PSI - Issue 75

Alberto Visentin et al. / Procedia Structural Integrity 75 (2025) 593–601 Alberto Visentin, Alberto Campagnolo,Vittorio Babini, Giovanni Meneghetti/ Structural Integrity Procedia (2025)

597

5

analysis. The solid submodel was reloaded and the displacements at the corresponding nodal locations computed from the global shell model results were imported and mapped using the Submodeling - Cut Boundary Constraint functionality with the Transfer Key Shell-Solid option available under the Static Structural analysis settings in Ansys® Mechanical. Eventually, the mesh density adopted in the solid submodel can differ from the global shell model, as the interpolation relies on element shape functions. • STEP 4 – Fatigue strength assessment according to the PSM : The solid submodel was solved in Ansys® Mechanical, after receiving the boundary displacements according to previous STEP 3. In this work, the “PSM App” fatigue analysis tool has been exploited for rapidly computing the equivalent peak stress ∆σ eq,peak distributions at the brace- and chord-side weld toes, as shown in Fig. 2 and Fig. 3.

∆σ = 33.20 MPa Linear elastic FE analysis: global solid model

Fatigue strength assessment according to the PSM

XY symmetry Plane: U Z = 0

Δσ eq,peak [ MPa ]

Global FE size d = t = 6.3 mm

PSMApp

Brace tube t = 6.3 mm

d

P

Max

Z X Y

U X = 0 U Y = 0 U Z = 0

Z X Y

Z X Y

Crack initiationpoint

Equivalent peak stress edge-contour plot

FE type: 10-node tetrahedral elements (SOLID187 of Ansys ® FE library)

U X = 0 U Y = 0 U Z = 0

Total number of FE nodes 1.85 ∙ 10 5 nodes (3 dof/node)

Chord tube T = 10 mm

Equivalent peak stress at the crack initiation point: ∆σ eq,peak

Fig. 2. Finite element analysis procedure to apply the PSM to a CHS-to-SHS welded joint starting from a solid model and a 10-node tetrahedral FE mesh. The “ PSM App ” fatigue analysis tool (see (Visentin et al. 2022, 2024)) has been employed to calculate the equivalent peak stress distributions along the weld toes automatically. The represented case study is derived from the specimen “ TA1 ” (see Table 1). 3. Case study: CHS-to-SHS full-penetration steel welded joints (Gandhi and Berge 1998) experimentally investigated the fatigue strength of Circular Hollow Section (CHS) braces welded to Square Hollow Section (SHS) chords made of API 2H grade structural steel (σ y = 355 MPa) and subjected to axial loading applied to the brace, the chord being constrained at its extremities (see Fig. 1). Different joints were fabricated by exploiting full-penetration welds and by combining various tube thicknesses ranging between 4 mm and 12.5 mm, as summarized in Table 1. The brace diameters varied by a factor of approximately three, leading to different levels of bending stress on the upper surface of the chord member. Constant amplitude axial fatigue loading was applied to the brace, with a load ratio equal to R = −0.36. In all considered joints (see Tables 1, 2), fatigue cracks initiated at the weld toe on the chord side near the saddle region, as reported in Fig. 1. For each joint in Table 1, a global solid model (referred as Model 1) has been generated by exploiting half of the full geometry with respect to XY symmetry plane. Loads and constraints have been applied according to Fig. 2. The geometry has been discretized by means of 10-node tetrahedral elements (SOLID187 of Ansys® FE library). According to PSM meshing rules (see (Meneghetti and Campagnolo 2020)), the input FE size must be equal to the thickness of the brace member, i.e. d = t (see Fig. 2) since full-penetration welds are involved and pure mode I local