Vapor depressurizing rate - API 521 requirements 1

[Guidoo]

According to API 521, 4th edition (March 1997), section 3.19.1:

"[...] Where fire is controlling, it may be appropriate to limit the application of vapor depressuring to facilities that operate at 250 psi (1724 kPa(g)) and above, where the size of the equipment and volume of the contents are significant. [...]

Does anyone know the background of this? Or have used this guideline before?

 

[SooCS]

There are many companies use this guideline. But I prefer to perform the thermal stress analysis to define the depressurizing duration. API RP 521 has indicated that the recommendation is based on 1" wall thickness vessel which is good. But if the vessel is less than 1", if you look at the Fig 1 of the RP, it takes less than 15 min to failed. For vessel with wall thicknes greater than 1", it seems to conservative but if the wall thickness of the associated pipings and instrumentation fittings its justified.

 

[Guidoo]

SooCS,

Thanks for your reply. But..., I am afraid that you misinterpreted my question.

I was not referring to the API RP 521 guideline that "pressure is to be reduced to 100 psig (6.9 barg) or 50% of design pressure, whichever is lower, in 15 minutes", but to the API RP 521 guideline that no depressurizing facilities need to be provided in case operating pressure is below 250 psig (17.24 barg).

In my opinion this guideline doesn't make sense. In case you have a large vessel, consequences of a vessel rupture due to external fire can be unacceptable, even with an operating pressure below 250 psig. API RP 521 also is not clear why the pressure of 250 psig was selected, and why operating pressure is used instead of relief pressure.

Following makes more sense (taken from NORSOK Standard S-001, Annex E):

"All pressure systems and piping segments, which during shut down contains more than 1.0 ton of produced hydrocarbons should be equipped with a depressurising system." I can imagine that when the system is vary small (e.g. less than 1 ton of hydrocarbons), the risk of a rupture can be acceptable.

The NORSOK standard Annex G also specifies that a time to rupture calculation should be performed.

I apologize if my original question was not clear enough.

 

[SooCS]

The way I interpreted API depressurization recommendation is for vessel, which contain mostly gas. Depressurization is optional, if it is fill with liquid. Though some companies would still have some system to bleed down the pressure in the event of a fire.

Norsok on the other hand requires depressurization for both oil and gas system with more than one ton inventory. It becomes too stringent for liquid system but too lenient for gas systems. A two m3 cube vessel would easily contain one tons of HC. But to hold one ton of gas, it required a very big vessel or very high pressure.

Both the API and Norsok have some flaw in them. That's why I prefer to perform the heat stress and consequence analysis to determine the depressurerization requirements.

 

[Guidoo]

The last couple of days I have been further studying the API RP521, as well as literature on this subject. Examples of this literature are:

"Review of the Response of Pressurised Process Vessels and Equipment to Fire Attack", HSE Offshore Technology Report OTO 2000 051, (2000).
"Size Depressurization and Relief Devices for Pressurized Segments Exposed to Fire", Chemical Engineering Progress, September 2002, p. 38-45.

(both can be downloaded from the Internet)

My conclusions are that the API RP 521 guidelines both on when to apply emergency depressuring systems (EDS), and what depressuring rate to use are inadequate. Use of these API guidelines may not prevent catastrophic vessel rupture, especially in confined areas such as offshore platforms. Main reason is that heat fluxes from local poolfires or jet fires can be significantly (e.g. 3.5 times) higher than heat flux from open pool fire (as assumed by API 521).

EDS should be applied when a risk analysis shows that system without EDS results in unacceptable risk. For this assessment, both likelihood and consequences of the vessel rupture due to fire have to be taken into account.

For determination of the depressuring rate, a detailed iterative calculation sequence has to be performed (as outlined in above mentioned literature). Problems are that currently no (validated) software tools are available, as well the required thermo-mechanical material properties. It may also be difficult to estimate the time dependent fire characteristics.

SooCS, could you please outline how you perform the thermal stress analysis? What heat flux do you use (fixed or time dependent, jet fire or pool fire)? What thermo-mechanical properties do you use? Or do you just use the figures 1 and 2 from API RP 521?

 

[Guidoo]

Thanks SooCS. Coming days, I will look at this in more detail, and I will let you know my experiences.

Thanks again, regards,

Guido