Additive Manufacturing (AM) for metals includes is a group of production methodst hat use a layer-by-layer approach to directly manufacture final parts. In recent years, the production rate and material quality of additive manufactured materials have improved rapidly which has gained increased interest from the industry to use AM not only for prototyping, but for serial production. AM offers a greater design freedom, compared to conventional production methods, which allows for parts with new innovative design. This is very attractive to the aerospace industry, in which parts could be designed to have reduced weight and improved performance contributing to reduced fuel consumption, increased payload and extended flight range. There are, however, challenges yet to solve before the potential of AM could be fully utilized in aerospace applications. One of the major challenges is how to deal with the poor fatigue behaviour of AM material with rough as-built surface.

The aim of this thesis is to increase the knowledge of how AM can be used for high performance industrial parts by investigating the fatigue behaviour of the titanium alloy Ti6Al4V produced with different AM processes. Foremost, the intention is to improve the understanding of how rough as-built AM surfaces in combination with AM built geometrical notches affects the fatigue properties.This was done by performing constant amplitude fatigue testing to compare different combinations of AM material produced by Electron Beam Melting(EBM) and Laser Sintering (LS) with machined or rough as-built surfaces with or without geometrical notches and Hot Isostatic Pressing (HIP) treatment. Furthermore, the material response can be different between constant amplitude and variable amplitude fatigue loading due to effects of overloads and local plastic deformations. The results from constant amplitude testing were used to predict the fatigue life for variable amplitude loading by cumulative damage approach and these predictions were then verified by experimental variable amplitude testing.

The constant amplitude fatigue strength of material with rough as-built surfaces was found to be 65-75 % lower, compared to conventional wrought bar, in which HIP treatments had neglectable influence on the fatigue strength. Furthermore, the fatigue life predictions with cumulative damage calculations showed good agreement with the experimental results which indicates that a cumulative damage approach can be used, at least for a tensile dominated load sequences, to predict the fatigue behaviour of additive manufactured Ti6Al4V.

Laser powder bed fusion (L-PBF) and electron beam powder bed fusion (E-PBF) are two of the most common additive manufacturing (AM) methods which both provide the engineer with a great freedom of design.This means that parts with light weight, multifunctional applications and improved performance could be achieved through innovative design solutions which have attracted a lot of interest from the aerospace industry.

This PhD project has focused on the following fatigue related areas forL-PBF and E-PBF Ti6Al4V material which all need to be addressed before AM can be fully introduced to critical aerospace applications: effect of geometry, roughness and loading on fatigue life, improved fatigue life through post processing, fatigue crack growth behaviour and fatigue prediction methods.

The results show that the rough as-built surface is the single most severe factor for fatigue but that the fatigue strength of at least L-PBF material can be improved to levels similar to conventionally manufactured material using surface post processing. Furthermore, the results verify that acumulative damage approach gives good accuracy in predicting fatigue life for variable amplitude loading and that fatigue crack growth rates using material data from standard specimens can be used for damage tolerancean alysis independent of part geometry and stress level.

The conclusion is therefore that the fatigue properties can be improved to acceptable levels and predicted using conventional methods. There are still some challenges to solve, however, especially within non-destructive testing before AM can be introduced to critical aerospace applications.