API 520 is the first reference you should review.  

Each manufacturer can determine their own flow coefficients (and areas if they aren't using the standard API orifice areas) to use.  The coefficients, and areas, are based on certified testing.

Do a search for Anderson Greenwood and look up their technical information, they have a lot of good information there.

As TD2K said, Anderson-Greenwood Crosby Engineering Handbook is a great reference to determine the area required.

http://www.andersongreenwood.com/docnet/library/enghnbk.pdf

If you need to know the area of a given Relief Valve, there is another reference, the National Board of Boiler Inspectors website has a document in PDF format called Pressure Relief Device Certifications, you can find the effective area of every relief valve in the world that has been capacity certified.  Check out www.nationalboard.org

If you need further assistance, try www.safetyrelief.com

J. Alton Cox
President
DeLuca Test Equipment
www.delucatest.com

Related to this topic, can anyone explain (or direct me to a source that will) which valve characteristics sahould be used to determine a relief valve's capacity - API orifice size & discharge coefficient or the "Actual" orifice size & discharge coefficient?

i know that typically manufacturers will express their product's capacity using API numbers, but in the user's application is it necessary to go with API or can "Actual" be used?  

any help is greatly appreciated.
kahlil

You should refer to the manual for the PSV in question.

But, in general, the manufacturer will publish API orifice areas and API discharge coefficients and ASME (or actual) orifice areas and ASME discharge coefficients.

You can use either set you want, I typically find the ASME area and coefficients give a slightly higher (and in some cases, quite a bit higher for the small orifice PSVs) relief capacity.  Though you can also argue that means they aren't as conservative.

DO NOT mix API areas with ASME discharge coefficients or vice versa.

API Numbers are minimum standards for Orifice Area.
ASME Numbers are actual numbers published by NBBI on line at www.nationalboard.org.  Example, H Orifice is API minimum 1.00" Diameter, Actual for typical PRV is 1.040" to 1.054".  The reason for this is that every five years PRV manufacturers must recertify their PRV capacity by Lab Testing.   Over the years, when flow tests have been less than the published numbers, the manufacturers have bored out the PRV nozzle to increase flow rather than reprint all their capacity charts.  All capacity calculations based on either set of numbers are conservative from the aspect that the coefficient of discharge established by Lab Testing is a maximum of .975.  In other words, no PRV is given credit for 100% of its actual capacity.  In addition, ASME Code provides for nameplate capacity being limited to 90% of actual measured/calculated flow.  Therefore, a "perfect" PRV design would still only be allowed by ASME Code to show 87.8% of its capacity on the Nameplate.  

J. Alton Cox
President
DeLuca Test Equipment
www.delucatest.com

Thanks jalton & td2k.  I do know that the A & Kd values are not to be mixed/interchanged.  

I have another question about the Kd. For natural gas i noticed that AGCO uses 0.975 (API) but that Farris uses 0.953   Should there be a difference? If so, why?

There is an ASME requirement for Pressure Relief Valves to "operate consistently".  The Kd Numbers (Max of .975) are the result of actual lab testing using the 9 valve method.  3 valves of 3 different sizes and 3 different set pressures are tested to determine the flow.  A graph is plotted and if any of the 9 capacity numbers falls outside of a plus or minus 5% band the design is considered inconsistent and the ASME will not issue a Code Stamp.  This value differs from one design to another depending on nozzle geometry, material, seat configuration, etc.  The numbers can be found on line at www.nationalboard.org under Pressure Relief Device Certifications.  This number is then multiplied by .9 to obtain Nameplate Capacity.

J. Alton Cox
President
DeLuca Test Equipment
www.delucatest.com

Crane Technical Publication 410 has a formula for Cv/area for control valves, which is equal to 37.99 / square root of loss coefficient of the control valve trim. [Equation 1]

Cv can be calculated using ISA S75.01 method which is independent of above. [Equation 2]

Equation 2 divided by Equation 1 will give you the area (which actually is the effective flow area) that you want.

This equates seffective flow area to the loss coefficient of the trim.  Knowing the geometry of the trim, you can calculate its loss coefficient, and therefore the effective flow area.