Sustained SIF in Dynamic COM# 1Max response Stress

ASME B 31.3 allows considering Sustained SIF as zero for qualifying static stress analysis.

Could Sustained SIF=Zero be used to qualify code stresses in dynamic COM #1 stress (Sustained + Dynamic case) for maximum response, when the system experiences failure by using Sustained SIF=1 in dynamic (Time history) case only.

Is this allowed as per B 31.3 piping code?

The 1987 interpretation , which still holds good , stands for SUS and SUS+OCC Cases . Your time history case falls under the second category,hence there is no reason why it should not be permitted by 31.3 to ignore SIF in this case.

Anindya Bhattacharya

I would like to add some clarification here...

It is a CAESAR II (C2) parameter that may be set to 0 or 1 (or 0.75) and not the SIF itself. The code-defined (in many cases Appendix D) SIF will be multiplied by this C2 value and this product will be used in formulating SUStained and OCCasional stresses with the added provision that this product is never taken as less than one.

The SIF reflects the strength of piping components in relation to fatigue failure of a circumferential butt weld. For example, if a butt weld fails after 10,000 cycles under a bending moment M, then a bend that fails (at the same number of cycles) under a moment of M/4 would have an SIF of 4.0. In this manner, the calculated stress is increased so that the same allowable stress can be used for all components. Over 50 years ago, Markl's work created these SIFs based on geometry and we still use them today.

But fatigue is not the measure of sustained and occasional loads in piping systems. Instead yielding or collapse is the mode of these failures. It so happens that the collapse load for an elbow is closely approximated by 3/4 SIF. Elbows, then, could use 0.75SIF for SUS and OCC stress calculations. But there is no similar jump from fatigue to collapse for other components (which do not show such a strength reduction). We are in a gray area here - B31.1 uses 0.75SIF for all components and B31.3 implies a full SIF but has two interpretations that indicate 0.0SIF and 0.75SIF (again with the product not less than 1.0).

So to add to your answer...

If your time history case is modeling some sort of impact load, that is not repeated, then collapse is the concern and using unintensified bending moments may be OK but watch out for the elbow stresses as their stresses may be underpredicted for the allowable stress used. Add these occasional stresses to the (unintensified) sustained stresses for evaluation with the allowable stress. If, instead, your dynamic load is cyclic, then fatigue is at issue and the full SIF should be used. These stresses should be evaluated against a fatigue limit or, in a pinch, the (adjusted) expansion allowable stress.