Weld Geometries and Fatigue data of Laser-Welded Thin-Walled T-Joints

Published: 25 November 2022| Version 1 | DOI: 10.17632/58wtyyby7j.1
Tobias Pawlowitz


The Fatigue analysis of laser-welded joints is significantly influencent by the weld geometry due to its stress raising effects. This especially applies to local approaches, which directly include the weld geometry in the fatigue analysis. The processing and incorporation of the weld shape three- oder two-dimensional is still part of the research. Here a statistical method was used to create fractile-based weld cross-sections from cross-sectional raw data. The algorithms for the creation of the fractile-based geometries are available, too. A fractile-based geometries was produced for each specimen of a sample. In addition, a fractile-based geometry is created for the total sample group of specimens. These fractile-based geometries are used to calculate notch stresses and predict the fatigue lives according to the notch stress approach with reference radii of 0.3mm and 0.05mm. The predicted fatigue lives are compared to actual results from fatigue experiments in oder to reveal the prediction quality. The dataset cotains more geometry data than used in the fatigue analysis, which could be employed in further calculations and analyses. The dataset includes four types of data. 1. Geometriecal data of 107 laser-welded thin-walled T-joints, i.e. 3D surface model as .stl-file as well as cross-sectional data of the specimens as .txt-files. 2. The MATLAB code (geometry analysis code) to create fractile-based geometries from the raw data (cross-sectional data, three dimensional laser scan data). 3. The fraktile-based geometries of a particular group of specimens out of the total amount of specimens. 4. The results of notch stress simulations, predicted fatigue life according to the notch stress approach and results from fatigue experiments.



Fachhochschule Kiel


Geometry, Welded Joint, Mild Steel, Fatigue Life