Residence times in liquid gas separator 7

[BigHornEngineer]

I have had two separate sources quote me a 5 minute minimum residence time for liquid in the bottom of a liquid/vapor separator. I have a vessel which does not quite meet this criteria, so I am wondering where the 5 minute rule of thumb came from. I have a lot of motivation to ignore this rule, but I don't want to do it if there is a reasonable basis behind it.

 

[MMcLean]

5 minutes seems quite conservative. I take it the column sump discharges to a pump suction?

I have used 3 minutes (minimum) from the NLL to empty before, based on the normal forward flow. This crieria was derived by some very experienced engineers on a project I worked on.


Mark McLean

WorleyParsons

 

[BigHornEngineer]

This vessel actually discharges using the pressure of the vessel through a control valve. The separation involves an aqueous solution which has been contacted with the plant fuel gas. The vessel rides on the fuel gas pressure and the liquid is discharged to a lower pressure vessel which is nearly atmospheric pressure.

It is essential to keep the fuel gas out of the downstream vessel because the aqueous solution is exposed to air. This may be the reason for the conservative numbers that I have been given. However, I believe that the dual control system that is installed on the vessel should be adequate because it closes the control valve in the case of a failure.

 

[Montemayor]

BigHorn:

There is nothing magical or sacred about any residence time value given to the sump of a 2-phase vessel. I have designed and built many separators and I have recited the process guidelines to many engineers. At the moment I’m working on a project involving multiple separators – vertical and horizontal, 2-phase and 3-phase. Since this subject still comes up from time to time, I’ll take this opportunity to share my experience in this area and hopefully assist in calming some concerns.

There are four zones involved in the process and mechanical design of a vapor-liquid separator:

1. Primary separation section. This is the area where the main portion of free liquid in the inlet stream is separated. It includes the inlet nozzle and any baffles or inertial effects used to change the direction of the gas or vapor involved.
2. The secondary or gravity section uses the force of gravity to segregate the liquid droplets from the main gas/vapor stream.
3. The tertiary or coalescing section employs vanes, wire mesh, or other coalescing tools to enhance the buildup of larger droplets and their separation from the gas/vapor stream.
4. The sump, or fourth zone, is the section of interest in this discussion. It acts as a receiver for all liquid removed in the other 3 sections.

The sump, like the other sections, has a definitive Scope of Work:

1. It serves as a surge volume during process upsets, flow changes or slugs;
2. It allows for liquid de-gassing – or in other words, allows for equilibrium to be established between the gas and the liquid phase;
3. It allows for a minimum liquid level necessary for process controls to function properly;
4. It furnishes a positive liquid seal between the process and the lower pressure system downstream of the drained liquid;
5. It offers the possibility of having decantation between two immiscible liquids; this is often the design case in special, 3-phase separators in the processing of crude oil streams;
6. It can fulfill special liquid hold-up process needs: this is often the design case in closed cycle processes such as refrigeration cycles where the separators are used to temporarily store the circuit liquid fluid during a shutdown or maintenance operation.

As you can appreciate from the above listing, a sump’s volumetric capacity (liquid residence time) depends primarily on your design and operating objectives and scope. Needless to say, the specific design depends on the fluids’ properties (densities, solubility, viscosities, etc.).

I would not concern myself with the magical 5 minute minimum residence time for liquid that you have been given. I would, instead, concern myself with the specific design/operating criteria that your process requires to operate safely and efficiently. After analyzing your process needs you may find that you really require MORE time than the 5 minutes. As a general rule, I always try to operate with a minimum sized sump – especially where corrosive or erosive fluids are involved. However, that is only a generality. For example, in the present project I’m on I am designing pre-settlement compartments prior to discharging the liquid because I expect that the wellhead flow lines feeding my separators will carry some well sand. Therefore, I want to have the ability to segregate and handle the sand without having it plug my liquid discharge line directly. Applications differ. Sometimes they differ in totally opposite directions and an engineer is wise to avoid employing “Rules of Thumb” without specific analysis on the application.

I do not agree with Mark McLean’s opinion that “5 minutes seems quite conservative”. This is a general statement that can't be used to judge your application because we don’t even know the identity of the fluids in question – much less the actual process and its operational requirements. If we are talking about a general estimate or an academic problem, fine. But you have led us to believe that you have an actual, operating separator and I can’t, in good engineering judgment, give you rules of thumb or generalities for you to apply. You need a decision based on factual data and you control that data because only you know the fluids, the application, and its sensitivities.

You have, in my opinion, resolved your own concern when you state “I have a lot of motivation to ignore this rule, but I don't want to do it if there is a reasonable basis behind it”. You should always base yourself on a logical, reasonable basis and I believe you're headed in the right direction - based on good, sound engineering judgment.

I hope this experience helps you out.

 

[25362]

If the liquid is to be used as a seal (for safety reasons) it might be worth installing level alarms as well as a "Low Low" liquid level alarm + a cutoff acting on the level controller or on other pertinent part of the process.

 

[25362]

See also

http://www.driedger.ca/ce6_v&t/CE6_V&T.html

 

[ankur2061]

As I understand, the retention time of any separator is the liquid surge time & is defined as the residence time between the controllable high liquid level (HLL) & the controllable low liquid level (LLL).

The liquid surge volume corresponds to the shell length between HLL & LLL (for vertical separators).

Surge requirements need to be set based on following criteria:
a. Adequate control during normal operations.
b. Continuity of process during routine process upsets.
c. Safe shutdown of system during major process upsets.

For general applications I have been using the following criteria for defining the various levels in vertical separators:

- HLL is kept 6 inches below feed distributor.
- NLL (Normal liquid level)is kept half-way between HLL & LLL.
- LLL is 6 to 15 iches from bottom tangent line depending on Level Control imstrument.

For pumped removal of liquid from a vapor-liquid separator bottom I have used 5 mins as surge time & it has worked out fine. For a gravity drain probably a lower residence time of 2-3 minutes may also suffice.

Additionally you can refer the following to get a fair idea of the design criteria used for sizing gas-liquid separators.

1. API 521
2. Norsok standard P-100
3. Perry's Handbook, 7th Edition, chapter 26, pg 26-36

 

[monaco8774]

It might be helpful for you to get a copy of API12j, this defines residence time for gas liquid seperation (aswell as two phase liquid seperations) and is an independent guide that can be referenced. For gas liquid sepetration it defines residence times based on the oil gravity as follows:

Gravity Residence
above 35 1 min
20-30 1-2 min
10-20 2-4 min

since the gravity driving force for liquid/ gas seperation is high usually you dont need to much residence for this purpose. As mentioned above however you have to also take other things into account. If you have a downstream vessel with air in it this seems pretty odd to me, usually you segregate hydrocarbons from air as much as possible. If a liquid seal is used then it should not contain dissolved gasses as these may come out into the downstream vessel when the pressure drops through the level control valve ?