Numerical modelling of surface roughness effect on the fatigue behavior of Ti-6Al-4V obtained by additive manufacturing

Abstract

Selective Laser Melting (SLM) is a powder bed fusion process which allows to build-up parts by successive addition of layers using 3D-CAD models. Among the advantages, the high degree of freedom for part design and the small loss of material explain the increasing number of Ti-6Al-4V parts obtained by this process. However, right after additive manufacturing, these parts contain defects (surface roughness, porosity, residual stresses) which significantly decrease the High Cycle Fatigue (HCF) life. In order to minimize the porosity and residual stresses, post-processing treatments like Hot Isostatic Pressing (HIP) and Stress Relieving (SR) are often conducted. But the reduction of the surface roughness by machining is very costly and not always possible, especially for parts with complex geometry. The aim of this work is to evaluate the effect of the surface roughness of Ti-6Al-4V parts produced by SLM on the HCF behavior and to propose a methodology to estimate this effect. Three sets of specimens were tested in tension-compression: Hot-Rolled (reference); SLM HIP machined; SLM HIP as-built. For each condition, microstructure characterization, observation of the fracture surface of broken specimens, surface analysis and volume analysis were carried out respectively by Optical Microscope (OM), Scanning Electron Microscope (SEM), 3D optical profilometer and 3D X-ray tomograph. Results of fatigue testing show a significant decrease of the HCF life mainly due to the surface roughness. Along with experimental testing, numerical simulations using FEM were conducted using the surface scans obtained by profilometry and tomography. Based on extreme values statistics of a non-local fatigue indicator parameter (FIP), a methodology is proposed to take into account the effect of the surface roughness on the HCF life.

Fatigue behaviour of additively manufactured Ti-6Al-4V alloy: The role of defects on scatter and statistical size effect

“…Regarding the effect of the as built surface on the fatigue behaviour, numerous studies [8][9][10][11][12] have shown that the as built surface is detrimental to the fatigue strength. Several approaches have been developed in the literature to model the effect of surface roughness defects.…”

“…Vayssette el al. [8] tried to estimate the fatigue strength via numerical simulations by using the finite element method and achieved satisfactory predictions. In the work of Nakatani et al [11], the classical Murakami approach is used in which the surface roughness defect size is measured by using the √ area parameter.…”

High Cycle Fatigue Performance of LPBF 304L Stainless Steel at Nominal and Optimized Parameters

“…They significantly influence the static and dynamic mechanical properties of a material including fatigue strength. Several studies have been carried out recently to assess the fatigue behavior of different AM materials, that is, AlSi10Mg [9][10][11][12], Ti6Al4V [13][14][15][16][17][18], Ni-based alloy [19], 15-5 PH stainless steel [20], steel [21], stainless steel [22]. Nasab et al [23] studied the effect of surface and subsurface defects on the fatigue behavior of AlSi10Mg.…”

What is going on with fatigue of additively manufactured metals?

“…(3) Modelling the material with the geometrical defects, usually by means of finite element software. This method can be done based on, eg , computed tomography 3D image or surface profile from a confocal microscope, obtaining the geometry, then importing it as a 3D model into a finite element software and applying loads …”