ASME B31.1-2024 Stress Intensification Factors (SIFs) — FAQ

Question 1

Where do B31.1 stress intensification factors come from now?

Accepted Answer

From ASME B31J. The 2024 edition deleted Mandatory Appendix D outright, stripped the Figure D-1 and Table D-1 references out of the Appendix G nomenclature, and points to B31J in three places: ¶104.8.1 defines the sustained indices from the B31J Table 1-1 i-factors, ¶104.8.3 defines ii, io and it as the B31J Table 1-1 factors, and ¶119.7.3 says that in the absence of more directly applicable data the flexibility and stress intensification factors shown in B31J may be used. The transition was gradual. In the 2016 edition Appendix D was the primary source with B31J footnoted as an alternative; the 2022 edition kept Appendix D but added a general note to its table permitting the B31J factors instead and re-pointed ¶119.7.3 at B31J; 2024 removed the appendix. So the code of record decides: a 2022 job may legitimately be on either source, a 2024 job is on B31J, and a 2016 job is on Appendix D unless the owner directs otherwise. The one thing no edition permits is mixing — B31J i-factors in the 2016 resultant-moment equation, or Appendix D single-i values fed into the 2024 direction-split equations, is a hybrid neither book describes.

Question 2

What is the difference between a sustained stress index (I) and a stress intensification factor (i)?

Accepted Answer

They are the same component data used at two different levels of severity. The stress intensification factor i is the fatigue-test value from B31J Table 1-1, and the 2024 displacement stress range equation — eq. (17) of Figure 104.8-1 — uses it unreduced: ii, io and it on the in-plane, out-of-plane and torsional moment ranges. The sustained stress index I is what ¶104.8.1 applies to weight and pressure: in the absence of more applicable data, Ii, Io and It are each the greater of 0.75 × the corresponding i and 1.00, and the axial index Ia is 1.00. The 0.75 is the Markl-era recognition that a sustained load produces a one-time stress rather than a fatigue range, and the floor of 1.00 means an index can never credit a component below straight pipe. The phrase more applicable data is a door, not a loophole: B31J's own General Note (d) gives sustained multipliers for branch sketches that differ from the 0.75i rule, and using them under B31.1 is a designer's justification to be written on the calculation cover sheet, not a software default to be left unexamined. Equation (16) for occasional loads uses the same I indices as sustained, with the sustained-plus-occasional moments.

Question 3

How do the 2024 equations use the factors differently from earlier editions?

Accepted Answer

Direction by direction. Figure 104.8-1 in 2024 writes the sustained stress as the square root of the sum of two squares: the first is the axial index times the magnitude of the longitudinal pressure stress plus the sustained axial force over the metal area, plus the SRSS of the indexed in-plane and out-of-plane moments over Z; the second is the indexed torsional moment over Z. The occasional equation adds the occasional forces and moments to those terms, and the displacement equation uses the same structure with ia, ii, io, it on the ranges. The 2016 edition, by contrast, put a single 0.75i on the resultant of all three moments for sustained stress and a single i on the resultant for expansion, with torsion buried inside the vector sum. That change is why a component with anisotropic factors — a welding tee, a reducing outlet, an unreinforced fabricated tee — can pass or fail differently under the two forms with nothing physical having changed, in either direction. It also means the 2024 equations carry an explicit axial-force term and the longitudinal pressure stress Slp, which ¶102.3.2(a)(3) lets you compute either as P·Do/4tn or from the inside-diameter form. Software set to the wrong code year applies the wrong equation family; the year, not the factor table alone, is what to check first.

Question 4

Do I have to model the branch-connection flexibility (k-factor) as well as the SIF?

Accepted Answer

B31.1's own wording is permissive: ¶119.7.3 requires flexibility calculations to take stress-intensifying conditions into account and says credit may be taken where extra flexibility exists in components. B31J, once you adopt it as the factor source, is not — its Table 1-1 note for the branch sketches states that the flexibility factors shall be applied as shown in its figures, with the branch spring placed at the run-pipe surface and the run spring at the centerline intersection. The engineering reason to follow B31J's instruction is that the i and k belong together: a softer branch connection redistributes moments through the whole system before any stress is checked, so applying B31J i-factors to a model whose tees are still rigid judges intensified stresses against moments the real system would never deliver. That combination is not reliably conservative — it usually overstates the moment at the tee itself but can understate equipment nozzle loads, restraint loads and natural frequencies elsewhere. On a 2024 job, set the stress software to B31J factors and B31J flexibilities, record both on the cover sheet, and treat the rigid-tee legacy model as a B31.1-2016 artifact rather than a conservative simplification.

