Fire relief load (vaporization+expansion) 2

[will48]

Colleagues,

When performing a fire case relief calculation, API521 equation for heat load against wetted area (Q=21000A^0.82, etc.) will give a relieving load due to vaporization. Do I also need to include an additional load due to gas expansion in the vapour space (Equation 8 in API521, W = 0.1406...), or is Equation 8 purely for 'dry' vessels with no liquid?

regards
W

 

[Latexman]

will48,

Do I also need to include an additional load due to gas expansion in the vapour space?

If you are looking at a vessel that is nominally full of liquid in the fire scenario, no. API allows one to consider the fraction of the nominal liquid height, thus making it good engineering practice to ignore what little area is left to expand.

If you are looking at a vessel in a fire scenario that normally has a liquid level lower than that specified by API, yes.

In most vessels, the liquid full case usually is the worst case for relief flow, but you have to think through your specific case thoroughly. At the higher temperatures of a very low liquid level or empty tank, gas expansion scenario, will the contents decompose?

You will have to provide more detail on your case for us to get more specific.

Good luck,
Latexman

 

[will48]

OK, thanks Latexman for getting the ball rolling.

Yes, the separator is typically 30%-60% liquid filled (oil + produced water). So my understanding is that the total relief load for fire case is vapour generation from heat input to wetted area (Eq.(3/4) in API521) + expansion of gas in the unwetted area (Eq.(8) in API521).

P.S. I realise that the latent heat will not be fixed as the fluid is multicomponent - but that is another thread.

Thanks
W

 

[Latexman]

will48,

Don't forget about non-routine operations that may give a worse case than routine operations. Like start-up, shutdown, chemical cleaning, water washing, steam out, etc.



Good luck,
Latexman

 

[tenac]

With a separator 30% to 60% liquid full I would NOT add up vapor generation from heat input to wetted area plus vapor expansion. Usual practice is to consider only vapor generation from heat input to wetted area.

As far as I know, either the vapor thermal expansion relief load or the boiling liquid vaporization relief load, but not both, should be used. It is a practice that has been used for many years. I don't know experimental studies where separate contributions of vapor thermal expansion and boiling liquid vaporization have been determined.

Of course, if there are operating scenarios without liquids or with a very low liquid level which would evaporate in a very short time, I would do thecalculation considering vapor expansion (only).

 

[Montemayor]

All:

Please refer to thread124-143554 and note my remarks about API RP 521.

For everyone's sake I hope I'm not coming across as a Depressurization Fanatic on a soap box, because I consider this too serious a subject to take lightly. From personal field experience I can attest to the downside effects related to the hope that a vessel's vapor content will act as a heat sink and absorb the heat input from an external pool fire.

My personal advice to anyone expecting this to happen in time to allow the vessel's PRV to activate and relieve internal pressure is: don't expect this to happen. There is a tremendous lag that exists between the exorbitant external skin temperature on the vessel (caused by a combination of radiation, convection, and conduction) and the internal vapor temperature caused primarily by natural convection currents. Without sufficient internal temperature, the vessel vapors are not able to generate the set PRV pressure in time. However, the vessel's external skin temperature continues to increase exponentially at the same time. This is the reason for API 521's concern and recommendations for depressurization of the vessel. What API 521 is trying to tell all of us is that the saving grace for a vessel caught in the grips of a surrounding external pool fire is the liquid content with its inherent latent heat acting as a heat sink. The moment this heat sink literally "evaporates", stress and mechanical failure of the vessel are imminent and you should be taking rapid and efficent steps to relieve the vessel to avoid a very serious, explosive event. This doesn't imply that the depressurized vessel is no longer exposed to failure - because it is - if the fire continues. The important mitigation that takes place is that a violent explosive failure is avoided. Failure to relieve any depressurized vessel from continuous exposure to a surrounding fire will ultimately result in its mechanical failure. Other than putting the fire out, we have little control on this effect. Our concern should be to avoid an explosive, pressurized failure - which is the only event we can control under the situation.

In my opinion liquid thermal expansion effects contributing a noticeable pressure contribution will be nil or nothing with respect to the pressure build up due to vaporization.