David:

You’ve given us more data than the previous post – but you’ve left out some more.

Do you mean that you referring to the Compressed Gas Association (CGA) guidelines and recommendations?  If so, are you dealing with compressed gases – one of them a liquefied, compressed gas (such as Ammonia)?  Or is one of them a dissolved gas – such as Acetylene?  What, specifically, is the identification of these two fluids, the containers (cylinders) and their capacity that are being used, and the existing relief devices installed on those containers?

Are you familiar with the “Handbook of Compressed Gases” as published by the CGA as well as with Recommended Practices API 520 and 521?  

Dear Montemayor

Yes I mean the Compressed Gas Association (CGA)
and the 2 gasses in question are liquefied compressed gas R32 and R125,

And I have a copy of "Handbook of Compressed Gases” as published by the CGA as well as Recommended Practices API 520 and 521,

I was just a little unsure in how to define the relieving temperature for the calculation as defined by the CGA.


David

David,

I don't have a copy of CGA S-1.2 but you best bet will be to find someone who has experience working with it.  Likely the answer to your question is there but you don't recognize it.

As I understand, the use of CGA S-1.2 would apply to containers designed to DOT specifications....
http://www.cganet.com/publication_detail.asp?id=S-1.2

I would be careful about trying to apply practices from API RP-520/521 to your system if in fact you have a DOT container but there may be some similarities that might help you understand CGA S-1.2 a little better.

Now, I feel like the little boy in the Emperor's New Clothes but can you and everyone else read your equation?  Unfortunately I can't.  If you need to display special characters you can find how to do this by clicking on the "Process TGML" link that you see at the bottom of the "Your Reply" window or the "Click Here for the full list of TGML tags" you see in the posting preview page, then look for the section covering special characters and you'll find another link " TGML Character Entity Reference" to the info you really need.  I believe this link gets you directly there...
http://www.tipmaster.com/includes/charentities.cfm

API does suggest an approach for determining the relief rate for supercritical fluids but only the reference to that method and not the method itself is found in API RP-521.
J. 0. Francis and W. E. Shackelton, “A Calculation of Relieving Requirements in the Critical Region,” 1985, Proceedings-Refining Department, Volume 64, American Petroleum Institute, pp. 179-182.

There was a recent article in Chemcial Engineering Progress that also covered the subject of "Rigorously size relief Valves for supercritical Fluids" and this can be downloaded from....
http://www.cepmagazine.org/pdf/080234.pdf

There were a couple of Letters to the Editor that followed that you should be aware of...
http://www.cepmagazine.org/pdf/010306.pdf
and another in the October 2002 issue but I don't have a link to that so I leave that up to you if you want it.

Anyway, I suspect the similarity between the CGA method and the other methods is likely that you need to perform an iterative calculation to find a point in time that the maximum relief rate occurs.  The API and CEP methods basically look for the maximum relief rate as a function of the heat input, change in fluid volume and change in fluid enthalpy (see equations 29 and 30 in the CEP article).  This is done over a period of time and requires you to have a way to estimate the needed fluid properties for the change in fluid temperature (I believe pressure is assumed constant during relief).  The point in time that produces the maximum relief rate will also give you the other parameters you need for actually sizing your relief area.

I'll be interested to know if the CGA method is anything like the others.

Good Luck!

Excellent reply from EGT01. The second letter in CEP he refers to is actually a correction to equation 34. It can be downloaded via: http://www.cepmagazine.org/pdf/100208.pdf

Thanks EGT01, I hadent noticed that my equation was not shown correctly here is the correct notation

Squrt ν/(ν(δh/δν))p

Dave