34 replies to this topic

[processengbd]

Sir,

We have been doing some prliminary design of three phase separator. Our to be build plant capcity is 20MMSCFD. So you can guass the plant is very small.

Now during desining horizontal three phase separator we faced some problem. We know that normal engineering practice state that minimum light for every level like from Liquid level high high (LLHH) to liquid level high (LLH) shuld be at least 6 inch.

But if we use this minimum hight then vessel diameter becomes too big and retention time becomes very very big. So from the economic point of view we dont need the vessel that big right?

As far as i can understand minimum six inch hight is given so that level controller can perform well. Is there any other reasons?

I have attached preliminary design scetch please have a look. I will very much appreciate your valuable suggestions.

 

 1.jpg   24.86KB   187 downloads

 

 2.jpg   26.2KB   113 downloads

[Dawood]

Dear Sir

 

I am a Process Engineer by profession and working in Oman.

 

We are Manufacturing and desiging 3 Phase Separators and its Internals as well. So if you are interested we can manufacture it for you.

Please let me know if you are interested.

Edited by Art Montemayor, 05 February 2013 - 09:56 AM.

[Art Montemayor]

Dawood:

 

I removed your email address.  As our Forums Director has previously stated, our members should not post their email addresses in our Forums unless they want to initiate a lot of trash, viruses, and other bad things in their email in-box.

 

Our Forums have no problem in you advising about your services and how you can help our members out with your engineering services or products.  You can advertise in the ChE Resources Website by contacting Chris Haslego.  If you prefer not to pay for advertising, then you can contact the indicated member through the messaging service.

 

Thank you

[Art Montemayor]

Processengbd:

 

If you have designed the 3-phase separator and would like to receive our members' comments and advice on your design, then please submit your calculations - preferably in an Excel spreadsheet - so that we can all see how you arrived at a 3 foot diameter.

[processengbd]

Thank you very much  Mr Art Montemayor. I have attached the spreadsheet.

Processengbd

Attached Files

•  3 Phase Separator.xlsx   18.01KB   1088 downloads

[Bobby Strain]

This spreadsheet is an excellent example of why we should not use this approach for any engineeering calculation. Who is going to go through all this stuff to see if it is correct? Art, maybe you can enlighten me.

 

Bobby

[thorium90]

well, its not that bad. Just go to Formulas, click Show Formulas and start scratching your head. At least thats what I did...

Edited by thorium90, 06 February 2013 - 12:07 PM.

[Art Montemayor]

 

Bobby:

 

It will be a pleasure to enlighten you on precisely why you are dead wrong in not going “through all this stuff”.

 

If you take less than one minute out of your life and open the spreadsheet you will see that the “retention time” (residence time) being used to arrive at a 3 ft diameter by 5 ft long 3-phase separator is 3 minutes.  A ridiculous value by any experienced engineer's measure.

 

If you are going to use your time to discuss, comment, and make recommendations on a sketch without knowing what is behind the engineering logic and basis of fabricating such a useless vessel, that is fine with me.  However, I wouldn’t put much faith on your comments.

 

How do you decide the size and dimensions of any separator you prescribe or recommend?  By using a summary computer printout without knowledge of the algorithm, equations, logic, or basis that is used for the answer?  After 53 years in the business I don’t trust ANY engineer – regardless of who she/he is – to dictate the size and dimension of any process equipment without backing it up with detailed accurate and acceptable calculations.  How is one to identify the responsibility and liability for the correct performance of the equipment?  Does one place it on the computer?

 

My present employer and all those that I have worked under in the past certainly would not accept that and I am sure no business investor would.  In the end, unless you own your own business, that’s who engineers have to respond to.

 

[thorium90]

guys like me stare at a spreadsheet for 10s and ponder for 10mins, guys like Art stare at a spreadsheet for 10s and can tell you at least 10mins worth of advice.

 

A quick search from the book :Petroleum and Gas Field Processing By Hussein K. Abdel-Aal, Mohamed Aggour, M. A. Fahim

"retention time usually range from 3 - 30min", " if lab data are not available, use a retention time of 10 min"

 

attached an excerpt. the equations in the spreadsheet suddenly look familiar. haha

Attached Files

•  separator sizing.pdf   274.7KB   985 downloads

Edited by thorium90, 06 February 2013 - 01:49 PM.

[processengbd]

Thank you very much for your valuable insight.

 

But I have followed GPSA for my calculation of retention time. For hydrocarbon water separation it states that 3 min retention time is enough.   Untitled.jpg   19.63KB   60 downloads  Untitled2.jpg   18.09KB   61 downloads

 

I have done some other calculation according to an AIChE (1) paper. Which gives me retention time calculation. There i found that it will be enough to give 3 min retention time. Please have a look.

 

 Untitled3.jpg   63.04KB   89 downloads  Untitled4.jpg   45.78KB   70 downloads  Untitled6.jpg   9.84KB   50 downloads

 

 

"Successfully Specify Three-Phase Separators" Wayne D. Monnery and William, Y. Svrcek, University Calgary

[Art Montemayor]

Typical retention (or residence) time used in the Hydrocarbon Industry applications - where the greatest majority of 3-phase separators are applied - is 15 to 30 minutes.  This is based on my experience.

