8 replies to this topic

[processengbd]

Client has provided some ASTEM D 86 Data without the specific gravity data? How can i characterize the data?

Edited by processengbd, 17 May 2014 - 06:05 AM.

[PingPong]

Pity that the assay does not detail composition of C4 and lighter, so you need to assume something, or ask client..

 

Assay gives quantities for gasoline, kero and gasoil. There must therefor also be gravities available for them, I am sure. Ask client.

 

If client cannot give further info, for fraction gravities you have two options:

 

- assume that the UOP-K (Watson K) of each pseudofraction is the same as that of the whole condensate,

 

- assume a UOP-K profile, but such that the sum of the pseudofractions yields the proper gravity of the condensate.

 

I would use the second option, even though that is more work.

[processengbd]

Pity that the assay does not detail composition of C4 and lighter, so you need to assume something, or ask client..

 

Assay gives quantities for gasoline, kero and gasoil. There must therefor also be gravities available for them, I am sure. Ask client.

 

If client cannot give further info, for fraction gravities you have two options:

 

- assume that the UOP-K (Watson K) of each pseudofraction is the same as that of the whole condensate,

 

- assume a UOP-K profile, but such that the sum of the pseudofractions yields the proper gravity of the condensate.

 

I would use the second option, even though that is more work.

Thank you for your reply.

 

 

 

Would you please suggest me a reference... How can i find UOP-K or watson K value of condensate? & on what basis i assume UOP-K profile?

 

I am doing characterization in HYSYS. WIll it be possible to do it in HYSYS or in PRO II for that matter. I can use both.

 

As you can see I am very new to this sorry.

 

Processengbd

[PingPong]

Yes, I see, and that is worrying if you want to design something that your company will have to guarantee.

 

UOP-K is calculated from the distillation data and specific gravity of the condensate. You can find that in many books on petroleum refining,  in the API Technical Data Book, and on many websites. I quickly estimate the UOP-K of that condensate to be 11.3 which is rather low.

 

Both HYSYS and PRO/II can be used for characterization.

 

Before you do anything final, ask the client for a breakdown of the C4 minus, and for the gravities of the Gasoline Cut, Kerosene and Gasoil fractions in the assay, and make sure you get their response in writing.

[Proceed123]

Processengbd,

 

The following is a path you can try for obtaining first-approach density estimates of condensate fractions for which their boiling point are reported in the D86: 

 

1. Click on this link: http://www.ijens.org...JET-IJENS.pdf .
2. Once there, go to Tables I and II. They show generalized set of properties (including SG at 60/60ºF) for pure hydrocarbon components with carbon number between C6 and C50 (they are further refinements of the original data published by Katz and Firoozabadi for condensate/crude oil systems back in 1978). 
3. For any boiling point datum of interest in your D86, you  may obtain (directly or by interpolation) the corresponding SG, based on tables I and II data. Alternatively, you may average the values obtained from the two sources.
4. For the heaviest fraction (i.e. that boiling between the D86 End Point and 100 vol%), you may i) mathematically extrapolate the distillation curve, ii) estimate what the boiling point is at 100 vol.%, iii) averaging the End Point and 100 vol% boiling point values and use the resultant average as representative mean average boiling point of the heaviest fraction. Then find from Tables I and II the corresponding SG values for the resultant mean boiling point.

Provided the condensate is not too light (i.e. API not lighter than, say, 43º and/or boiling point at ~55 vol.% not lower than, say, 160ºC, you might estimate the Watson K of the overall stock from its viscosity data (viscosity at two temperatures will be required) via the algorithm published in the article New Equations Predict Boiling Point from Viscosity, Gravity (Gomez, J.V., Oil & Gas Journal, Oct. 7, 1996, pp. 92-98).

 

If you are allowed to disclose any other data that is provided by the condensate assay, I might be able to come up with something else.

 

Good luck.

[Proceed123]

Hum! I do not know why, but it seems the link does not work when posted like here (although I am positive it is active). 

 

I am copying-and-paste it again: 

 

http://www.ijens.org/104902-1818%20IJET-IJENS.pdf 

 

If it still does not respond, you may write to me and I will send it to you via email note. 

 

Regards. 

[PingPong]

Both your links don't work because they contain rubbish at the end (after .pdf).

Copy them into notepad and you will see what I mean.

 

Correct link: http://www.ijens.org... IJET-IJENS.pdf

 

Info in link does however not help. Condensate does not consist of pure paraffins (alkanes) but also contains naphthenes and aromatics.

 

Watson K shall be calculated from D86 data and SG, and in this case that gives K_W = 11.3 which, in my opinion, is rather low, so I doubt the assay data, but without the gravities of the gasoline, kerosene and gasoil cuts I cannot for sure prove that they are wrong.

 

The fact that there are formulas floating around that imply a relation between K_W and viscosities does not mean that that is a proper way to determine K_W

The accuracy of such formulas is lousy.

 

By the way: that also applies to formulas for estimating the MW

 

In the mean time processengbd deleted the attachment with the assay data, so I think he does not intend to further communicate on this subject, probably because he is not allowed for confidentiality reasons.

Edited by PingPong, 20 May 2014 - 05:23 AM.

[Proceed123]

PingPong; 

 

I suggest you not to dismiss the K_W = 11.3 too quickly. 

 

There are nasty condensates out there. Personally, I am most familiar with a couple of Middle East ones with values in the 10.8 to 11.3 range (not to mention their residua' values!). I am not talking about doubtful published data from obscure sources; I am talking about actual, double-checked assays, performed by a world-renowned testing company back in mid-2013. I was personally involved at all stages so I can tell you not all condensates are created equal. 

 

Yes, condensates do not consist purely of n-paraffins; that's why I clarified in advance the path to be aimed to estimate first approach values (which in the face of lacking data might be informative to Processengbd). Density-wise, I have tried that approach with samples of the nasty stuff mentioned above and realized it can provide reasonable values (not for final design purposes, but good enough as to give a flavor of the material nature). 

 

The other approach I suggested (in case some other key properties of the condensate were available) was that of 'internal consistency' checking, which usually works fine (depending upon of the reliability of the data, of course). 

Edited by Proceed123, 20 May 2014 - 08:09 AM.

[Zubair Exclaim]

A humble suggestion ask  your client to provide you with complete data .. that is his responsibility and not  a hard thing to do …..

assumption and calculation from k factors can be close in proximity and but can never be completely accurate

If you form your simulation on a weak basis it will produce a less reliable out put.