Skip to main navigation Skip to search Skip to main content

Ultrasonic Fatigue Behaviour of Additively Manufactured Ti-6Al-4V and Inconel 718: Influence of Build Orientation

Research output: Contribution to journalConference articlepeer-review

Abstract

This study investigates build orientation effects on ultrasonic fatigue testing (UFT) calibration and mechanical behaviour of additively manufactured Ti-6Al-4V and Inconel 718 specimens. A statistical framework quantifies uncertainty propagation from measurement systems through calibration to stress-life estimation, which variability is a result from Additive Manufacturing (AM) specific variability including anisotropy, surface texture, and microstructural heterogeneity. Five specimens per material-orientation combination were tested at 20 kHz using Digital Image Correlation (DIC) and laser displacement measurements. Probabilistic calibration curves revealed orientation-dependent differences of 5-9% in stress-displacement relationships, with horizontal builds exhibiting higher effective stiffness than vertical builds. Monte Carlo simulations demonstrated that ±12.5% calibration uncertainty may propagate to two-order-of-magnitude variations (106 to 108 cycles) in predicted very high cycle fatigue (VHCF) life. Early fracture analysis identified build defects as dominant failure mechanisms. The framework provides confidence intervals essential for reliable VHCF characterization and industrial qualification of AM components in fatigue-critical applications.
Original languageEnglish
Pages (from-to)153-161
JournalProcedia Structural Integrity
Volume82
Early online date1 May 2026
DOIs
Publication statusE-pub ahead of print - 1 May 2026
Event8th International Conference on Structural Integrity and Durability (ICSID2025) - Dubrovnik, Croatia
Duration: 16 Sept 202519 Sept 2025

Fingerprint

Dive into the research topics of 'Ultrasonic Fatigue Behaviour of Additively Manufactured Ti-6Al-4V and Inconel 718: Influence of Build Orientation'. Together they form a unique fingerprint.

Cite this