AGA 8

[sysengineer]

I can use AGA 8 to calculate the compressibility of natural gas from flowing temperature and pressure and the mole fractions.

Which equation of state is then used to calculate density at base conditions in a typical flow computer?

Thanks
Dan

 

[BigInch]

Compressibility is the measure of deviation of a gas' behavior from that of an ideal gas. If you have calculated the compressibility factor, in effect then you have made the non-ideal gas ideal, so you simply use the ideal gas equation of state and adjust for any non-ideal behavior by including the compressibility term.

I hate Windowz 8!!!!

 

[sysengineer]

Thanks, so is this method used in modern flow computers? I can't find any text that describes how standard volume is calculated in a flow computer. They all seem to use the various AGA algorithms but they don't state how density is derived.

 

[ScotchWhisky]

Typically ISO6976 is used in the UK to calculate standard density, hence std volume

 

[BigInch]

AGA8 is for relatively dry natural gas. I believe it is only valid for natural gases with specific gravities within a rather narrow range. Specific gravity and density can be calculated by knowing the percentage of each kind of gas making up your natural gas stream and multiplying by molecular weight of each gas.

I hate Windowz 8!!!!

 

[BigInch]

AGA8 I believe also has limits for N, H2S and CO2 content.

I hate Windowz 8!!!!

 

[sysengineer]

FYI - Just had some feedback from FloBoss regarding their AGA 8 algorithm.

"The molar density of gas is calculated from AGA 8 equation 24 by using the Brent's Method in all cases."

This doesn't make a lot of sense to me I don't have AGA 8 and I'm not familiar with Brent's Method.

 

[BigInch]

Not sure either, but it appears to be a method for finding roots of a polynominal equation. Using various equations of state to find the compressibility factor often depends on finding the roots of a cubic equation, such as Pv3 - (Pb + RT)v2 + av - ab = 0 from the van der Waals equation. I think they are using the "Brent method" to find the roots.

http://www.et.byu.edu/~rowley/ChEn2...e_Systems/Van_der_Waals_Equation_of_State.htm

http://www.et.byu.edu/~rowley/ChEn2.../Single_Phase_Systems/Van_der_Waals_Roots.htm

The feedback you have received is probably from someone that's looking at the algorithm, but isn't exactly sure what's going on there. They are only giving you half (or less) of the whole story. Once you have the gas conditions located in relation to the critical point, you can determine the density. I haven't looked at this for years, but I think that AGA8 also prescribes the compressibility factors to use in an associated table. They may be substituting an equation to do that, as it is probably within the accuracy of the AGA table, and more easily calculated by their flow computer, rather than doing lookups of table values.

Continue on with the Brigham Young University tutorial to see other equations of state you might be able to use.

http://www.et.byu.edu/~rowley/ChEn2...le_Phase_Systems/Other_Equations_of_State.htm

I have always found the BWRS has been accurate enough to suit my purposes, but I haven't been making flow computers for sales gas purposes either.



I hate Windowz 8!!!!

 

[BigInch]

Google "AGA8 equation of state" and "AGA8 compressibility calculation" for more info

This may be an online solver that works within the acceptable limits of AGA8 using gas composition

http://www.mysolv.com/AGA8CalcIn.asp

I hate Windowz 8!!!!

 

[RCameron]

ISO 12213-2 is the key. [

http://www.cngspw.com/Doc/WebNote/2...u6469u5C14u4EE3u6570u8FD0u7B97u5206u67901.pdf

]

In the most readable order (which is the reverse order of the calculation process):
[ul]
[li]Equation B.10[/li]
[li]Equation B.11[/li]
[li]Equation B.13[/li]
[li]Equation B.14[/li]
[li]The various constants can be found in Table B.2[/li]
[/ul]

The general confusion over density in AGA8 (and consistent lack of description in textbooks and papers) is explained by Figure B.1 - AGA8 uses a guess and check method for density. Excel has a "Goal Seek" or Solver tool that can reduce an error-calculation cell to a value near zero. If you're using other mathematics software...then you already know what to do.

I'm also partial to

http://www.ijcea.org/papers/4-A504.pdf

and

http://bib.irb.hr/datoteka/293349.FMI_Cp_JT_Ie_2007.pdf

for more detailed thermodynamics if you have a more complicated applications than simply "gas in a tank/reservoir/non-restricted pipe".

Hope this helps!

 

[apetri]

AGA 8 has been originally designed for densities and then extended for derived properties as speed of sound,
the core includes a iterative procedure which solves a volume root,
with a tool as PRODE PROPERTIES you can compare AGA vs. other ISO formulations, or EOS as SRK, PR , LK, BWRS and GERG formulations,
AGA is very accurate and competitive against GERG formulations.

About calculating density for natural gas at standard conditions
do not forget that you are usually well above the critical point
for some components and ideal gas rule should be corrected with
compressibility factor (calculated with AGA or equivalent method)