2 replies to this topic

[Skyline]

Dear all,

It's my first time using Aspen 7.2 to find the properties of crude oil at 330 C and 10.38 kg/cm2g. The result is as I attached. I found 3 phases stream: 1 vapor phase and 2 liquid phase.
The vapor phase has 0.1453 vapor/phase fraction
The liquid phase no 1 has 0.0051 vapor/phase fraction
The liquid phase no 2 has 0.8496 vapor/phase fraction

What does it mean? I usually use HYSYS 3.2, and never found this kind of results. I thought that for liquid phase, the vapor fraction has to be 0. The assay distillation is as follow:
SG 0.8584
IBP 43 C
5% 116 C
10% 159 C
20% 208 C
30% 240 C
40% 296 C
50% 338 C









Thank you for your help

Best regards,


EY

Attached Files

•  untitled.JPG   187.21KB   7 downloads

[Technical Bard]

What thermo package are you using? You may need to adjust the flash setting to force it to only look for one hydrocarbon liquid phase. The problem here is that the solver is finding that splitting off some of the oil into a separate phase results in a lower Gibbs free energy - but it's not a real possibility.

Your assay is incomplete. (338°C at 50% - what does the curve past that look like?) This is bad because it the simulator will extrapolate and give eroneous results. This could be part of the problem, especially since your oil is light (gravity similar to diesel) and the flash condition you have selected is relatively hot.



The phase fraction of the two liquid phases is indicating what fraction of the total stream is in that phase. The three values add to 1.000

Dear all,

It's my first time using Aspen 7.2 to find the properties of crude oil at 330 C and 10.38 kg/cm2g. The result is as I attached. I found 3 phases stream: 1 vapor phase and 2 liquid phase.
The vapor phase has 0.1453 vapor/phase fraction
The liquid phase no 1 has 0.0051 vapor/phase fraction
The liquid phase no 2 has 0.8496 vapor/phase fraction

What does it mean? I usually use HYSYS 3.2, and never found this kind of results. I thought that for liquid phase, the vapor fraction has to be 0. The assay distillation is as follow:
SG 0.8584
IBP 43 C
5% 116 C
10% 159 C
20% 208 C
30% 240 C
40% 296 C
50% 338 C

Thank you for your help

Best regards,


EY

Edited by Technical Bard, 22 July 2011 - 12:41 AM.

[Skyline]

What thermo package are you using? You may need to adjust the flash setting to force it to only look for one hydrocarbon liquid phase. The problem here is that the solver is finding that splitting off some of the oil into a separate phase results in a lower Gibbs free energy - but it's not a real possibility.

Your assay is incomplete. (338°C at 50% - what does the curve past that look like?) This is bad because it the simulator will extrapolate and give eroneous results. This could be part of the problem, especially since your oil is light (gravity similar to diesel) and the flash condition you have selected is relatively hot.



The phase fraction of the two liquid phases is indicating what fraction of the total stream is in that phase. The three values add to 1.000

Dear all,

It's my first time using Aspen 7.2 to find the properties of crude oil at 330 C and 10.38 kg/cm2g. The result is as I attached. I found 3 phases stream: 1 vapor phase and 2 liquid phase.
The vapor phase has 0.1453 vapor/phase fraction
The liquid phase no 1 has 0.0051 vapor/phase fraction
The liquid phase no 2 has 0.8496 vapor/phase fraction

What does it mean? I usually use HYSYS 3.2, and never found this kind of results. I thought that for liquid phase, the vapor fraction has to be 0. The assay distillation is as follow:
SG 0.8584
IBP 43 C
5% 116 C
10% 159 C
20% 208 C
30% 240 C
40% 296 C
50% 338 C

Thank you for your help

Best regards,


EY

Dear Tech Bard,

Thank you for your explanation. Just as you said, the assay wasn't complete. I tried to fix this problem by following your explanation: I make sure that the extrapolation only for lower end only, and not for the higher end. It results in two phase stream: liquid and vapor only. Looks like the extrapolation causes this problem. Thank you very much for the explanation.

Best regards,



EY