2 replies to this topic

[sumwunrandom]

Hello fellow Chemical Engineers,

 

If one is determining a gas breakthrough scenario through an existing control valve with a high dP (such that the flow will be choked), is it more appropriate to use the 'real' Cp/Cv (e.g. calculated by HYSYS) or the 'ideal' Cp/Cv?

 

I have compared results for both and found that using the 'real' ratio gives a 12 % higher flowrate. For the purpose of my calculation (determining amount of gas breakthrough), the 'real' ratio is more 'conservative'.

 

I would compare this to PSV sizing in which API 520 notes that 'ideal' ratio should be used as it *typically* results in a more conservative design, and that if you are not careful, using 'real' ratio can result in undersized orifices. I have trawled through Annex B of API520, posts in this forum and ENGTIPS (incl https://www.cheresou...2-psv-sizing/),where caveats to the API 520 statement are mentioned (such as that 'ideal' ratios are not always appropriate for a select few mixtures, or when gas is near critical point). I also note ' Using the ideal gas specific heat ratio for relief-valve sizing"; by Aubry Shackelford, Berwanger, Inc.; Chemical Engineering, November 2003.'

 

However, I can't seem to find any information anywhere on CONTROL VALVE SIZING specifically. This includes reading through control valve supplier 'technical bulletins' and the such.

 

Any thoughts?

[breizh]

Hi,

I believe it's based on ideal. Consider the document attached and annexes.

Probably good to ring valve suppliers.

EDIT: More convenient to use ideal when you don't have access to a simulator.

 

Note: Make sure you compare apple to apple,

Compare Xt calculated using the ratio K gas /K air at atmospheric conditions for ideal and at process condition for real.

 

Good luck. 

Breizh 

Attached Files

•  ISA_750101_SPBd.pdf   268.51KB   1 downloads

[Alpe]

MASONEILAN / ISA formulations probably have been empirically defined (and measured),

for example the distinction between a subcritical flow (velocity < speed of sound) and supercritical is based on a value = 0.5 which does not correspond to reality in most cases (accordind some papers the values for the critical pressure ratio in valves can be in the range  0.2 … 0.7).

 

There are alternatives, for example ISO 6358 etc. 

 

I did similar evaluations with the models for control / safety valves included in PRODE PROPERTIES (see prode.com) and I see differences.

 

As suggested by Breizh probably with Masoneilan / ISA formulations one should use ideal values and take care of the limits...