VESSEL CONNECTION FOR RELIEF 6

[LOWTEMPMAN]

Does the ASME Code allow the connection to a relief valve for a given pressure vessel to be alternatively placed in the relatively large gas outlet pipe rather than on top of the vessel itself? Many refrigeration systems are being converted from CFC ozone depleting refrigerants to non ODP refrigerants. In some cases the new refrigerant requires a new larger relief valve in order to satisfy the sizing methodology in ANSI B9.1 Safety Code and thus a larger connection on the vessel. Cutting and welding a new relief connection on the vessel is costly and time consuming and requires an "R" stamp and approval by an ASME Code inspector to my knowledge. Does anyone have any information on this approach? Thanks.

 

[chaulklate]

The PSV does not have to be directly attached to the vessel provided there are no intervening stop valves between the vessel and the PSV. Refer to ASME Section VIII Div.1 UG-125(g)(2) for clarification.

 

[jte]

For VIII Division 1 applications, see also UG-135 and Appendix M.

To directly answer your question, UG-135(a) states that Pressure relief devices intended for use in compressible fluid service shall be connected to the vessel in the vapor space above any contained liquid or to piping connected to the vapor space in the vessel which is to be protected…

Intervening block valves are explicitly permitted by UG-135(d) within its restrictions, including Appendix M-5. It is common practice to install block valves which are car sealed open (CSO) on banks of PSV's. Say 3 PSV's are required. Four are installed with CSO block valves. The CSO provides an "administrative control" which allows one at a time to be closed for PSV service. Section I is another story...

jt

 

[pleckner]

Be forewarned about the 3% Rule when making this change. Teeing into a large pipe is perfectly acceptable and often done but the larger relief rate may very well cause excessive pressure drop between the vessel, with its still smaller nozzle, and the PSV.

 

[LOWTEMPMAN]

Thank you for the responses. They have all been quite helpful. I am not familiar with the 3% Rule noted in pleckner's reply. Thus I do not know how to interpret the remainder of the posted response. With regard to the comment about the original smaller nozzle for the relief connection, my intent would be that it would be abandoned in favor of reconnecting the new larger relief connection to the original "large" separate gas outlet used to conduct the normal operational gas flow out of the vessel.

 

[Aconnell]

Hi Lowtempman

I believe the 3% rule that pleckner is refering to is that nonrcoverable pressure losses from the vessel to the PRV should not exceed 3%. There is some discussion about this in API 520 "Sizing, Selection, and Installation of Pressure-Relieving Devices in Refineries", Part II Installation.

With regards to the block valves although the ASME code allows them some juristrictions do not. For example in Alberta where I work the BPV authority does not allow them, even though the code does.

Hope this helps

 

[pleckner]

Yes, as Aconnell stated, this is a general rule that limits the non-recoverable pressure losses to 3% of the PSV set pressure.

A PSV opens on pressure but something must allow the PSV to close and that pressure is called the blowdown pressure. When the pressure at the inlet to the PSV reaches blowdown, the PSV closes. If the pressure drop between the inlet of the vessel and the inlet of the PSV is such that the blowdown pressure at the PSV is reached prematureley, then the PSV sill shut and you will develope a cycle of opening and closing. This is called chattering and will damage the PSV.

The good news is that the larger connection should minimize this but you must check it at the PSVs rated (stamped) flow.

 

[jte]

Aconnell-

Good point about checking with the jurisdiction.

I do most of my work in a jurisdiction which is generally (reasonably) conservative but CSO block valves are allowed. This allows for units to keep producing gasoline without a PSV maintenance schedule driving the overall unit turnaround run length.

I guess another point to make is to keep in touch with the jurisdiction - make sure the info one is working with is current. We had some significant changes (not PSV related) to the vessel regulations recently which make some previously standard procedures obsolete. Gotta keep track of the changes!

jt

 

[NozzleTwister]

LOWTEMPMAN,

Let me give you some things to think about from a Piping Engineering (stress/support) point of view.

Most of my experience has been in refineries and I have analyzed many relief systems where the relief valve inlet line comes right off of the vessel overhead line ... and it can be a headache.

Most of the tall columns that I've dealt with, the relief line tied into the overhead line way down the side of the tower for easier maintenance access. This can be a problem because of the displacements at the tie-in relative to the tower makes the relief line hard to support. Many times the relief outlet line has to be supported with springs (which is not recommended for relief lines) or from the overhead line. The relief system may have to ride with the overhead line and flexibility provided for the relief outlet line. This situation will also increase the support loads on the overhead line supports and if you are modifying an existing system, you will need to determine if the existing pipe support attachments and vessel support and attachment will handle this increased load.

I recommend talking with your piping engineer so you can agree on the best tie-in location and support system. From a pipes stress point of view the new relief valve will be located near the vessel outlet at a location where the relief discharge line can be independently supported from the overhead line.

Good luck,


NozzleTwister
Houston, Texas

 

[LOWTEMPMAN]

This was my first post on this message board and I am completely overwhelmed and pleasantly surprised by the number of responses and their level of expertise. All of the responses have been quite helpful as I have previously noted. I should probably mention that I am a consulting engineer in the refrigeration industry. The application I have described refers to a low temperature refrigerant 22 horizontal low pressure pump receiver approximately 60" dia x 20 ft long for a petrochemical plant. In the refrigeration industry we are guided,in addition to the ASME Code, of course, by ANSI/ASHRAE B9.1 Saftey Code for Refrigeration. This safety code within the last 10 years or so has adopted the chemical piping code requirement of restricting the back pressure in the discharge piping of a relief valve. The Safety Code now requires that the discharge piping from a relief valve that relieves to the atmosphere be sized such that the back pressure at the relief valve outlet be limited to no more than 15% of the relief valve setting.[It was 25% in the old Code.] This change was made to help prevent valve chatter. I mention this to point out that it relates to the valve chatter as mentioned in pleckner's response. While I am at it, I may as well mention that we often use dual relief valves on refrigerant service to facilitate repair and maintenance. The dual relief assembly includes two individual relief valves and a non shut off 3 way transfer valve between the vessel and the relief valves. In this manner we do not have to resort to car sealed block valves.

 

[EGT01]

Just for reference...

Std 15-2004 Safety Standard for Refrigeration Systems (Supercedes ANSI B9.1, Safety Code for Mechanical Refrigeration)

http://www.ashrae.org/publications/detail/14890

 

[LOWTEMPMAN]

EGT01 is 100% correct. B9.1 became Std 15-2004 many years ago. My reference to B9.1 shows you how old I am. [Why I had a temp memory lapse is a good question].

 

[dcasto]

You may also want to make the second PSV a smaller on and keep the original in place. The shouldn't be any material compatability with the old relief valve. This will save some time and money.