Fatigue strength: effect of welding type and joint design executed in Ti-6Al-4V structures

Abstract

Fatigue strength: effect of welding type and joint design executed in Ti-6Al-4V structuresBackground: This study evaluated the fatigue strength of Ti-6Al-4V-machined structures submitted to laser (L)-welding and TIG (TIG)-welding procedures, varying the joint designs.Materials and methods: Seventy dumbbell rods were machined in Ti-6Al-4V alloy with central diameters of 3.5 mm. The specimens were sectioned and welded using TIG or L and three joint designs {'I' design, varying welding distances [0.0 mm (I00) or 0.6 mm (I06)], or 'X' [X] design}. The combinations of variables created six groups, which, when added to the intact group, made a total of seven groups (n = 10). L was executed as follows: 360 V/8 ms (X) and 390 V/9 ms (I00 and I06), with focus and frequency regulated to zero. TIG was executed using 2: 2 (X) and 3: 2 (I00 and I06) as welding parameters. Joints were finished, polished and submitted to radiographic examination to be analysed visually for the presence of porosity. The specimens were then subjected to mechanical cyclic tests, and the number of cycles completed until failure was recorded. The fracture surface was examined using a scanning electron microscope.Results: The Kruskal-Wallis and Dunn test (alpha = 0.05) indicated that the number of cycles resisted for fracture was higher to X for both welding procedures. To L, I06 was as resistant as X. The Mann-Whitney U-test (alpha = 0.05) indicated that L joints were more resistant than TIG to I00 and I06. Spearman's correlation coefficient (alpha = 0.05) indicated a negative correlation between the number of cycles and presence of porosity.Conclusion: Thus, to weld Ti-6Al-4V structures, the best condition is X, independent of the welding method employed.