6 replies to this topic

[pradeep14892]

Dear All,

 

I have a big confusion regarding gas-liquid separators.

 

1) According to NORSOK The specification for maximum allowable liquid entrainment from the scrubber, shall be set in agreement with

the downstream equipment vendor and the operating company. A design margin (overlap) shall be included.

A typical general specification of maximum liquid entrainment has historically been 13 litre/MSm3 (0.1 US

gallon/Million SCF).

 

2) Consider a 2 phase horizontal separator. I am required to design it to remove particles of size say 150 microns and higher. I  calculate the corresponding terminal velocity and and size the separator giving it adequate gas area to satisfy the droplet settling criteria. I HAVE NOT CHOSEN ANY OUTLET DEVICE.

 

NOW HOW DO I ENSURE THAT THE CARRY-OVER CRITERIA IS MET??

 

Since i do not know the feed particle size dristribution. 

 

what if all the liquid droplets are less than 150 microns in size (assume).Then i will have a total liquid carryover at gas outlet.

 

What am i not understanding properly here?? How do i design a separator for the required carry-over.

 

Waiting for ur inputs.

Edited by pradeep14892, 18 September 2015 - 10:59 AM.

[breizh]

Hi ,

Consider the document attached to support your work.

 

Good luck,

Breizh

[PingPong]

NOW HOW DO I ENSURE THAT THE CARRY-OVER CRITERIA IS MET??

......

How do i design a separator for the required carry-over.

You are quite right: you cannot really make a design for a specific carry-over of, say, that 13 litre/MSm3 (0.1 USgallon/Million SCF), or any other number.

 

Engineering companies design based on experience. They use correlations to size separators and scrubbers, based on the specific service, or specific process.

 

If liquid carry-over is very critical and has to be minimised, they include some form of mist eliminator like in the Amistco brochure that breizh posted, and moreover often add a special device on the inlet nozzle to minimise liquid entrainment by decreasing the momentum of the inlet feed stream, and moreover obtain an evenly distribution of the vapor flow towards the mist eliminator.

See also: http://www.koch-glit...ductCatalog.pdf

Edited by PingPong, 19 September 2015 - 06:05 AM.

[Bobby Strain]

PingPong,

     You said "Engineering companies design based on experience." That is mostly not true. They all use a "poke & hope" spreadsheet. They occasionally get it right. But for the most part, the engineering companies don't have a clue as to how to size a separator. Shell and ExxonMobil, though, do know how they should be sized. They give their criteria to whichever engineering company is working on their project. Shell is very conservative in their design. And the specialty providers of production separators can supply separators that work.

 

Bobby

[PingPong]

With "engineering companies" I meant companies that do basic process designs, either open art or licensed, or specialized vendors that design, build and supply separators. I did not mean detailed engineering companies that may already be challenged when sizing vents and drains.

 

Those designers all use formulas that history has proven to be satisfactory. Whether those formulas are in spreadsheets or other software does not really matter, except ...... that it hides the actual procedure. As a result of the increased use of spreadsheets and other software, most younger engineers hardly understand anymore what they are really doing. That is not only the case for sizing separators, but also for sizing columns, sizing or specifying exchangers, specifying compressors, expanders, et cetera.

 

I sized many separators in the past decades, all of them were in refineries or petrochemical units, some were for Shell projects using DEP. None were upstream or offshore, so maybe it is done differently there.

[gegio1960]

interesting discussion....especially the "poke & hope" philosophy.

someone considers demister a "piping item", not worthy of attention by the process engineers (only a "yes" or "no" on the datasheet)

someone wants a reduced section demister....in a square shape.

other can discuss for weeks of 100 or 150 mm height.

someone is specialized in demisters: they have one for each single design.

recently I've also red a spec where the type of wire to be utilized for the demister is minutely described.

most part of the demisters foreseen by the designers are removed at the first opportunity in field.

good luck!

[Zauberberg]

It is impossible to predict the actual (absolute) amount of carryover in kg/hr for any separator design, since the extent of generation of liquid droplets upstream of the separator, as well as particle size distribution, are unknown and depend on dozens of factors (mixed stream velocity, piping configuration, properties of gas and liquid phase, etc.).

 

The best approach is to design for certain removal efficiency of particles of given size, and that will always work - unless you have some really weird system where a lot of liquid gets atomized and floods the vessel internals. Te charts in the article uploaded by Breizh are very useful and we have used them several times to check the separator designs in new projects.