18 replies to this topic

[Shariq Khan]

Dear All,

I was trying to solve a problem of Shell DEP 31.22.05.12 but I got stuck in the calculation of vdl (settling velocity of light droplet) and vdh (settling velocity of heavy droplet) because my results are not matching with the values mentioned in DEP. I am also attaching the PDF of shell DEP and excel file which I prepared in order to solve the example. I will be thankful if anyone can help.
 Shell Sep.xlsx   38.55KB   1016 downloads

 31220512.pdf   2.64MB   1257 downloads

Regards,

Shariq

 

Edited by Shariq Khan, 15 May 2013 - 12:00 AM.

[Dacs]

Without going through your documents in detail, I think you're asking on how DEP was able to get the values for vdl and vdh, which looks like liquid phase horizontal velocity.

 

In that case, it simply means how fast your liquid moves with respect to the levels in the drum.

 

For example, let's say your oil portion is between LLL and HLL, then the oil horizontal velocity (which I presume vdl in this case) is Oil Volumetric Flowrate / Liquid Cross-Section Area. Liquid Cross-Section Area is defined by the region made by the oil section, which basically means the cylinder cross-section area between LLL and HLL.

 

[Shariq Khan]

Hi Dacs,

 

Yes I was not able to get the values for vdl and vdh. vdl is the settling velocity of light phase droplet and vdh is the settling velocity of heavy phase droplet.

 

I am also attaching the appendix containing the formulas for solving the example for better understanding.

 

 Appendix-IV.pdf   2.52MB   677 downloads

 

 

 

 

[M.Sadiq]

@Shariq

 

I think you indicated the values of V_{dh, set} and V_{dl, set} as i highlighted in RED BOLD in the attached image. Sometime ago, i also got stuck at this step. The Fomula is so confusing V_{dh, set} = Ql / (L_set*D*0.8) to which i was unable to predict the values of V_{dh, set} at Position I,II,III and IV. Can any one please resolve the issue.

Attached Files

•  Shell_R1.png   206.78KB   77 downloads

[Dacs]

You're analyzing it the wrong way.

 

The equations that you showed are usually used when designing a vessel (when you're required a vessel dimension to do disengagement given feed conditions and target particle diameter)

 

But the problem shown is essentially rating a vessel. Difference is, you already have a vessel and you just need to check whether this vessel is sufficient in size to disengage a given feed.

 

In this sense, you have to use the dimensions given in the vessel and essentially back calculate the settling velocity given your data.

 

When you do that, you'll be able to match the settling velocities given in your example. I was able to.

 

Good luck

Edited by Dacs, 15 May 2013 - 03:13 AM.

[M.Sadiq]

Yup, the vessel is rated as per specified dimensions in the Example Problem (ID=1.5 m and  L=5 m). Shell specified the formula V_{dh, set} = Ql / (L_set*D*0.8) before proceeding the example problem. Please share the updated Excel sheet so i will understand the logic that you able to calculate the values at different positions.

[Dacs]

I didn't touch the spreadsheet (I did the calculation by hand).

 

Anyway, just to guide you on how to do this:

Let's work on the light liquid (oil) side. You got Vl,ax, which is the horizontal velocity of oil.

You also have Lset, which is the effective settling length. Divide Lset by VL,ax will give you the horizontal residence time (of oil).

When doing separator design, the goal is your horizontal residence time should be greater than your vertical residence time. Since you already have the vessel dimension, what you can do is to set both equally. In this way, you can back calculate the minimum particle diameter your existing vessel can disengage.

Going back to the analysis, what do you disengage in oil? Obviously it's water. So you have to consider the settling velocity of water particles in oil. Where does that happen? Obviously in the oil portion in the vessel.

 

Given the horizontal residence time and the vertical distance between the oil-water interface and the feed (because water has to come from the feed which is on the topmost portion of the vessel), you can calculate the equivalent settling velocity.

 

You don't need to use the equation given in DEP because you are doing vessel rating. You already have the vessel dimension at hand so what you can do is to back calculate everything to get the effective settling velocity (and ultimately the minimum particle diameter that can be possibly disengaged).

 

[M.Sadiq]

@Dacs

 

You may be right but my concern is; How 'Shell DEP' calculate the following values of V_{dh, set} at Position I,II,III and IV.

