Depends of the process and the fluid. I don´t know if this is mandatory, but I have seen some "good practices" in companies, for example who handle uninhibited 1,3 butadiene, where is recommended to install a PRV in trapped isolated piping with lenght equal or more than 20 meters.
You should check API 521. I dont have it right here - but i think there is a minimum volume for gas. For gas in _most_ circumstances its only fire that could cause overpressure.
Liquids: Well expansion is expansion and solar heating could in many cases cause the pressure to increase significantly if its a blocked in volume. In this case i think that a thermal relief valve would be recommended in any case.
Thank you very much for the answers. Thermal expansion in a trapped liquid is discussed only briefly in API RP 520 and 521.
My situation comprises a trapped liquid piping, with lenght above 100ft between inventory valves, which will be closed in case of fire, for example.
The fluid is cracked naphta with low boiling point.
Does anyone had a similar situation??
I dont really think the length is of any importance with regards to choosing: Its a liquid. Thermal expansion will cause a high increase in pressure with a low dT. So unless your ambient is always similar or lower then i would say that you need a thermal relief valve.
API 520/521 and ASME code technically wouldn't apply because you are working with piping rather than vessels. However, I've worked at places where similar situations can (and have) occurred. One thing to point out is that you aren't really talking about thermal expansion from a blocked in situtation (ie, solar radiation heating trapped liquid in a loading/offloading line, for example). If you are concerned about the effects of a fire, you've got a bit of a special situation, and there are provisions in the piping codes that require overpressure protection. We've taken several approaches to preventing trapped liquid from overpressurizing line sections. My preference, where feasible, would be to drain the line-automatically blow nitrogen, etc., in order to prevent the line from being liquid full in the first place. This isn't always feasible, so if it's possible to carseal valves open to ensure a relief pathway back to a tank, etc., that's another approach that I've used. Finally, and I must say that this is my least favored approach, is to install either a thermal relief compensator (essentially a bladder tank with a nitrogen pad) or a relief valve, as the last option. Compensators can bleed nitrogen, and if you install a relief valve, well, you've got to make sure you handle the discharge safely, which isn't always the easiet thing to do. I know that other options out there are to drill small (1/4" or less) holes through block valves, once again to provide a relief path-again, not something I favor (you put a block valve there for a reason...) but that's an option that some use.
As Morten mentioned, it's really not a function of line length; rather, it's the potential risk involved with an overpressure in the line (contents, hazard, expansion coefficients, etc.).
I've worked with companies that will not install thermal relief valves on above ground piping even if the piping is steam traced and others who do want them, usually on longer above ground lines with 'tight' shut-off valves.
You don't need much 'flow' to protect against overpressure in this case but if your isolation valves are truly 'bubble tight', the increase in pressure can be up to 100 psi per deg F. Naphtha won't be that high but a relatively small temperature rise can still result in a lot of pressure.
