CAPACITY OF 2 PHASE SEPARATOR

[SSSHARMA]

We are having a vertical separator of dia 1424mm and height 4790mm . The heads are torispherical. The design pressure is 10 kg/cm2.This separator is planned to be used for separation of crude oil & gas at 6kg/cm2 (operating Pressure). The sp. gravity of crude is 0.8194 abd that of gas is 1.0554(relative to air). Prsently there is no water in the crude.We would like to know the maximum OIL ,WATER and GAS this separator can handle safely with a retention time of approx. 3 minutes . we are also anticipating a water cut of say 90% in next few years . Thus the capacity in terms of only oil,only water and a mixture of 50-50 may be considered .

 

[22082002]

I suggest one article, 'Design two-pahse separators within the right limits' by Svrcek and Monnery, in Chem.Eng.Progress in Oct 93 issue.

Regards,
SAA

 

[SSSHARMA]

Hello SAA,
Thanks for reference , it would be appreciated if you can refer any software, or any site available on net on this . The design book can also be referred to solve this problem .
regerds,
sssharma

 

[TD2K]

SSSharma, you should find that article referenced or look at some other sources such as the GPSA engineering data book, API 521, etc to understand how the theory of separator design. You can also request information from companies like Koch/Glitch or ACS who manufacture demisting devices and have information on the theory of sizing. Simply using a piece of software when you don't understand the theory leaves you open to a mistake that you won't catch (and it will be your fault, not the software, you are ultimately responsible for ensuring any results out of software are realistic).

The maximum capacity depends on several factors. Typically, for a vertical vessel the diameter is sized for a maximum gas phase superfical velocity which is a function of the gas density, the liquid density and an empircial factor but there is nothing sacred about the velocity number you come up with. If you have a lot of very fine droplets (and that number is VERY difficult to estimate up front), a separator can have problems. If the separator is ahead of a recip compressor, I'm going to want to use a lower velocity than if I'm removing bulk liquid prior to cooling the gas in a heat exchanger. If some liquid carryover isn't a problem (or isn't as much as a problem as in other services), you can accept a higher velocity and save $$ by building a smaller vessel or in your case, increasing throughput. Software won't consider this for you.

A horizontal separator is more complicated because of the iterative nature of the design. Essentially (and API 521 does a good job explaining this), you size the diameter so that all liquid particles above a certain diameter will settle out of the gas to the liquid phase before the gas carrying the liquid travels through the vessel and reaches the exit. However, each time you change liquid height or the diameter, the gas velocity also changes and you need to redo the calculations to ensure you will down to your droplet size cut-off.

The capacity is definitely not a single design point that everything will be fine if you run less than it and problems will happen if you run over, it's much more subtle.

 

[waydamo]

As one of the authors of the article mentioned, there is enough info there or in GPSA. A good reference to consult are the books by Manning and Thmopson, Pennwell Publishers. When rating separators, I would suggest understanding the concepts as this is much trickier than design. As stated, there are several factors and some subjectivity on the parameters. We also did a report for PTAC Canada which discusses max velocities before carryover. There is also a paper on re-entrainment velocities by Viles, JPT, May/93. I don't know of any reliable software except the spreadsheet I set up from our article that I use for consulting (unfortunately I don't give it away).