 

                                                                                                  Position-I            Position-II              Position-III            Position-IV

Heavy phase Droplet Settling Velocity                   V_{dh, set}                   2.1                     2.4                         2.9                           2.2

Light phase Droplet Settling Velocity                      V_{dl, set}                    1.4                     1.2                         1.1                           1.3

[Dacs]

Do what I said in my last post and you'll be able to get it

[Shariq Khan]

Hi Dacs,

 

Thanks for your response. Please tell the equation you are using to calculate settling velocity?

[M.Sadiq]

@Dacs

 

Thanks for your valuable responses. I have calculated V_dh,set at different Position using the logic  you have specified in above responses. The file is attached.

 

However, it is still unclear that what is the logic behind the SHELL DEP formula and how one can implement this formula for the rating of Vessel.

 

 

 

@shariq

 

Please find the attached spreadsheet for clarity.

Attached Files

•  Vh, set.xlsx   8.39KB   403 downloads

[Art Montemayor]

Shariq:

• If you work for Shell, ask your supervisor.
• If you don't work for Shell but are doing a design for Shell using their specifications, then ask your Shell  representative.
• If you don't work for Shell and obtained a Shell DEP from others to use in a design, then you are using Shell property for your benefit without permission.  If this is so, then you have no business toying with other people's specifications without permission and you are using tools you know little about.

Which is it?

[Dacs]

The reason that I did not give out any calculation details is we're treading with information from a proprietary source. I just gave the benefit of the doubt that you have proper access to Shell DEP and I just helped you to understand how things are done.

That said, I cannot comment on DEP itself because I'm not supposed to have that information. I can help you in capacity of whatever information that can be legally provided.

Shell DEP would probably have been formulated from their experience operating their installations. I never used it before so I can't comment more on it. Not to mention it's proprietary.

And as an advice, you better get hold of references (or someone with more experience in your organization) that gives out in detail on how to carry out separator sizing. I can go on lengths in explaining the rationale of my answer but I may not be able to get across the message to you effectively.

What I can mention with confidence is that gravity settling is not something considered a trade secret so the approach on sizing would be the same, using Shell DEP or otherwise.

If you have questions about separators in general, I'll try to do my best to answer. But try to avoid touching stuff that we're not supposed to touch

Edited by Dacs, 15 May 2013 - 08:01 AM.

[Bobby Strain]

Seems as though everyone in the world has access to Shell's DEP. But their use is limited to Shell's work, or Shell's authorization, which is clearly stated in the DEP. You can visit my site, which you can find by Googling my name. There you will find software and links that will give you all you need. And, maybe you can find something of use at CheCalc.com.

 

Bobby Strain

[Shariq Khan]

@Dacs:

 

Thanks for your valuable responses. Nobody is forcing you to share proprietary material but you can share general information about separators and the reason for asking settling velocity equation was just to confirm and nothing else. Hope you understand. 

[breizh]

You may find some interest reading this resource :

 

http://www.red-bag.c...sel-sizing.html

 

 

Breizh

Edited by breizh, 29 June 2014 - 07:28 PM.

[Dacs]

Shariq:

 

Thing is, we're discussing something based on a document that's not generally publicly accessible. You may have access on it, but I don't.

 

Anyway, I think I already gave my answer to your query. If you have clarifications, let me know.

[pika_doh9]

Hey guys, I am doing this calculation too and I am very stuck on step 4, the control time calculation

I just cannot understand how that are calculating Acon for each of the positions

Acon (m2) 0.505 0.520 0.505
ΔVhd (m3) 0.143 0.152 0.143

can anyone help with this?

[Bobby Strain]

It's calculated using ACos, which is a geometry function. You can derive the formulas from basic calculus and geometry. If you visit my website (Google my name) you will find an application that calculates various aspects of horizontal cylindrical drums. Including partial volumes. And it has been verified to be accurate. It's compiled, and, unlike Excel formulas, nobody but me can change the formulas. You will also find software to rate horizontal separators. Shell's design restrictions will give a larger vessel than required. Be aware, too, that Shell regularly updates their design guides.

 

Bobby

Edited by Bobby Strain, 30 June 2014 - 06:57 PM.