REGULATOR FAIL OPEN CAPACITY PER ASTM F1795

[Bambie]

Can the fail open capacity as defined in ASTM F1795 be determined from regulator performance curves?
The regulator manufacturer lists a Cv but the calculated compressible gas flow is much larger (4x) than the 'Flow Performance Curves' with similar inlet and outlet pressures.
I suspect the the 'regulated' flows are not a conservative indication of fail open capacity but I wonder if the Cv conversion is too conservative.

 

[SNORGY]

Hi Bambie.

I am dealing with the same issue at the moment. I have, for now, asked for the regulator vendors to provide the "wide open" discharge coefficient. For a gas regulator, in the absence of superseding data, for P1/P2 > 2, I assume sonic flow and Cd = 0.72.

 

[Bambie]

Snorgy, would you mind sharing your calculations with me?

 

[Latexman]

Search out Fisher's website. In their regulator section I believe they have technical information that will guide you through the calculation.

Also, your vendor may be able to supply this calculation to you.

Good luck,
Latexman

Need help writing a question or understanding a reply? forum1529

 

[SNORGY]

Bambie,

Tomorrow, what I have, I will share.

Tonight, I regret to admit that I spilled too much beer on it...

 

[Bambie]

Snorgy,

Without an area (A) or coefficient of discharge (Cd) for the regulator, I calculated 314 SCFM N2 using Crane 410 and Cv = .2, P1 = 2000 psig and P2 = 100 psig. The Manufacturer suggested that 205 SCFM is more accurate but I haven't seen their numbers.

 

[Bambie]

Snorgy,

Here are my calculations.

 

[Latexman]

Bambie,

I believe you have forgotten to use the fact that the flow will choke at the "critical pressure", Pc. So, dP = 2000 - Pc. Also, is Crane's equation you are using for compressible flow? It needs to be. I don't recall Crane having a Cv based equation for compressible flow, but I am 5000 miles from my copy at the moment. That's why I suggested you search Fisher's web site. I know they have the right equations for the problem.

Good luck,
Latexman

Need help writing a question or understanding a reply? forum1529

 

[Latexman]

Try with the vendor of your regulator first, but here is the Fisher technical document I was referring to:

http://www.documentation.emersonprocess.com/intradoc-cgi/idc_cgi_isapi.dll?IdcService=GET_XML_HIGHLIGHT_INFO&QueryText=xTrigger_Field+%3csubstring%3e+%60Marketing_Regs%60++%3cAND%3e++xBrand+%3csubstring%3e+%60Fisher+Regulators%60++%3cAND%3e++xLanguage+%3csubstring%3e+%60English%60++%3cAND%3e++dDocTitle+%3csubstring%3e+%60sizing%60&SortField=dDocTitle&SortOrder=Asc&dDocName=D351798X012_11&HighlightType=PdfHighlight]Fisher Valve Sizing Calculations[/url]

Good luck,
Latexman

Need help writing a question or understanding a reply? forum1529

 

[SNORGY]

Bambie:

Sorry, I was away from home this weekend.

The equation I use (or resort to) is per the link attached. You need some idea of the orifice size and, as I mentioned, the discharge coefficient.

For your conditions, my initial calculation for a 1/4" port (orifice) was on the order of 2300 SCFM nitrogen *at the onset of flow*. But, I assumed a 1/4" hole and C=0.72 in the above.

In any event, the 312 SCFM at that pressure differential sounds low to me.

 

[SNORGY]

See also:

thread124-322376

 

[steamdog]

Full Disclosure: I work for the local Fisher Representative. The full open flow coefficient for a specific regulator (used so you can do the flow calculation - either choked or not) is not a calculated number but actually comes from empirical data from the flow lab at Fisher. So you will find two different regulator models with the same flow coefficient may have a different wide open flow coefficient. So that you cannot assume what the full open coefficient is, you should get this from the manufacturer.

 

[Bambie]

Steamdog,
ASTM F1795 only states that fail-open capacity must be established accurately and defined clearly - not that it must be empirically established.
Snorgy and Latexman,
I re-worked my calculations in accordance with CGA E-4-2006 for sonic flow conditions and the actual mixture under consideration (90%Argon 10%Methane).
I would appreciate any comments, thanks.

 

[Latexman]

Looks good! And, it's close to the manufacturer's number. Not bad, for a sparky. Run with it!

Good luck,
Latexman

Need help writing a question or understanding a reply? forum1529

 

[Bambie]

Latexman,
I made a correction for % gas content by volume - not much change.