Valve Stem Orientation on Relief Valve Header

[every_name_is_taken]

I have been asked to endorse the installation of a gate valve with its stem below the horizontal plane. This valve will be in a header that is a common relief path for PSVs that relieve to a flare system. The aim of installing it in this fashion is to eliminate the hazard of the gate separating from the stem and dropping into the flow path. I have been an advocate of modifying the piping to accept a plug or ball valve rather than install a gate valve with the stem below horizontal, but it is obviously cheaper and faster to just rotate the valve than do pipe modifications (just horizontal is not an option due to adjacent piping). I'm appealing to the collective wisdom and experience of the forum for opinions.

I am opposed to installing the gate valve in this fashion because I think it will provide a pocket, which can never be cleaned, for the nasty, corrosive sediment that collects in flare lines, and I think it's likely that the packing would be adversely affected by being in contact with this substance.

I know that API 520 permits the installation of gate valves in this way (and so do the people pushing for this option), but that doesn't mean that it's a good idea in every case. Are there other industry documents that speak more clearly to the orientation of gate valves in dirty services? Do gate valves exist that are manufactured to survive in this condition? Is it even a problem at all, and should I just accept it?

Your thoughts would be appreciated.

 

[georgeverghese]

Agree, if you cannot orient this valve to be on a horizontal plane, then a ball valve would be a better idea to avoid corrosion due to wet corrosive condensate on the gate valve stem and packing.

Where is this block valve - on the exit of a single PSV ? Do you really need a block valve on th exit of each PSV ? A better option is to have a single isolation block valve on the exit of a system of PSV's, with this entire system being shutdown for access to any PSV within this system. This removes the anxiety of many PSV exit block valves and keeping track of the associated locks, since you would then have a simpler system with only one block valve on the common exit of a system.

This reduces operational flexibility somewhat, but if the operations people are happy with this, its a much safer system to operate. Usually, this concept is outlined in the plant isolation, sparing and maintenance philosophy document developed by process systems engineers during FEED for plant Operations Group approval for the project.

 

[gerhardl]

I also do agree!

To find the best possible recommendation, apart from generally state that both ball and plug valves are better in this situation, it should be interesting to have the flow conditions, pressure and diameter, the reason for the valve and placement, and if the valve is just a shut off valve for service, or something else.

You might also have an option using an 'o-port stargate' type knifegate valve, short building length, both shut and open without pockets where dirt can assemble.

http://stainlessvalveco.com/

or equal types.