Towards a fundamental understanding of the effects of surface conditions on fatigue resistance for safety-critical AM applications
Introduction
There is strong evidence that surface conditions in as-built (AB) specimens have a considerable influence on the fatigue properties of Additive Manufactured (AM) materials [1], [2], [3], [4], [5]. AB specimens are known to exhibit dramatic reductions of some 40–50% of fatigue strength compared with those machined or polished [2], [3], [4], [5], [6]. The most detrimental effect is attributed to the surface roughness, which cannot be improved by machining post AM, as machining operation exposes large subsurface pores [1]. Although post AM processing methods such as HIP or heat treatment may relieve residual stress, improve microstructure or reduce pores in the bulk, such treatments have little impact on improving surface conditions. Surface roughness has now been recognised as one of the major limiting factors in the application of AM technology in fracture-critical components or structures [7], [8].
Conventional engineering treatments of the effects of surface roughness on high cycle fatigue (HCF) strength consider the surface as a series of micro-notches. A stress concentration factor, defined as where is the nominal stress, has been used for characterisation purposes. To overcome the difficulties in measuring the actual geometrical parameters needed to calculate the stress concentration factor for notches, Neuber [9] and Arola & Williams [10] proposed simplified analytical solutions to obtain stress concentration factors based on surface roughness parameters. These models assume a periodic and homogeneous surface typically from machining, but not applicable to surfaces produced by AM approaches, where more irregular surface profiles may be obtained associated with specific AM procedures. More recently, Ås et al [11] proposed a new method using automated FE simulation of surface geometries, obtained from white light interferometry, to predict fatigue life of a physical model; whilst Vayssette et al [12] utilised 3D topography measurements to estimate the HCF strength in AM parts. Although the advances in computed tomography have contributed to more realistic physical characterisation of surface conditions of AM parts, the essential approach to fatigue life estimation remains unchanged, i.e. stress parameters, or variants of a stress component, have been used as “fatigue indicators” [12] to describe fatigue lives in HCF regime.
Such approaches have been accepted for the purposes of fatigue life-time management, but they offer little insight into the mechanisms of fatigue damage, from which more physical models might be developed towards informing material design and life management. This has become highly desirable due to the advent of new AM routes which promise to produce net-to-shape parts, to have a greater control of materials properties through processing towards a unified design and analysis process. To achieve the full potential of AM technology for net-shape production of high quality parts, a fundamental understanding of the effects of surface conditions in AB specimens on fatigue properties must be developed, so that insights may be gained towards design and fatigue life management of AM components and structures for safety-critical applications.
The objectives of this study are: (i) To investigate the material behaviour of idealised notches under cyclic loading conditions; (ii) to identify the cyclic deformation mechanisms in the HCF regime; (iii) to examine the cyclic deformation behaviour in selected critical features from the surface profile measurements of AM specimens and (iv) to inform design strategies against fatigue damage in AM parts. Surface roughness measurements and HCF experiments as well as tensile testing were performed on 17-4 PH stainless steel specimens produced by selective laser melting (SLM) [13]. Selected idealised notch sizes and critical surface features were studied in the material under applied HCF loading conditions using the finite element approach, and the evolution of cyclic damage and the mode of deformation were monitored. In addition, FE simulations were also carried out on the same notch features in stainless steel 316L [14] and a nickel-based superalloy, RR1000 [15], using elastic-plastic and visco-plastic material models, respectively. These additional exercises aim at removing potential influence of particular material model types on the simulated material behaviour, so that more generic results may be obtained.
Section snippets
The FE models
The FE analyses were conducted using ABAQUS (6.14-1). Due to symmetry, only half of gauge section of the specimen (Fig. 1(a)) was modelled. The geometry of the basic FE model, together with the loading and the boundary conditions, is shown in Fig. 1(b), to which selected notch sizes of a semi-circular shape (Fig. 1(c)) were introduced. All nodes at bottom of the model were fixed in the vertical direction, and the node at the left lower corner was also fixed in the horizontal direction to
Experimental results
A typical 2D surface profile from the surface roughness measurement (2.2) is shown in Fig. 2(a), from which three critical points (A, B and C) were identified for the FE analysis. The engineering stress-strain curves were obtained from the monotonic tensile testing, from which an average material constitutive behaviour was assumed. Fig. 3(a) shows a typical engineering stress-strain curve, and simulated results from the tabulated experimental data (Table 1) of 17-4 PH stainless steel.
