Vacuum Serivce Test Procedure

[aggieengineerrmb]

Does anyone know of a good procedure to follow if you were going to "rate" a valve as suitable for vacuum service? I'd like to send some to an independent lab for testing, but need a good procedure to follow.

I can easily pull a vacuum against a closed ball and see just how low it will go, but I'm not sure if there is some "acceptable" leak rate that is allowable. If I pull down to 20 microns and it holds perfectly tight, no problem with rating it to 20 microns, but what would happen if I go down to 10 microns and I get a small leak? Does it have to be perfect?

 

[NGiLuzzu]

Aggieengineerrmb,

the pressure in a vacuum system is the result of an equilibrium between the leakage (intrusions from outside and/or through the valve seat) and the pumping system capability to remove the flow resulting from leakage itself.

Soo it seems correct to define fixed and repeatable test parameters, and also to distinguish between external tightness (regarding leakage through packing, body flanges, etc. or so-called fugitive emissions) and seat tightness.


For example, when testing for fugitive emissions with "hood technique" (as per Shell MESC SPE 77/307 Specification and/or ASME CODE: see below), we can easily reach in a few hours and maintain a pressure level below 0.1 mbar_a (= 10^-4 bar_a) within valves up to 24", even with our laboratory vacuum pumps. This is called a "medium vacuum grade" and it covers almost all the usual industrial vacuum applications (higher vacuum grades are used in Laboratories and special applications).

The allowable leaks are expressed in the same terms as other fugitive emissions tests (see Threads on this subject within this Forum and Shell MESC SPE 77/312), i.e. in terms of mbar*l/s (or equivalent flow rate units) measured by a mass spectrometer.

I would suggest to take a look at ASME BOILER & PRESSURE VESSEL CODE, Section V "Nondestructive Examination", Article 10 "Leak Testing", Appendix IX "Hood Technique".


The simplest way to measure seat tightness, instead, may be to establish a certain vacuum grade on a side of the closed valve, a certain pressure on the other, exclude the vacuum pump and then see how much time is needed for pressure in the vacuum side to increase; that is to measure the vacuum seat leakage in terms of "pressure recovery rate".
For example: you may say that your valve passes the test if it maintains a 100 mbar_a vacuum against atmospheric pressure (or more) for at least 10 min.

For more severe service conditions and requirements, the above mentioned "hood technique" may be adapted for seat leakage testing too.


Hope this helps, 'NGL