2 replies to this topic

[abhiped]

Hi all,
I am a beginner in this area and would appreciate some help. I skimmed through the forum topics and couldn't find one similar to my case (or maybe I couldn't identify). Anyways here goes:
I am trying to calculate the relief load (fire case) for a high pressure separator. I have two specific problems
1.) Calculating the latent heat of vaporization
2.) Relieving temperature.

The relieving pressure is about 3600 psia. The contents of the separator are a lot of heavy and light hydrocarbons and also N2, H2S, H2 and H20. Its a mixed phase with 17% vapor. If I just consider the liquid phase, how do I go about calculating the relieving temperature and latent heat of vaporization?
Let me know if I should give more information.

thanks

[jprocess]

Dear Sir,
When having a hydrocarbon mixture,calculation of relieving temperature and latent heat of vaporization for fire case relief calculation are not easy in my idea because I have not ever found a reliable method!
For relieving temperature:
1.API 521 method: using ideal gas equation and using operating temperature and pressure and relieving pressure to obtain the relieving temperature.
My idea: For lots of instances I found it unreliable and results values are far different from simulators like HYSYS.
2.Using simulators like HYSYS to obtain the relieving T @ vapor fraction=1 and relieving pressure
My idea:Is my prefered method and accurate enough.
3.Using HYSYS depressuring utility with Relief as selected valve:
My idea:I do not know that how much the peak flow rate and relieving temperature are reliable?
For Latent heat calculation:
1.I guess that for such a high pressure that you have for your problem the contents of vessel will reach its critical point before the psv reach its set pressure.In these cases the API 521 recommends a value of 50 Btu/lb for latent heat.
My idea: I have not any sense or background form this value which is proposed by API and using this value may overload your HP flare network.This value(with a minimun of 40 Btu/lb) is also recommended in a famous article:
"Fires,Vessels and the Pressure Relief Valve", Wing Y.Wong MAY 2000
2.The above mentioned article recommends flashing of 5-30% of liquid contents using simulators like HYSYS and using the average(or minimum) value of first and second flash liquid products.
My idea: I have seen this method in some other guidelines and seems to be good enough.
3.Mr Phil Leckner recommends flashing of 1% of liquid contents to obtain the value.
My idea: The resultant value by this method is far different from hysys reported value for a stream at properties page.
4.Using average MW of mixture and curve which is in appendix D of API 520.
My idea: The curve is essentially for pure components and using it for mixtures will result unreliable and as I found low values.
Your valuable comments are appreciated.
Best of Luck.

[JoeWong]

abhiped,

You may read some valuable responses by Mr. Montemayor (HERE), Phil (HERE & HERE) & Latexman (HERE)

JoeWong