (MMscfd) Leakage (SCFM)

1 100 102 219 10 0 850

2 200 190 389 10 0 850
3 200 207 418 10 4.5 850

The post-slam testing also included a drift test and a 10,000 psi flapper test, which were successfully
passed. The TRSV was then disassembled, and no abnormalities were found.
In summary, a very robust validation program was successfully performed on the 9 5/8-in. 10,000 psi
rated TRSV, as described in the following list:

• API 14A validation testing

• Gas slam testing in the USA

SPE-181275-MS 9

◦ Multiple gas slams up to 224MM scfd and up to 43 ft/sec

◦ Drift test after gas slams

◦ 10,000 psi flapper gas test

• Gas slam testing in the UK

◦ Multiple gas slams up to 418MM scfd and up to 207 ft/sec

◦ Drift test after gas slams

◦ 10,000 psi flapper gas test

• High pressure hydraulic piston chamber integrity test

◦ 25,000 psi test on a production valve with no deformation

◦ 40,000 psi test on prototype valve with no deformation

◦ Piston seals previously validated to 25,000 psi gas at 450°F

The completion of this testing enabled the project to move into the next phase, which included building
and delivering the production versions of the 9 5/8-in. 10,000 psi TRSV for well installations.

Field Implementation of 9 5/8 in. 10,000 psi TRSV

The new TRSV was deployed in the completion designs for the Hasbah and Arabiyah offshore gas fields.
The valve is function tested immediately after the installation. Typically, the valve is opened and closed two
or three times; the number of strokes to open the valve and the oil return from the control line when it is
closed are recorded, as shown in Fig. 10. The data before and after cleanup are then compared to ensure
consistency. Finally, a negative test is performed by closing the TRSV and the bleeding pressure above it; the
wellhead pressure (WHP) is then recorded against time to confirm that the valve can reliably hold pressure,
as shown in Fig. 11. During the field start-up and before putting the wells on production, some TRSVs were
found to be leaking above the API acceptable rate of 15 scf/min. Fortunately, this issue was addressed with
troubleshooting by cycling (opening and closing) the valve several times until an acceptable leak rate was
achieved. This solution was implemented with no need to flow the well. Table 3 provides an example; in
this particular case, the valve was cycled four times, then it was closed and the pressure above was bled off.
The leak rate was calculated every five minutes, then it was calculated in a 30 minute duration; the leak rate
significantly reduced with time, and an acceptable leak rate was achieved in less than 45 minutes.
10 SPE-181275-MS

Figure 10—A chart showing a typical TRSV function test.

Figure 11—TRSV negative test.

Table 3—TRSV troubleshooting results.

Cycle Duration, min Leak Rate SCF/min

1 5 90
2 5 29
3 5.25
4 30 9.6
SPE-181275-MS 11

Conclusion
Saudi Aramco is the first oil operator worldwide to successfully use the 10,000 psi 9 5/8 in. TRSV in its gas
fields. This was achieved by careful qualification testing; the valve passed all required tests, including API
14A validation testing and gas slam testing in the USA and UK. After installation, the valve was tested, in
both functionality and ability to hold pressure to confirm reliability. During field start-up, the self-equalizing
feature worked normally, the high leak rate was fixed by exercising the valves, and no other issues were
encountered.

Acknowledgements
The authors are grateful to Saudi Aramco and Halliburton for allowing this paper to be published. Special
appreciation goes to Technical Review Committee (TRC) of Northern Area Production Engineering & Well
Services Department (NAPE&WSD) for their insightful comments and feedback. The authors would like
to also thank Abdulrahman Ahmari, Rotimi Adesegha and Mohammad Ali for their support.