What is a reasonable back-flow through a check valve? 3

[almostbroken]

I vaguely rememeber something perhaps in an older version of API-520 that suggested a reasonable flowrate through a leaking check valve was 10% of the normal forward flow. Has anyone seen anything like this?

Almost Broken

 

[zdas04]

Why do we always want numbers for stuff like this? Why not ask "why are we putting a check valve here?"

If "here" is a full-flow bypass around a compressor, then leakage impacts system effeciency but doesn't usually have a major impact on system performance so quite a bit of leakage is often acceptable.

If "here" is downstream of a centrifugal pump, then any leakage below the threshold of rotating the pump backwards it probably OK. Your 10% number probably works here (and that is likely where it came from).

If "here" is downstream of a meter run, then leakage below the measurement threshold is OK (a smaller number than the pump case).

If "here" is under a PSV on a vessel that can operate in a vacuum then the acceptable leakage is even less (since people have so little tollerance for air in pressurized systems).

David

 

[ornerynorsk]

Right on Zdas04! A problem well defined is a problem half solved.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[moltenmetal]

David's suggestions are as usual very practical. Except of course when a check valve is installed backwards by accident- then Murphy's Law says that it will seal tighter than the hubs of hell...

 

[zdas04]

That's been my experience as well. Also, when you are "just going to stab on a threaded valve" behind a single check they leak approximately 200% of the flow you'd get through an open full-port ball valve.

David