10 replies to this topic

[Chemie]

Hi All,

I am trying to figure out a way out of a confusion for sizing a vertical separator. I am having two independent inputs from two sources. Both the streams are at same conditions but different phase fractions.

To be more specific the streams are two phase condensate flows after flashing across a control valve. I am planning to put two different nozzles in the separator for inlet.

There is stream 1 with low liquid content and high flow where as stream 2 with relatively high liquid content and relatively low flow.

My confusion is

1. Will it be OK if we are having two individual nozzles in a separator? If yes should where should I locate the nozzle for flow with high liquid content.
2. I am uncomfortable with this idea: Can I combine two streams upstream of the separator and provide only one inlet nozzle? The reason that I am uncomfortable is that there might be some chances of liquid hammering in the line to separator.

Can anybody give me experienced advise and clear my confusion

Chemie

[Radionise]

Hi Chemie,

First of all, I'm not an industrially experienced engineer, no, not yet. But I'd hope to help you perhaps by expressing my view on the options that you mentioned.

1. Honestly, I am not sure if there is any double inlet nozzles two phase separator out there. However, wouldn't an extra nozzle means more internal parts (i.e. vortex breakers, baffles etc..)? Wouldn't having an extra nozzle also complicates the sizing calculations? Because I think you then may need to calculate out separate set of diameter and height of the vessels by referring to the different inlet nozzle flow conditions. Subsequently you may need to compare and see which set is of height and diameter can fulfil all the liquid retention time and gas capacity constraints. Potentially, there are quite a lot to do in the calculations. (Somebody please correct me if I'm wrong here by the way). So, with the extra internals and complicated calculation methods, there may be issues of cost and time constraint to do so?

2. In regards to the matter above, combining both inlet streams seem to be a good idea. And I assume that you'd select a T-junction to do the job. However, when combining two streams of two-phase flows at a T-junction, the potential of flow pattern changes is very very high, in fact it bound to happen. You would notice for instance,
i. slug/plug flow after the mixing, which would cause sudden in-rush of liquid or gas surge into your separator.
ii. backward flow of the liquid phase at the T-junction.

I've been involved in some bit of these research about T-junction and two phase flow at my current university during my master course. As a whole, prediction of the flow pattern downstream of the mixing is very very complicated. For example, combining two stratified flow will not even necessary result in stratified flow in the downstream of the mixing. In fact, plug or even slug flow is likely to happen.

Hopefully, other forumers would come in to contribute a much more industrial oriented experience on this too!

[joerd]

I certainly like the two-nozzle option better for process reasons - like the one you yourself pointed out - although it could cost more than a single line. The sizing of the separator is done on the combined vapor and liquid flow rates, the nozzles should be sized based on the individual feeds.
The nozzle with the high liquid fraction should be either at the same elevation, but at 90° orientation from the other. Alternatively, you can place it about 1 ft below the other nozzle. It could give you some more entrainment of liquids into the vapor stream, so allow for sufficient space above the inlets.
This is not uncommon for condensate KO drums.

[Chemie]

Hi

Thanks for your valuable inputs. I was concerned about the extra stress put on the vessel if we put extra nozzle. But after some considerations and your valuable inputs I am of the opinion of having two nozzles put at a different elevation.

Yes, Joerd I will be putting the high liquid fraction nozzle below about a feet(just an approximate number) and then check the vapour zone.

Thanks guys again for the valuable time and inputs you provided.

Chemie

[panduru]

As an engineer outside the field of chE; I get a random thought:

Is it possible to use a pair of concentric pipes to form a [compound] inlet nozzle? Do such multiple-nozzles exist in practice?

Has the behavior of such nozzles been studied and design methods established?

[Chemie]

panduru,

I have not heard of any concentric pipes used for two phase flows within little field experience I have. But I suspect erosion problem with using such type of system in vessel inlet nozzles where phase separation would be there.

I would certainly like to know about new developments in these regards and more experienced opinions from the forum.

Chemie

[Radionise]

Hi Guest,

I've always liked to ask this question, are there any other inlet types which are similar to Schoepentoetor, either in terms of design or in their purposes as primary stage separation? Schoepentoetor is a proprietary design by Shell and is being used a lot in the design of their separators on offshore platforms.

Lately, research has been done on the usage of T-junctions as novel phase separators (for primary separation), and the separation effect has proven to be quite good. And if I'm not mistaken it is already currently being used in some offshore platforms but i don't know where.

Cheers.

[joerd]

You can buy the Schoepentoeter type inlet devices from Koch-Otto York if you insiste. They sell them as "vane-type inlet device" or something like that. These things give you the smallest nozzle size for a given flow, because they allow the highest inlet impulse (rho-v2). However, in many cases (onshore), a simple baffle, half-open pipe or no device at all gives a cheaper design.

[turnersteve]

Better late than never I hope.

You should combine the streams upstream of the separator thus only one nozzle. The liquid handling is not a problem if the correct choice of inlet device is added inside the separator on the inlet nozzle.

Have a look at the attached document that hopefully discusses inlet devices in a little more detail and compares the most common ones.

Two different nozzles is not the answer

Attached Files

•  Inlet_20devices_20overview_2012_9_05.pdf   214.25KB   352 downloads

[processji]

I understand the inlet baffle or a proprietery device reduces inlet momentum and enables primary separation.I am trying to understand the impact of surface tension in water oil separation.Can someone throw light on this.