API 2008/2014 New Sizing Method...

[processeng2014]

Does anyone have any experience with purchasing relief valves with real-gas isentropic flow equations vs ideal flow (cp/cv)?

Do vendors (like pentair and farris) support the new methods outlined by API?

 

[processeng2014]

I apologize, I am referring to API 520 Part 1

 

[chemlite]

in my experience in most cases API correlation for gas/vapor based on ideal gas cp/cv returns values not too different from rigorous models,
there are old threads in this forum (2011 ?) reporting results comparing values calculated with rigorous models (the isentropic nozzle model included in Prode Properties) vs. API ideal cp/cv and a method discussed in Hydrocarbon Processing magazine, in many cases results are not too different (except in critical area where a rigorous methos is preferable)
note that you shouldn't enter real cp/cv in correlations based on ideal cp/cv

 

[Latexman]

The Safety Relief Valve forum would have been a better place to ask this question. You may want to put a link in the PSV forum pointing to this question to see if you get some more support.

After sizing a PSV using real gas methods, one knows the flow area and coefficient that is required. If the vendor checks the sizing using ideal gas methods, and comes up with a different flow area/coefficient, then there needs to be a technical discussion with the vendor. In my experience, this difference due only to sizing methods is rare. The flow would have to be on the edge of the PSVs capacity for this to happen. I don't remember a single case where this has happened to me.

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.