15 replies to this topic

[Nikolay_]

Hello,

I can't find the reason of gap between final boiling point of top stream and initial boiling point of bottom stream in HYSYS. Also I observe significant discrepancy of IBP of feed and top stream.

I have modeled distillation column with one feed stream, breath from condenser, top and bottom stream.

The composition of feed was defined as mixture of pure components by DHA (Straight run gasoline). In order to get curves I've used the Boiling Point utility. The results I obtained are below.

 1.jpg   132.33KB   2 downloads

 

Does anybody have any suggestion to resolve it?

 

Best regards,

Nikolai

 

[sgkim]

Hi Nikolai, 

 

IBP in D86 indicates the temperature when first drop is collected from the condenser, while 1% temperature indicates that when 1% of total volume is collected.  Discrepancy in IBP in feed and distillate from a distillation column overhead is quite natural.  

 

D86 apparatus collects evaporates, and quite much amount of light ends may remain in the heating pot. But the distillation column collects most light ends as possible.  To get narrower distillation ranges in either top and bottom products more number of stage or higher reflux would be required.    

 

~Stefano

[Zauberberg]

Nikolai, see referenced materials:

 

https://www.ch2m.com...appendexes_.pdf

http://nptel.ac.in/c...es/103102022/11

 

as well as Handbook of Petroleum Processing by D.S.Jones.

Edited by Zauberberg, 21 March 2017 - 09:59 AM.

[PingPong]

@Nikolai

The IBP and FBP of TBP or D86 predicted by a process simulator (like Hysys) are of no value.

The IBP or FBP determined by actual TBP or D86 distillations in the lab are of little value.

 

So I suggest you do not worry about them.

[Nikolay_]

Thank you all!

The definition of the Gap-Overlap and its typical values were very helpful.

[Nikolay_]

Nikolai, see referenced materials:

 

https://www.ch2m.com...appendexes_.pdf

http://nptel.ac.in/c...es/103102022/11

 

as well as Handbook of Petroleum Processing by D.S.Jones.

Hello,

 

Could you please send me example Factor read calculation (for case "Overhead to Top Sidestream Product") that mentioned at figure 3.A.2. of Handbook of Petroleum Processing by D.S.Jones? I confused with its reflux ratio formula. As I know the Reflux ratio is calculated as reflux flow/distillate flow. In the Handbook formula of reflux ratio is different.

[Zauberberg]

You can access Jones' book here: http://ajaysingh.in/..._Processing.pdf

[PingPong]

I confused with its reflux ratio formula. As I know the Reflux ratio is calculated as reflux flow/distillate flow. In the Handbook formula of reflux ratio is different.

Read carefully what it says:

This Reflux ratio is defined as:

Hot overflow in gallons per hour leaving the lightest product draw-off tray

-----------------------------------------------------------------------------------------------------
Total product gallons per hour entering the lightest product draw-off tray

It is a kind of Reflux Ratio defined as L/D, but it is based on flowrates in liquid gallons, not moles. The mentioned product is the distillate D.

 

Those graphs are known as Packie's curves. In the old days engineers used them to design crude distillation columns, but nowadays, with process simulators widely available, nobody uses them anymore for design purposes. They are still useful to get a rough idea of the relation between 5% - 95% ASTM gap and F=(L/D)*N

In which N is number of actual trays in the section between two products, not theoretical trays.

 

Note that the curves are based on 5% - 95% gaps, not 6% - 95% gaps. That 6% is a misreading by the OCR software that was used to digitize the old graphs.

You can see that misreading also in the text above the curves where it says

OVHO instead of OVHD and TCP instead of TOP.

Edited by PingPong, 22 March 2017 - 04:19 AM.

[Nikolay_]

You can access Jones' book here: http://ajaysingh.in/..._Processing.pdf

Thank you for link.

Can you explain the nature of appearance the gap of product on ASTM D86? Where do components disappear? Is it error of method?

 

Kind regards,

Nikolai

[Nikolay_]

 

I confused with its reflux ratio formula. As I know the Reflux ratio is calculated as reflux flow/distillate flow. In the Handbook formula of reflux ratio is different.

Read carefully what it says:

This Reflux ratio is defined as:

Hot overflow in gallons per hour leaving the lightest product draw-off tray

-----------------------------------------------------------------------------------------------------
Total product gallons per hour entering the lightest product draw-off tray

It is a kind of Reflux Ratio defined as L/D, but it is based on flowrates in liquid gallons, not moles. The mentioned product is the distillate D.

 

Those graphs are known as Packie's curves. In the old days engineers used them to design crude distillation columns, but nowadays, with process simulators widely available, nobody uses them anymore for design purposes. They are still useful to get a rough idea of the relation between 5% - 95% ASTM gap and F=(L/D)*N

In which N is number of actual trays in the section between two products, not theoretical trays.

