Safety Relief Valve Cv / Kv values 3

[valveoholic]

Hi, I'm in need of some help please.

Can anybody educate me on how I go about calculating Cv coefficient values for relief valves.

I believe that Kv is the metric equivalent to Cv.

I know the calculation differs depending on the state of the medium that is to flow through the valves, and it is also related to flow and pressure (upstream and downstream).

I have orifice diameters of my valves:

10", 16", 18", 20", 24", 26", 28" and 30".

And I know my relieving pressures for each are:

500 millibar (7.25 psig)

There will be no back pressure to my knowledge.

The design of the valves are as such:

Entry through bottom run of standard long radius equal tee, exit through branch (90 degree angle style).


From some initial searches from this site I have found:

Cv = 29.9 * d^2 / (K)^0.5

Although for the above I needed “K” the resistance (or loss) coefficient:

K = 2.1478E+9 * (d)^4 / (Cv)^2

These equations are both great but I need to know "K" for one, and "Cv" for the other.

(Neither of which I know yet)

I then found the equation that calculated K using:

K = f * L /D

This I thought may have been the answer, but I have had trouble calculating “f” (fully developed turbulent flow) to crack this one. The good thing is I know internal lengths and diameters, so I figure I’m almost there J

Please let me know if I am complicating things, or barking up the wrong tree

Can anyone please help me with this topic and advise me what I am missing in order to do the Cv calculations (i.e. state of medium, density of medium, flow rates ?)

Any help would be greatly appreciated.

This is my first post to, so if I have made any dumb errors I apologise.

Greg

 

[MortenA]

You normally dont calculate the Cv value for a relief valve (even though you could i guess.

You calculate an orifice size.

Are you confusing Cv (heat capacity) that is used in sizing for gas with cv - the capacity coefficient for a control valve?

Best regards

Morten

 

[pleckner]

@MortenA hit upon part of the answer. When you say "relief valve", the first thought that comes to most of us is you are talking about a Pressure Relief Valve or a Safety Relief Valve. For these devices, one does not use the concept of Cv because these are in actuality isentropic nozzles, not isenthalpic valves.

If on the other hand you are talking about "relief valves" in the sense of a pressure reducing valve or the like that are used to reduce a pressure buildup in a process line, then yes, you would want to calculate them using the concept of Cv. For these and other types of process valves I strongly suggest you use this link to get the world famous Fisher Control Valve Handbook (it's free).

http://www.documentation.emersonprocess.com/groups/public/documents/book/cvh99.pdf

This book has an example calculation (step-by-step) for sizing both liquid and gas valves. The equations are ISA, which means that they are universal and good for any type of valve.

 

[JimCasey]

Or if you are lazy like I am, access one of the sizing programs available from various Control Valve companies. Flowserve, Masoneilan, Fisher, Cashco, all make a program available.

Plug in the parameters you know, and the program should cough up an >>effective<< Cv for the valve in question at the service conditions in question. I'd start with a value of 0.9 for Fl (That's Bif F, Little L) If not prompted for Fl then select a valve with comparable geometry to your valve in question. MortenA is unquestionably right in that these are normally rated by flow cross-sectional area in conformance with the code methods. I'm not sure I agree with Pleckner about the isentropic Nozzle...Adiabatic, Isenthalpic, Throttling Valves frequently discharge to atmosphere with a critical pressure drop, and that's what the SRV also does. Definitely evaluat the SRV on the basis of choked flow, though.That's why I suggested the program. It >Should< be dialed in the the program and doing it longhand is messy and a bit circular.

 

[pleckner]

PSVs such as those used to protect ASME coded vessels, which is what I'm refering to, are indeed isentropic nozzles, NOT valves (which are considered isenthalpic devices). I can understand, if you lack experience in this area, how one can confuse the fact that these are not valves but nozzles (after all, we call them Pressure Relief/Safety Valves, don't we?) but in the end, and there is plenty of documentation that backs this up, they are isentropic nozzles. And note that the critical flow equation that we all use from API is derived from the isentropic flow equation and has nothing to do with the concept of Cv (sorry, I'm repeating myself).

Using a program is all well and good but if one wants to understand what the programs do that JimCasey mentions in his post above, get the copy of the Fisher Handook I link above. You will then learn how it is done (for valves, NOT PSVs) and then you can start using these programs for future calculations but now with an understanding of what is going on.

For use of the PSV calculations, you will need to get a hold of a PSV vendor catalog (Farris, Anderson-Greenwood, etc.) or better yet, a copy of API 520 and the latest copy of API Standard 521.

 

[sailoday28]

I find it most interesting that we can have a real life adiabatic flow process that is without friction. Simply speaking how can the nozzle design be such that it is isentropic?

 

[pleckner]

A PSV more closely resembles a nozzle than it does a valve. I invite anyone interested to pick up a copy of API 520, or any of the vendor cataloges I mentioned above, where there are scehmatics of what they look like. To take into account that there really is friction (as well as other dynamics going on), a coefficient of discharge is introduced into the equation to account for this non-ideality of the isentropic nozzle. This coefficient of discharge, which is determined in ASME certified labs, is a ratio of the actaul flow measured through the PSV to the theoretically calculated flow through an isentropic nozzle. For those who are not experienced in this field, you can consider this coefficient of discharge to be analagous to a Cv but it is not the same beast by definition.

 

[valveoholic]

I want to thank everyone who replied to my post, I really appreciated the help.

Greg