API 520 v API 2000 vent sizing for fire relief cases 7

[TrevorP]

I first posted this in the API forum, but didn't get any response there, so am posting it here to see if anyone who frequents this forum can cast any more light on it.

I recently sized a relief device for the fire case for an existing vessel using API520. The relief device manufacturer came back to me and advised that the relief device I had specified was sized wrongly according to API 2000. As it turns out, he was right, since the set pressure was below 15 psig. However, this got me looking at the difference between the two standards - surely they cover the same ground (excluding pressure), so shouldn't they be seamless between 14.99 psig and 15 psig??
It appears that they are not - API 2000 has different equations based on the heat input from a fire (It appears to be more comprehensive). One other area of contention is that the conversion factor from kg/hr to scfh is curious to say the least (I supplied a kg/hr figure to the manufacturer based on Q/[λ] and the manufacturer converted from there - this did not tie in with API 2000's direct calculation for SCFH). Needless to say, I accepted the manufacturer's sizing, however I do find the differences to be curious.
I accept that I may not be using the latest version of either of the standards so may require an upgrade, but can anyone shed any light on the apparent above discrepancies?

 

[Montemayor]

Trevor:

I’m going to attempt to “shed some light” on the differences between:

1) API Standard 2000 – “Venting Atmospheric and Low Pressure Storage Tanks;
2) API Recommended Practice 520 – “Sizing, Selection and Installation of Pressure Relieving Devices in Refineries”.

You don’t define what type of vessel you sized a relief device for under the Fire Case. Presumably, from the relief device manufacturer’s comments regarding API 2000, it must be a Storage Tank – API 650 or API 620? That would explain why API 520 does not apply. Process pressure vessels are subject to the recommendations found in API 520 (parts 1 & 2) as well as to API 521. Storage tanks, as explicitly stated, are covered by the API 2000 Standard. There is a definite, engineering reason for distinguishing between the two types of generic vessels when designing for pressure (or vacuum) relief. I’m also going to presume you are a Chemical Engineer by training and have been subjected to the same poor or deficient training that I also received in regards to vessel mechanical design while in university. I say this not as a critique, but as a basis for the fact that the two types of vessels are mechanically designed from a different base of logic. As a result, the relief devices for each type are different in design and performance. One must appreciate the inherent limitations that reside within the API storage tank design, be it API 650 or 620. The same applies to the “conservation vent” type of relief devices used on these storage tanks. I suspect these inherent characteristics – or lack of detailed knowledge of them – are what is giving you problems in understanding their differences.

Note that one document is a “Standard”, while the second is “Recommended Practice”. There is a subtle difference in methodology as well as in the intent. They are not competing documents; rather they were (& are) intended to serve entirely different applications.

The reliance on accurate fluid density at relief conditions is important for either application. This data is difficult to predict for Storage Tanks whose contents have varying boiling points (& varying densities) due to their mixed composition. This is particularly true of liquid hydrocarbon storage. When confronted with this situation (& the fact that the tank is often used for storing a variety of different liquids), I have resorted to applying the vapor density of pure Hexane for the worse case fire scenario. From my experience, I have noted little major change in the API 2000 Standards.

I hope these comments are of some help.


Art Montemayor
Spring, TX

 

[johngerhart]

If you pick up a copy of Reliefs Systems Handbook by Cyril Perry , the differences are explained

 

[TrevorP]

Art,
Many thanks for your detailed response. I can certainly understand and agree with your response relating to design standards for vessels with different design pressures.

To give a bit more background, this particular case was for a batch distillation vessel, with a design pressure of 1 Barg (14.5 psig). I did not specify the tank itself, but am merely putting it into service for a different duty, and need to uprate the relief device as a result of doing so.
This vessel is certainly a pressure vessel, designed and constructed to a pressure vessel code (BS5500, now superceded by PD5500).
It is not US based, nor is it in the petroleum industry. Nevertheless, API 520/521 and API 2000 are commonly held up as the standard to use outside of the petroleum industry to assess fire case relief (It is certainly common practice to use these standards within much of the chemical industry, anyway). The vessel contents are mostly a hydrocarbon mixture, anyway (Mostly aromatics).
This vessel must have a pressure relief device rather than a conservation vent, yet, in theory, should be sized to API 2000, as the design pressure is less than 15 psig. This makes a big difference to the size of the relief device. The vessel is small (10 m3 / 1760 USgall), and so the heat input as specified by API 2000 is much higher than API 520. The total differences make a difference of 2 standard orifice sizes (From a J to an L).

