For B31 Codes, we determine sustained, occasional and expansion stresses relative to allowable stresses.

My approach has been to: include operating case with the relief valve force for calculation of occasional stress; include operating case without the relief valve force for calculating the expansion stress.

Realizing that a relief event will normally result in a displacement (rotation) of the relief valve, also displacements of the relief stack or discharge piping may occur, and understanding that displacements result in secondary stresses (fatigue), I am concerned that my approach does not adequately address fatigue as a result of the occasional load events.

I think I need to include an expansion stress case based on operating cases with the relief valve force (OPE+F)-(SUS). Is this the approach other analysts use to address fatigue for relief valve events?

Ken,

In my opinion, relief valve discharge loading should be treated as occasional as :

1) Lexically speaking it is occasional.
2) It is not secondary as it is not displacement driven and I don't see a "seismic anchor movement" type scenario in relief valve loading.
3) Regarding fatigue , if you envision that as a problem and since it is stress controlled fatigue, ty to keep the computed stress less than endurance limit.

Regards

Anindya,
Thank you for providing input, I've been thinking about how to reply...
I agree that relief valve load is occasional load, with primary stresses.
I also envision a relief valve event as one in which many hundreds of load cycles can occur before the system is stabilized, and such events can occur dozens of times in the life a system. So, for that reason, there seems to also be a fatigue issue.
Perhaps in evaluating an actual system, in order to meet the occasional load allowable stress of 1.33Sh, the system will be under the endurance limit, and fatigue is not normally a factor.
I will be looking at a relief system again soon and will look at the stresses with this discussion in mind.

Regards,