 

What you are using as GPSA information is based on 2-phase liquid-vapor separators, a horse of a different color.  In 3-phase separators you are dependent on the heavier of two liquids in an emulsion to separate out by gravity, with the water phase gravitating to the bottom and crude oil floating on top.  This is analogous to a decanting Unit Operation.  You could never logically expect that all it would take is 3 minutes for two liquids to separate completely by gravity.  It just won't happen in the real world.

[processengbd]

Thank you very much for your explanation.But Mr Art Montemayor i have retention time data for three phase separator. It says that if you have oil and water mixture you could have 3-5 minute retention time.  Untitled5.jpg   111.25KB   50 downloads

 

 I was also very skeptic about three minute retention time. So i calculated theoretical retention time necessary for the separation of heavy liquid out of light liquid and light liquid out of heavy liquid. I found out that (according to theoretical calculation ) three minute time is sufficient.  Untitled6.jpg   9.84KB   34 downloads

 

 Mr Art Montemayor the calculation sheet i showed you is for my personal calculation. We have different in house spreadsheet will you have the time to analyze it? It does only calculate the retention time for two compartment at different level. I would be great full if you give some comment. (Macro was used to calculate the different sentimental volume)

 

Mr Art Montemayor I think experience is the main thing in designing. Specially separator. I very much appreciate your comment.

Attached Files

•  condensate flash.xlsm   1.75MB   498 downloads

Edited by processengbd, 06 February 2013 - 10:14 PM.

[Bobby Strain]

This is really a liquid-liquid separator. Gas separation requires minimum space. So, your sizing should focus on separating the two liquid phases. And, like Art suggests, check for difficult separation if the fluid has passed through a control valve. And you need to check to see which phase is most likely continuous. If you don't know, design for either phase continuous. For separating two liquid phases, one looks for an L/D which provides for longer settling length. And, pay attention to the liquid velocity. Maybe your spreadsheet does all of this. If not, consult your company's design procedure.

 

Good luck,

Bobby

[processengbd]

Thank you very much Bobby Strain for your comment. Yes basically a liquid liquid separator. So vapor space calculation is not that important or separator design controlling factor. This is natural gas dehydration process. Oil and water separated from natural gas will come to this three phase separator. Yes through a level control valve of initial two phase separator. As per GPSA L/D should be within 2.5-5. AIChE paper gives more specific L/D in different pressure range.i would be glad if you have some comment on it.  Untitled8.jpg   20.62KB   34 downloads

[Bobby Strain]

Your separator of 3' dia. x 5' should be quite adequate. You need only 12" vapor space, so liquid can occupy the rest. Now you only need to apportion the space for normal level and maximum level for oil and water.

 

Bobby

[processengbd]

Thanks Bobby. Do you have any comment about L/D ratio? I wanted to know the condition's or criteria which controls selection of L/D ratio.

Processengbd

[chem101]

Retention time in Three Phase separator depends on operating temperature and densities of
liquids. if difference between (liquid) densities is close to 10 lb/ft3 then
typically 5 minutes applied for retention time. if difference between densities
is greater than 15 - 20lb/ft3 then 15 -30 minutes.

 

3 minutes is usually used when your gas volume is higher than liquid and
governing factor in sizing separator is gas,

 

[Bobby Strain]

The article you have by Monnery has a good explanation. The software at my site is designed to provide dimensions to meet the separation criteria and also provide the lowest weight (cost) vessel. You won't find this feature in the spreadsheets that you get. I also have an article on flare ko drums written about 15 years ago, and it explains the procedure. And it is described in the API relief guides, too. If you send me a message from my website (google my name to find it) with your e-mail address, I will send the article.

 

Bobby Strain

Edited by Bobby Strain, 09 February 2013 - 10:47 AM.

[processengbd]

Thanks chem101 for your comment. Yes normal intuition says that density difference is the driving force of liquid liquid separation. That would mean that when we have two fluid which has similar densities will take more time to separate. Can you please tell me the name of the reference? 

 

Thanks Bobby for your advice. But I cannot see the algorithm you used. So I don't feel very confident. I have set you my mail id.

[chem101]

I don't have any artical or book where i can refer you to. But I have worked with an engineer who had extensive experience with separation process at NATCO (the company is bought by Cameron).

He had given us lecture on separation. where he mention the difference in densities. Also if you have access to API 12J. in  API 12J there is recommendation for rentetion time based on Oil Gravity and temperature. NATCO have conducted lot of Research on separation process in lab as well as in fields.

 

The artical you  mention about seprator in my opinion is general artical it does not talk about specific liquid.

 

Thank you

Chem101 

Edited by chem101, 11 February 2013 - 02:36 PM.

[breizh]

Consider this resource , it might help 

 

Breizh

[processengbd]

Thanks Chem101.

 

Moez Please create another thread. Then post your query.

 

processengbd

[prgerber]

Could you share the password?  @processengbd

 

Thank you.

[Benoo]

Hi Chem101,

 

There is something in your post that I don't understand, basically the higher the difference in densities, the faster the separation of two liquids but what you say is that when the difference is higher we need longer retention time even though the separation is going to be faster. That doesn't make sense. Please advise.

 

Regards,

[ali_fz]

May be this new topic be useful for 3-phase drum sizing:

 

http://www.cheresour...ng-spreadsheet/