 

[CHD01]

I will just add to what has been discussed. Technically, API 520 and API 521 applies to pressure vessels built to ASME Section VIII Code with MAWP of 15 psig or greater; and API 2000 applies to Atmospheric and low pressure storage tanks for operation from 1/2 ounces/in2 (0.8304 INWC) to 15 psig (1.034 barg); except that it does not apply to floating roof tanks. The atmospheric storage tank may be built according to API650; if the MAWP increases significantly above atmospheric, then construction to API620 is required. What is significant? I don't have API650 available right now, but I believe it is +2 INWC above atmospheric or greater. I can check this for you if needed, its been awhile since I last had the need to look this up.

Standard relief valves (which you must be using since you talk of an L orifice) NORMALLY are ONLY used on an ASME Section VIII designed pressure vessel for MAWP of 15 psig or more. There are some true relief valves that can be purchased slightly below 15 psig - buts its special! I would use a true vent valve in most circumstances (be careful some vent manufactures refer to their vent valves as RELIEF VALVES when they are really a pressure vent). If you use a L orifice standard relief valve the storage tank had better be designed to ASME Section VIII. However, you can use Section VIII standard as a design criteria for what is nomally used as an atmospheric storage tank. In this case I would have the tank ASME Code stamped. If it is not designed to ASME Code, then I would not use the L orifice valve. Why. because the relief valve flow capacity is based on an orifice with a specified % overpressure (normally 10%, not 20% or 21%) while the true vent valve required orifice is based on the actual delta p across the vent device so that the tank relieving pressure does not exceed tank MAWP plus any allowable overpressure. Increasingly I like to size these vents to eliminate any overpressure above MAWP. Further, the relief valve is sonic (normally) and the vent is sub-sonic.



The more you learn, the less you are certain of.

 

[TrevorP]

Thanks for your input CHD01.

I'm beginning to see where all this is coming from now.

API520 and API2000, although adopted globally, do have at their core the fact that they are US standards (or RP). As such, the vessel codes are different between those with a design pressure below and those with a design pressure above 15 psig. Those above 15 psig are designed (exclusively?) to ASME VIII, and there are certain criteria that allows for some conservatism in the design, thereby allowing for some more flexibility in the API520 standard. Those below 15 psig are designed to different standards - such as API620, API650 etc., which has a less stringent design criteria, so the relief device design has to be more conservative. Hence the differences between API2000 and API520.

In Europe, the PED regulations specify a cut-off point at 0.5 Barg. Under these regulations, various design codes are allowed (Including, PD5500 and ASME VIII amongst others). These design codes are all broadly similar, and have at their origin, various national standards. All will require hydrostatic testing at a pressure substantially higher than the design pressure (I think this is defined under PED as 150% of design). Hence, I believe, API520 might actually be suitable for vessels designed to these codes, but to tie in with ASME VIII, it is specifically stated as being only suitable for vessels above 15 psig.

No relief device manufacturer has told me that a relief device set at below 15 psig is a special, only that it cannot be ASME stamped. This is not an issue in Europe, it is a legal requirement that they are CE marked at pressures above 0.5 Barg, and ASME stamping is irrelevant. In the US, I assume it is the other way round, the ASME stamping is a requirement, the CE marking irrelevant.

I think the learning point for me (and for other engineers specifying such equipment), is to ensure that pressure vessels are in future designed to a pressure above 15 psig to avoid potential conflict with API520. As I stated earlier, this is an existing vessel, so I'm stuck with it.

 

[CHD01]

15 psig and greater must be designed to ASME Code. Less than 15 psig is normally designed to API650 or API620, but may also be designed to ASME Code as a storage tank. Standard and certified relief valves (D orifice or larger designed for sonic flow) must be used for 15 psig and greater, vents designed foe sub-sonic flow are used below 15 psig. Relief valves below 15 psig requires consultation with the relief valve manufacturer for assurance of capacity and operation. Really, I would say that designing for above 15 psig avoids conflict with the ASME Code ( a requirement in USA); I view API 520 and API 521 as recommended practice which supplement the ASME Code(not code requirements per se although you should follow the RP) for installation of relief systems. 1.5X the MAWP did used to be the ASME requirement for a hydrostatic test, but now its more complicated and depending on the allowable stress could be as low as 1.2X.

Sounds like you have a good understanding of how the standards interact; I wish I understood international standards better than I do; although I did get some exposure in Londonderry and Dublin to British standards for pressure vessels.

The more you learn, the less you are certain of.