Question 5

Did the elbow factors change, or only the branch connections?

Accepted Answer

Both, but for different reasons. For branch connections B31J is a different level of resolution: up to twelve factors at one tee — in-plane, out-of-plane and torsional i and k for the run and for the branch — where Appendix D gave one pair, plus stated applicability limits on diameter and thickness ratios and correction rules when a factor would otherwise fall outside them. For elbows the B31J closed forms are close cousins of the old appendix's h-based expressions, but the treatment still changes under B31.1: Appendix D applied a single i to every moment direction, whereas B31J distinguishes in-plane from out-of-plane intensification and sets the torsional factor separately, and the 2024 equations consume those three values separately. B31J-2023 also revised the sketch 1.1 elbow SIF equations relative to the 2017 edition, so a factor computed from a 2017 form is not automatically the current one. Two rules keep this straight: read the factor from the B31J edition your contract invokes, by sketch number, and confirm with the software vendor which B31J edition its factor library implements — the code year toggle and the B31J edition are two different settings.

Question 6

What about components B31J does not cover — valves, trunnions, base ells, closely spaced fittings?

Accepted Answer

The Code's answer is the phrase it uses in every SIF clause: more directly applicable data. B31J is the default source, and where a geometry is not in Table 1-1 or falls outside a sketch's applicability limits, the factor comes from somewhere more applicable — a B31J Nonmandatory Appendix A fatigue test, a numerical analysis of the actual geometry, published test data for the fitting, or a defensible comparison with a tabulated geometry that the designer is prepared to sign for. B31.1 ¶104.7.2 gives the general route for specially designed components (service experience, experimental stress analysis, proof test, or detailed stress analysis such as finite element), and B31J's general notes leave the judgment for combined geometries — a trunnion on an elbow, a branch fitting welded to anything but straight pipe — explicitly with the designer. Two practical cautions. A stress program's library value for a component it has no test basis for is an assumption wearing a number, so record the source next to the factor. And an SIF is never less than 1.0 and a flexibility factor is never less than 1.0: any factor library returning less is extrapolating outside its data.

Question 7

Does the free SIF calculator on this site give B31.1-2024 values?

Accepted Answer

No, and it says so on its own page. The free calculator implements the legacy Appendix D closed forms for elbows and B16.9 welding tees, which is the right tool for three jobs: a code-of-record check on a B31.1-2016 or B31.3-2018-and-earlier system, verifying what a stress program actually applied at an elbow or tee, and watching how a wall change moves the flexibility characteristic before committing it to the model. It is not a B31.1-2024 source, because 2024 has no Appendix D. For 2024 work the factors are B31J Table 1-1 by sketch, with in-plane, out-of-plane and torsional values for the run and branch legs; the PiperSIF tool on the Pro tier embeds the B31J-2023 catalog with its applicability limits and cites the sketch, and a pipe stress program set to B31J does the same inside the model. Whichever route you use, the calculation cover sheet should carry three lines — the B31.1 edition, the SIF source and its edition, and whether B31J flexibilities were applied — because a reviewer opening the package in five years should not have to reverse-engineer any of them.

Paragraph	What it sets
¶104.8.1	Sustained stress indices: I_a = 1.00; I_i, I_o, I_t = the greater of 0.75 × the B31J Table 1-1 i-factor and 1.00, absent more applicable data.
¶104.8.3	Displacement stress range uses the full i_i, i_o, i_t of B31J Table 1-1; i_a = 1.0 for elbows, bends and miters, i_o (or i) for other components.
Figure 104.8-1	Equations (15), (16) and (17) — the sustained, occasional and displacement stress equations, each an SRSS of an axial-plus-bending term and a torsion term.
¶104.8.4	Moment conventions per B31J Figures 1-1 and 1-2; the most conservative combination when dynamic results carry no sign; eq. (17) reused to compute S_i for equivalent cycles.
¶119.7.3	Flexibility calculations shall account for stress-intensifying conditions; absent more directly applicable data, the B31J factors may be used; credit may be taken for extra component flexibility.
Mandatory Appendix D	Deleted in the 2024 edition; Appendix G nomenclature references to Figure D-1 and Table D-1 removed with it.