 

Note that the curves are based on 5% - 95% gaps, not 6% - 95% gaps. That 6% is a misreading by the OCR software that was used to digitize the old graphs.

You can see that misreading also in the text above the curves where it says

OVHO instead of OVHD and TCP instead of TOP.

 

Thank you!

 

I have 26030 m3/h of vapor from top column and 199 m3/h of reflux (to 1st plate). Is the Reflux Ratio of 26030/199=130? The value is questionable.

[Nikolay_]

 

I confused with its reflux ratio formula. As I know the Reflux ratio is calculated as reflux flow/distillate flow. In the Handbook formula of reflux ratio is different.

Read carefully what it says:

This Reflux ratio is defined as:

Hot overflow in gallons per hour leaving the lightest product draw-off tray

-----------------------------------------------------------------------------------------------------
Total product gallons per hour entering the lightest product draw-off tray

It is a kind of Reflux Ratio defined as L/D, but it is based on flowrates in liquid gallons, not moles. The mentioned product is the distillate D.

 

Those graphs are known as Packie's curves. In the old days engineers used them to design crude distillation columns, but nowadays, with process simulators widely available, nobody uses them anymore for design purposes. They are still useful to get a rough idea of the relation between 5% - 95% ASTM gap and F=(L/D)*N

In which N is number of actual trays in the section between two products, not theoretical trays.

 

Note that the curves are based on 5% - 95% gaps, not 6% - 95% gaps. That 6% is a misreading by the OCR software that was used to digitize the old graphs.

You can see that misreading also in the text above the curves where it says

OVHO instead of OVHD and TCP instead of TOP.

 

Does anybody have original Packie's curves (scan)?

[PingPong]

I have 26030 m3/h of vapor from top column and 199 m3/h of reflux (to 1st plate). Is the Reflux Ratio of 26030/199=130? The value is questionable.

You should read carefully what was said: product flowrate is to be converted into liquid volume.

 

Also the product is the distillate vapor plus liquid, not the gross overhead of the column.

 

Can you explain the nature of appearance the gap of product on ASTM D86? Where do components disappear? Is it error of method?

Components do not disappear. No error of method.

 

Suppose you want to produce a naphtha and kerosene based on a TBP cutpoint of 180 ^oC between them. Only with an infinite number of trays and infinite reflux ratio would it be possible to have no 180⁺ components appearing in the naphtha and no 180^- components in the kerosene. In reality it is not a problem to have some 180⁺ components in the naphtha and some 180^- components in the kerosene, so nobody uses a huge amount of trays or huge RR as that would be uneconomical. As a result there will be an overlap in the TBP curves of the naphtha and the kerosene, and consequently also an overlap in the ASTM D86 curves. The IBP of the kero will be lower than the FBP of the naphtha, however the D86 5% of the kero could nevertheless be higher than the D86 95% of the naphtha. In that case we say there is a 5%-95% ASTM gap. But it could also be that there is no 5%-95% gap but an overlap (D86 95% of the naphtha higher than D86 5% of the kero). Whether your column gives a 5%-95% ASTM gap or overlap between adjacent products depends on its design and operation.

Edited by PingPong, 22 March 2017 - 09:51 AM.

[Nikolay_]

Thus, in order to get economical process should I chose the number of trays and reflux ratio appropriate minimum gap or some overlap (5-95 ASTM D86)?

 

[Nikolay_]

What do you think about attached example? Reflux ratio was calculated with mass basic.  2011_chaudhuri_u_r_fundamentals_of_petroleum_and_petroch.pdf   1.24MB   23 downloads

[PingPong]

The Packie method dates back to 1941. In those days everybody worked with liquid barrels, sometimes also for vapors. So Packie defined his RR based on liquid barrels.

But if the distillate D is all liquid it makes of course no difference whether you calculate L/D based on volume or on mass, because L and D will have the same specific gravity (density).

If the distillate is partly vapor plus partly liquid, or all vapor, then there will be some difference whether you calculate L/D based on liquid volumes or on mass.

 

Packie is history. His method was never very accurate, but his graphs are still useful to see the relative impact of changing L/D and/or N on the 5%-95% gap or overlap. A process simulator is much more accurate than Packie to calculate the real gap or overlap.

 

Some books, like the one by Chaudhury, demonstrate Packie's method, but you should realize that all those books are not suitable to design a CDU without further learning in practice from experienced design engineers. Those books are usually written by persons who never designed anything that actually got built and performed as it was intended.

[Nikolay_]

Thank you for useful explanations.

 

Best regards,

Nikolai