Hi all, I'm working on some PSV' 1

[pb2102sm]

Hi all,

I'm working on some PSV's contingency calcs for a client. At some point I had the need to work with Net Expansion Factor for compressible flow (see attached picture). Can someone please tell how to handle/approach when we have K values outside the specified range, for example lower then 1.2 or higher than 100 (see tables for k=1.3 and k=1.4).

Just for clarification, I'm trying to find the corresponding Y and -P/P'1 when the K is outside the specified range

Thanks in advance for the help.
Paul

 

[Latexman]

Since K[sub]i[/sub] = 1 for the expansion at the end of the line, which has to be included in [&sum;]K, I don’t see [&sum;]K being << 1.2 in a real situation. For [&sum;]K = 1-1.2, extrapolate the graph.

For [&sum;]K > 100 the line is probably long, which suggests to me there is time for heat transfer and the flow may be closer to isothermal, than adiabatic. This method does not apply for isothermal flow.

Good Luck,
Latexman

 

[The Obturator]

What is it exactly that you are trying to determine relative to a Pressure-relief Valve?


*** Per ISO-4126, the generic term
'Safety Valve' is used regardless of application or design ***

*** 'Pressure-relief Valve' is the equivalent ASME/API term ***

 

[pb2102sm]

Thanks Latexman for the input, I appreciate.

The Obturator, this goes into the Darcy's formula for compressible flow, for sub-sonic flow conditions. You then get that flow for sizing. (of course it depends on case by case... you don't always need this info).

 

[The Obturator]

In that case it's best if you use the recommendations in API-520 Section 5.6.4 - (Pressure-relief Valve) Sizing for Subcritical flow. It probably does not answer you query on K being outside the range, but it is more relevant to PRV design.


*** Per ISO-4126, the generic term
'Safety Valve' is used regardless of application or design ***

*** 'Pressure-relief Valve' is the equivalent ASME/API term ***

 

[Latexman]

Me thinks pb2102sm is estimating the mass (W) or volumetric (V) flow rate of a credible scenario, which they can then use Equations 16 to 21 in API-520, Part I, Section 5.6.4 to determine the nozzle/orifice area. But, I could be wrong.

Good Luck,
Latexman

 

[Snickster]

I believe you can find ratio of specific heat factors down to 1.0 and factors above 2 are very rare if non-existent. What the hell has a factor of 100 are you serious. If you are doing a PSV calc only one of the inputs is the ratio of specific heats and if it is above 2.0 I would say you have a very special gas like from an alternate universe.

 

[Latexman]

Snickster, it’s K (velocity head loss), not k (Cp/Cv).

Good Luck,
Latexman