separator design consideration

[kloroform]

hi all,

i need to design a two phase separator to separate natural gas 60 MMSCFD, 200 F, 400 psig from 1000 barrel water per day. i have some questions to be asked :

1. how much separator that i need to design ? i mean, do i need to design 2 units of separator (@ 30 MMSCFD) or is it enough to design only 1 separator with 60 MMSCFD capacity ?

2. let say if i only design 1 separator (60 MMSCFD), does it need the backup separator ? (so, i will design 2 identical separator with 60 MMCSFD capacity). what is the consideration ?

3. i suppose i will design the vertical separator in this case since the gas and liquid ratio is high, am i right ? because i can't find the reference that states the number of gas liquid ratio .

thanks

 

[pennpiper]

Look here:

https://www.google.com/search?q=two...v&sa=X&ei=LNQZVcXwMI3asASt5IH4Dg&ved=0CCwQsAQ

and here

http://petrowiki.org/Separator_sizing

Sometimes its possible to do all the right things and still get bad results

 

[georgeverghese]

Pressure vessels are usually not spared. The duty of the vessel may need to be split amongst a number of vessels if the dimensions of one vessel alone become impractical for fabrication or for transport or lifting.
Run the sizing calcs for both vertical and horizontal options and choose which one is more compact or suitable - in some locations, there may limitations on footprint, in other locations, there may be headroom limits.

 

[EmmanuelTop]

1) It is common to design for a single separator but this may differ for cases when you could have transportation issues due to equipment size. For 30 MMSCFD @ 400 psig vessel, transportation will likely not be an issue. Another thing is distribution of load between two parallel separators - one can hardly expect equal loads even if the piping around the separators is absolutely symmetrical. So in real operation one separator could be taking much more load and, if designed for half the load, it will effectively act as an undersized vessel. This can result in carryover, mechanical damage of internals, and other problems.

2) Provision of Backup separator will depend on your operational, maintenance, and availability requirements. You need to verify what is the planned shutdown cycle for the plant where this separator will be located and then see if you could perform all the required maintenance and calibration/certification activities (e.g. PSV) during that time frame and at given frequency of planned shutdowns. In some countries, pressure test and PSV calibration are required to be performed each 2 years. This means you have to isolate the equipment and do the necessary work each 2 years, whether you planned to shutdown the system or not. Then you can plug the cost of the backup separator into RAM model and compare against the value of increased availability (no production shutdowns, no lost/deferred production). This is usually a cost vs. benefit thing.

3) Some guidance on separator design available online is pasted below:

http://www.red-bag.com/engineering-guides/249-bn-eg-ue109-guide-for-vessel-sizing.html

http://det-mop.ir/ips/pr/e-pr-880.pdf

http://razifar.com/cariboost_files/...20Type_20Selection_20and_20Design_20Rules.pdf

http://www.chemwork.org/chem/vessel/Vessel Sizing.html

Dejan IVANOVIC
Process Engineer, MSChE

 

[EmmanuelTop]

Here are some pages where you can read about equipment sparing philosophy (e.g. Norsok standard) and a few examples how this works in design of some real plants.

http://www.standard.no/pagefiles/1132/p-100.pdf

http://www.gasification.org/uploads/eventLibrary/31ALDERSON.pdf

http://www.ivt.ntnu.no/ept/fag/tep4215/innhold/LNG Conferences/2005/SDS_TIF/050186PR.pdf

Dejan IVANOVIC
Process Engineer, MSChE

 

[georgeverghese]

A key parameter for separator design is vapor liquid separation eff and vapor flow turndown. These will be influenced by the limitations of the equipment downstream of this separator.

 

[georgeverghese]

If your downstream equipment cannot accomodate even trace amounts of liquid carryover in the gas exit stream from this separator, then a gas filter coalescer may be required.

Some companies sell special gas cyclones that can be fitted into vessels to increase vapor liquid separation efficiency.

On the liquid side, surge handling volumes are some times included in the level gas NLL to LAH - this is common in cases where feed from this separator is going to columns, or upstream applications where slugs can surge into these vessels.

If this separator is feeding a pump, it is good practice to ensure that the liquid residence time in this vessel can degas the liquid even at minimum normal liquid level - typically we use the rising velocity of a 200micron gas bubble as the criterion for this.