Valve Leakage Rates While Staying Online

[RRE]

Valve vendors specify the leakage rates for their products using ANSI standard FCI 70-2. This spec is used to determine what the Control Valve Seat Leakage is based on valve classes. The leakage rates measured in ml/min or bubbles/min an be found in the table. The higher the valve Class, the lower the leakage rate. I am also aware that there are tests that can be used to measure some of these leakage rates based on ISO 5208.

In the petroleum industry, what do operators do to determine valve leakage rates for SIS rated valves WITHOUT taking the valves out of service to do bench test or perform some other test (ie acoustic or other) ?? Just relying on the vendor leakage rate specs is NOT enough, especially if valve class is lower rated Class (ie. less than or equal to class IV). Is there a method (ie calculation method or other) to derive at these SIS valve leakage rates without taking the valve out of service that will comprise daily production, for example ??? Has anyone developed such a tool?? Kind regards

 

[Danlap]

Hi,

FCI 70-2 or some manufacturer use IEC 60534-4, is an agreed condition of valve (leakage rate, etc.) being tested at between 0-50 degree Celcius using either clean water (may use corrosion inhibitor) or air or nitrogen.
In other words this is an ideal condition fresh from the factory whether its new or after being refurbished by the OEM itself.

You should compare apple to apple. Ensure the valve parameter (and if possible the medium) as close as possible as the Factory condition, in order to make it less subjective
For "acoustic" and or "frequency / hissing converter to ppm" is to provide indication of leakage for some point. Do remember that the Seat (om which you cannot look at since it is inside) is a full circle which have a lots of possible leakage point. Which one point can propagates vibration to other nearby points

In special circumstances, how operator do it:
a. Ensure:
[ul]
[li]that End user may allow you to close the valve[/li]
[li]you have the most complete Data-sheet[/li]
[li]have some Online Diagnostic tool (SMART, etc.)[/li]
[li]A block valve located downstream of the control valve --> condition: known volume between Block and control valve; reliable pressure transmitter / gauge[/li]
[/ul]

b. Bypass the DCS by means of using Diagnostic tool. Ensure all parameter (Seating force, hysterisis, friction, etc.) falls within tolerance from the value given by OEM
c. If b values are confirmed, close the valve. Wait and see the pressure built up. With a known volume, pressure built up and time, you will be able to calculate the leakage rate.

Nobody says its gonna be easy, precise and not time consuming.
Control valve is for controlling. I assume by closing the valve (to determine leakage rate), then you switch its function to On/Off or Block valve. If you have another block valve upstream of the control valve, I don't see any reason why you cannot test it on the test bench.
In normal case, yes preferably do it off-line.

PS: The medium density along with its temperature also will show different value of leakage shall compared to factory condition (water/air/nitrogen)

Regards,
MR

All valves will last for years, except the ones that were poorly manufactured; are still wrongly operated and or were wrongly selected