Well Integrity Technical Section

We at Cairn India are trying to formulate a policy for an acceptable Leak Rate for X-Mas tree valves. While API 14B talks about Leak rate for SSSV, API 14H talks about SSV. In 14B, while calculating Leak rate, the factor "V" is taken as the Tubing Volume above the SSSV. Extrapolating the same formula for X-Mas tree vlaves, will the Volume factor "V" be
a) Volume above the concerned valve. Then there will be different leak rate depending on the position of the valve (LMV / UMV / Swab). Length of X-Mass tree will also be important
b) Closed volume of the valve seat - then it becomes size specific - 3" valve will have a different leak rate as against a 4" or 5" valve

Is there any specific guideline on this issue? Thanks in advance for your valuable inputs

-------------------------------------------
Abhijit Chaudhuri
Advisor
Cairn India Ltd.
Ahmedabad
-------------------------------------------

Abhijit,

The volume is always the 'closed in volume being used in the test' which will be different between separate valves, brands, and sizes.  You wan't the 'high pressure' side to be unrestricted so that the pressure does not decline if there is a leak.  In other words, LMV open when testing UMV-PWV test volume.

API 14H has an example calculation in Appendix A that may be useful for you to consider.  Basically you;

1. Shut-in the valve being tested,
2. Bleed off the volume being tested as much as possible,
3. Shut-in the volume being tested, and
4. Monitor the pressure rise over a few minutes. 

My procedures have the Operator record the time and pressure every minute or so for a 15 min test as leak rate will decrease as the test volume pressure increases (i.e. less and less differential pressure across the test valve).

Knowing the volume of the 'closed in volume being used in the test' you can calculate how much material entered the volume.  For gas and oil wells I typically use the compressibility calculations and for water I use incompressible fluid calculations. In most oil and gas wells that are shut-in, there is gas at surface and the gas is the material leaking across the valve gate and effecting the tests.

We performed these calculations for each tree type and developed a guideline for Operators that they could use for Normal tree valve tests.  For example; 7" LMV (Lower Master Valve) test against a closed UMV (Upper Master Valve),  if pressure rise in first 5 minutes of the test is less than xxx kPa/min the valve passes the acceptable leak rate.  This guidance was written on the Workorder as part of the test procedure so that the Operator knew when a valve had passed the test and did not have to do any more work.  It also meant the Well Integrity team  did not have to calculate the leak rate for every valve on every test.  If the pressure increase was greater than the xxx kPa/mn they retested and if still not acceptable the well integrity team got a copy of the test data and looked at the valve test to see if we had a problem.   This simplified the test for the Operators who only had to do simple math [(final pressure - starting pressure)/time in minutes] which we included a place for on the preventative maintenance work order.

The biggest challenges were; locations with multiple valves where one of them has a known leak so the pressure rise is not 'unique'; and 'zero leak rate' tests when the valves needed to be used for isolation for facility work and containment would be broken downstream of the valve.  Both of these typically required more attention from the well integrity team.

I hope this helps Let me know if you have any questions,

-------------------------------------------
Dan Gibson
Completion / Well Integrity Advisor
Gibson Petroleum Consultants
FulshearTX
-------------------------------------------

Abhijit,

Another good post from Dan Gibson; I would just like to add a few points about valve testing:

- Allowable leak rates are based on the size of a leak which, if ignited, could be put out with a hand held fire extinguisher;

- The test measures the ability of the valve to reduce an external leak to below this level;

- On this basis, leak rates should be measured from working pressure (e.g. CITHP) to atmosphere.

As Dan says, leak rates will decrease with the pressure differential across the valve.  If the downstream pressure is bled to a pressure above atmospheric (e.g. flowline pressure) then the allowable leak rate should be reduced (roughly in proportion to the pressure differential).

Another problem is to ensure that the length of the test is appropriate, particularly if you measure the average leak rate.  As the pressure differential decreases with time the leak rate decreases, so if you leave a test running long enough, any valve will pass.  Ideally, we should measure the initial leak rate (this will be the case when the pressure build up is small in relation to the pressure differential).

One way to overcome these problems is to bleed off the pressure above the valve to be tested, and then vent the volume through a gas meter so that any leakage past the valve registers on the meter.  However, it is not always possible or desirable to vent gas, and a pressure build up test is the only real alternative.

 

Let me know if anything is unclear, or if I can help further.

-------------------------------------------
Keith Munday
Well Integrity Advisor
Petrofac Integrated Energy Services
-------------------------------------------







Original Message:
Sent: Feb 03,2014 03:20 AM
From: Dan Gibson
Subject: Leak Rate Calculations for X-Mas tree Valves

Abhijit,

The volume is always the 'closed in volume being used in the test' which will be different between separate valves, brands, and sizes.  You wan't the 'high pressure' side to be unrestricted so that the pressure does not decline if there is a leak.  In other words, LMV open when testing UMV-PWV test volume.

API 14H has an example calculation in Appendix A that may be useful for you to consider.  Basically you;

1. Shut-in the valve being tested,
2. Bleed off the volume being tested as much as possible,
3. Shut-in the volume being tested, and
4. Monitor the pressure rise over a few minutes. 

My procedures have the Operator record the time and pressure every minute or so for a 15 min test as leak rate will decrease as the test volume pressure increases (i.e. less and less differential pressure across the test valve).

Knowing the volume of the 'closed in volume being used in the test' you can calculate how much material entered the volume.  For gas and oil wells I typically use the compressibility calculations and for water I use incompressible fluid calculations. In most oil and gas wells that are shut-in, there is gas at surface and the gas is the material leaking across the valve gate and effecting the tests.

We performed these calculations for each tree type and developed a guideline for Operators that they could use for Normal tree valve tests.  For example; 7" LMV (Lower Master Valve) test against a closed UMV (Upper Master Valve),  if pressure rise in first 5 minutes of the test is less than xxx kPa/min the valve passes the acceptable leak rate.  This guidance was written on the Workorder as part of the test procedure so that the Operator knew when a valve had passed the test and did not have to do any more work.  It also meant the Well Integrity team  did not have to calculate the leak rate for every valve on every test.  If the pressure increase was greater than the xxx kPa/mn they retested and if still not acceptable the well integrity team got a copy of the test data and looked at the valve test to see if we had a problem.   This simplified the test for the Operators who only had to do simple math [(final pressure - starting pressure)/time in minutes] which we included a place for on the preventative maintenance work order.

The biggest challenges were; locations with multiple valves where one of them has a known leak so the pressure rise is not 'unique'; and 'zero leak rate' tests when the valves needed to be used for isolation for facility work and containment would be broken downstream of the valve.  Both of these typically required more attention from the well integrity team.

I hope this helps Let me know if you have any questions,

-------------------------------------------
Dan Gibson
Completion / Well Integrity Advisor
Gibson Petroleum Consultants
FulshearTX
-------------------------------------------