A finite-element (FE) based method for numerically predicting fatigue life of MAG-welded thin-sheet structures has been established and tested. The method uses nodal forces and moments calculated along the weld line, together with an analytical expression for the structural stress at the weld toe. The calculated stress is used together with an experimentally determined Wöhler, or S-N, curve. A “stiff” welded joint with structural stress dominated by membrane forces is found to have a steeper S-N curve than a “flexible” joint with structural stress dominated by bending moment. All test results were seen to lie close to one of two different S-N curves. The proportion of bending stress over total structural stress could be used for choosing the appropriate S-N curve.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Fatigue Life Prediction of MAG-Welded Thin-Sheet Structures


    Weitere Titelangaben:

    Sae Technical Papers


    Beteiligte:

    Kongress:

    International Body Engineering Conference & Exposition ; 1998



    Erscheinungsdatum :

    29.09.1998




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Print


    Sprache :

    Englisch


    Schlagwörter :


    Fatigue Life Prediction of MAG-Welded Thin-Sheet Structures

    Fermer, M. / Andreasson, M. / Frodin, B. et al. | British Library Conference Proceedings | 1998


    Fatigue life prediction of MAG-welded thin-sheet structures

    Fermer,M. / Andreasson,M. / Frodin,B. et al. | Kraftfahrwesen | 1998


    Fatigue life prediction for welded steel sheet structures

    Galtier, André / Seto, Atsushi / Yoshida, Yuichi | SAE Technical Papers | 2003


    Fatigue Life Prediction for Welded Steel Sheet Structures

    Yoshida, Y. / Seto, A. / Galtier, A. et al. | British Library Conference Proceedings | 2003


    THE FATIGUE LIFE PREDICTION OF THIN-WALLED WELDED JOINTS

    Sladky, M. / Machac, M. / Papuga, J. et al. | DataCite | 2024