Influence of material data input on accuracy of fatigue assessments for beam weldments

Gorash, Yevgen and Comlekci, Tugrul and MacKenzie, Donald (2016) Influence of material data input on accuracy of fatigue assessments for beam weldments. Proceedings in Applied Mathematics and Mechanics, PAMM, 16 (1). 90857. ISSN 1617-7061 (In Press)

[img]
Preview
Text (Gorash-etal-PAMM2016-Material-data-input-on-accuracy-of-fatigue-assessments)
Gorash_etal_PAMM2016_Material_data_input_on_accuracy_of_fatigue_assessments.pdf
Accepted Author Manuscript

Download (647kB)| Preview

    Abstract

    This numerical study investigates the effects of fatigue material data and finite element types on accuracy of residual life assessments under HCF conditions. The bending of cross-beam connections is simulated in ANSYS Workbench for three different combinations of beam profiles. The weldments are made of the high-strength steel grades C350LO and C450LO according to AS3678. The stress analysis of weldments is implemented with solid and shell elements using linear material and geometry consideration. The stress distributions are transferred to the embedded fatigue code nCode DesignLife. For both variants of FE-mesh, the nominal stress in the weld toes is extracted by splitting the total stress into membrane and bending components and filtering out non-linear component. Considering the effects of bending, size and mean stress, failure locations and fatigue life are predicted using the Volvo method and rules from ASME BPV Code. Three different pairs of experimental S-N curves (stiff and flexible) are used as material data input for fatigue analysis. The obtained numerical predictions are compared to the experimental results for shell FE-models. The predictions using the S-N curves for an equivalent steel demonstrate the best accuracy proving the fact that specific material data input is more effective than a generic data.