instrument valve seat leak standard

[davidlee97]

I am working on verification for many our valves. For seat leakage (internal leak), as my experience, API, AWWA, FM etc. they have the standards for leak rate and test specification. But my current company product line is instrument valves with low flow (small Cv). We use HE sniffer to detect external leak (10e-5) and seat leakage is check by bubble in water or liquid leak detector. The time duration is variant depend on operator, also the connect tube size is different too. So, is there standard or common rule for leakage? Out instrument valve are for petrochemical, refinery and so on.

 

[gerhardl]

Although attached link are for control valves and regulators there are some comments and pointers for small Cv valves that could possibly be to your interest.

As a thought wouldnt the best procedure be to be at least at par with competitors, or hopefully better by proving you can go below leakage at known tests, if possible?

http://www.isa.org/InTechTemplate.c...Management/ContentDisplay.cfm&ContentID=30548

 

[bcd]

For testing instrument valves, you can refer to standard MSS SP-99. It is a standard for instrument valves. SP-99 seat leak test can be with gas or liquid, and the duration with either is 15 seconds.

bcd

 

[reysor]

Can someone tell me if it is acceptable for ultrasound device to confirm whether a valve leaks or not without taking the valve off line and perform a bench leak teast?
I know that frequency signiture can be obtained upstream and downstream of the valve as turbulence of the throttled gas is created. The only thing one surely can not tell with this method is the leakage rate specified by API598.

I need to hear your ideas.

 

[JLSeagull]

API STD 598 requires zero leakage for resiliant seats. ASTM E 1002 ultrasonic leak testing applies to gas leaking to the atmosphere. I doubt that such testing would provide suitable leakage rate details for the metal seated rates specified for API STD 598 or FCI/IEC seat leakage.

 

[NGiLuzzu]

I agree with JLSeagull, even if I never could have an experimental confirmation of this feeling (and, of course, I would be curious to see, ehm... to hear about some kind of correlation between any acoustic measure and actual leak rate).

In more general terms, EN 1779 standard (Edition 1999 + Amendment 1 dated 2003) about Non-destructive testing - Leak testing - Criteria for method and technique selection may be helpful when choosing how to detect a gas leakage based on the type and shape of the object under test and on the entity of the leakage itself; moreover, the typical accuracy ranges for each method (e.g. tracer gas, vacuum, pressure change, bubbles, etc...) are also given and the influence of pressure, temperature and nature of gas on leakage is also studied.
Unluckily, this standard excludes hydrostatic, ultrasonic or electromagnetic methods (for which other reference documents must be applied).

See also thread408-150132 within this Forum...

Coming back to reysor's question, I would also suggest the pressure change method: of course this is feasible if there is another valve tight enough (or if it is possible to blind the pipe off) downstream the instrument valve under examination...
In any case, I believe that it should be the plant on the field to dictate the allowable leak rate and not just a standard on the paper. In other terms: why taking the valve off the line if you can't even notice the leakage? ;-)


Hope this helps, 'NGL

 

[davidlee97]

reysor's question should be inline testing. My first thinking is by pressure change. However, a shutoff valve have to put downstream. The system have to bear no flow downstream sometimes when testing the valve.