Equation of state, density relative to air

[ortabe]

Hi,

Background:
AGA PAR NX-19 is an equation of state applicable to some compositions of natural gas for estimating the compressibility z. The required input is:
p, T, mol% CO2, mol% N2 and d
where d is the density relative to air (I assume this is at 1 atm). d is often not known or at least not provided.

Case:
From a known d (0.65) for a given composition, the calculated sum(comp1_mol% * MW1 + comp2_mol% * MW2 ...)/avg_MW_air = 0.62, which is off by 5%.

Question:
What would be an accurate way to determine d when only the composition is known? Is the above under Case 'good enough'?

 

[Unotec]

School?

<<A good friend will bail you out of jail, but a true friend
will be sitting beside you saying ” Damn that was fun!” - Unknown>>

 

[katmar]

I would do it the way you have done it - using the ratio of molecular weights. For most gases, including methane, CO2 and hydrogen, the deviation from the ideal gas density at "standard" conditions is less than 1%. I am sufficiently surprised by the 5% error you have found to be skeptical of your data source.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[ortabe]

No, just structured. Though the chemical part of natural gas compression is not my strongest point.

We have Design2 but it doesn't integrate well into some of the in-house tools. I'm looking for a reasonable alternative to speed tasks such as orifice sizing and generally get a better understanding equations of state. Properties of gases and liquids by Prausnitz et al describes many equations of state but most of them require large databanks of coefficients in order to be useful.

So it would be interesting to know whether or not the mol weight approach suffices or perhaps there are other equations of state that are of interest.

 

[ortabe]

katmar,

Thanks for your input. I thought 5% to be a bit much as well. Perhaps there's an error in my program, I'll need to go through it again.

 

[katmar]

I googled for a NG composition and the first one I found was

http://www.uniongas.com/aboutus/aboutng/composition.asp

The average MW method you proposed gave a gas density ratio within 0.5% for their data. If you are looking at normal process design calcs then I believe you are safe with this method. If you are doing transfer of custody work then the contracts should specify the calculation methods.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com