3 replies to this topic

[jkobs]

Hello all,

I am trying to calculate a vertical separator's gas capacity(mmscfd) for a given diameter and length and varying pressure. I am confused on the K-value for the souders-brown equation: Vt = K * sqrt((pl-pg)/pg)
I have found a source online saying to derate the k value .01 for every 100 psi over 100psi starting with .35. This seems like it works, but when I have the pressure near or above 1100psi, the gas capacity starts decreasing because the k-value is so low.(around .25 - .21) It does not make sense to me that the capacity would decrease with a higher pressure.

On another note, I have found an equation that can be used to find a k value: K = sqrt((4*g*Dp)/(3*C'))
I don't think this equation can be used because to find a C' value, you need to know a Reynolds number, and to find a Reynolds number you need to know a Vt. Vt is what I am looking for in the first place.

If anyone can help I would greatly appreciate it.

Thanks,

Jeff

[Art Montemayor]

Jeff:

Are you trying to apply an existing separator? Or are you fiddlin’ around with the pressure on a given separator just to find out how the variables affect the separator’s apparent capacity? Regardless of what you are trying to find out, you should not cite engineering relationships without referencing the source and the site where you obtained the information. It’s hard enough reading about everyone’s favorite equation and constants without knowing where, why, and by whom the relationship came about. If you don’t have that information, how can you feel secure in using it in an engineering design?

Refer to the attached Excel workbook I put together some years ago on this very topic and notice that I always make a specific, detailed, and explained identification of the sources I employ. Perhaps this will help to clear out a lot of the fog generated on the value of “K”. I always guide my judgment by the explanation given by John Campbell in his book: use the experience factor to decide how much confidence you can rely on when selecting a “K”.

Attached Files

•  Art's Liq Vapor SeparatorRev1.xls   147.5KB   315 downloads

[jkobs]

Thank you for the response. We are going to start fabricating separators at certain sizes and I am trying to get a good estimation of the capacities of each. I feel secure about the equations I am using as they are from section 7 of the GSPA Engineering Data book.(Eqn's 7-8 & 7-11)

What I am trying to make sense of is the fact that at higher operating pressures the gas capacity seems to decrease, which is opposite of what I would think should happen. Looking at just the math, the K value is the reason for this if I follow the guideline of derating the k value .01 for every 100psi over 100psi. (Found on p. 93 of Oilfield Processing Of Petroleum: Crude Oil by Manning and Thompson) I was wondering if that is generally the case or if I am doing something wrong.

My second question is can I pick a general(or average) drag coefficient to find my K value? In section 7 of the GSPA Engineering Data Book, the drag coefficient is found by determining the reynolds number which is found by knowing the terminal(or critical) velocity of the gas. But to find the terminal velocity, you need to know the K value.

[jkobs]

Your spreadsheet has answered my question. The gas capacity for a vertical separator does decrease with pressures at or above 1200psig for a constant diameter and length.

After looking at your horizontal separator sizing, I am confused because you used the same k values. I thought that different, generally higher, k values were used for horizontals. Can you shed some light on this.

Thank you for your help