Fatigue
Discussion
Conventionally, when cyclic stress levels are relatively low such that fatigue failure occurs between 104 and 106 or more, High-cycle fatigue (HCF) regime is assumed. In this regime, material deformation is believed to be predominantly elastic, and fatigue life may be described by a Stress-Life, or S-N curve approach. As a recent review [8] illustrated, this is the default method for the evaluation of fatigue strength of AM materials when comparing with materials produced by conventional means.
Conclusions
Two distinct deformation modes are found in HCF regime: Shakedown and strain ratchetting. The former appears to occur at low stress levels; whilst the latter occurs at high stress levels. The trend appears to be independent of loading type (asymmetric vs symmetric) or material types.
Author Contributions
MCM performed the monotonic tensile testing and surface roughness measurements; CL conducted the fatigue experiment; GWZ carried out the FE analysis with the assistance of BL. JT conceived the idea and formulated the plans for the experiments and the FE simulations. The research was conducted at the University of Portsmouth.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
GWZ was supported by a Visiting Scholarship from China Scholarship Council (No. 201808420331).
References: (21)
- et al.
Effects of defects in laser additive manufactured Ti-6Al-4V on fatigue properties
Phys Procedia
(2014) - et al.
Influence of defects on mechanical properties of Ti–6Al–4V components produced by selective laser melting and electron beam melting
Mater Des
(2015) - et al.
Mechanical behavior of additive manufactured, powder-bed laser-fused materials
Mater Sci Eng A-Struct
(2016) Additive manufacturing in the context of structural integrity
Int J Fatigue
(2017)- et al.
Additive manufacturing of fatigue resistant materials: Challenges and opportunities
Int J Fatigue
(2017) - et al.
Estimating the fatigue stress concentration factor of machined surfaces
Int J Fatigue
(2002) - et al.
Surface roughness characterization for fatigue life predictions using finite element analysis
Int J Fatigue
(2008) - et al.
Numerical modelling of surface roughness effect on the fatigue behavior of Ti-6Al-4V obtained by additive manufacturing
Int J Fatigue
(2019) - et al.
A study of cyclic behaviour of a nickel-based superalloy at elevated temperature using a viscoplastic-damage model
Int J Fatigue
(2011) - et al.
Ratchetting strain as a damage parameter in controlling crack growth at elevated temperature
Eng Fract Mech
(2009)
Cited by (6)
Analysis of prospective SIF and shielding effect for cylindrical rough surfaces obtained by L-PBF
2021, Engineering Fracture MechanicsCitation Excerpt :An as-built L-PBF surface can be seen as a random distribution of micro-notches [22] introducing large stress concentrations and interacting with each other in order to accommodate the local stress field around them. Several methods are proposed in literature to model the effect of the surface quality based on the fatigue behaviour of as-built AM parts, mainly in relation to: stress concentration analysis [23–25], equivalent crack model [5,26–29], FE analyses [30–33]. Another approach for the surface refers to Fracture Mechanics concepts, similar to those adopted for fatigue limit estimation in the presence of defects (rough surfaces are treated as “defects”).
In-situ sensing, process monitoring and machine control in Laser Powder Bed Fusion: A review
2021, Additive ManufacturingCitation Excerpt :The use of an inert gas atmosphere during processing can limit or mitigate some of these oxidation effects. Surface roughness defects are a major factor in the fatigue and crack nucleation of where small surface peaks and valleys facilitate the initiation and subsequent propagation of cracks through the piece [15–18]. Finally, balling – the formation of solidified balls of material on the part surfaces after melting – often occurs where the powder used in a PBF process has high surface tension in the liquid phase.
Killer notches: The effect of as-built surface roughness on fatigue failure in AlSi10Mg produced by laser powder bed fusion
2020, Additive ManufacturingCitation Excerpt :This study is of primary importance for two different reasons: accurate assessment for predictive modelling and general rating of surface quality. As for the modelling and assessment, there are different approaches for numerical modelling surface effects (in terms of fatigue stress concentration [32–36] or cracks [19,37–42]), but these approaches were applied thus far for conservative predictions without a ‘sample by sample’ verification. As for the measurement of surface quality, the different approaches for fatigue modelling use generalized roughness parameters to describe the surface quality, but undoubtedly the concept of notch stress concentration is only related to “valleys” and depressions in the profile.
Fatigue strength assessment of “as built” AlSi10Mg manufactured by SLM with different build orientations
2020, International Journal of FatigueSignificant Advancements in Numerical Simulation of Fatigue Behavior in Metal Additive Manufacturing-Review
2022, Applied Sciences (Switzerland)