4 replies to this topic

[Taran]

In the past a client of ours has sized token relief valves (for protecting downstream pipework against overpressure due to regulator leakage) on the basis of 10% of wide open flow through the upstream regulators. Generally this has given 1" x 2" valves, which have seemed appropriate. However, for larger regulators a larger relief valve is obviously required which is inappropriate given the mode of valve leakage is the same between the smaller and larger regulators and has nothing to do with wide open flow. Specifically, when comparing a Fisher 627 and a pilot operated valve: The wide open flow through a similar sized pilot operated valve is ~4 times that of the Fisher, yet their leakage rates would be very similar.

We were wondering what other token relief sizing basis's other engineers / companies have used. A figure of 2% of wide open flow has been suggested. Any thoughts?

[Art Montemayor]

Taran:

I note your use of the term "token" and I am mystified as to why you address a potential hazardous condition with only a "token" device. I am not being critical of your description (as of yet), but rather I am curious to Find out if the over-pressure scenario(s) you describe are real and credible.

Am I correct in assuming that you have a situation where your client is potentially exposing his piping system to an over-pressure that could cause pipe failure or bursting and protecting against this possibility with only a "token", estimated (or guessed-at) capacity? I interpret your description of the operation as one where you have a process gas regulator feeding downstream piping. The regulator's feed pressure is higher than the downstream pipe pressure rating and should the regulator fail in the open position, all the downstream piping would be exposed to the upstream higher pressure and subject to rupture. Is this correct? I'll wait your reply.

[Taran]

Art,

Sorry to confuse you, but I am applying sound engineering practice. In New Zealand, pipelines are designed to the standard AS2885 which requires at least two pressure protection devices. ie either an active regulator with full flow relief valve; or an active regulator with a slamshut valve; or an active regulator with a monitor regulator that controls the pressure if the active regulator fails open. However, the code makes reference to considering the possibility of overpressure if both valves are closed but are leaking a small amount of gas into the downstream section: hence the installation of a "token" relief valve. Unfortunately the code doesn't offer any guidelines as to how the valve should be sized, hence my question to the forum.

Ben,

Agree with your comments and its a similar situation here, ie a leaking valve. However, still no closer to achieving a suitable sizing basis! I have asked a Fisher valve supplier, but they are obviously coy about the amount of leakage their valves a susceptible to.

Taran

[gvdlans]

In my opinion, use of two regulators in series for overpressure protection is normally not accepted (at lease outside the Australia/NZ area...).

Norwegian standard Norsok P-001, section 4.4.1.1. describes use of HIPPS (High Integrity Pressure Protection System) to protect subsea pipelines. Here, a small PSV should be installed to prevent a valve leakage from overpressuring the system. The maximum acceptable leakage rate through the valve should be in accordance with the valve vendor, but should not exceed 50% of the PSV capacity. Normally the maximum leakage through a valve is defined via its TSO (Tight Shutoff) class.

But with a HIPPS we are talking about slamshut valves, not about regulators. For control valves, it is normally said that they cannot remain TSO because of the wear that occurs when it is trying to control pressure...