Over the past couple of weeks, I have had one particular issue come up several times with respect to elastic analysis to Part 5. There seems to be widespread misunderstanding about how to apply load cases to Protection Against Plastic Collapse and Protection Against Failure from Cyclic Loading – Ratcheting. So, I wanted to write this post to (hopefully) clear up some of this misunderstanding.
Article 5.2, Protection Against Plastic Collapse, describes what design load cases and design load case combinations must be used. For an elastic analysis, ALL of the load case combinations in Table 5.3 must be evaluated. That part seems to be pretty straight forward. One item that seems to frequently missed, however, is the little notes below the Table – specifically Note 3:
Loads listed herein shall be considered to act in the combinations described above; whichever produces the most unfavorable effect in the component being considered. Effects of one or more loads not acting shall be considered.
For example, Design Load Combination 1 is P+Ps+D. If there is something in Ps or D that somehow counteracts P, or vice versa, then you are obligated to also consider the case where one or more of those loads are not acting (equal to zero). So, while it may appear that this Design Load Combination is about the deadload, the design static head, and the design pressure, it also requires you to check:
- Empty, no pressure, deadload only
- Static head, but no internal pressure, and deadload
- Internal pressure, but empty, and deadload
- External pressure, empty, and deadload
- External pressure, static head, and deadload
That’s not to necessarily say that each and every permutation has to have its own finite element analysis. Indeed, applying a healthy dose of engineering judgement is both required and encouraged. Some people may call this “by inspection”. However, as you continue further down the list and start adding more loads, then this list of permutations grows more slowly. Another important item to remember is that, in confirming Protection Against Plastic Collapse, you are really only checking Pm (which should be checked using hand-calculations), PL, and PL+Pb. P+Q is used in checking ratcheting and P+Q+F is used in checking fatigue.
Which brings me to the topic of ratcheting. Article 5.5.6 provides the rules for confirming Protection Against Failure From Cyclic Loading: Ratcheting using the elastic stress analysis method. Contrasted to Protection Against Plastic Collapse, where you have a wide variety of design loading combinations, in Ratcheting you only have operating load ranges. To repeat: operating load ranges. Under normal (and planned-for abnormal) operation, each component will undergo a load range. That may be from a specified low internal pressure to a high internal pressure, external pressure to internal pressure, or no pressure to internal pressure. Thermally, you can also have operating load ranges.
The focus here is on the load range, and the resulting stress range. Unlike Protection Against Plastic Collapse, which is only interested in total stress values, here we are interested in the stress ranges. Significantly, we want to make sure that we are on the look out for stress reversals. So, when we calculate the stress ranges, we have to ensure that we perform the stress difference calculations at the component level, calculating the component stress ranges, before rolling the component stress ranges up into an equivalent (von Mises) stress range.
So, in summary, not only do you have to evaluate all of the Design Load Combinations to confirm Protection Against Plastic Collapse, but you also are obligated to check ALL of the permutations with some of the design loads being equal to zero (or in the case of pressure, the design external pressure). However, when evaluating protection Against Failure From Cyclic Loading: Ratcheting, we perform the stress range calculations based on the using the operating load ranges.
This Post Have been publish with the Authorization of his Author Trevor Seipp. The original post can be found here