This presentation was made at CAASE18, The Conference on Advancing Analysis & Simulation in Engineering. CAASE18 brought together the leading visionaries, developers, and practitioners of CAE-related technologies in an open forum, to share experiences, discuss relevant trends, discover common themes, and explore future issues.

Resource Abstract

The mechanical behavior of complex assemblies depends on the constitutive response of the fasteners joining the components. The modeling of all components and threaded fasteners in detail is often too computationally demanding and rarely practical. Instead, reduced-order models like springs, beam elements, or plugs are commonly used. Linear reduced-order models are usually straightforward and accurate in the elastic regime, but the emergence of plastic deformation creates a strong nonlinear mechanical response. Hence, the accuracy of reduced-order models for threaded fasteners at medium to high strains is dictated by the constitutive behavior of the bulk material. Although the specification of metallic threaded fasteners typically includes alloy grade, manufacturing procedures often work harden the fastener material and increase the uncertainty of the material properties. These effective properties for a given fastener can be obtained by performing fastener tensile tests, but this is a time-consuming process that quickly becomes infeasible as the number of fasteners of interest continues to grow. Hence, an assumed hardening curve must be used to model the plastic response of the fastener. This assumed constitutive behavior will introduce epistemic uncertainty into simulations involving fastened joints. By identifying trends in fastener constitutive behavior the material properties for a fastener model can be more confidently assumed from limited test data.

This investigation explores the modeling of fasteners with various sizes by developing finite element models in which material parameters were calibrated to match tension test data on stainless steel A286 fasteners. Sizes ranging from #0 to #6 UNF are considered with two models of varying geometric fidelity: a smooth plug of elements, and a higher-fidelity model including thread geometry. The material calibrations were similar for the plug and the threaded models, consistent with previous findings that the material nonlinearities dominate the tensile response of fasteners. The material model for a smooth annealed steel A286 specimen was calibrated to compare the original material to the fastener material after manufacturing. This investigation reveals that the fasteners have higher yield stresses than the smooth specimen, indicating various degrees of work hardening during manufacturing. This conclusion is extended to the analysis by estimating the fasteners’ hardening curves through a shift to the hardening curve of the annealed specimen to match the yield stress, and these curves are compared to independent calibrations of the same fastener. This study indicates that for the fasteners considered, when given the hardening curve of the original material, knowledge of the load versus displacement response of the fastener is unnecessary for calibration; the yield stress of the fastener is the essential piece of information required to estimate its hardening curve.

Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.