1 reply to this topic

[davidbarros]

Hello everyone,

Im a chemist and work for a refining company in Colombia.

We've recently purchased a tbp unit wich i'm trying to put into service. we've never used this kind of units before, we only do d86 and d1160 distillations.

i've already created a distillation program and done some runs in the unit. In order to determine if the cuts obtained are ok i performed a d86 distillations of the cuts obtained, but the temperatures are very different that the ones from the tpb.

so i found a way to convert d86 curves to tbp curves but i don't know if they apply to the entire d86 distillation of crudes or the distillation of the fractions obtained. in other words i want to check if the initial and end point of the fractions obtained in a tbp run are ok, so i apply the convertions the wauquier or the raizi and the api technical book to the data obtained on d86 of that cut, for example de light nafta should be between 140°C and 190°C but if i do the d86 its not even close and if i apply the conversions it's not getting better.

so i hope you could point me in the right direction, i really need this unit to be working asap.

thank's for your help.

PS: Excuse me if i wrote something wrong, my english is not the best.

 

[S.AHMAD]

Dave

1. I am not a chemist so I cannot give comments the way you want the answer to be. However, my suggestion as a process engineer who has been working in oil refinery for some solid years. No doubt that process engineers are the main users of the TBP and very well verse with the conversion from TBP to D86 and D86 to TBP but I myself never run a TBP test.

2. My suggestion is that you work out with the CDU process engineer to conduct a unit testrun. A simple testrun, where what you need is sufficient sample of crude oil and products. You also need the product yields. For the overhead LPG and flared gas, obtained the composition using gas chromatography. For the crude and product run TBP and ASTM D86 and density. Using rhe products TBP and the yield, you can re-construct (sometime we called it reconstitute) the crude TBP and compare with the test crude TBP.

3. The reconstituted TBP should be the same as (coincides with) the crude TBP. If not the same, the accuracy of the yields normally suspected. Try to obtain the products yield using tank gauging.

4. Convert ASTM D86 to TBP and compare the results. I suspect some inaccuracy of the two results. Probably the 50% points are very close but the end points (IBP and FBP) are not very satisfactory. Let us know your findings.

Edited by S.AHMAD, 29 December 2013 - 02:36 AM.