Three phase separation 1

[sbalaut]

We have a three separator (HC, HC-Water vapor, Water). We are facing difficulties in separation of water and hydrocarbon. There is carryover of HC with water. Is there is any effect of gas velocity on separation of water and hydrocarbon in liquid phase.

 

[Chemitofreak]

It is very difficult to analyse the situation, because you have not provided any information, you are just specifying that HC is carried along with water.

Liquid-liquid separation due to gravity is governed Stoke's Law

There are various factors that affect liquid-liquid separation. Some of them are listed below
1. Particle size or droplet size
2. Viscosity of HC and water
3. Is HC present as finely dispersed liquid.
4. Density difference between HC and water.
5. Internals provided for separation

Please check the calculations or provide more details.

Revert in case of any query.





Dinesh S SHELATKAR
Process Engineer

 

[khaledc2010]

its important to check your separator capacity mainly for water HC separation section and relocate the height of this section this will be affected by the separator configuration and effluent composition and proporties

 

[EmmanuelTop]

Interface level is between two liquid layers, so unless both layers are very thin I don't think that gas velocity can affect separation between the two liquid phases and cause carryover of oil in water. You probably need to look at other things such as:

1) Malfunctioning level controls (this is quite common in separators handling two different liquid phases);
2) Undersized separator (compare the actual G/L/L rates versus design rates and calculate actual phase velocities);
3) Possibility of clogging somewhere in the oil phase circuit, including the downstream piping;
4) The actual ratio of water to oil is far away from the design value - or even inverted - making this type of separator unsuitable for current operating conditions/flows;
5) Any other issue we are not aware of.

Here is a DEP available online to assist you with calculations, but make sure you do proper data collection in the field before you start crunching figures. Bad data in = wrong conclusions out. Proven over and over again.

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

Dejan IVANOVIC
Process Engineer, MSChE

 

[georgeverghese]

Is this 3phase separator in an offshore location and receiving feed from a submarine carbon steel pipeline?

 

[25362]

Since no information has been given, I suggest to use the 'search post' fixture at the top of this page asking for "three-phase separator". There are 90 entries. For example, one of them: thread127-29283.

 

[sbalaut]

thanks all for the reply.
dear george, the subject separator is not located in offshore as u mentioned in your post.
This is FCC main fractionator reflux drum (temp: 60-70 C), there is no internal as such for separation of three phases. we only have elevated liquid hydrocarbon outlet nozzle as water content is very high.

 

[LittleInch]

sbalaut,

If you don't supply more information, there's not much more anyone can do.

We need some data such as:
Size of separator, liquid (H/C and water) and gas flows, operating pressure, residence time, liquid levels in the separator, measurement / control over water / HC level, how much carryover are you getting?, and the other data requested baove.

What is the difference between design flow rates and actual flow rates?

If the water cut has increased a lot, then you could just be vortexing the H/C flow into the water outlet which is flowing at a much higher flow than the design or simply not have enough residence time or margin between water/ H/C level and the wate routlet nozzle.

Who knows - we can't see your vessel.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[EmmanuelTop]

These overhead receivers have common interface level problems. If a float type, one small upset resulting in change of interface layer location is usually sufficient to affect accuracy of the measurement.

Other potential culprits stand as well - as mentioned in previous posts.

Dejan IVANOVIC
Process Engineer, MSChE

 

[georgeverghese]

Interface level detection is always a challenge - 3phase separators can be configured with bucket and weir type internals or double overweir internals to avoid liquid interface detection. That may be a longer term plan for this reflux drum. Also a plate pack in the separator may help.

Interim options may be to
(a) change the interface level detector to a direct mount dp cell with impulse lines which are diaphragm sealed
(b)check with plant chemists / corrosion engineers if there are some emulsion stabilising chemicals being injected into FCC overheads (for corrosion management?). If this is the case, reduce the injection rate if possible, or find some other corrosion inhibitor.

Obviously, flux rates for water and hydrocarbon should be within limits in this sep; if flux rates are higher than permissible foe either phase, the plate pack may help.

 

[golpesar88]

Hi
with these informations you gave I think that it is a matter of operation at first! If there is HC in water, it means that your system works good! The inlet flow water content is very low! check from when this happened? From the first begining day of plant startup or from a specific date
1.Check level transmiters calibration.especially stand pipe if its cloged or not.
2.check Water outlet control valve operation and calibration.
3. Check mechanical drawings for water nozzle elevation and where inlet nozzle placed(if there is)
4. If every thing is ok, just close water outlet valve and let drum get some water level. Instruct operator that water is a good thing.
5. Also check if there ahould be some wire mesh inside drum and if its installed properly.