Hello all,

I have been lurking here around this thread since it was started and I think it is sad that this discussion seems to have drifted away (from the technical issues) to where some of us are showing disrespect for others. I would like to congratulate everyone who participated in the discussion of this thread. I think it is good for all of us to try to understand and “be comfortable” with what we are doing for a living. I especially thank those who, although English is not your first language, have made the effort to participate in this English speaking discussion board. This was very unselfish of you and by doing this you have brought to the forum, your ideas which we would not have had the benefit of considering if you would not have posted. This discussion has been truly international and that is a good thing.

But back to the point at issue. After thinking about the statement of the question, perhaps flexMen may simply be saying that he understands that when calculating stresses the Code stress limit for the material at the highest design temperature will a limiting factor in the final design of any system. However, I think it may be that flexMen would prefer that CAESAR II do a different comparison for each load (combination) for each temperature (T1, T2, and T3). That would show that the seismic event would have to occur at one of the higher temperatures before Code stress limits would be exceeded. From the COADE point of view I can see where they could be reluctant to do that because that might be confusing if some person who is not a piping engineer were to just quickly look through the calculated tabular data. But then again, maybe I simply do not understand flexMen's point.

As for what I am reading from my esteemed colleagues in this thread, my sense of the discussion so far is that many of us are thinking the following. flexMen is designing one system (not three) and the materials of construction will be chosen only one time and the system will be built (with the understanding that it can accommodate all credible loading conditions). That piping system will be expected to provide a reliable service life of some finite number of years. My understanding of the service scenario is that the piping system at issue will from time to time be operating at its “normal operating temperature and pressure” (something less severe than the “design conditions”) and it will be subject to normal “in-service” deterioration. For the sake of discussion, allow me to speculate that the highest calculated bending stress in the system (at a branch connection of course) is at 75 percent of the Code stress limit. Given enough years of service without “upsets” the system would (since the Code covers fatigue fairly well) likely eventually fail due to corrosion (I won’t be around that many years). But, due diligence dictates that the “upset” conditions be considered and included in our analysis even though it is at least likely that the piping system at issue will never experience a seismic event.

Consider the piping system at a time when the temperature is 25 degrees F (I do not know if this is an operating condition or a shut-down condition but no matter). The system then experiences an earthquake of a non-trivial magnitude. The resulting calculated bending stress are well below the Code limit of 1.33 x Sh at the 25 degrees F temperature. The highest bending stress at our branch connection is 125 percent of Sh and there is no yielding of the material at the branch connection or at any other location in the piping (and I hope that no hangers were broken). Should the designer say “ah ha, I was right, the earthquake happened and my piping system is not damaged”?

A year later the same piping system is operating at “normal operating temperature and pressure” and let us (again, for the sake of discussion) assume the temperature is 550 degrees F. Stretching probability (and credibility) “a bit”, the piping system experiences an earthquake of the same magnitude as the event of the previous year. This time although the calculated bending stress at our branch connection is (just for discussion) exactly the same at it was due to the previous event, that bending stress is 138 percent of Sh (at temperature) or 104 percent of yield this time and so there is yielding of the material at the branch connection and maybe at other locations in the piping. Now remember, these are primary stresses and after the earthquake “rings down” there is still a pressure loading at the same temperature (no shakedown to elastic response for these loadings). What if it were thin wall pipe and the hoop (circumferential) stresses (twice the longitudinal stresses in our sustained case) were up at about 80 percent of allowable? With the material now yielding due to bending stresses....OOOps! Now, considering the above scenario, what Code occasional stress limit should the piping system be designed for? 1.33 times Sh at 25 degrees or 1.33 times Sh at 550 degrees? So, if I am understanding the responses of my esteemed colleagues, that is the essence of the answer you are posing for flexMen’s question.

I would be interested in knowing if I have gotten the question right (if you will comment please, flexMen – please expand upon my statement of the scenario) and I would be interested in knowing if I am understanding the answers correctly (please comment my esteemed colleagues).

And, if you are evaluating stress range per the new B31.3 Appendix P, what do you do with the above scenario? Hey, just wondering.

Regards, (and play nice), John.