Pressure Drop Equations for Compresible Fluids 1

[Lijantropo]

Good Morning,

I have been reading about the expressions used to calculate the pressure drop in gas lines and mainly this thread:

thread 378-242585

Mr Montemayor uses for his calculations -in the spreadsheet- the expresion given in "Transport Processes and Unit Operations" by Geankoplis.

Mr Katmar said that Weymouth and Panhandle equations are old and they have a difficulty for beginners: the choose of the efficiency factor E. He recommends to use equation 2.7 from BigInch's document (the document appears in that thread).

Well my question is about the AGA equations. The GPSA Engineer Data Book (12th Edition) and the Exxon Company
recommends to use that equations calculations; however in the Journal of Petroleum Science and Engineering 14-1996 (page 137-158), the authors said that AGA Equations introduce erros with the elevations changes.

Now I am using AGA equations for my calculations but I would like to know if it is better to change the expressions?

(Usually I do not work with gas, so I do not have experience about the accuracy of these equations.)

Thank you,

Lij

 

[zdas04]

The only equation I use for lines that are predominately gas is the AGA equation. The other empirical equations are simplified versions of this equation. The simplified versions have more simplifying assumptions which was very importatant in the days of slide rules, and is less important in the days of MathCad.

Errors introduced with elevation changes on a single-phase gas line are absolutely not within the accuracy of the data gathered and are safely ignored. This is true of AGA and of the dominant "simplified" versions.

If I have significant liquid in a gas line then I use the Duckler correlation for two phase flow and the Flannigan correlation for positive elevation changes.

I've compared these three equations to a lot of actual field data and they consistently stand up very well. All of the pipeline models that have passed my "calibration test" start with these three equations. The "true" multi phase flow processes have not matched field data as well as the AGA equation with adjustments for friction and elevation in the high-liquid case.

David

 

[Lijantropo]

Thank you Mr. David
I will continue with AGA equation.


Regards,
Lij

 

[Lijantropo]

Mr David:
the AGA equations also uses an efficiency factor "E". Usually I take 0.90, but I am not sure about this (I think this is the point of Mr Katmar).

Do you have any reference about "E"?

Regards,
Lij

 

[zdas04]

For new pipe not yet in the ground, I use 0.95. For pipe that has been in the ground "a while" (indeterminate time, gut feel), I use 0.80. If I have a data set then I adjust efficiency to match actual pressure drop in the known segments. Generally at that point I'm calibrating a pipeline model to determine an overall condition.

David

 

[Lijantropo]

Thank you Mr. David.

Regards,
Lij