I have a relief valve on the discharge of a compressor in a flammable gas system (60% methane, 40% carbon dioxide). The valves will relieve to atmosphere. Do I need a flame arrestor on the vent line? What document governs?

Thanks,

Mike

mvancleave,

I guess you need to find out the LEL for methane and then calculate the concentration of methane which is realeased from this PSV in to the atmosphere. if this value is well below the LEL for methane then probably you could do without a flame arrestor........but hang on!!!!!................I wonder why aren't you routing this relief discaharge stream to a flare???.......................................

Cheers

I wouldn't expect to see a flame arrestor on the outlet of a PSV. First, it's an additional pressure drop that needs to be considered as well as a possible source for plugging. Secondly, when a PSV relieves the velocities in the discharge piping are quite high and would likely blow apart a flame arrestor. The velocities are also well above any sort of flame speeds (except for a detonation), typical PSV outlet piping is anywhere between 10% to 50% of sonic.

I've seen atmospheric PSVs on a crude and vacuum tower that were routed to atmosphere, neither had flame arrestors. If you've seen horizontal propane storage bullets, their PSVs do not have flame arrestors on them.

I do not think it is so black and white...

It is definitely not required to have a flame arrestor when you relief flammable gas to atmosphere. For example, nothing about this is mentioned in API RP 520 or 521. I know that some operating companies normally do have flame arrestors in these cases where others don't.

Best reference I could find was Cyril F. Parry, "Relief Systems Handbook", published by Institution of Chemical Engineers (1992), section 4.4:

"When venting flammable vapours direct to atmosphere, it is prudent to consider providing a flame arrestor at the discharge exit. Some companies restrict the installation of flame arrestors to open vents on low pressure storage tanks. With a common venting system a flame arrestor can prevent a flashback in the manifolds, so in these circumstances it is worth considering. The major consideration is the cleanliness of the discharge. Since the principle of the flame arrestor depends on a multiplicity of fine passages, any dirt or debris in the discharge will tend to block the arrestor an a blocked arrestor can be more dangerous than a possible flashback. Where the situation is not clearcut, a risk analysis should be performed to asses the hazard potential with and without the flame arrestor. Providing that the discharge is clean and the buil-up backpressure of the arrestor is acceptable, then provision of a flame arrestor can provide some extra protection"

I would recommend you to get a copy of Parry's book since there is more good stuff about venting directly to atmosphere, e.g. about safe location. You should also check API RP 521 section 5.4.4 on vent stacks. Built-up backpressure of vent piping (incl flame arrestor if applicable) should be within acceptable limits (e.g. 10% of set pressure for conventional relief valve).

There should be no potential ignition sources near vent stack exit as per Area Classification (see API RP 505, NFPA 497 or IP15).