I am analyzing the relief system for some propane horizontal vessels, and found they are using multiport manifold with 3 or 4 relief valves (similar to REGO's).

In one of them, the inlet connection to the manifold is 4" and each of the 4 individual relief valve has a 2 1/2" inlet. Even considering that only three of them will relieve simultaneously (the 4th is closed by the manifold), the inlet area to the manifold is less that the total area of the relieving valves.

Questions:

1) Is this situation acceptable under good engineering practice? API 521 says that the first consideration to design inlet piping is to provide full inlet area as a minimum, however, it looks that is only a recommendation. Being these valves in the market, makes me believe that this could be acceptable.

2) Some of the valves are rated under UL and not ASME: Is is right that ASME code requires valves rated unde ASME only or there are exceptions that allows the use of valves rated under UL?

3) If the valves where rated at 20% accumulation, is it allowed to recalculate their capacity at 21% (fire case) or they have to be "re-rated" under ASME by the manufacturer?

4) Anybody has experience calculating the pressure drop in the multiport manifolds? It looks that manufacturers don't have information available.

Any comment will be appreciated.

[font="Times New Roman"]
Luigi:

These are good questions and comments on your part.

Is the subject installation an industrial, commercial, or residential one? What is the size of the propane horizontal vessels? I have seen many similar installations on commercial installations (like restaurants, business offices, etc.) and they have, as you infer, used RegO safety relief devices. If the installation is supposed to be designed according to ASME standards and codes, then the relief devices should follow suit and also be sized and designed according to ASME.

Most, if not all of the LPG installations I have been involved in have employed multiple-installed relief valves. However, the scope of work for the installation called for multiple relief devices – BUT each PSV was rated for full capacity relief. The intent here is to avoid a situation where you have a leaky PSV and cannot suffer a shutdown of the installation. Therefore, what is conventionally done is that multiple PSVs are installed (each capable of full relief) and each has the capability to be isolated with a block valve in order to remove it and repair it while the tank is under operation. Some tanks have two PSVs connected to a special, 3-way valve that selectively routs the vessel gas to one of the two PSVs – but not to both at the same time. This type of valve saves a lot of space, piping, and money in having a spare PSV ready to go into service. Is that what your installation has as a scope of work?

If your installation does not have spare PSV capability and depends on all the installed, manifolded PSVs for proper relief capacity and the inlet nozzle connection is reduced, then you are not in compliance with ASME and I seriously doubt if you have proper capacity capability for relief. If this is the case, then you have a serious cause for serious concern about the safety of the installation, in my opinion.

[/font]

Art: Thanks for your answer. I would like to give additional information in hope you could give me more specifics on my questions.

This is an industrial installation that has to comply with ASME code. Vessels are between 50,000 and 100,000 gallons.

The valves I am talking about are, in fact, like those you refered to: For example, there is one manifold (4" supplied by the manufacturer) that has 4 relief valves on it (2 1/2" inlet each), and the manifold allows to block only one at the time. The capacity of that arrangement is considered based on three of the valves relieving at the same time.


Here is a location with information about these valves (look for multiport valves):
http://www.regoproducts.com/Old/Web%20L-500/SectionD.pdf

In my particular case, the inlet area of 3 of the 2 1/2" valves is more than the area of the 4" inlet connection to the manifold.

Art, from your last paragraph that I quote:

"If your installation does not have spare PSV capability and depends on all the installed, manifolded PSVs for proper relief capacity and the inlet nozzle connection is reduced, then you are not in compliance with ASME"

is it intended to say that an ASME requirement is to have an inlet piping at least equal to the relief valve(s) inlet area? I found that "recommendation" in API 521 but not in ASME.

It seems extrange that the manufacturer sales relief valve packages (including the manifold) that does not satisfy the API 521 recommendation (and would be worse if there is an ASME requirement). Even more extrange that the valves can be ASME rated.