Horizontal three phase separator sizing check

[eriskawk]

Dear members,

Greetings

I am a junior engineer, doing a separator sizing check for the separators that our Client has. I need to determine the maximum capacity of the separator can accomodate in case our Client want to maximize their production in the future.

I am still new to this separator sizing so, I would like to ask for your assistance. It is a horizontal 3 phase separator completed with demister.

All I know is that basically there are 2 methods to size a separator; K factor method and settling velocity using droplet size.

I tried to calculate it using inhouse spreadsheet but the result appear to be garbage. Before I start determine the maximum capacity, initially I put the flowrate, vessel data including dimension, weir, level setting, and everything. I also had already specify the criteria based on Client's criteria and GPSA/API 12J.

But it gave me error result, saying that the settling velocity isn't sufficient enough, that is very far from the criteria given. It doesn't make sense since our Client's current operation is running on that data. I believe there is something wrong with the calculation but I have no idea of what/where.

I really appreciate if you can enlighten me, review and correct my spreadsheet calculation attached.

Here are the input data:
gas: 123037.78 lb/h (59.88 MMscfd)
condensate: 965.32 lb/h
water: 145.7 lb/h
dimension 2200 x 6000 mm, weir height 1150 mm
Level setting for both side:
HHLL: 1100 mm
HLL: 800 mm
NLL: 650 mm
LLL: 500 mm
LLLL: 200 mm

retention time: 3-5 minutes
droplet size: 90-10 micron


Thank you in advance.
May God grant you joy!

 

[LittleInch]

Well there's something very wrong with your input data.

There's no way the the phases are equal in mass flow.

No mention of operating pressure which will make a massive difference.

Get some proper data first.

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

 

[pierreick]

Hi,
Consider the resource attached to support your work
Good luck
Pierre

 

[eriskawk]

Well there's something very wrong with your input data.

There's no way the the phases are equal in mass flow.

No mention of operating pressure which will make a massive difference.

Get some proper data first.

Dear LittleInch,

Thank you for your correction. I have typed them wrong in this forum. But I can assure the mass flow input data on the spreadsheet are correct.

The operating pressure is 492 psig; T = 120 F which has been considered in the process simulation prior to obtaining the fluid mass flow data. raw P is not necessary for the calculation since there's no equation inside it dependent on P data.

Cheers,
EK

 

[LittleInch]

I had a look at the spreadsheet and had difficulty working out what was input and output.

And working out someone else's spread sheet is always a complete nightmare unless it is really well set out.

Which particular part are you finding is giving you wrong answer?

You can edit your OP. Click the edit button in the bottom right hand side.

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

 

[eriskawk]

I had a look at the spreadsheet and had difficulty working out what was input and output.

And working out someone else's spread sheet is always a complete nightmare unless it is really well set out.

Which particular part are you finding is giving you wrong answer?

You can edit your OP. Click the edit button in the bottom right hand side.

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

Dear LittleInch,

That's the kind of you. Really appreaciate that.

The input are the ones with blue font colour.

When I put 59.88 MMscfd (as it is the initial input) in the yellow coloured column as captured below, the 'Horizontal Gas Velocity - calculated and Conds. in Gas Droplet Fall Times gives me error warning

I think its due to the droplet size that is too small.
Is it common to use droplet size below 100 micron?
I saw some other calculations, they generally use 100 - 300 micron size.

Cheers,
EK

 

[eriskawk]

Hi,
Consider the resource attached to support your work
Good luck
Pierre

Dear pierreick,

Thank you! It'll help.

Cheers,
EK

 

[LittleInch]

Like I said, sorting out some else's spreadsheet without lots of back ground explanation is the work of the devil.

When I open the spreadsheet with your numbers in it it seems to imply that the max gas flow is 6 MMscfd??

I can't work out how it calculates actual velocity without knowing the pressure, but I think this might be where you need to use the ACTUAL gas density at your operating pressure and temperature in cell G13.

A calculation in cell Q60 simply divides lbs/hr by the gas density to get ft3/hr.

So you appear to have input the standard gas density at atmospheric conditions.

Put in 12.3 ( approx density at 490 psi) and it all goes green!

Like I said - Other peoples spreadsheet....

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

 

[pierreick]

Hi,
Get a copy of Gas-liquid and liquid-liquid separators by Maurice Stewart and Ken Arnold [Gulf equipment Guides]
Good luck
Pierre

 

[eriskawk]

Like I said, sorting out some else's spreadsheet without lots of back ground explanation is the work of the devil.

When I open the spreadsheet with your numbers in it it seems to imply that the max gas flow is 6 MMscfd??

I can't work out how it calculates actual velocity without knowing the pressure, but I think this might be where you need to use the ACTUAL gas density at your operating pressure and temperature in cell G13.

A calculation in cell Q60 simply divides lbs/hr by the gas density to get ft3/hr.

So you appear to have input the standard gas density at atmospheric conditions.

Put in 12.3 ( approx density at 490 psi) and it all goes green!

Like I said - Other peoples spreadsheet....

Dear LittleInch,

yes it said so, but it's supposed to be more than the current gas flow. as the current gas flow is 59.88 MMscfd.


anyway thanks for the hint!
I have resolved it.

you were right, one of the problems here is incorrect density data.

the fluid properties I input, including mass flow, density, and viscosity are taken from simulation data. It was the simulation data that weren't tabulated well, leading to strange result as we had. The gas density is supposed to be 1.365 lb/ft3

As summary here are the problems:
- density data is not correct
- Q30 check formula is wrong, it should be dependent on cell P32 instead of P34

Now I can goalseek the max capacity very well.


Cheers,
EK

 

[pierreick]

Hi,
To support your work , you may find useful this website

http://www.bobby-strain-group.com/

where the owner offers resource for your application .
Good Luck
Pierre